OpenE2K
/
binutils-gdb
12
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
binutils-gdb/opcodes/i386-opc.tbl

4092 lines
544 KiB
Plaintext
Raw Normal View History

2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// i386 opcode table.
Update year range in copyright notice of binutils files
2020-01-01 08:57:01 +01:00
// Copyright (C) 2007-2020 Free Software Foundation, Inc.
Change source files over to GPLv3.
2007-07-05 11:49:03 +02:00
//
// This file is part of the GNU opcodes library.
//
// This library is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 3, or (at your option)
// any later version.
//
// It is distributed in the hope that it will be useful, but WITHOUT
// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
// or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
// License for more details.
//
// You should have received a copy of the GNU General Public License
// along with GAS; see the file COPYING. If not, write to the Free
// Software Foundation, 51 Franklin Street - Fifth Floor, Boston, MA
// 02110-1301, USA.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: pre-process opcodes table before parsing Instead of expanding macro-like constructs in i386-gen, have the C pre- processor do this for us. Besides being a prerequisite for the next template folding steps, this also paves the way for removing various hidden dependencies between #define-s in i386-opc.h and plain literal numbers used in i386-opc.tbl. The #undef of None is solely to leave the generated i386-tbl.h entirely unchanged.
2018-07-19 08:28:29 +02:00
#define OPCODE_I386_H
#include "i386-opc.h"
#undef None
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
#define Amd64 ISA64=AMD64
#define Intel64 ISA64=INTEL64
#define Intel64Only ISA64=INTEL64ONLY
x86: introduce operand type "class" Many operand types, in particular the various kinds of registers, can't be combined with one another (neither in templates nor in register entries), and hence it is not a good use of resources (memory as well as execution time) to represent them as individual bits of a bit field.
2019-11-08 09:03:23 +01:00
#define Reg8 Class=Reg|Byte
#define Reg16 Class=Reg|Word
#define Reg32 Class=Reg|Dword
#define Reg64 Class=Reg|Qword
x86: re-do "shorthand" handling Now that the opcode table gets preprocessed, undo parts of commit dc821c5f9a ("x86: replace Reg8, Reg16, Reg32, and Reg64"): Have the preprocessor handle the expansion there, while making the expansions explicit in i386-gen and the register table.
2019-10-30 09:07:40 +01:00
x86: eliminate ImmExt abuse Drop the remaining instances left in place by commit c3949f432f ("x86: limit ImmExt abuse), now that we have a way to specify specific GPRs. Take the opportunity and also introduce proper 16-bit forms of applicable SVME insns as well as 1-operand forms of CLZERO.
2019-11-12 09:08:32 +01:00
#define Acc Instance=Accum
#define RegC Instance=RegC
#define RegD Instance=RegD
#define RegB Instance=RegB
#define ShiftCount RegC|Byte
#define InOutPortReg RegD|Word
x86: introduce operand type "instance" Special register "class" instances can't be combined with one another (neither in templates nor in register entries), and hence it is not a good use of resources (memory as well as execution time) to represent them as individual bits of a bit field. Furthermore the generalization becoming possible will allow improvements to the handling of insns accepting only individual registers as their operands.
2019-11-12 09:07:34 +01:00
x86: re-do "shorthand" handling Now that the opcode table gets preprocessed, undo parts of commit dc821c5f9a ("x86: replace Reg8, Reg16, Reg32, and Reg64"): Have the preprocessor handle the expansion there, while making the expansions explicit in i386-gen and the register table.
2019-10-30 09:07:40 +01:00
#define FloatAcc Acc|Tbyte
x86: introduce operand type "class" Many operand types, in particular the various kinds of registers, can't be combined with one another (neither in templates nor in register entries), and hence it is not a good use of resources (memory as well as execution time) to represent them as individual bits of a bit field.
2019-11-08 09:03:23 +01:00
#define FloatReg Class=Reg|Tbyte
x86: re-do "shorthand" handling Now that the opcode table gets preprocessed, undo parts of commit dc821c5f9a ("x86: replace Reg8, Reg16, Reg32, and Reg64"): Have the preprocessor handle the expansion there, while making the expansions explicit in i386-gen and the register table.
2019-10-30 09:07:40 +01:00
x86: convert SReg from bitfield to enumerator This is to further shrink the operand type representation.
2019-11-08 09:04:09 +01:00
#define SReg Class=SReg
x86: convert Control/Debug/Test from bitfield to enumerator This is to further shrink the operand type representation.
2019-11-08 09:04:53 +01:00
#define Control Class=RegCR
#define Debug Class=RegDR
#define Test Class=RegTR
x86: convert RegSIMD and RegMMX from bitfield to enumerator This is to further shrink the operand type representation.
2019-11-08 09:05:36 +01:00
#define RegMMX Class=RegMMX
#define RegXMM Class=RegSIMD|Xmmword
#define RegYMM Class=RegSIMD|Ymmword
#define RegZMM Class=RegSIMD|Zmmword
x86: re-do "shorthand" handling Now that the opcode table gets preprocessed, undo parts of commit dc821c5f9a ("x86: replace Reg8, Reg16, Reg32, and Reg64"): Have the preprocessor handle the expansion there, while making the expansions explicit in i386-gen and the register table.
2019-10-30 09:07:40 +01:00
x86: convert RegMask and RegBND from bitfield to enumerator This is to further shrink the operand type representation.
2019-11-08 09:06:24 +01:00
#define RegMask Class=RegMask
#define RegBND Class=RegBND
x86: fold individual Jump* attributes into a single Jump one ..., taking just 3 bits instead of 5. No two of them are used together.
2019-11-14 08:47:44 +01:00
#define JumpByte Jump=JUMP_BYTE
#define JumpDword Jump=JUMP_DWORD
#define JumpAbsolute Jump=JUMP_ABSOLUTE
#define JumpInterSegment Jump=JUMP_INTERSEGMENT
x86: fold Size{16,32,64} template attributes Only one of them can be set at a time, which means they can be expressed by a single 2-bit field instead of three 1-bit ones.
2018-10-10 08:41:52 +02:00
#define Size16 Size=SIZE16
#define Size32 Size=SIZE32
#define Size64 Size=SIZE64
x86: Replace IgnoreSize/DefaultSize with MnemonicSize Since an instruction template can't have both IgnoreSize and DefaultSize, this patch replaces IgnoreSize and DefaultSize with MnemonicSize. gas/ * config/tc-i386.c (match_template): Replace ignoresize and defaultsize with mnemonicsize. (process_suffix): Likewise. opcodes/ * i386-gen.c (opcode_modifiers): Replace IgnoreSize/DefaultSize with MnemonicSize. * i386-opc.h (IGNORESIZE): New. (DEFAULTSIZE): Likewise. (IgnoreSize): Removed. (DefaultSize): Likewise. (MnemonicSize): New. (i386_opcode_modifier): Replace ignoresize/defaultsize with mnemonicsize. * i386-opc.tbl (IgnoreSize): New. (DefaultSize): Likewise. * i386-tbl.h: Regenerated.
2020-03-03 20:24:16 +01:00
#define IgnoreSize MnemonicSize=IGNORESIZE
#define DefaultSize MnemonicSize=DEFAULTSIZE
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
// RegMem implies a ModR/M byte
#define RegMem Modrm|RegMem
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
#define IsStringEsOp0 IsString=IS_STRING_ES_OP0
#define IsStringEsOp1 IsString=IS_STRING_ES_OP1
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
#define AddrPrefixOpReg AddrPrefixOpReg|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf
x86: XOP VPHADD* / VPHSUB* are VEX.W0 Also avoid introducing further uses of VexW=1, by introducing and using VexW0 at this occasion. Move the marker past all #define-s.
2018-11-06 11:38:47 +01:00
#define VexW0 VexW=VEXW0
#define VexW1 VexW=VEXW1
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
#define VexWIG VexW=VEXWIG
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
#define Vex128 Vex=VEX128
#define Vex256 Vex=VEX256
#define VexLIG Vex=VEXScalar
#define EVex128 EVex=EVEX128
#define EVex256 EVex=EVEX256
#define EVex512 EVex=EVEX512
#define EVexLIG EVex=EVEXLIG
#define EVexDYN EVex=EVEXDYN
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
// The EVEX purpose of StaticRounding appears only together with SAE. Re-use
// the bit to mark commutative VEX encodings where swapping the source
// operands may allow to switch from 3-byte to 2-byte VEX encoding.
#define C StaticRounding
x86: XOP VPHADD* / VPHSUB* are VEX.W0 Also avoid introducing further uses of VexW=1, by introducing and using VexW0 at this occasion. Move the marker past all #define-s.
2018-11-06 11:38:47 +01:00
### MARKER ###
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Move instructions.
x86: consolidate Disp<NN> handling a little In memory operand addressing, which forms of displacement are permitted besides Disp8 is pretty clearly limited - outside of 64-bit mode, Disp16 or Disp32 only, depending on address size (MPX being special in not allowing Disp16), - in 64-bit mode, Disp32s or Disp64 without address size override, and solely Disp32 with one. Adjust assembler and i386-gen to match this, observing that templates already get adjusted before trying to match them against input depending on the presence of an address size prefix. This adjustment logic gets extended to all cases, as certain DispNN values should also be dropped when there's no such prefix. In fact behavior of the assembler, perhaps besides the exact diagnostics wording, should not differ between there being templates applicable to 64-bit and non-64-bit at the same time, or there being fully separate sets of templates, with their DispNN settings already reduced accordingly. This adjustment logic further gets guarded such that there wouldn't be and Disp<N> conversion based on address size prefix when this prefix doesn't control the width of the displacement (on branches other than absolute ones). These adjustments then also allow folding two MOV templates, which had been split between 64-bit and non-64-bits variants so far. Once in this area also - drop the bogus DispNN from JumpByte templates, leaving just the correct Disp8 there (compensated by i386_finalize_displacement() now setting Disp8 on their operands), - add the missing Disp32S to XBEGIN. Note that the changes make it necessary to temporarily mark a test as XFAIL; this will get taken care of by a subsequent patch. The failing parts are entirely bogus and will get replaced.
2019-12-27 09:22:03 +01:00
mov, 2, 0xa0, None, 1, 0, D|W|No_sSuf|No_qSuf|No_ldSuf, { Disp16|Disp32|Disp64|Unspecified|Byte|Word|Dword|Qword, Acc|Byte|Word|Dword|Qword }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
mov, 2, 0x88, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|HLEPrefixOk=3, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// In the 64bit mode the short form mov immediate is redefined to have
// 64bit value.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
mov, 2, 0xb0, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32 }
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
mov, 2, 0xc6, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf|HLEPrefixOk=3|Optimize, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
mov, 2, 0xb8, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|Optimize, { Imm64, Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// The segment register moves accept WordReg so that a segment register
// can be copied to a 32 bit register, and vice versa, without using a
// size prefix. When moving to a 32 bit register, the upper 16 bits
// are set to an implementation defined value (on the Pentium Pro, the
// implementation defined value is zero).
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
mov, 2, 0x8c, None, 1, 0, RegMem|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { SReg, Reg16|Reg32|Reg64 }
x86: fold SReg{2,3} They're the only exception to there generally being no mix of register kinds possible in an insn operand template, and there being two bits per operand for their representation is also quite wasteful, considering the low number of uses. Fold both bits and deal with the little bit of fallout. Also take the liberty and drop dead code trying to set REX_B: No segment register has RegRex set on it. Additionally I was quite surprised that PUSH/POP with the permitted segment registers is not covered by the test cases. Add the missing pieces.
2019-07-16 09:30:29 +02:00
mov, 2, 0x8c, None, 1, 0, D|Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { SReg, Word|Unspecified|BaseIndex }
mov, 2, 0x8e, None, 1, 0, Modrm|IgnoreSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Reg16|Reg32|Reg64, SReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Move to/from control debug registers. In the 16 or 32bit modes
// they are 32bit. In the 64bit mode they are 64bit.
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
mov, 2, 0xf20, None, 2, Cpu386|CpuNo64, D|RegMem|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Control, Reg32 }
mov, 2, 0xf20, None, 2, Cpu64, D|RegMem|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Control, Reg64 }
mov, 2, 0xf21, None, 2, Cpu386|CpuNo64, D|RegMem|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Debug, Reg32 }
mov, 2, 0xf21, None, 2, Cpu64, D|RegMem|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Debug, Reg64 }
mov, 2, 0xf24, None, 2, Cpu386|CpuNo64, D|RegMem|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Test, Reg32 }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
movabs, 2, 0xa0, None, 1, Cpu64, D|W|No_sSuf|No_ldSuf, { Disp64|Unspecified|Byte|Word|Dword|Qword, Acc|Byte|Word|Dword|Qword }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
movabs, 2, 0xb8, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Imm64, Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention XSAVE, EPT and MOVBE. * config/tc-i386.c (cpu_arch): Add .movbe and .ept. (md_show_usage): Add .movbe and .ept. * doc/c-i386.texi: Add movbe and ept to -march=. Document .movbe and .ept. gas/testsuite/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run movbe, movbe-intel, inval-movbe, ept, ept-intel, inval-ept, x86-64-movbe, x86-64-movbe-intel, x86-64-inval-movbe. x86-64-ept, x86-64-ept-intel and x86-64-inval-ept. * gas/i386/arch-10.s: Add movbe and invept. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/ept.d: New file * gas/i386/ept-intel.d: Likewise. * gas/i386/ept.s: Likewise. * gas/i386/inval-ept.l: Likewise. * gas/i386/inval-ept.s: Likewise. * gas/i386/inval-movbe.l: Likewise. * gas/i386/inval-movbe.s: Likewise. * gas/i386/movbe.d: Likewise. * gas/i386/movbe-intel.d: Likewise. * gas/i386/movbe.s: Likewise. * gas/i386/x86-64-inval-ept.l: Likewise. * gas/i386/x86-64-inval-ept.s: Likewise. * gas/i386/x86-64-inval-movbe.l: Likewise. * gas/i386/x86-64-inval-movbe.s: Likewise. * gas/i386/x86-64-ept.d: Likewise. * gas/i386/x86-64-ept-intel.d: Likewise. * gas/i386/x86-64-ept.s: Likewise. * gas/i386/x86-64-movbe.d: Likewise. * gas/i386/x86-64-movbe-intel.d: Likewise. * gas/i386/x86-64-movbe.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (MOVBE_Fixup): New. (Mo): Likewise. (PREFIX_0F3880): Likewise. (PREFIX_0F3881): Likewise. (PREFIX_0F38F0): Updated. (prefix_table): Add PREFIX_0F3880 and PREFIX_0F3881. Update PREFIX_0F38F0 and PREFIX_0F38F1 for movbe. (three_byte_table): Use PREFIX_0F3880 and PREFIX_0F3881. * i386-gen.c (cpu_flag_init): Add CPU_MOVBE_FLAGS and CPU_EPT_FLAGS. (cpu_flags): Add CpuMovbe and CpuEPT. * i386-opc.h (CpuMovbe): New. (CpuEPT): Likewise. (CpuLM): Updated. (i386_cpu_flags): Add cpumovbe and cpuept. * i386-opc.tbl: Add entries for movbe and EPT instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-05-02 18:53:40 +02:00
// Move after swapping the bytes
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movbe, 2, 0x0f38f0, None, 3, CpuMovbe, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
movbe, 2, 0x0f38f1, None, 3, CpuMovbe, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64, Word|Dword|Qword|Unspecified|BaseIndex }
gas/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention XSAVE, EPT and MOVBE. * config/tc-i386.c (cpu_arch): Add .movbe and .ept. (md_show_usage): Add .movbe and .ept. * doc/c-i386.texi: Add movbe and ept to -march=. Document .movbe and .ept. gas/testsuite/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run movbe, movbe-intel, inval-movbe, ept, ept-intel, inval-ept, x86-64-movbe, x86-64-movbe-intel, x86-64-inval-movbe. x86-64-ept, x86-64-ept-intel and x86-64-inval-ept. * gas/i386/arch-10.s: Add movbe and invept. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/ept.d: New file * gas/i386/ept-intel.d: Likewise. * gas/i386/ept.s: Likewise. * gas/i386/inval-ept.l: Likewise. * gas/i386/inval-ept.s: Likewise. * gas/i386/inval-movbe.l: Likewise. * gas/i386/inval-movbe.s: Likewise. * gas/i386/movbe.d: Likewise. * gas/i386/movbe-intel.d: Likewise. * gas/i386/movbe.s: Likewise. * gas/i386/x86-64-inval-ept.l: Likewise. * gas/i386/x86-64-inval-ept.s: Likewise. * gas/i386/x86-64-inval-movbe.l: Likewise. * gas/i386/x86-64-inval-movbe.s: Likewise. * gas/i386/x86-64-ept.d: Likewise. * gas/i386/x86-64-ept-intel.d: Likewise. * gas/i386/x86-64-ept.s: Likewise. * gas/i386/x86-64-movbe.d: Likewise. * gas/i386/x86-64-movbe-intel.d: Likewise. * gas/i386/x86-64-movbe.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (MOVBE_Fixup): New. (Mo): Likewise. (PREFIX_0F3880): Likewise. (PREFIX_0F3881): Likewise. (PREFIX_0F38F0): Updated. (prefix_table): Add PREFIX_0F3880 and PREFIX_0F3881. Update PREFIX_0F38F0 and PREFIX_0F38F1 for movbe. (three_byte_table): Use PREFIX_0F3880 and PREFIX_0F3881. * i386-gen.c (cpu_flag_init): Add CPU_MOVBE_FLAGS and CPU_EPT_FLAGS. (cpu_flags): Add CpuMovbe and CpuEPT. * i386-opc.h (CpuMovbe): New. (CpuEPT): Likewise. (CpuLM): Updated. (i386_cpu_flags): Add cpumovbe and cpuept. * i386-opc.tbl: Add entries for movbe and EPT instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-05-02 18:53:40 +02:00
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Move with sign extend.
// "movsbl" & "movsbw" must not be unified into "movsb" to avoid
// conflict with the "movs" string move instruction.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movsbl, 2, 0xfbe, None, 2, Cpu386, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg8|Byte|Unspecified|BaseIndex, Reg32 }
movsbw, 2, 0xfbe, None, 2, Cpu386, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg8|Byte|Unspecified|BaseIndex, Reg16 }
movswl, 2, 0xfbf, None, 2, Cpu386, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex, Reg32 }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
movsbq, 2, 0xfbe, None, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg8|Byte|Unspecified|BaseIndex, Reg64 }
movswq, 2, 0xfbf, None, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg16|Word|Unspecified|BaseIndex, Reg64 }
movslq, 2, 0x63, None, 1, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg32|Dword|Unspecified|BaseIndex, Reg64 }
x86: replace adhoc (partly wrong) ambiguous operand checking for MOVSX/MOVZX For these to get treatment consistent with other operand size checking the special logic shouldn't live in md_assemble(), but process_suffix(). And there's more logic involved than simply zapping the suffix. Note however that MOVS[BW]* and MOVZ[BW]* still won't be fully consistent, due to the objection to fold MOVS* templates just like was done for MOVZ* in c07315e0c6 ("x86: allow suffix-less movzw and 64-bit movzb"). Note further that it is against my own intentions to have MOVSX/MOVZX silently default to a byte source in AT&T mode. This should happen only when the destination register is a 16-bit one. In all other cases there is an ambiguity, and the user should be warned. But it was explicitly requested for this to be done in a way inconsistent with everything else. Note finally that the assembler change points out (and this patch fixes) a wrong Intel syntax test introduced by bc31405ebb2c ("x86-64: Properly encode and decode movsxd"): When source code specifies a 16-bit destination register, disassembly expectations shouldn't have been to find a 32-bit one.
2020-02-14 14:27:28 +01:00
movsx, 2, 0xfbe, None, 2, Cpu386, W|Modrm|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg8|Reg16|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
movsx, 2, 0x63, None, 1, Cpu64, Modrm|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, Reg32|Reg64 }
x86-64: Properly encode and decode movsxd movsxd is a 64-bit only instruction. It supports both 16-bit and 32-bit destination registers. Its AT&T mnemonic is movslq which only supports 64-bit destination register. There is also a discrepancy between AMD64 and Intel64 on movsxd with 16-bit destination register. AMD64 supports 32-bit source operand and Intel64 supports 16-bit source operand. This patch updates movsxd encoding and decoding to alow 16-bit and 32-bit destination registers. It also handles movsxd with 16-bit destination register for AMD64 and Intel 64. gas/ PR binutils/25445 * config/tc-i386.c (check_long_reg): Also convert to QWORD for movsxd. * doc/c-i386.texi: Add a node for AMD64 vs. Intel64 ISA differences. Document movslq and movsxd. * testsuite/gas/i386/i386.exp: Run PR binutils/25445 tests. * testsuite/gas/i386/x86-64-movsxd-intel.d: New file. * testsuite/gas/i386/x86-64-movsxd-intel64-intel.d: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64-inval.l: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64-inval.s: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64.d: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64.s: Likewise. * testsuite/gas/i386/x86-64-movsxd-inval.l: Likewise. * testsuite/gas/i386/x86-64-movsxd-inval.s: Likewise. * testsuite/gas/i386/x86-64-movsxd.d: Likewise. * testsuite/gas/i386/x86-64-movsxd.s: Likewise. opcodes/ PR binutils/25445 * i386-dis.c (MOVSXD_Fixup): New function. (movsxd_mode): New enum. (x86_64_table): Use MOVSXD_Fixup and movsxd_mode on movsxd. (intel_operand_size): Handle movsxd_mode. (OP_E_register): Likewise. (OP_G): Likewise. * i386-opc.tbl: Remove Rex64 and allow 32-bit destination register on movsxd. Add movsxd with 16-bit destination register for AMD64 and Intel64 ISAs. * i386-tbl.h: Regenerated.
2020-01-27 13:38:10 +01:00
movsxd, 2, 0x63, None, 1, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, Reg32|Reg64 }
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
movsxd, 2, 0x63, None, 1, Cpu64, Amd64|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, Reg16 }
x86-64: Properly encode and decode movsxd movsxd is a 64-bit only instruction. It supports both 16-bit and 32-bit destination registers. Its AT&T mnemonic is movslq which only supports 64-bit destination register. There is also a discrepancy between AMD64 and Intel64 on movsxd with 16-bit destination register. AMD64 supports 32-bit source operand and Intel64 supports 16-bit source operand. This patch updates movsxd encoding and decoding to alow 16-bit and 32-bit destination registers. It also handles movsxd with 16-bit destination register for AMD64 and Intel 64. gas/ PR binutils/25445 * config/tc-i386.c (check_long_reg): Also convert to QWORD for movsxd. * doc/c-i386.texi: Add a node for AMD64 vs. Intel64 ISA differences. Document movslq and movsxd. * testsuite/gas/i386/i386.exp: Run PR binutils/25445 tests. * testsuite/gas/i386/x86-64-movsxd-intel.d: New file. * testsuite/gas/i386/x86-64-movsxd-intel64-intel.d: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64-inval.l: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64-inval.s: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64.d: Likewise. * testsuite/gas/i386/x86-64-movsxd-intel64.s: Likewise. * testsuite/gas/i386/x86-64-movsxd-inval.l: Likewise. * testsuite/gas/i386/x86-64-movsxd-inval.s: Likewise. * testsuite/gas/i386/x86-64-movsxd.d: Likewise. * testsuite/gas/i386/x86-64-movsxd.s: Likewise. opcodes/ PR binutils/25445 * i386-dis.c (MOVSXD_Fixup): New function. (movsxd_mode): New enum. (x86_64_table): Use MOVSXD_Fixup and movsxd_mode on movsxd. (intel_operand_size): Handle movsxd_mode. (OP_E_register): Likewise. (OP_G): Likewise. * i386-opc.tbl: Remove Rex64 and allow 32-bit destination register on movsxd. Add movsxd with 16-bit destination register for AMD64 and Intel64 ISAs. * i386-tbl.h: Regenerated.
2020-01-27 13:38:10 +01:00
movsxd, 2, 0x63, None, 1, Cpu64, Intel64|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Unspecified|BaseIndex, Reg16 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: allow suffix-less movzw and 64-bit movzb ... just like is already the case for 16- and 32-bit movzb: I can't see why omitting suffixes on this (and movs{b,w,l}) is not allowed, when it is allowed for all other instructions where the suffix is redundant with (one of) the operands.
2016-07-01 09:01:41 +02:00
// Move with zero extend.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movzb, 2, 0xfb6, None, 2, Cpu386, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg8|Byte|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
movzw, 2, 0xfb7, None, 2, Cpu386, Modrm|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex, Reg32|Reg64 }
Remove Intel syntax comments on movsx and movzx Since movsx and movzx are valid mnemonic in AT&T syntax, remove Intel syntax comments on movsx and movzx to avoid confusing other readers. * i386-opc.tbl (movsx): Remove Intel syntax comments. (movzx): Likewise.
2020-02-14 14:40:19 +01:00
// The 64-bit variant is not particularly useful since the zero extend
// 32->64 is implicit, but we can encode them.
x86: replace adhoc (partly wrong) ambiguous operand checking for MOVSX/MOVZX For these to get treatment consistent with other operand size checking the special logic shouldn't live in md_assemble(), but process_suffix(). And there's more logic involved than simply zapping the suffix. Note however that MOVS[BW]* and MOVZ[BW]* still won't be fully consistent, due to the objection to fold MOVS* templates just like was done for MOVZ* in c07315e0c6 ("x86: allow suffix-less movzw and 64-bit movzb"). Note further that it is against my own intentions to have MOVSX/MOVZX silently default to a byte source in AT&T mode. This should happen only when the destination register is a 16-bit one. In all other cases there is an ambiguity, and the user should be warned. But it was explicitly requested for this to be done in a way inconsistent with everything else. Note finally that the assembler change points out (and this patch fixes) a wrong Intel syntax test introduced by bc31405ebb2c ("x86-64: Properly encode and decode movsxd"): When source code specifies a 16-bit destination register, disassembly expectations shouldn't have been to find a 32-bit one.
2020-02-14 14:27:28 +01:00
movzx, 2, 0xfb6, None, 2, Cpu386, W|Modrm|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg8|Reg16|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Push instructions.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
push, 1, 0x50, None, 1, CpuNo64, No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
push, 1, 0xff, 0x6, 1, CpuNo64, Modrm|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32|Word|Dword|Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
push, 1, 0x6a, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8S }
push, 1, 0x68, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16|Imm32 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
push, 1, 0x6, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { SReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// In 64bit mode, the operand size is implicitly 64bit.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
push, 1, 0x50, None, 1, Cpu64, No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
push, 1, 0xff, 0x6, 1, Cpu64, Modrm|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg64|Word|Qword|Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
push, 1, 0x6a, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Imm8S }
push, 1, 0x68, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Imm16|Imm32S }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
push, 1, 0xfa0, None, 2, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { SReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
pusha, 0, 0x60, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Pop instructions.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
pop, 1, 0x58, None, 1, CpuNo64, No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pop, 1, 0x8f, 0x0, 1, CpuNo64, Modrm|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32|Word|Dword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
pop, 1, 0x7, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { SReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// In 64bit mode, the operand size is implicitly 64bit.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
pop, 1, 0x58, None, 1, Cpu64, No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pop, 1, 0x8f, 0x0, 1, Cpu64, Modrm|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg64|Word|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
pop, 1, 0xfa1, None, 2, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { SReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
popa, 0, 0x61, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Exchange instructions.
// xchg commutes: we allow both operand orders.
// In the 64bit code, xchg rax, rax is reused for new nop instruction.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
xchg, 2, 0x90, None, 1, 0, CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64, Acc|Word|Dword|Qword }
xchg, 2, 0x90, None, 1, 0, CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Acc|Word|Dword|Qword, Reg16|Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xchg, 2, 0x86, None, 1, 0, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk=2, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
xchg, 2, 0x86, None, 1, 0, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk=2, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex, Reg8|Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// In/out from ports.
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
in, 2, 0xe4, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Acc|Byte|Word|Dword }
in, 2, 0xec, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { InOutPortReg, Acc|Byte|Word|Dword }
Remove CheckRegSize from instructions with 0, 1 or fixed operands. 2010-10-14 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Remove CheckRegSize from instructions with 0, 1 or fixed operands. * i386-tbl.h: Regenerated.
2010-10-14 23:37:30 +02:00
in, 1, 0xe4, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
in, 1, 0xec, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { InOutPortReg }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
out, 2, 0xe6, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Acc|Byte|Word|Dword, Imm8 }
out, 2, 0xee, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Acc|Byte|Word|Dword, InOutPortReg }
Remove CheckRegSize from instructions with 0, 1 or fixed operands. 2010-10-14 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Remove CheckRegSize from instructions with 0, 1 or fixed operands. * i386-tbl.h: Regenerated.
2010-10-14 23:37:30 +02:00
out, 1, 0xe6, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
out, 1, 0xee, None, 1, 0, W|No_sSuf|No_qSuf|No_ldSuf, { InOutPortReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Load effective address.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
lea, 2, 0x8d, None, 1, 0, Modrm|Anysize|No_bSuf|No_sSuf|No_ldSuf, { BaseIndex, Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Load segment registers from memory.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
lds, 2, 0xc5, None, 1, CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { DWord|Fword|Unspecified|BaseIndex, Reg16|Reg32 }
les, 2, 0xc4, None, 1, CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { DWord|Fword|Unspecified|BaseIndex, Reg16|Reg32 }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
lfs, 2, 0xfb4, None, 2, Cpu386, Amd64|Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Fword|Unspecified|BaseIndex, Reg16|Reg32 }
lfs, 2, 0xfb4, None, 2, Cpu64, Intel64|Modrm|No_bSuf|No_sSuf|No_ldSuf, { Dword|Fword|Tbyte|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
lgs, 2, 0xfb5, None, 2, Cpu386, Amd64|Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Fword|Unspecified|BaseIndex, Reg16|Reg32 }
lgs, 2, 0xfb5, None, 2, Cpu64, Intel64|Modrm|No_bSuf|No_sSuf|No_ldSuf, { Dword|Fword|Tbyte|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
lss, 2, 0xfb2, None, 2, Cpu386, Amd64|Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Fword|Unspecified|BaseIndex, Reg16|Reg32 }
lss, 2, 0xfb2, None, 2, Cpu64, Intel64|Modrm|No_bSuf|No_sSuf|No_ldSuf, { Dword|Fword|Tbyte|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Flags register instructions.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
clc, 0, 0xf8, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cld, 0, 0xfc, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cli, 0, 0xfa, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
clts, 0, 0xf06, None, 2, Cpu286, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cmc, 0, 0xf5, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
lahf, 0, 0x9f, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
sahf, 0, 0x9e, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
pushf, 0, 0x9c, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
pushf, 0, 0x9c, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { 0 }
popf, 0, 0x9d, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
popf, 0, 0x9d, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { 0 }
stc, 0, 0xf9, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
std, 0, 0xfd, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
sti, 0, 0xfb, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Arithmetic.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
add, 2, 0x0, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
add, 2, 0x83, 0x0, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
add, 2, 0x4, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
add, 2, 0x80, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
inc, 1, 0x40, None, 1, CpuNo64, No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
inc, 1, 0xfe, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
sub, 2, 0x28, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
sub, 2, 0x83, 0x5, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
sub, 2, 0x2c, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
sub, 2, 0x80, 0x5, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
dec, 1, 0x48, None, 1, CpuNo64, No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
dec, 1, 0xfe, 0x1, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
sbb, 2, 0x18, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sbb, 2, 0x83, 0x3, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
sbb, 2, 0x1c, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
sbb, 2, 0x80, 0x3, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cmp, 2, 0x38, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
cmp, 2, 0x83, 0x7, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
cmp, 2, 0x3c, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
cmp, 2, 0x80, 0x7, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
test, 2, 0x84, None, 1, 0, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Unspecified|Byte|Word|Dword|Qword|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
test, 2, 0x84, None, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Reg8|Reg16|Reg32|Reg64 }
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
test, 2, 0xa8, None, 1, 0, W|No_sSuf|No_ldSuf|Optimize, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
test, 2, 0xf6, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf|Optimize, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: optimize AND/OR with twice the same register It seems to be not uncommon for people to use AND or OR in this form for just setting the status flags. TEST, which doesn't write to any register other than EFLAGS, ought to be preferred. Make the change only for -O2 and above though, at least for now.
2019-07-01 08:35:08 +02:00
and, 2, 0x20, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86-64: also optimize ANDQ with immediate fitting in 7 bits The same reasoning applies here as did/does for immediates fitting in 31 bits.
2019-06-25 09:23:48 +02:00
and, 2, 0x83, 0x4, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
and, 2, 0x24, None, 1, 0, W|No_sSuf|No_ldSuf|Optimize, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
and, 2, 0x80, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: optimize AND/OR with twice the same register It seems to be not uncommon for people to use AND or OR in this form for just setting the status flags. TEST, which doesn't write to any register other than EFLAGS, ought to be preferred. Make the change only for -O2 and above though, at least for now.
2019-07-01 08:35:08 +02:00
or, 2, 0x8, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
or, 2, 0x83, 0x1, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
or, 2, 0xc, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
or, 2, 0x80, 0x1, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/testsuite/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * gas/i386/i386.s: Add tests for fnstsw and fstsw. * gas/i386/inval.s: Likewise. * gas/i386/x86_64.s: Likewise. * gas/i386/intel.s: Use word instead of dword on ss. * gas/i386/x86-64-inval.s: Add tests for fnstsw, fstsw, in and out. * gas/i386/prefix.s: Remove invalid fstsw. * gas/i386/inval.l: Updated. * gas/i386/intelbad.l: Likewise. * gas/i386/i386.d: Likewise. * gas/i386/x86_64.d: Likewise. * gas/i386/x86-64-inval.l: Likewise. * gas/i386/prefix.d: Updated. gas/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * config/tc-i386.c (_i386_insn): Update comment. (operand_type_match): Also clear unspecified. (operand_type_register_match): Likewise. (parse_operands): Initialize unspecified. (i386_intel_operand): Likewise. (match_template): Check memory and accumulator operand size. (i386_att_operand): Clear unspecified on register operand. (intel_e11): Likewise. (intel_e09): Set operand size and clean unspecified for "XXX PTR". opcodes/ 2008-01-12 H.J. Lu <hongjiu.lu@intel.com> PR gas/5534 * i386-gen.c (operand_type_init): Add Dword to OPERAND_TYPE_ACC32. Add Qword to OPERAND_TYPE_ACC64. (opcode_modifiers): Remove CheckSize, Byte, Word, Dword, Qword and Xmmword. (operand_types): Add Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. (set_bitfield): Make Mmword an alias of Qword. Make Oword an alias of Xmmword. * i386-opc.h (CheckSize): Removed. (Byte): Updated. (Word): Likewise. (Dword): Likewise. (Qword): Likewise. (Xmmword): Likewise. (FWait): Updated. (OTMax): Likewise. (i386_opcode_modifier): Remove checksize, byte, word, dword, qword and xmmword. (Fword): New. (TBYTE): Likewise. (Unspecified): Likewise. (Anysize): Likewise. (i386_operand_type): Add byte, word, dword, fword, qword, tbyte xmmword, unspecified and anysize. * i386-opc.tbl: Updated to use Byte, Word, Dword, Fword, Qword, Tbyte, Xmmword, Unspecified and Anysize. * i386-reg.tbl: Add size for accumulator. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-01-12 17:05:42 +01:00
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
xor, 2, 0x30, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk|Optimize, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xor, 2, 0x83, 0x6, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
xor, 2, 0x34, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
xor, 2, 0x80, 0x6, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// clr with 1 operand is really xor with 2 operands.
x86-64: Also optimize "clr reg64" "clr reg" is an alias of "xor reg, reg". We can encode "clr reg64" as "xor reg32, reg32". gas/ * config/tc-i386.c (optimize_encoding): Also encode "clr reg64" as "xor reg32, reg32". * testsuite/gas/i386/x86-64-optimize-1.s: Add "clr reg64" tests. * testsuite/gas/i386/x86-64-optimize-1.d: Updated. opcodes/ * i386-opc.tbl: Add Optimize to clr. * i386-tbl.h: Regenerated.
2018-03-08 15:41:34 +01:00
clr, 1, 0x30, None, 1, 0, W|Modrm|No_sSuf|No_ldSuf|RegKludge|Optimize, { Reg8|Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
adc, 2, 0x10, None, 1, 0, D|W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
adc, 2, 0x83, 0x2, 1, 0, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
adc, 2, 0x14, None, 1, 0, W|No_sSuf|No_ldSuf, { Imm8|Imm16|Imm32|Imm32S, Acc|Byte|Word|Dword|Qword }
adc, 2, 0x80, 0x2, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8|Imm16|Imm32|Imm32S, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
neg, 1, 0xf6, 0x3, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
not, 1, 0xf6, 0x2, 1, 0, W|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
aaa, 0, 0x37, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
aas, 0, 0x3f, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
daa, 0, 0x27, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
das, 0, 0x2f, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
aad, 0, 0xd50a, None, 2, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
aad, 1, 0xd5, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
aam, 0, 0xd40a, None, 2, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
aam, 1, 0xd4, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Conversion insns.
// Intel naming
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
cbw, 0, 0x98, None, 1, 0, Size16|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cdqe, 0, 0x98, None, 1, Cpu64, Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cwde, 0, 0x98, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cwd, 0, 0x99, None, 1, 0, Size16|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cdq, 0, 0x99, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cqo, 0, 0x99, None, 1, Cpu64, Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// AT&T naming
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
cbtw, 0, 0x98, None, 1, 0, Size16|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cltq, 0, 0x98, None, 1, Cpu64, Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cwtl, 0, 0x98, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cwtd, 0, 0x99, None, 1, 0, Size16|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cltd, 0, 0x99, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
cqto, 0, 0x99, None, 1, Cpu64, Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Warning! the mul/imul (opcode 0xf6) must only have 1 operand! They are
// expanding 64-bit multiplies, and *cannot* be selected to accomplish
// 'imul %ebx, %eax' (opcode 0x0faf must be used in this case)
// These multiplies can only be selected with single operand forms.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
mul, 1, 0xf6, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
imul, 1, 0xf6, 0x5, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
imul, 2, 0xfaf, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Unspecified|Word|Dword|Qword|BaseIndex, Reg16|Reg32|Reg64 }
imul, 3, 0x6b, None, 1, Cpu186, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Imm8S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
imul, 3, 0x69, None, 1, Cpu186, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Imm16|Imm32|Imm32S, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// imul with 2 operands mimics imul with 3 by putting the register in
// both i.rm.reg & i.rm.regmem fields. RegKludge enables this
// transformation.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
imul, 2, 0x6b, None, 1, Cpu186, Modrm|No_bSuf|No_sSuf|No_ldSuf|RegKludge, { Imm8S, Reg16|Reg32|Reg64 }
imul, 2, 0x69, None, 1, Cpu186, Modrm|No_bSuf|No_sSuf|No_ldSuf|RegKludge, { Imm16|Imm32|Imm32S, Reg16|Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
div, 1, 0xf6, 0x6, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
div, 2, 0xf6, 0x6, 1, 0, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
idiv, 1, 0xf6, 0x7, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
idiv, 2, 0xf6, 0x7, 1, 0, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
rol, 2, 0xd0, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rol, 2, 0xc0, 0x0, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rol, 2, 0xd2, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rol, 1, 0xd0, 0x0, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ror, 2, 0xd0, 0x1, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ror, 2, 0xc0, 0x1, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ror, 2, 0xd2, 0x1, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ror, 1, 0xd0, 0x1, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcl, 2, 0xd0, 0x2, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcl, 2, 0xc0, 0x2, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcl, 2, 0xd2, 0x2, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcl, 1, 0xd0, 0x2, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcr, 2, 0xd0, 0x3, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcr, 2, 0xc0, 0x3, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcr, 2, 0xd2, 0x3, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
rcr, 1, 0xd0, 0x3, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sal, 2, 0xd0, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sal, 2, 0xc0, 0x4, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sal, 2, 0xd2, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sal, 1, 0xd0, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shl, 2, 0xd0, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shl, 2, 0xc0, 0x4, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shl, 2, 0xd2, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shl, 1, 0xd0, 0x4, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shr, 2, 0xd0, 0x5, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shr, 2, 0xc0, 0x5, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shr, 2, 0xd2, 0x5, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
shr, 1, 0xd0, 0x5, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sar, 2, 0xd0, 0x7, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Imm1, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sar, 2, 0xc0, 0x7, 1, Cpu186, W|Modrm|No_sSuf|No_ldSuf, { Imm8, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sar, 2, 0xd2, 0x7, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { ShiftCount, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
sar, 1, 0xd0, 0x7, 1, 0, W|Modrm|No_sSuf|No_ldSuf, { Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
shld, 3, 0xfa4, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Imm8, Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
shld, 3, 0xfa5, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { ShiftCount, Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
shld, 2, 0xfa5, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
shrd, 3, 0xfac, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Imm8, Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
shrd, 3, 0xfad, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { ShiftCount, Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
shrd, 2, 0xfad, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Control transfer instructions.
Support Intel MPX gas/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386.c (BND_PREFIX): New. (struct _i386_insn): Add new field bnd_prefix. (add_bnd_prefix): New. (cpu_arch): Add MPX. (i386_operand_type): Add regbnd. (md_assemble): Handle BND prefixes. (parse_insn): Likewise. (output_branch): Likewise. (output_jump): Likewise. (build_modrm_byte): Handle regbnd. (OPTION_MADD_BND_PREFIX): New. (md_longopts): Add entry for 'madd-bnd-prefix'. (md_parse_option): Handle madd-bnd-prefix option. (md_show_usage): Add description for madd-bnd-prefix option. * doc/c-i386.texi: Document mpx/.mpx and -madd-bnd-prefix. gas/testsuite/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/i386/mpx-add-bnd-prefix.s: New. * gas/i386/mpx-add-bnd-prefix.d: New. * gas/i386/mpx-inval-1.l: New. * gas/i386/mpx-inval-1.s: New. * gas/i386/mpx.d: New. * gas/i386/mpx.s: New. * gas/i386/x86-64-mpx-add-bnd-prefix.d: New. * gas/i386/x86-64-mpx-add-bnd-prefix.s: New. * gas/i386/x86-64-mpx-addr32.d: New. * gas/i386/x86-64-mpx-addr32.s: New. * gas/i386/x86-64-mpx-inval-1.l: New. * gas/i386/x86-64-mpx-inval-1.s: New. * gas/i386/x86-64-mpx-inval-2.l: New. * gas/i386/x86-64-mpx-inval-2.s: New. * gas/i386/x86-64-mpx.d: New. * gas/i386/x86-64-mpx.s: New. * gas/i386/nops.d: Adjust to MPX changes. * gas/i386/nops.s: Likewise. * gas/i386/x86-64-nops.d: Likewise. * gas/i386/x86-64-nops.s: Likewise. * gas/i386/ilp32/x86-64-nops.d: Likewise. * gas/i386/i386.exp: Run new MPX tests. include/opcode/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386.h (BND_PREFIX_OPCODE): New. opcodes/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis.c (BND_Fixup): New. (Ebnd): New. (Ev_bnd): New. (Gbnd): New. (BND): New. (v_bnd_mode): New. (bnd_mode): New. (MOD enum): Add new entries. (PREFIX enum): Likewise. (dis tables): Replace XX with BND for near branch and call instructions. (prefix_table): Add new entries. (mod_table): Likewise. (names_bnd): New. (intel_names_bnd): New. (att_names_bnd): New. (BND_PREFIX): New. (prefix_name): Handle BND_PREFIX. (print_insn): Initialize names_bnd. (intel_operand_size): Handle new modes. (OP_E_register): Likewise. (OP_E_memory): Likewise. (OP_G): Likewise. * i386-gen.c (cpu_flag_init): Add CpuMPX. (cpu_flags): Add CpuMPX. (operand_type_init): Add RegBND. (opcode_modifiers): Add BNDPrefixOk. (operand_types): Add RegBND. * i386-init.h: Regenerate. * i386-opc.h (CpuMPX): New. (CpuUnused): Comment out. (i386_cpu_flags): Add cpumpx. (BNDPrefixOk): New. (i386_opcode_modifier): Add bndprefixok. (RegBND): New. (i386_operand_type): Add regbnd. * i386-opc.tbl: Add BNDPrefixOk to near jumps, calls and rets. Add MPX instructions and bnd prefix. * i386-reg.tbl: Add bnd0-bnd3 registers. * i386-tbl.h: Regenerate.
2013-07-24 17:47:25 +02:00
call, 1, 0xe8, None, 1, CpuNo64, JumpDword|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk, { Disp16|Disp32 }
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
call, 1, 0xe8, None, 1, Cpu64, Amd64|JumpDword|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk, { Disp16|Disp32S }
x86: drop further pointless/bogus DefaultSize - 64-bit CALL permitting just a single operand size doesn't need it. - FLDENV et al should never have had it. It remains suspicious that a number of 64-bit only insns continue to have the attribute, despite this being intended for .code16gcc handling only.
2020-01-30 11:33:53 +01:00
call, 1, 0xe8, None, 1, Cpu64, Intel64|JumpDword|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk, { Disp32S }
x86: make JumpAbsolute an insn attribute ... instead of an operand one: There's only ever one operand here anyway.
2019-11-14 08:47:03 +01:00
call, 1, 0xff, 0x2, 1, CpuNo64, Modrm|JumpAbsolute|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk|NoTrackPrefixOk, { Reg16|Reg32|Unspecified|BaseIndex }
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
call, 1, 0xff, 0x2, 1, Cpu64, Amd64|Modrm|JumpAbsolute|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk|NoTrackPrefixOk, { Reg16|Reg64|Unspecified|BaseIndex }
x86: drop further pointless/bogus DefaultSize - 64-bit CALL permitting just a single operand size doesn't need it. - FLDENV et al should never have had it. It remains suspicious that a number of 64-bit only insns continue to have the attribute, despite this being intended for .code16gcc handling only.
2020-01-30 11:33:53 +01:00
call, 1, 0xff, 0x2, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk|NoTrackPrefixOk, { Reg64|Unspecified|BaseIndex }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
// Intel Syntax remaining call instances.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
call, 2, 0x9a, None, 1, CpuNo64, JumpInterSegment|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16, Imm16|Imm32 }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
call, 1, 0xff, 0x3, 1, 0, Amd64|Modrm|JumpAbsolute|DefaultSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Dword|Fword|BaseIndex }
call, 1, 0xff, 0x3, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Fword|Tbyte|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
lcall, 2, 0x9a, None, 1, CpuNo64, JumpInterSegment|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16, Imm16|Imm32 }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
lcall, 1, 0xff, 0x3, 1, 0, Amd64|Modrm|JumpAbsolute|DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
lcall, 1, 0xff, 0x3, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: fold two JMP templates Now that the AMD64 check in match_template() applies to 64-bit code only, the non-64-bit and the Amd64 template can be folded, as being otherwise compatible with one another. (Oddly enough the same doesn't apply to CALL, due to the suffixes it permits, while JMP doesn't allow for any.)
2020-02-12 16:19:52 +01:00
jmp, 1, 0xeb, None, 1, 0, Amd64|Jump|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk, { Disp8|Disp16|Disp32|Disp32S }
Replace CpuAMD64/CpuIntel64 with AMD64/Intel64 AMD64 vs CpuIntel64 ISA should be handled similar as AT&T vs Intel syntax. Since cpu_flags isn't sorted by position, we need to check the whole cpu_flags array for the maximum position when verifying CpuMax. gas/ PR gas/20154 * config/tc-i386.c (cpu_flags_match): Don't set cpuamd64 nor cpuintel64. (match_template): Check Intel64/AMD64 ISA. opcodes/ PR gas/20154 * i386-gen.c (cpu_flags): Remove CpuAMD64 and CpuIntel64. (opcode_modifiers): Add AMD64 and Intel64. (main): Properly verify CpuMax. * i386-opc.h (CpuAMD64): Removed. (CpuIntel64): Likewise. (CpuMax): Set to CpuNo64. (i386_cpu_flags): Remove cpuamd64 and cpuintel64. (AMD64): New. (Intel64): Likewise. (i386_opcode_modifier): Add amd64 and intel64. (i386-opc.tbl): Replace CpuAMD64/CpuIntel64 with AMD64/Intel64 on call and jmp. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2016-05-27 17:02:56 +02:00
jmp, 1, 0xeb, None, 1, Cpu64, Intel64|Jump|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk, { Disp8|Disp32S }
x86: make JumpAbsolute an insn attribute ... instead of an operand one: There's only ever one operand here anyway.
2019-11-14 08:47:03 +01:00
jmp, 1, 0xff, 0x4, 1, CpuNo64, Modrm|JumpAbsolute|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk|NoTrackPrefixOk, { Reg16|Reg32|Unspecified|BaseIndex }
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
jmp, 1, 0xff, 0x4, 1, Cpu64, Amd64|Modrm|JumpAbsolute|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk|NoTrackPrefixOk, { Reg16|Reg64|Unspecified|BaseIndex }
x86: make JumpAbsolute an insn attribute ... instead of an operand one: There's only ever one operand here anyway.
2019-11-14 08:47:03 +01:00
jmp, 1, 0xff, 0x4, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|BNDPrefixOk|NoTrackPrefixOk, { Reg64|Unspecified|BaseIndex }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
// Intel Syntax remaining jmp instances.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
jmp, 2, 0xea, None, 1, CpuNo64, JumpInterSegment|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16, Imm16|Imm32 }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
jmp, 1, 0xff, 0x5, 1, 0, Amd64|Modrm|JumpAbsolute|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Dword|Fword|BaseIndex }
jmp, 1, 0xff, 0x5, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Fword|Tbyte|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
ljmp, 2, 0xea, None, 1, CpuNo64, JumpInterSegment|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16, Imm16|Imm32 }
x86-64: Intel64 adjustments for insns dealing with far pointers AMD and Intel differ in their handling of far indirect branches as well as LFS/LGS/LSS: AMD CPUs ignore REX.W while Intel ones honors it. (Note how the latter three were hybrids so far, while far branches were fully AMD-like.)
2020-02-12 16:19:03 +01:00
ljmp, 1, 0xff, 0x5, 1, 0, Amd64|Modrm|JumpAbsolute|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
ljmp, 1, 0xff, 0x5, 1, Cpu64, Intel64|Modrm|JumpAbsolute|No_bSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
Support Intel MPX gas/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386.c (BND_PREFIX): New. (struct _i386_insn): Add new field bnd_prefix. (add_bnd_prefix): New. (cpu_arch): Add MPX. (i386_operand_type): Add regbnd. (md_assemble): Handle BND prefixes. (parse_insn): Likewise. (output_branch): Likewise. (output_jump): Likewise. (build_modrm_byte): Handle regbnd. (OPTION_MADD_BND_PREFIX): New. (md_longopts): Add entry for 'madd-bnd-prefix'. (md_parse_option): Handle madd-bnd-prefix option. (md_show_usage): Add description for madd-bnd-prefix option. * doc/c-i386.texi: Document mpx/.mpx and -madd-bnd-prefix. gas/testsuite/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/i386/mpx-add-bnd-prefix.s: New. * gas/i386/mpx-add-bnd-prefix.d: New. * gas/i386/mpx-inval-1.l: New. * gas/i386/mpx-inval-1.s: New. * gas/i386/mpx.d: New. * gas/i386/mpx.s: New. * gas/i386/x86-64-mpx-add-bnd-prefix.d: New. * gas/i386/x86-64-mpx-add-bnd-prefix.s: New. * gas/i386/x86-64-mpx-addr32.d: New. * gas/i386/x86-64-mpx-addr32.s: New. * gas/i386/x86-64-mpx-inval-1.l: New. * gas/i386/x86-64-mpx-inval-1.s: New. * gas/i386/x86-64-mpx-inval-2.l: New. * gas/i386/x86-64-mpx-inval-2.s: New. * gas/i386/x86-64-mpx.d: New. * gas/i386/x86-64-mpx.s: New. * gas/i386/nops.d: Adjust to MPX changes. * gas/i386/nops.s: Likewise. * gas/i386/x86-64-nops.d: Likewise. * gas/i386/x86-64-nops.s: Likewise. * gas/i386/ilp32/x86-64-nops.d: Likewise. * gas/i386/i386.exp: Run new MPX tests. include/opcode/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386.h (BND_PREFIX_OPCODE): New. opcodes/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis.c (BND_Fixup): New. (Ebnd): New. (Ev_bnd): New. (Gbnd): New. (BND): New. (v_bnd_mode): New. (bnd_mode): New. (MOD enum): Add new entries. (PREFIX enum): Likewise. (dis tables): Replace XX with BND for near branch and call instructions. (prefix_table): Add new entries. (mod_table): Likewise. (names_bnd): New. (intel_names_bnd): New. (att_names_bnd): New. (BND_PREFIX): New. (prefix_name): Handle BND_PREFIX. (print_insn): Initialize names_bnd. (intel_operand_size): Handle new modes. (OP_E_register): Likewise. (OP_E_memory): Likewise. (OP_G): Likewise. * i386-gen.c (cpu_flag_init): Add CpuMPX. (cpu_flags): Add CpuMPX. (operand_type_init): Add RegBND. (opcode_modifiers): Add BNDPrefixOk. (operand_types): Add RegBND. * i386-init.h: Regenerate. * i386-opc.h (CpuMPX): New. (CpuUnused): Comment out. (i386_cpu_flags): Add cpumpx. (BNDPrefixOk): New. (i386_opcode_modifier): Add bndprefixok. (RegBND): New. (i386_operand_type): Add regbnd. * i386-opc.tbl: Add BNDPrefixOk to near jumps, calls and rets. Add MPX instructions and bnd prefix. * i386-reg.tbl: Add bnd0-bnd3 registers. * i386-tbl.h: Regenerate.
2013-07-24 17:47:25 +02:00
ret, 0, 0xc3, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|RepPrefixOk|BNDPrefixOk, { 0 }
ret, 1, 0xc2, None, 1, CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf|RepPrefixOk|BNDPrefixOk, { Imm16 }
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
ret, 0, 0xc3, None, 1, Cpu64, Amd64|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|RepPrefixOk|BNDPrefixOk, { 0 }
ret, 1, 0xc2, None, 1, Cpu64, Amd64|DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|RepPrefixOk|BNDPrefixOk, { Imm16 }
x86-64: honor vendor specifics for near RET While vendors agree about default operand size (64 bits) and hence unavilability of a 32-bit form, AMD honors a 16-bit operand size override (0x66) while Intel doesn't.
2020-01-30 11:36:33 +01:00
ret, 0, 0xc3, None, 1, Cpu64, Intel64|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|RepPrefixOk|BNDPrefixOk, { 0 }
ret, 1, 0xc2, None, 1, Cpu64, Intel64|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64|RepPrefixOk|BNDPrefixOk, { Imm16 }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
lret, 0, 0xcb, None, 1, 0, DefaultSize|No_bSuf|No_sSuf|No_ldSuf, { 0 }
lret, 1, 0xca, None, 1, 0, DefaultSize|No_bSuf|No_sSuf|No_ldSuf, { Imm16 }
gas/testsuite/ 2008-08-28 Jan Beulich <jbeulich@novell.com> * gas/i386/intel.s: Add retf. * gas/i386/intel.{d,e}: Adjust. * gas/i386/opcode-intel.d: Replace lret with retf. opcodes/ 2008-08-28 Jan Beulich <jbeulich@novell.com> * i386-dis.c (dis386): Adjust far return mnemonics. * i386-opc.tbl: Add retf. * i386-tbl.h: Re-generate.
2008-08-28 17:59:32 +02:00
// Intel Syntax.
retf, 0, 0xcb, None, 1, 0, DefaultSize|No_bSuf|No_sSuf|No_ldSuf, { 0 }
retf, 1, 0xca, None, 1, 0, DefaultSize|No_bSuf|No_sSuf|No_ldSuf, { Imm16 }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
enter, 2, 0xc8, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm16, Imm8 }
enter, 2, 0xc8, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Imm16, Imm8 }
leave, 0, 0xc9, None, 1, Cpu186|CpuNo64, DefaultSize|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
leave, 0, 0xc9, None, 1, Cpu64, DefaultSize|No_bSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: allow opcode templates to be templated In order to reduce redundancy as well as the chance of things going out of sync (see a later patch for an example), make the opcode table generator capable of recognizing and expanding templated templates. Use the new capability for compacting the general purpose conditional insns.
2020-03-09 10:12:14 +01:00
<cc:opc, o:0, no:1, b:2, c:2, nae:2, nb:3, nc:3, ae:3, e:4, z:4, ne:5, nz:5, be:6, na:6, nbe:7, a:7, \
s:8, ns:9, p:a, pe:a, np:b, po:b, l:c, nge:c, nl:d, ge:d, le:e, ng:e, nle:f, g:f>
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Conditional jumps.
x86: allow opcode templates to be templated In order to reduce redundancy as well as the chance of things going out of sync (see a later patch for an example), make the opcode table generator capable of recognizing and expanding templated templates. Use the new capability for compacting the general purpose conditional insns.
2020-03-09 10:12:14 +01:00
j<cc>, 1, 0x7<cc:opc>, None, 1, 0, Jump|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|BNDPrefixOk, { Disp8|Disp16|Disp32|Disp32S }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// jcxz vs. jecxz is chosen on the basis of the address size prefix.
x86: consolidate Disp<NN> handling a little In memory operand addressing, which forms of displacement are permitted besides Disp8 is pretty clearly limited - outside of 64-bit mode, Disp16 or Disp32 only, depending on address size (MPX being special in not allowing Disp16), - in 64-bit mode, Disp32s or Disp64 without address size override, and solely Disp32 with one. Adjust assembler and i386-gen to match this, observing that templates already get adjusted before trying to match them against input depending on the presence of an address size prefix. This adjustment logic gets extended to all cases, as certain DispNN values should also be dropped when there's no such prefix. In fact behavior of the assembler, perhaps besides the exact diagnostics wording, should not differ between there being templates applicable to 64-bit and non-64-bit at the same time, or there being fully separate sets of templates, with their DispNN settings already reduced accordingly. This adjustment logic further gets guarded such that there wouldn't be and Disp<N> conversion based on address size prefix when this prefix doesn't control the width of the displacement (on branches other than absolute ones). These adjustments then also allow folding two MOV templates, which had been split between 64-bit and non-64-bits variants so far. Once in this area also - drop the bogus DispNN from JumpByte templates, leaving just the correct Disp8 there (compensated by i386_finalize_displacement() now setting Disp8 on their operands), - add the missing Disp32S to XBEGIN. Note that the changes make it necessary to temporarily mark a test as XFAIL; this will get taken care of by a subsequent patch. The failing parts are entirely bogus and will get replaced.
2019-12-27 09:22:03 +01:00
jcxz, 1, 0xe3, None, 1, CpuNo64, JumpByte|Size16|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
jecxz, 1, 0xe3, None, 1, 0, JumpByte|Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
jrcxz, 1, 0xe3, None, 1, Cpu64, JumpByte|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Disp8 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// The loop instructions also use the address size prefix to select
// %cx rather than %ecx for the loop count, so the `w' form of these
// instructions emit an address size prefix rather than a data size
// prefix.
x86: consolidate Disp<NN> handling a little In memory operand addressing, which forms of displacement are permitted besides Disp8 is pretty clearly limited - outside of 64-bit mode, Disp16 or Disp32 only, depending on address size (MPX being special in not allowing Disp16), - in 64-bit mode, Disp32s or Disp64 without address size override, and solely Disp32 with one. Adjust assembler and i386-gen to match this, observing that templates already get adjusted before trying to match them against input depending on the presence of an address size prefix. This adjustment logic gets extended to all cases, as certain DispNN values should also be dropped when there's no such prefix. In fact behavior of the assembler, perhaps besides the exact diagnostics wording, should not differ between there being templates applicable to 64-bit and non-64-bit at the same time, or there being fully separate sets of templates, with their DispNN settings already reduced accordingly. This adjustment logic further gets guarded such that there wouldn't be and Disp<N> conversion based on address size prefix when this prefix doesn't control the width of the displacement (on branches other than absolute ones). These adjustments then also allow folding two MOV templates, which had been split between 64-bit and non-64-bits variants so far. Once in this area also - drop the bogus DispNN from JumpByte templates, leaving just the correct Disp8 there (compensated by i386_finalize_displacement() now setting Disp8 on their operands), - add the missing Disp32S to XBEGIN. Note that the changes make it necessary to temporarily mark a test as XFAIL; this will get taken care of by a subsequent patch. The failing parts are entirely bogus and will get replaced.
2019-12-27 09:22:03 +01:00
loop, 1, 0xe2, None, 1, CpuNo64, JumpByte|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
loop, 1, 0xe2, None, 1, Cpu64, JumpByte|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Disp8 }
loopz, 1, 0xe1, None, 1, CpuNo64, JumpByte|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
loopz, 1, 0xe1, None, 1, Cpu64, JumpByte|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Disp8 }
loope, 1, 0xe1, None, 1, CpuNo64, JumpByte|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
loope, 1, 0xe1, None, 1, Cpu64, JumpByte|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Disp8 }
loopnz, 1, 0xe0, None, 1, CpuNo64, JumpByte|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
loopnz, 1, 0xe0, None, 1, Cpu64, JumpByte|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Disp8 }
loopne, 1, 0xe0, None, 1, CpuNo64, JumpByte|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp8 }
loopne, 1, 0xe0, None, 1, Cpu64, JumpByte|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Disp8 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Set byte on flag instructions.
x86: allow opcode templates to be templated In order to reduce redundancy as well as the chance of things going out of sync (see a later patch for an example), make the opcode table generator capable of recognizing and expanding templated templates. Use the new capability for compacting the general purpose conditional insns.
2020-03-09 10:12:14 +01:00
set<cc>, 1, 0xf9<cc:opc>, 0x0, 2, Cpu386, Modrm|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg8|Byte|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// String manipulation.
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
cmps, 0, 0xa6, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
cmps, 2, 0xa6, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
scmp, 0, 0xa6, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
scmp, 2, 0xa6, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
ins, 0, 0x6c, None, 1, Cpu186, W|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
ins, 2, 0x6c, None, 1, Cpu186, W|No_sSuf|No_qSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { InOutPortReg, Byte|Word|Dword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
outs, 0, 0x6e, None, 1, Cpu186, W|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
outs, 2, 0x6e, None, 1, Cpu186, W|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { Byte|Word|Dword|Unspecified|BaseIndex, InOutPortReg }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
lods, 0, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: string insns don't allow displacements Remove the misleading indicators from the table.
2017-11-14 08:40:48 +01:00
lods, 1, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
lods, 2, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
slod, 0, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: string insns don't allow displacements Remove the misleading indicators from the table.
2017-11-14 08:40:48 +01:00
slod, 1, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86: drop stray CheckRegSize uses They are relevant only when multiple operands permit registers: operand_type_register_match() returns true if either operand is not a register one. IOW grep -i CheckRegSize i386-opc.tbl | grep -Ev "(Reg[8136]|Acc).*,.*(Reg|Acc)" should produce no output.
2017-12-15 09:13:54 +01:00
slod, 2, 0xac, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
movs, 0, 0xa4, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
movs, 2, 0xa4, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
smov, 0, 0xa4, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
smov, 2, 0xa4, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
scas, 0, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
scas, 1, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
scas, 2, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
ssca, 0, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
ssca, 1, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ssca, 2, 0xae, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex, Acc|Byte|Word|Dword|Qword }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
stos, 0, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
stos, 1, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
stos, 2, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { Acc|Byte|Word|Dword|Qword, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
ssto, 0, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
ssto, 1, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Byte|Word|Dword|Qword|Unspecified|BaseIndex }
ssto, 2, 0xaa, None, 1, 0, W|No_sSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { Acc|Byte|Word|Dword|Qword, Byte|Word|Dword|Qword|Unspecified|BaseIndex }
x86/Intel: fix operand checking for MOVSD The dual purpose mnemonic (string move vs scalar double move) breaks the assumption that the isstring flag would be set on both the first and last entry in the current set of templates, which results in bogus or missing diagnostics for the string move variant of the mnemonic. Short of mostly rewriting i386_index_check() and its interaction with the rest of the code, simply shrink the template set to just string instructions when encountering the second memory operand, and run i386_index_check() a second time for the first memory operand after that reduction.
2016-07-01 08:56:13 +02:00
xlat, 0, 0xd7, None, 1, 0, No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString, { 0 }
x86: string insns don't allow displacements Remove the misleading indicators from the table.
2017-11-14 08:40:48 +01:00
xlat, 1, 0xd7, None, 1, 0, No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString, { Byte|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Bit manipulation.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
bsf, 2, 0xfbc, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|RepPrefixOk, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
bsr, 2, 0xfbd, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|RepPrefixOk, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
bt, 2, 0xfa3, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
bt, 2, 0xfba, 0x4, 2, Cpu386, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Imm8, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
btc, 2, 0xfbb, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
btc, 2, 0xfba, 0x7, 2, Cpu386, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
btr, 2, 0xfb3, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
btr, 2, 0xfba, 0x6, 2, Cpu386, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
bts, 2, 0xfab, None, 2, Cpu386, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg16|Reg32|Reg64, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
bts, 2, 0xfba, 0x5, 2, Cpu386, Modrm|No_bSuf|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Imm8, Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Interrupts & op. sys insns.
// See gas/config/tc-i386.c for conversion of 'int $3' into the special
// int 3 insn.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
int, 1, 0xcd, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
int3, 0, 0xcc, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
into, 0, 0xce, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
iret, 0, 0xcf, None, 1, 0, DefaultSize|No_bSuf|No_sSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// i386sl, i486sl, later 486, and Pentium.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
rsm, 0, 0xfaa, None, 2, Cpu386, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
bound, 2, 0x62, None, 1, Cpu186|CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32, Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
hlt, 0, 0xf4, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
nop, 1, 0xf1f, 0x0, 2, CpuNop, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// nop is actually "xchg %ax,%ax" in 16bit mode, "xchg %eax,%eax" in
// 32bit mode and "xchg %rax,%rax" in 64bit mode.
gas/testsuite/ * gas/i386/rep-suffix.s: Add 'rep nop' case. * gas/i386/x86-64-rep-suffix.s: Likewise. * gas/i386/rep-suffix.d: Updated. * gas/i386/x86-64-rep-suffix.d: Likewise. * gas/i386/ilp32/x86-64-rep-suffix.d: Likewise. opcodes/ * i386-opc.tbl: Add RepPrefixOk to nop. * i386-tbl.h: Regenerate.
2012-07-02 20:12:28 +02:00
nop, 0, 0x90, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|RepPrefixOk, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Protection control.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
arpl, 2, 0x63, None, 1, Cpu286|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16, Reg16|Word|Unspecified|BaseIndex }
lar, 2, 0xf02, None, 2, Cpu286, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
lgdt, 1, 0xf01, 0x2, 2, Cpu286|CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Fword|Unspecified|BaseIndex }
lgdt, 1, 0xf01, 0x2, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Tbyte|Unspecified|BaseIndex }
lidt, 1, 0xf01, 0x3, 2, Cpu286|CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Fword|Unspecified|BaseIndex }
lidt, 1, 0xf01, 0x3, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Tbyte|Unspecified|BaseIndex }
lldt, 1, 0xf00, 0x2, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex }
lmsw, 1, 0xf01, 0x6, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex }
lsl, 2, 0xf03, None, 2, Cpu286, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
ltr, 1, 0xf00, 0x3, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex }
sgdt, 1, 0xf01, 0x0, 2, Cpu286|CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Fword|Unspecified|BaseIndex }
sgdt, 1, 0xf01, 0x0, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Tbyte|Unspecified|BaseIndex }
sidt, 1, 0xf01, 0x1, 2, Cpu286|CpuNo64, Modrm|No_bSuf|No_sSuf|No_qSuf|No_ldSuf, { Fword|Unspecified|BaseIndex }
sidt, 1, 0xf01, 0x1, 2, Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Tbyte|Unspecified|BaseIndex }
x86: don't emit REX.W for SLDT and STR Just like for other selector register reads, they're unnecessary and should hence be avoided.
2018-06-01 08:40:38 +02:00
sldt, 1, 0xf00, 0x0, 2, Cpu286, Modrm|No_bSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
sldt, 1, 0xf00, 0x0, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
smsw, 1, 0xf01, 0x4, 2, Cpu286, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
smsw, 1, 0xf01, 0x4, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
x86: don't emit REX.W for SLDT and STR Just like for other selector register reads, they're unnecessary and should hence be avoided.
2018-06-01 08:40:38 +02:00
str, 1, 0xf00, 0x1, 2, Cpu286, Modrm|No_bSuf|No_sSuf|No_ldSuf|NoRex64, { Reg16|Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
str, 1, 0xf00, 0x1, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
verr, 1, 0xf00, 0x4, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex }
verw, 1, 0xf00, 0x5, 2, Cpu286, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Word|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Floating point instructions.
// load
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fld, 1, 0xd9c0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fld, 1, 0xd9, 0x0, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fld, 1, 0xd9c0, None, 2, CpuFP, IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Intel Syntax
x86: drop stray FloatMF The flag is supposed to be used in templates which allow for both a "short" and a "long" format memory operand. Drop it from templates not matching this pattern. In the control/status word cases it was (ab)used in place of the intended IgnoreSize.
2019-08-07 10:46:52 +02:00
fld, 1, 0xdb, 0x5, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Tbyte|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fild, 1, 0xdf, 0x0, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
fild, 1, 0xdf, 0x5, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: don't accept FI{LD,STP,STTP}LL in Intel syntax mode As of commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic") these have been accepted even with "qword ptr" operand size specifier, but in 64-bit mode they're now wrongly having a REX prefix (with REX.W set) emitted in this case. These aren't Intel syntax mnemonics, so rather than fixing code generation, let's simply reject them. As a result, the Qword attribute can then be dropped, too.
2020-03-06 08:53:56 +01:00
fildll, 1, 0xdf, 0x5, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fldt, 1, 0xdb, 0x5, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Tbyte|Unspecified|BaseIndex }
fbld, 1, 0xdf, 0x4, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Tbyte|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// store (no pop)
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fst, 1, 0xddd0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fst, 1, 0xd9, 0x2, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fst, 1, 0xddd0, None, 2, CpuFP, IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fist, 1, 0xdf, 0x2, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// store (with pop)
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fstp, 1, 0xddd8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fstp, 1, 0xd9, 0x3, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fstp, 1, 0xddd8, None, 2, CpuFP, IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Intel Syntax
x86: drop stray FloatMF The flag is supposed to be used in templates which allow for both a "short" and a "long" format memory operand. Drop it from templates not matching this pattern. In the control/status word cases it was (ab)used in place of the intended IgnoreSize.
2019-08-07 10:46:52 +02:00
fstp, 1, 0xdb, 0x7, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Tbyte|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fistp, 1, 0xdf, 0x3, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
fistp, 1, 0xdf, 0x7, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: don't accept FI{LD,STP,STTP}LL in Intel syntax mode As of commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic") these have been accepted even with "qword ptr" operand size specifier, but in 64-bit mode they're now wrongly having a REX prefix (with REX.W set) emitted in this case. These aren't Intel syntax mnemonics, so rather than fixing code generation, let's simply reject them. As a result, the Qword attribute can then be dropped, too.
2020-03-06 08:53:56 +01:00
fistpll, 1, 0xdf, 0x7, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fstpt, 1, 0xdb, 0x7, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Tbyte|Unspecified|BaseIndex }
fbstp, 1, 0xdf, 0x6, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf, { Tbyte|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// exchange %st<n> with %st0
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fxch, 1, 0xd9c8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fxch %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fxch, 0, 0xd9c9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// comparison (without pop)
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcom, 1, 0xd8d0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fcom %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fcom, 0, 0xd8d1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fcom, 1, 0xd8, 0x2, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcom, 1, 0xd8d0, None, 2, CpuFP, IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ficom, 1, 0xde, 0x2, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// comparison (with pop)
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcomp, 1, 0xd8d8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fcomp %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fcomp, 0, 0xd8d9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fcomp, 1, 0xd8, 0x3, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcomp, 1, 0xd8d8, None, 2, CpuFP, IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ficomp, 1, 0xde, 0x3, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fcompp, 0, 0xded9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// unordered comparison (with pop)
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fucom, 1, 0xdde0, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fucom %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fucom, 0, 0xdde1, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fucomp, 1, 0xdde8, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fucomp %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fucomp, 0, 0xdde9, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fucompp, 0, 0xdae9, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
ftst, 0, 0xd9e4, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fxam, 0, 0xd9e5, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// load constants into %st0
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fld1, 0, 0xd9e8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldl2t, 0, 0xd9e9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldl2e, 0, 0xd9ea, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldpi, 0, 0xd9eb, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldlg2, 0, 0xd9ec, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldln2, 0, 0xd9ed, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fldz, 0, 0xd9ee, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Arithmetic.
// add
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fadd, 2, 0xd8c0, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fadd %st(i), %st
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fadd, 1, 0xd8c0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for faddp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fadd, 0, 0xdec1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fadd, 1, 0xd8, 0x0, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fiadd, 1, 0xde, 0x0, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
faddp, 2, 0xdec0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatAcc, FloatReg }
faddp, 1, 0xdec0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for faddp %st, %st(1)
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
faddp, 0, 0xdec1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
faddp, 2, 0xdec0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// subtract
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsub, 1, 0xd8e0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fsub, 2, 0xd8e0, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fsubp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsub, 0, 0xdee1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic|ATTSyntax, { 0 }
fsub, 0, 0xdee9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsub, 2, 0xd8e0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|D|FloatR, { FloatReg, FloatAcc }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fsub, 1, 0xd8, 0x4, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fisub, 1, 0xde, 0x4, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubp, 2, 0xdee0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatAcc, FloatReg }
fsubp, 1, 0xdee0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsubp, 0, 0xdee1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubp, 2, 0xdee8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatAcc, FloatReg }
fsubp, 1, 0xdee8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsubp, 0, 0xdee9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// subtract reverse
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubr, 1, 0xd8e8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fsubr, 2, 0xd8e8, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fsubrp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsubr, 0, 0xdee9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic|ATTSyntax, { 0 }
fsubr, 0, 0xdee1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubr, 2, 0xd8e8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|D|FloatR, { FloatReg, FloatAcc }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fsubr, 1, 0xd8, 0x5, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fisubr, 1, 0xde, 0x5, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubrp, 2, 0xdee8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatAcc, FloatReg }
fsubrp, 1, 0xdee8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsubrp, 0, 0xdee9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fsubrp, 2, 0xdee0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatAcc, FloatReg }
fsubrp, 1, 0xdee0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fsubrp, 0, 0xdee1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// multiply
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fmul, 2, 0xd8c8, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fmul, 1, 0xd8c8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fmulp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fmul, 0, 0xdec9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fmul, 1, 0xd8, 0x1, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fimul, 1, 0xde, 0x1, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fmulp, 2, 0xdec8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatAcc, FloatReg }
fmulp, 1, 0xdec8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fmulp, 0, 0xdec9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fmulp, 2, 0xdec8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// divide
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdiv, 1, 0xd8f0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fdiv, 2, 0xd8f0, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fdivp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdiv, 0, 0xdef1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic|ATTSyntax, { 0 }
fdiv, 0, 0xdef9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdiv, 2, 0xd8f0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|D|FloatR, { FloatReg, FloatAcc }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fdiv, 1, 0xd8, 0x6, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fidiv, 1, 0xde, 0x6, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivp, 2, 0xdef0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatAcc, FloatReg }
fdivp, 1, 0xdef0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdivp, 0, 0xdef1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivp, 2, 0xdef8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatAcc, FloatReg }
fdivp, 1, 0xdef8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdivp, 0, 0xdef9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// divide reverse
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivr, 1, 0xd8f8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fdivr, 2, 0xd8f8, None, 2, CpuFP, D|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg, FloatAcc }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for fdivrp
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdivr, 0, 0xdef9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic|ATTSyntax, { 0 }
fdivr, 0, 0xdef1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Ugh|ATTMnemonic, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivr, 2, 0xd8f8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|D|FloatR, { FloatReg, FloatAcc }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
fdivr, 1, 0xd8, 0x7, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Dword|Qword|Unspecified|BaseIndex }
fidivr, 1, 0xde, 0x7, 1, CpuFP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivrp, 2, 0xdef8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatAcc, FloatReg }
fdivrp, 1, 0xdef8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdivrp, 0, 0xdef9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTMnemonic|ATTSyntax, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fdivrp, 2, 0xdef0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatAcc, FloatReg }
fdivrp, 1, 0xdef0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fdivrp, 0, 0xdef1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { 0 }
f2xm1, 0, 0xd9f0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fyl2x, 0, 0xd9f1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fptan, 0, 0xd9f2, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fpatan, 0, 0xd9f3, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fxtract, 0, 0xd9f4, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fprem1, 0, 0xd9f5, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fdecstp, 0, 0xd9f6, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fincstp, 0, 0xd9f7, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fprem, 0, 0xd9f8, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fyl2xp1, 0, 0xd9f9, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fsqrt, 0, 0xd9fa, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fsincos, 0, 0xd9fb, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
frndint, 0, 0xd9fc, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fscale, 0, 0xd9fd, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fsin, 0, 0xd9fe, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fcos, 0, 0xd9ff, None, 2, Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fchs, 0, 0xd9e0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fabs, 0, 0xd9e1, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// processor control
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fninit, 0, 0xdbe3, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
finit, 0, 0xdbe3, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
x86: drop stray FloatMF The flag is supposed to be used in templates which allow for both a "short" and a "long" format memory operand. Drop it from templates not matching this pattern. In the control/status word cases it was (ab)used in place of the intended IgnoreSize.
2019-08-07 10:46:52 +02:00
fldcw, 1, 0xd9, 0x5, 1, CpuFP, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
fnstcw, 1, 0xd9, 0x7, 1, CpuFP, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
fstcw, 1, 0xd9, 0x7, 1, CpuFP, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { Word|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fnstsw, 1, 0xdfe0, None, 2, Cpu287|Cpu387, IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Word }
x86: drop stray FloatMF The flag is supposed to be used in templates which allow for both a "short" and a "long" format memory operand. Drop it from templates not matching this pattern. In the control/status word cases it was (ab)used in place of the intended IgnoreSize.
2019-08-07 10:46:52 +02:00
fnstsw, 1, 0xdd, 0x7, 1, CpuFP, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fnstsw, 0, 0xdfe0, None, 2, Cpu287|Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fstsw, 1, 0xdfe0, None, 2, Cpu287|Cpu387, IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { Acc|Word }
x86: drop stray FloatMF The flag is supposed to be used in templates which allow for both a "short" and a "long" format memory operand. Drop it from templates not matching this pattern. In the control/status word cases it was (ab)used in place of the intended IgnoreSize.
2019-08-07 10:46:52 +02:00
fstsw, 1, 0xdd, 0x7, 1, CpuFP, Modrm|IgnoreSize|No_bSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { Word|Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fstsw, 0, 0xdfe0, None, 2, Cpu287|Cpu387, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
fnclex, 0, 0xdbe2, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fclex, 0, 0xdbe2, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
x86: drop further pointless/bogus DefaultSize - 64-bit CALL permitting just a single operand size doesn't need it. - FLDENV et al should never have had it. It remains suspicious that a number of 64-bit only insns continue to have the attribute, despite this being intended for .code16gcc handling only.
2020-01-30 11:33:53 +01:00
// Short forms of fldenv, fstenv, fsave, and frstor use data size prefix.
fnstenv, 1, 0xd9, 0x6, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
fstenv, 1, 0xd9, 0x6, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf|FWait, { Unspecified|BaseIndex }
fldenv, 1, 0xd9, 0x4, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
fnsave, 1, 0xdd, 0x6, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
fsave, 1, 0xdd, 0x6, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf|FWait, { Unspecified|BaseIndex }
frstor, 1, 0xdd, 0x4, 1, CpuFP, Modrm|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
// 8087 only
fneni, 0, 0xdbe0, None, 2, Cpu8087, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
feni, 0, 0xdbe0, None, 2, Cpu8087, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
fndisi, 0, 0xdbe1, None, 2, Cpu8087, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fdisi, 0, 0xdbe1, None, 2, Cpu8087, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
// 287 only
fnsetpm, 0, 0xdbe4, None, 2, Cpu287, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fsetpm, 0, 0xdbe4, None, 2, Cpu287, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|FWait, { 0 }
frstpm, 0, 0xdbe5, None, 2, Cpu287, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
ffree, 1, 0xddc0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// P6:free st(i), pop st
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
ffreep, 1, 0xdfc0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
gas/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * tc-i386.c (cpu_arch): Add .8087, .287, .387, .no87, .nommx, .nosse, and .noavx. (cpu_flags_and_not): New. (set_cpu_arch): Check whether sub-architecture specified is a feature disable. (md_parse_option): Likewise. (parse_real_register): Don't return floating point register when x87 functionality is disabled. (md_show_usage): Add new sub-options. * doc/c-i386.texi: Update with new command line sub-options. gas/testsuite/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * gas/i386/8087.[ds]: New. * gas/i386/287.[ds]: New. * gas/i386/387.[ds]: New. * gas/i386/no87.[ls]: New. * gas/i386/no87-2.[ls]: New. * gas/i386/i386.exp: Run new tests. * gas/i386/att-regs.s: Also check FPU register access. * gas/i386/intel-regs.s: Likewise. * gas/i386/att-regs.d: Adjust expectations. * gas/i386/intel-regs.d: Likewise. opcodes/ 2009-07-24 Jan Beulich <jbeulich@novell.com> * i386-dis.c (fgrps): Correct annotation for feni/fdisi. Add frstpm. * i386-gen.c (cpu_flag_init): Add FP enabling flags where needed. (cpu_flags): Add Cpu8087, Cpu287, Cpu387, Cpu687, and CpuFISTTP. (set_bitfield): Expand CpuFP to Cpu8087|Cpu287|Cpu387. * i386-opc.h (Cpu8087, Cpu287, Cpu387, Cpu687, CpuFISTTP): Define. (union i386_cpu_flags): Add cpu8087, cpu287, cpu387, cpu687, and cpufisttp. * i386-opc.tbl: Qualify floating point instructions by their respective CpuXXX flag. Fix fucom{,p,pp}, fprem1, fsin, fcos, and fsincos to be avilable only on 387. Fix fstsw ax to be available only on 287+. Add f{,n}eni, f{,n}disi, f{,n}setpm, and frstpm. * i386-init.h, i386-tbl.h: Regenerate.
2009-07-24 17:41:20 +02:00
fnop, 0, 0xd9d0, None, 2, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
fwait, 0, 0x9b, None, 1, CpuFP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Opcode prefixes; we allow them as separate insns too.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
addr16, 0, 0x67, None, 1, Cpu386|CpuNo64, Size16|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
addr32, 0, 0x67, None, 1, Cpu386, Size32|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
aword, 0, 0x67, None, 1, Cpu386|CpuNo64, Size16|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
adword, 0, 0x67, None, 1, Cpu386, Size32|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
data16, 0, 0x66, None, 1, Cpu386, Size16|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
data32, 0, 0x66, None, 1, Cpu386|CpuNo64, Size32|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
word, 0, 0x66, None, 1, Cpu386, Size16|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
dword, 0, 0x66, None, 1, Cpu386|CpuNo64, Size32|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
lock, 0, 0xf0, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
wait, 0, 0x9b, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
cs, 0, 0x2e, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
ds, 0, 0x3e, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
es, 0, 0x26, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
fs, 0, 0x64, None, 1, Cpu386, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
gs, 0, 0x65, None, 1, Cpu386, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
ss, 0, 0x36, None, 1, CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rep, 0, 0xf3, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
repe, 0, 0xf3, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
repz, 0, 0xf3, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
repne, 0, 0xf2, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
repnz, 0, 0xf2, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
ht, 0, 0x3e, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
hnt, 0, 0x2e, None, 1, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex, 0, 0x40, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexz, 0, 0x41, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexy, 0, 0x42, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexyz, 0, 0x43, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexx, 0, 0x44, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexxz, 0, 0x45, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexxy, 0, 0x46, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rexxyz, 0, 0x47, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64, 0, 0x48, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64z, 0, 0x49, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64y, 0, 0x4a, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64yz, 0, 0x4b, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64x, 0, 0x4c, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64xz, 0, 0x4d, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64xy, 0, 0x4e, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex64xyz, 0, 0x4f, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.b, 0, 0x41, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.x, 0, 0x42, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.xb, 0, 0x43, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.r, 0, 0x44, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.rb, 0, 0x45, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.rx, 0, 0x46, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.rxb, 0, 0x47, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.w, 0, 0x48, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wb, 0, 0x49, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wx, 0, 0x4a, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wxb, 0, 0x4b, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wr, 0, 0x4c, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wrb, 0, 0x4d, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wrx, 0, 0x4e, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
rex.wrxb, 0, 0x4f, None, 1, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
X86: Add pseudo prefixes to control encoding Many x86 instructions have more than one encodings. Assembler picks the default one, usually the shortest one. Although the ".s", ".d8" and ".d32" suffixes can be used to swap register operands or specify displacement size, they aren't very flexible. This patch adds pseudo prefixes, {xxx}, to control instruction encoding. The available pseudo prefixes are {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Pseudo prefixes are preferred over the ".s", ".d8" and ".d32" suffixes, which are deprecated. gas/ * config/tc-i386.c (_i386_insn): Add dir_encoding and vec_encoding. Remove swap_operand and need_vrex. (extra_symbol_chars): Add '}'. (md_begin): Mark '}' with LEX_BEGIN_NAME. Allow '}' in mnemonic. (build_vex_prefix): Don't use 2-byte VEX encoding with {vex3}. Check dir_encoding and load. (parse_insn): Check pseudo prefixes. Set dir_encoding. (VEX_check_operands): Likewise. (match_template): Check dir_encoding and load. (parse_real_register): Set vec_encoding instead of need_vrex. (parse_register): Likewise. * doc/c-i386.texi: Document {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Remove ".s", ".d8" and ".d32" * testsuite/gas/i386/i386.exp: Run pseudos and x86-64-pseudos. * testsuite/gas/i386/pseudos.d: New file. * testsuite/gas/i386/pseudos.s: Likewise. * testsuite/gas/i386/x86-64-pseudos.d: Likewise. * testsuite/gas/i386/x86-64-pseudos.s: Likewise. opcodes/ * i386-gen.c (opcode_modifiers): Replace S with Load. * i386-opc.h (S): Removed. (Load): New. (i386_opcode_modifier): Replace s with load. * i386-opc.tbl: Add {disp8}, {disp32}, {swap}, {vex2}, {vex3} and {evex}. Replace S with Load. * i386-tbl.h: Regenerated.
2017-03-09 18:58:46 +01:00
// Pseudo prefixes (opcode_length == 0)
{disp8}, 0, 0x0, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
{disp32}, 0, 0x1, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
{load}, 0, 0x2, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
{store}, 0, 0x3, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
x86: Add {vex} pseudo prefix There are 2-byte VEX prefix and 3-byte VEX prefix. 2-byte VEX prefix can't encode all operands. By default, assembler tries 2-byte VEX prefix first. {vex3} can be used to force 3-byte VEX prefix. This patch adds {vex} pseudo prefix and keeps {vex2} for backward compatibility. gas/ * config/tc-i386.c (_i386_insn): Replace vex_encoding_vex2 with vex_encoding_vex. (parse_insn): Likewise. * doc/c-i386.texi: Replace {vex2} with {vex}. Update {vex} and {vex3} documentation. * testsuite/gas/i386/pseudos.s: Replace 3 {vex2} tests with {vex}. * testsuite/gas/i386/x86-64-pseudos.s: Likewise. opcodes/ * i386-opc.tbl: Add {vex} pseudo prefix. * i386-tbl.h: Regenerated.
2020-01-17 16:07:55 +01:00
{vex}, 0, 0x4, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
X86: Add pseudo prefixes to control encoding Many x86 instructions have more than one encodings. Assembler picks the default one, usually the shortest one. Although the ".s", ".d8" and ".d32" suffixes can be used to swap register operands or specify displacement size, they aren't very flexible. This patch adds pseudo prefixes, {xxx}, to control instruction encoding. The available pseudo prefixes are {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Pseudo prefixes are preferred over the ".s", ".d8" and ".d32" suffixes, which are deprecated. gas/ * config/tc-i386.c (_i386_insn): Add dir_encoding and vec_encoding. Remove swap_operand and need_vrex. (extra_symbol_chars): Add '}'. (md_begin): Mark '}' with LEX_BEGIN_NAME. Allow '}' in mnemonic. (build_vex_prefix): Don't use 2-byte VEX encoding with {vex3}. Check dir_encoding and load. (parse_insn): Check pseudo prefixes. Set dir_encoding. (VEX_check_operands): Likewise. (match_template): Check dir_encoding and load. (parse_real_register): Set vec_encoding instead of need_vrex. (parse_register): Likewise. * doc/c-i386.texi: Document {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Remove ".s", ".d8" and ".d32" * testsuite/gas/i386/i386.exp: Run pseudos and x86-64-pseudos. * testsuite/gas/i386/pseudos.d: New file. * testsuite/gas/i386/pseudos.s: Likewise. * testsuite/gas/i386/x86-64-pseudos.d: Likewise. * testsuite/gas/i386/x86-64-pseudos.s: Likewise. opcodes/ * i386-gen.c (opcode_modifiers): Replace S with Load. * i386-opc.h (S): Removed. (Load): New. (i386_opcode_modifier): Replace s with load. * i386-opc.tbl: Add {disp8}, {disp32}, {swap}, {vex2}, {vex3} and {evex}. Replace S with Load. * i386-tbl.h: Regenerated.
2017-03-09 18:58:46 +01:00
{vex2}, 0, 0x4, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
{vex3}, 0, 0x5, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
{evex}, 0, 0x6, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
x86: Add {rex} pseudo prefix Add {rex} pseudo prefix to generate a REX byte for integer and legacy vector instructions if possible. Note that this differs from the rex prefix which generates REX prefix unconditionally. gas/ * config/tc-i386.c (_i386_insn): Add rex_encoding. (md_assemble): When i.rex_encoding is true, generate a REX byte if possible. (parse_insn): Set i.rex_encoding for {rex}. * doc/c-i386.texi: Document {rex}. * testsuite/gas/i386/x86-64-pseudos.s: Add {rex} tests. * testsuite/gas/i386/x86-64-pseudos.d: Updated. opcodes/ * i386-opc.tbl: Add {rex}, * i386-tbl.h: Regenerated.
2018-02-22 15:18:27 +01:00
{rex}, 0, 0x7, None, 0, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
x86: Add -O[2|s] assembler command-line options On x86, some instructions have alternate shorter encodings: 1. When the upper 32 bits of destination registers of andq $imm31, %r64 testq $imm31, %r64 xorq %r64, %r64 subq %r64, %r64 known to be zero, we can encode them without the REX_W bit: andl $imm31, %r32 testl $imm31, %r32 xorl %r32, %r32 subl %r32, %r32 This optimization is enabled with -O, -O2 and -Os. 2. Since 0xb0 mov with 32-bit destination registers zero-extends 32-bit immediate to 64-bit destination register, we can use it to encode 64-bit mov with 32-bit immediates. This optimization is enabled with -O, -O2 and -Os. 3. Since the upper bits of destination registers of VEX128 and EVEX128 instructions are extended to zero, if all bits of destination registers of AVX256 or AVX512 instructions are zero, we can use VEX128 or EVEX128 encoding to encode AVX256 or AVX512 instructions. When 2 source registers are identical, AVX256 and AVX512 andn and xor instructions: VOP %reg, %reg, %dest_reg can be encoded with VOP128 %reg, %reg, %dest_reg This optimization is enabled with -O2 and -Os. 4. 16-bit, 32-bit and 64-bit register tests with immediate may be encoded as 8-bit register test with immediate. This optimization is enabled with -Os. This patch does: 1. Add {nooptimize} pseudo prefix to disable instruction size optimization. 2. Add optimize to i386_opcode_modifier to tell assembler that encoding of an instruction may be optimized. gas/ PR gas/22871 * NEWS: Mention -O[2|s]. * config/tc-i386.c (_i386_insn): Add no_optimize. (optimize): New. (optimize_for_space): Likewise. (fits_in_imm7): New function. (fits_in_imm31): Likewise. (optimize_encoding): Likewise. (md_assemble): Call optimize_encoding to optimize encoding. (parse_insn): Handle {nooptimize}. (md_shortopts): Append "O::". (md_parse_option): Handle -On. * doc/c-i386.texi: Document -O0, -O, -O1, -O2 and -Os as well as {nooptimize}. * testsuite/gas/cfi/cfi-x86_64.d: Pass -O0 to assembler. * testsuite/gas/i386/ilp32/cfi/cfi-x86_64.d: Likewise. * testsuite/gas/i386/i386.exp: Run optimize-1, optimize-2, optimize-3, x86-64-optimize-1, x86-64-optimize-2, x86-64-optimize-3 and x86-64-optimize-4. * testsuite/gas/i386/optimize-1.d: New file. * testsuite/gas/i386/optimize-1.s: Likewise. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.s: Likewise. * testsuite/gas/i386/x86-64-optimize-1.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. opcodes/ PR gas/22871 * i386-gen.c (opcode_modifiers): Add Optimize. * i386-opc.h (Optimize): New enum. (i386_opcode_modifier): Add optimize. * i386-opc.tbl: Add "Optimize" to "mov $imm, reg", "sub reg, reg/mem", "test $imm, acc", "test $imm, reg/mem", "and $imm, acc", "and $imm, reg/mem", "xor reg, reg/mem", "movq $imm, reg" and AVX256 and AVX512 versions of vandnps, vandnpd, vpandn, vpandnd, vpandnq, vxorps, vxorpd, vpxor, vpxord and vpxorq. * i386-tbl.h: Regenerated.
2018-02-27 16:36:33 +01:00
{nooptimize}, 0, 0x8, None, 0, 0, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
X86: Add pseudo prefixes to control encoding Many x86 instructions have more than one encodings. Assembler picks the default one, usually the shortest one. Although the ".s", ".d8" and ".d32" suffixes can be used to swap register operands or specify displacement size, they aren't very flexible. This patch adds pseudo prefixes, {xxx}, to control instruction encoding. The available pseudo prefixes are {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Pseudo prefixes are preferred over the ".s", ".d8" and ".d32" suffixes, which are deprecated. gas/ * config/tc-i386.c (_i386_insn): Add dir_encoding and vec_encoding. Remove swap_operand and need_vrex. (extra_symbol_chars): Add '}'. (md_begin): Mark '}' with LEX_BEGIN_NAME. Allow '}' in mnemonic. (build_vex_prefix): Don't use 2-byte VEX encoding with {vex3}. Check dir_encoding and load. (parse_insn): Check pseudo prefixes. Set dir_encoding. (VEX_check_operands): Likewise. (match_template): Check dir_encoding and load. (parse_real_register): Set vec_encoding instead of need_vrex. (parse_register): Likewise. * doc/c-i386.texi: Document {disp8}, {disp32}, {load}, {store}, {vex2}, {vex3} and {evex}. Remove ".s", ".d8" and ".d32" * testsuite/gas/i386/i386.exp: Run pseudos and x86-64-pseudos. * testsuite/gas/i386/pseudos.d: New file. * testsuite/gas/i386/pseudos.s: Likewise. * testsuite/gas/i386/x86-64-pseudos.d: Likewise. * testsuite/gas/i386/x86-64-pseudos.s: Likewise. opcodes/ * i386-gen.c (opcode_modifiers): Replace S with Load. * i386-opc.h (S): Removed. (Load): New. (i386_opcode_modifier): Replace s with load. * i386-opc.tbl: Add {disp8}, {disp32}, {swap}, {vex2}, {vex3} and {evex}. Replace S with Load. * i386-tbl.h: Regenerated.
2017-03-09 18:58:46 +01:00
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// 486 extensions.
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
bswap, 1, 0xfc8, None, 2, Cpu486, No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xadd, 2, 0xfc0, None, 2, Cpu486, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
cmpxchg, 2, 0xfb0, None, 2, Cpu486, W|CheckRegSize|Modrm|No_sSuf|No_ldSuf|IsLockable|HLEPrefixOk, { Reg8|Reg16|Reg32|Reg64, Reg8|Reg16|Reg32|Reg64|Byte|Word|Dword|Qword|Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
invd, 0, 0xf08, None, 2, Cpu486, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
wbinvd, 0, 0xf09, None, 2, Cpu486, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
invlpg, 1, 0xf01, 0x7, 2, Cpu486, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// 586 and late 486 extensions.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
cpuid, 0, 0xfa2, None, 2, Cpu486, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Pentium extensions.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
wrmsr, 0, 0xf30, None, 2, Cpu586, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
rdtsc, 0, 0xf31, None, 2, Cpu586, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
rdmsr, 0, 0xf32, None, 2, Cpu586, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop {X,Y,Z}MMWORD_MNEM_SUFFIX They aren't really useful (anymore?): The conflicting operand size check isn't applicable to any insn validly using respective memory operand sizes (and if they're used wrongly, another error would result), and the logic in process_suffix() can be easily changed to work without them. While re-structuring conditionals in process_suffix() also drop the CMPXCHG8B special case in favor of a NoRex64 attribute in the opcode table.
2018-03-08 08:52:27 +01:00
cmpxchg8b, 1, 0xfc7, 0x1, 2, Cpu586, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|IsLockable|NoRex64|HLEPrefixOk, { Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Pentium II/Pentium Pro extensions.
x86: Accept Intel64 only instruction by default Commit d835a58baae720 disabled sysenter/sysenter in 64-bit mode by default. By default, assembler should accept common, Intel64 only and AMD64 ISAs since there are no conflicts. gas/ PR gas/25516 * config/tc-i386.c (intel64): Renamed to ... (isa64): This. (match_template): Accept Intel64 only instruction by default. (i386_displacement): Updated. (md_parse_option): Updated. * c-i386.texi: Update -mamd64/-mintel64 documentation. * testsuite/gas/i386/i386.exp: Run x86-64-sysenter. Pass -mamd64 to x86-64-sysenter-amd. * testsuite/gas/i386/x86-64-sysenter.d: New file. opcodes/ PR gas/25516 * i386-gen.c (opcode_modifiers): Replace AMD64 and Intel64 with ISA64. * i386-opc.h (AMD64): Removed. (Intel64): Likewose. (AMD64): New. (INTEL64): Likewise. (INTEL64ONLY): Likewise. (i386_opcode_modifier): Replace amd64 and intel64 with isa64. * i386-opc.tbl (Amd64): New. (Intel64): Likewise. (Intel64Only): Likewise. Replace AMD64 with Amd64. Update sysenter/sysenter with Cpu64 and Intel64Only. Remove AMD64 from sysenter/sysenter. * i386-tbl.h: Regenerated.
2020-02-10 17:37:22 +01:00
sysenter, 0, 0xf34, None, 2, Cpu64, Intel64Only|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
sysenter, 0, 0xf34, None, 2, Cpu686|CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
sysexit, 0, 0xf35, None, 2, Cpu64, Intel64Only|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
sysexit, 0, 0xf35, None, 2, Cpu686|CpuNo64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: Add CpuCMOV and CpuFXSR There are separate CPUID feature bits for fxsave/fxrstor and cmovCC instructions. This patch adds CpuCMOV and CpuFXSR to replace Cpu686 on corresponding instructions. gas/ * config/tc-i386.c (cpu_arch): Add .cmov and .fxsr. (cpu_noarch): Add nocmov and nofxsr. * doc/c-i386.texi: Document cmov and fxsr. opcodes/ * i386-gen.c (cpu_flag_init): Add CpuCMOV and CpuFXSR to CPU_I686_FLAGS. Add CPU_CMOV_FLAGS, CPU_FXSR_FLAGS, CPU_ANY_CMOV_FLAGS and CPU_ANY_FXSR_FLAGS. (cpu_flags): Add CpuCMOV and CpuFXSR. * i386-opc.tbl: Replace Cpu686 with CpuFXSR on fxsave, fxsave64, fxrstor and fxrstor64. Replace Cpu686 with CpuCMOV on cmovCC. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2018-08-11 23:37:14 +02:00
fxsave, 1, 0xfae, 0x0, 2, CpuFXSR, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
fxsave64, 1, 0xfae, 0x0, 2, CpuFXSR|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
x86: Add CpuCMOV and CpuFXSR There are separate CPUID feature bits for fxsave/fxrstor and cmovCC instructions. This patch adds CpuCMOV and CpuFXSR to replace Cpu686 on corresponding instructions. gas/ * config/tc-i386.c (cpu_arch): Add .cmov and .fxsr. (cpu_noarch): Add nocmov and nofxsr. * doc/c-i386.texi: Document cmov and fxsr. opcodes/ * i386-gen.c (cpu_flag_init): Add CpuCMOV and CpuFXSR to CPU_I686_FLAGS. Add CPU_CMOV_FLAGS, CPU_FXSR_FLAGS, CPU_ANY_CMOV_FLAGS and CPU_ANY_FXSR_FLAGS. (cpu_flags): Add CpuCMOV and CpuFXSR. * i386-opc.tbl: Replace Cpu686 with CpuFXSR on fxsave, fxsave64, fxrstor and fxrstor64. Replace Cpu686 with CpuCMOV on cmovCC. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2018-08-11 23:37:14 +02:00
fxrstor, 1, 0xfae, 0x1, 2, CpuFXSR, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
fxrstor64, 1, 0xfae, 0x1, 2, CpuFXSR|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
rdpmc, 0, 0xf33, None, 2, Cpu686, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// official undefined instr.
Add Cpu186 to ud1/ud2/ud2a/ud2b. 2010-08-05 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Add Cpu186 to ud1/ud2/ud2a/ud2b. * i386-tbl.h: Regenerated.
2010-08-06 03:03:17 +02:00
ud2, 0, 0xf0b, None, 2, Cpu186, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// alias for ud2
Add Cpu186 to ud1/ud2/ud2a/ud2b. 2010-08-05 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Add Cpu186 to ud1/ud2/ud2a/ud2b. * i386-tbl.h: Regenerated.
2010-08-06 03:03:17 +02:00
ud2a, 0, 0xf0b, None, 2, Cpu186, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// 2nd. official undefined instr.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ud1, 2, 0xfb9, None, 2, Cpu186, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
Add ud1 to x86. gas/testsuite/ 2010-08-05 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run arch-4. * gas/i386/arch-4.d: New. * gas/i386/arch-4.s: Likewise. * gas/i386/intel.d: Replace ud2a/ud2b with ud2/ud1. * gas/i386/opcode-intel.d: Likewise. * gas/i386/opcode-suffix.d: Likewise. * gas/i386/opcode.d: Likewise. opcodes/ 2010-08-05 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (dis386_twobyte): Replace ud2a/ud2b with ud2/ud1. * i386-opc.tbl: Add ud1. Remove Cpu686 from ud2/ud2a/ud2b. * i386-tbl.h: Regenerated.
2010-08-06 02:52:57 +02:00
// alias for ud1
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ud2b, 2, 0xfb9, None, 2, Cpu186, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
x86: correct UDn Make the assembler recognize UD0, supporting only the newer form expecting a ModR/M byte. Make assembler and disassembler properly emit / expect a ModR/M byte for UD1. For the testsuite, as arch-4 already tests all UDn, avoid producing a huge delta for other tests using UD2B by making them use UD2 instead.
2017-11-23 10:59:48 +01:00
// 3rd official undefined instr (older CPUs don't take a ModR/M byte)
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ud0, 2, 0xfff, None, 2, Cpu186, Modrm|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
x86: allow opcode templates to be templated In order to reduce redundancy as well as the chance of things going out of sync (see a later patch for an example), make the opcode table generator capable of recognizing and expanding templated templates. Use the new capability for compacting the general purpose conditional insns.
2020-03-09 10:12:14 +01:00
cmov<cc>, 2, 0xf4<cc:opc>, None, 2, CpuCMOV, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcmovb, 2, 0xdac0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovnae, 2, 0xdac0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmove, 2, 0xdac8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovbe, 2, 0xdad0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovna, 2, 0xdad0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovu, 2, 0xdad8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovae, 2, 0xdbc0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovnb, 2, 0xdbc0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovne, 2, 0xdbc8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmova, 2, 0xdbd0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovnbe, 2, 0xdbd0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcmovnu, 2, 0xdbd8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
fcomi, 2, 0xdbf0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fcomi, 0, 0xdbf1, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcomi, 1, 0xdbf0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fucomi, 2, 0xdbe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fucomi, 0, 0xdbe9, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fucomi, 1, 0xdbe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fcomip, 2, 0xdff0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fcomip, 0, 0xdff1, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcomip, 1, 0xdff0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fcompi, 2, 0xdff0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fcompi, 0, 0xdff1, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fcompi, 1, 0xdff0, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fucomip, 2, 0xdfe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fucomip, 0, 0xdfe9, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fucomip, 1, 0xdfe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
fucompi, 2, 0xdfe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg, FloatAcc }
x86: drop stray ShortForm attributes This attribute is meaningless when there are no operands to encode.
2020-02-11 11:20:05 +01:00
fucompi, 0, 0xdfe9, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
fucompi, 1, 0xdfe8, None, 2, Cpu687, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { FloatReg }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Pentium4 extensions.
x86: fix (dis)assembly of certain SSE2 insns in 16-bit mode MOVNTI was wrongly assembled with a 66h prefix. Add IgnoreSize to address this. It and the scalar to/from integer conversion insns also were also wrongly using Ev / Gv, leading to 16-bit register names being printed when 32-bit ones were meant. Clone the 32-bit SSE2 test to cover both assembler and disassembler.
2019-06-25 09:25:26 +02:00
movnti, 2, 0xfc3, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoAVX, { Reg32|Reg64, Dword|Qword|Unspecified|BaseIndex }
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
clflush, 1, 0xfae, 0x7, 2, CpuClflush, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
lfence, 0, 0xfaee8, None, 3, CpuSSE2, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { 0 }
mfence, 0, 0xfaef0, None, 3, CpuSSE2, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { 0 }
gas/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * gas/config/tc-i386.c (cpu_arch): Add corei7, .clflush and .syscall. (i386_align_code): Handle PROCESSOR_COREI7. (md_show_usage): Add corei7, clflush and syscall. (i386_target_format): Replace cpup4 with cpuclflush. * gas/config/tc-i386.h (processor_type): Add PROCESSOR_COREI7. * doc/c-i386.texi: Document corei7, clflush and syscall. gas/testsuite/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/arch-10.s: Add clflush and syscall. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (cpu_flag_init): Replace CpuP4 and CpuK6 with CpuClflush and CpuSYSCALL, respectively. Remove CpuK8. Add CPU_COREI7_FLAGS, CPU_CLFLUSH_FLAGS and CPU_SYSCALL_FLAGS. (cpu_flags): Remove CpuP4, CpuK6 and CpuK8. Add CpuClflush and CpuSYSCALL. (lineno): Removed. (set_bitfield): Take an argument, lineno. Don't report lineno on error if it is -1. (process_i386_cpu_flag): Take an argument, lineno. (process_i386_opcode_modifier): Likewise. (process_i386_operand_type): Likewise. (output_i386_opcode): Likewise. (opcode_hash_entry): Add lineno. (process_i386_opcodes): Updated. (process_i386_registers): Likewise. (process_i386_initializers): Likewise. * i386-opc.h (CpuP4): Removed. (CpuK6): Likewise. (CpuK8): Likewise. (CpuClflush): New. (CpuSYSCALL): Likewise. (CpuMMX): Updated. (i386_cpu_flags): Remove cpup4, cpuk6 and cpuk8. Add cpuclflush and cpusyscall. * i386-opc.tbl: Update movnti, clflush, lfence, mfence, pause, syscall and sysret. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2009-01-10 18:25:52 +01:00
// Processors that do not support PAUSE treat this opcode as a NOP instruction.
pause, 0, 0xf390, None, 2, Cpu186, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// MMX/SSE2 instructions.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
emms, 0, 0xf77, None, 2, CpuMMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// These really shouldn't allow for Reg64 (movq is the right mnemonic for
// copying between Reg64/Mem64 and RegXMM/RegMMX, as is mandated by Intel's
// spec). AMD's spec, having been in existence for much longer, failed to
// recognize that and specified movd for 32- and 64-bit operations.
x86: Set Vex=1 on VEX.128 only vmovd and vmovq AVX "VMOVD xmm1, r32/m32", "VMOVD r32/m32, xmm2", "VMOVQ xmm1, r64/m64" and "VMOVD r64/m64, xmm2" can only be encoded with VEX.128. Set Vex=1 on VEX.128 only vmovd and vmovq. gas/ PR gas/23665 * testsuite/gas/i386/avx-scalar.s: Remove vmovq and vmovd tests. * testsuite/gas/i386/x86-64-avx-scalar.s: Likewise. * testsuite/gas/i386/avx-scalar-intel.d: Updated. * testsuite/gas/i386/avx-scalar.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar.d: Likewise. * testsuite/gas/i386/i386.exp: Run avx-scalar2 and x86-64-avx-scalar2. * testsuite/gas/i386/avx-scalar-2.d: New file. * testsuite/gas/i386/avx-scalar-2.s: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-2.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-2.s: Likewise. opcodes/ PR gas/23665 * i386-dis.c (vex_len_table): Update VEX_LEN_0F6E_P_2 and VEX_LEN_0F7E_P_2 entries. * i386-opc.tbl: Set Vex=1 on VEX.128 only vmovd and vmovq. * i386-tbl.h: Regenerated.
2018-09-17 18:31:07 +02:00
movd, 2, 0x666e, None, 1, CpuAVX, D|Modrm|Vex=1|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Reg32|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
movd, 2, 0x666e, None, 1, CpuAVX|Cpu64, D|Modrm|Vex=1|VexOpcode=0|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64|SSE2AVX, { Reg64|BaseIndex, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movd, 2, 0x660f6e, None, 2, CpuSSE2, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
movd, 2, 0x660f6e, None, 2, CpuSSE2|Cpu64, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64|BaseIndex, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movd, 2, 0xf6e, None, 2, CpuMMX, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, RegMMX }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
movd, 2, 0xf6e, None, 2, CpuMMX|Cpu64, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64|BaseIndex, RegMMX }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// In the 64bit mode the short form mov immediate is redefined to have
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
// 64bit displacement value. We put the 64bit displacement first and
// we only mark constants larger than 32bit as Disp64.
x86: drop stray W The flag is used to indicate opcodes which can be switched between byte and word/dword/qword forms (in a "canonical" way). Obviously it's quite odd then to see it on insns not allowing for byte operands in the first place. As a result the opcode bytes need to be adjusted accordingly, which includes comparisons done in optimize_encoding(). To make re-introduction of such issues less likely have i386-gen diagnose it (in a generally non-fatal way for now).
2019-10-30 09:05:46 +01:00
movq, 2, 0xa1, None, 1, Cpu64, D|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Disp64|Unspecified|Qword, Acc|Qword }
movq, 2, 0x89, None, 1, Cpu64, D|Modrm|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|HLEPrefixOk=3, { Reg64, Reg64|Unspecified|Qword|BaseIndex }
movq, 2, 0xc7, 0x0, 1, Cpu64, Modrm|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|HLEPrefixOk=3|Optimize, { Imm32S, Reg64|Qword|Unspecified|BaseIndex }
x86: drop ShortForm attribute It is very simple to derive from other template properties, and hence there's little point wasting storage for it.
2020-02-11 11:20:55 +01:00
movq, 2, 0xb8, None, 1, Cpu64, Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Imm64, Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
movq, 2, 0xf37e, None, 1, CpuAVX, Load|Modrm|Vex=1|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
movq, 2, 0x66d6, None, 1, CpuAVX, Modrm|Vex=1|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM }
movq, 2, 0x666e, None, 1, CpuAVX|Cpu64, D|Modrm|Vex=1|VexOpcode=0|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64|SSE2AVX, { Reg64|Unspecified|BaseIndex, RegXMM }
movq, 2, 0xf30f7e, None, 2, CpuSSE2, Load|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Unspecified|Qword|BaseIndex|RegXMM, RegXMM }
movq, 2, 0x660fd6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { RegXMM, Unspecified|Qword|BaseIndex|RegXMM }
movq, 2, 0x660f6e, None, 2, Cpu64, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64|Unspecified|BaseIndex, RegXMM }
movq, 2, 0xf6f, None, 2, CpuMMX, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Unspecified|Qword|BaseIndex|RegMMX, RegMMX }
movq, 2, 0xf6e, None, 2, Cpu64, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64|Unspecified|BaseIndex, RegMMX }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// The segment register moves accept Reg64 so that a segment register
// can be copied to a 64 bit register, and vice versa.
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
movq, 2, 0x8c, None, 1, Cpu64, D|RegMem|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { SReg, Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Move to/from control debug registers. In the 16 or 32bit modes they
// are 32bit. In the 64bit mode they are 64bit.
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
movq, 2, 0xf20, None, 2, Cpu64, D|RegMem|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Control, Reg64 }
movq, 2, 0xf21, None, 2, Cpu64, D|RegMem|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Debug, Reg64 }
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
// Real MMX/SSE instructions.
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
packssdw, 2, 0x666b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
packssdw, 2, 0x660f6b, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
packssdw, 2, 0xf6b, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
packsswb, 2, 0x6663, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
packsswb, 2, 0x660f63, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
packsswb, 2, 0xf63, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
packuswb, 2, 0x6667, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
packuswb, 2, 0x660f67, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
packuswb, 2, 0xf67, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddb, 2, 0x66fc, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddb, 2, 0x660ffc, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddb, 2, 0xffc, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddw, 2, 0x66fd, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddw, 2, 0x660ffd, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddw, 2, 0xffd, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddd, 2, 0x66fe, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddd, 2, 0x660ffe, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddd, 2, 0xffe, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddq, 2, 0x66d4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddq, 2, 0x660fd4, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddq, 2, 0xfd4, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddsb, 2, 0x66ec, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddsb, 2, 0x660fec, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddsb, 2, 0xfec, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddsw, 2, 0x66ed, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddsw, 2, 0x660fed, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddsw, 2, 0xfed, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddusb, 2, 0x66dc, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddusb, 2, 0x660fdc, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddusb, 2, 0xfdc, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
paddusw, 2, 0x66dd, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
paddusw, 2, 0x660fdd, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
paddusw, 2, 0xfdd, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pand, 2, 0x66db, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pand, 2, 0x660fdb, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pand, 2, 0xfdb, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pandn, 2, 0x66df, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pandn, 2, 0x660fdf, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pandn, 2, 0xfdf, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pcmpeqb, 2, 0x6674, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpeqb, 2, 0x660f74, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpeqb, 2, 0xf74, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pcmpeqw, 2, 0x6675, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpeqw, 2, 0x660f75, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpeqw, 2, 0xf75, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pcmpeqd, 2, 0x6676, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpeqd, 2, 0x660f76, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpeqd, 2, 0xf76, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpgtb, 2, 0x6664, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpgtb, 2, 0x660f64, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpgtb, 2, 0xf64, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpgtw, 2, 0x6665, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpgtw, 2, 0x660f65, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpgtw, 2, 0xf65, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pcmpgtd, 2, 0x6666, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpgtd, 2, 0x660f66, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pcmpgtd, 2, 0xf66, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmaddwd, 2, 0x66f5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmaddwd, 2, 0x660ff5, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmaddwd, 2, 0xff5, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmulhw, 2, 0x66e5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmulhw, 2, 0x660fe5, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmulhw, 2, 0xfe5, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmullw, 2, 0x66d5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmullw, 2, 0x660fd5, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmullw, 2, 0xfd5, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
por, 2, 0x66eb, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
por, 2, 0x660feb, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
por, 2, 0xfeb, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psllw, 2, 0x6671, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psllw, 2, 0x66f1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psllw, 2, 0x660f71, 0x6, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psllw, 2, 0x660ff1, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psllw, 2, 0xf71, 0x6, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psllw, 2, 0xff1, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pslld, 2, 0x6672, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
pslld, 2, 0x66f2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pslld, 2, 0x660f72, 0x6, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
pslld, 2, 0x660ff2, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pslld, 2, 0xf72, 0x6, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
pslld, 2, 0xff2, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psllq, 2, 0x6673, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psllq, 2, 0x66f3, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psllq, 2, 0x660f73, 0x6, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psllq, 2, 0x660ff3, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psllq, 2, 0xf73, 0x6, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psllq, 2, 0xff3, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psraw, 2, 0x6671, 0x4, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psraw, 2, 0x66e1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psraw, 2, 0x660f71, 0x4, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psraw, 2, 0x660fe1, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psraw, 2, 0xf71, 0x4, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psraw, 2, 0xfe1, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psrad, 2, 0x6672, 0x4, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psrad, 2, 0x66e2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psrad, 2, 0x660f72, 0x4, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psrad, 2, 0x660fe2, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psrad, 2, 0xf72, 0x4, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psrad, 2, 0xfe2, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psrlw, 2, 0x6671, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psrlw, 2, 0x66d1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psrlw, 2, 0x660f71, 0x2, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psrlw, 2, 0x660fd1, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psrlw, 2, 0xf71, 0x2, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psrlw, 2, 0xfd1, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psrld, 2, 0x6672, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psrld, 2, 0x66d2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psrld, 2, 0x660f72, 0x2, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psrld, 2, 0x660fd2, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psrld, 2, 0xf72, 0x2, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psrld, 2, 0xfd2, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psrlq, 2, 0x6673, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psrlq, 2, 0x66d3, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psrlq, 2, 0x660f73, 0x2, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psrlq, 2, 0x660fd3, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psrlq, 2, 0xf73, 0x2, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMMX }
psrlq, 2, 0xfd3, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubb, 2, 0x66f8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubb, 2, 0x660ff8, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubb, 2, 0xff8, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubw, 2, 0x66f9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubw, 2, 0x660ff9, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubw, 2, 0xff9, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubd, 2, 0x66fa, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubd, 2, 0x660ffa, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubd, 2, 0xffa, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubq, 2, 0x66fb, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubq, 2, 0x660ffb, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubq, 2, 0xffb, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubsb, 2, 0x66e8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubsb, 2, 0x660fe8, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubsb, 2, 0xfe8, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubsw, 2, 0x66e9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubsw, 2, 0x660fe9, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubsw, 2, 0xfe9, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubusb, 2, 0x66d8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubusb, 2, 0x660fd8, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubusb, 2, 0xfd8, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psubusw, 2, 0x66d9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psubusw, 2, 0x660fd9, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psubusw, 2, 0xfd9, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpckhbw, 2, 0x6668, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpckhbw, 2, 0x660f68, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
punpckhbw, 2, 0xf68, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpckhwd, 2, 0x6669, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpckhwd, 2, 0x660f69, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
punpckhwd, 2, 0xf69, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpckhdq, 2, 0x666a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpckhdq, 2, 0x660f6a, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
punpckhdq, 2, 0xf6a, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpcklbw, 2, 0x6660, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpcklbw, 2, 0x660f60, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
punpcklbw, 2, 0xf60, None, 2, CpuMMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpcklwd, 2, 0x6661, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpcklwd, 2, 0x660f61, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
punpcklwd, 2, 0xf61, None, 2, CpuMMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
punpckldq, 2, 0x6662, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpckldq, 2, 0x660f62, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
punpckldq, 2, 0xf62, None, 2, CpuMMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pxor, 2, 0x66ef, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pxor, 2, 0x660fef, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pxor, 2, 0xfef, None, 2, CpuMMX, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSE instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: use template for SSE floating point comparison insns These all follow an almost common pattern, with the exception of being commutative. This exception can be easily taken care of.
2020-03-09 10:13:04 +01:00
<sse_frel:imm:comm, eq:0:C, lt:1:, le:2:, unord:3:C, neq:4:C, nlt:5:, nle:6:, ord:7:C>
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
addps, 2, 0x58, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
addps, 2, 0xf58, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
addss, 2, 0xf358, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
addss, 2, 0xf30f58, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
andnps, 2, 0x55, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
andnps, 2, 0xf55, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
andps, 2, 0x54, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
andps, 2, 0xf54, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: use template for SSE floating point comparison insns These all follow an almost common pattern, with the exception of being commutative. This exception can be easily taken care of.
2020-03-09 10:13:04 +01:00
cmp<sse_frel>ps, 2, 0xc2, <sse_frel:imm>, 1, CpuAVX, Modrm|<sse_frel:comm>|Vex128|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>ps, 2, 0xfc2, <sse_frel:imm>, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>ss, 2, 0xf3c2, <sse_frel:imm>, 1, CpuAVX, Modrm|<sse_frel:comm>|VexLIG|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>ss, 2, 0xf30fc2, <sse_frel:imm>, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
cmpps, 3, 0xc2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
cmpps, 3, 0xfc2, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cmpss, 3, 0xf3c2, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cmpss, 3, 0xf30fc2, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
comiss, 2, 0x2f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
comiss, 2, 0xf2f, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvtpi2ps, 2, 0xf2a, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegXMM }
cvtps2pi, 2, 0xf2d, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegXMM, RegMMX }
i386: Update VexW field for VEX instructions 1. Mark VEX.W0 VEX instructions with VexW=1. 2. Mark VEX.W1 VEX instructions with VexW=2. 3. Remove VexW=1 from WIG VEX instructions. * i386-opc.tbl: Add VexW=1 to VEX.W0 VEX movd, cvtsi2ss, cvtsi2sd, pextrd, pinsrd, vcvtsi2sd, vcvtsi2ss, vmovd, vpextrd and vpinsrd. Add VexW=2 to VEX.W1 VEX movd, movq, pextrq, pinsrq, vmod, vmovq, vpextrq and vpinsrq. Remove VexW=1 from WIG VEX movq and vmovq. * i386-tbl.h: Regenerated.
2018-09-13 00:12:59 +02:00
cvtsi2ss, 2, 0xf32a, None, 1, CpuAVX|CpuNo64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
cvtsi2ss, 2, 0xf32a, None, 1, CpuAVX|Cpu64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX|ATTSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
cvtsi2ss, 2, 0xf32a, None, 1, CpuAVX|Cpu64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX|IntelSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvtsi2ss, 2, 0xf30f2a, None, 2, CpuSSE|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
cvtsi2ss, 2, 0xf30f2a, None, 2, CpuSSE|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
cvtsi2ss, 2, 0xf30f2a, None, 2, CpuSSE|Cpu64, Modrm|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvtss2si, 2, 0xf32d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvtss2si, 2, 0xf30f2d, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvttps2pi, 2, 0xf2c, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegXMM, RegMMX }
cvttss2si, 2, 0xf32c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvttss2si, 2, 0xf30f2c, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
divps, 2, 0x5e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
divps, 2, 0xf5e, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
divss, 2, 0xf35e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
divss, 2, 0xf30f5e, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
ldmxcsr, 1, 0xae, 0x2, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex }
x86: avoid SSE check for LDMXCSR/STMXCSR Neither touches any XMM register, so the check is pointless. It is imo even questionable whether in SSE2AVX mode the two should be converted to their AVX counterparts.
2018-03-08 08:35:48 +01:00
ldmxcsr, 1, 0xfae, 0x2, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { Dword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
maskmovq, 2, 0xff7, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegMMX, RegMMX }
maxps, 2, 0x5f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
maxps, 2, 0xf5f, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
maxss, 2, 0xf35f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
maxss, 2, 0xf30f5f, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
minps, 2, 0x5d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
minps, 2, 0xf5d, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
minss, 2, 0xf35d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
minss, 2, 0xf30f5d, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movaps, 2, 0x28, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movaps, 2, 0xf28, None, 2, CpuSSE, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
movhlps, 2, 0x12, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, RegXMM }
movhlps, 2, 0xf12, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movhps, 2, 0x16, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex, RegXMM }
movhps, 2, 0x17, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Qword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movhps, 2, 0xf16, None, 2, CpuSSE, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
movlhps, 2, 0x16, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, RegXMM }
movlhps, 2, 0xf16, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movlps, 2, 0x12, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex, RegXMM }
movlps, 2, 0x13, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Qword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movlps, 2, 0xf12, None, 2, CpuSSE, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
movmskps, 2, 0x50, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { RegXMM, Reg32|Reg64 }
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
movmskps, 2, 0xf50, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { RegXMM, Reg32|Reg64 }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
movntps, 2, 0x2b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Xmmword|Unspecified|BaseIndex }
movntps, 2, 0xf2b, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Xmmword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movntq, 2, 0xfe7, None, 2, CpuSSE|Cpu3dnowA, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegMMX, Qword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
movntdq, 2, 0x66e7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Xmmword|Unspecified|BaseIndex }
movntdq, 2, 0x660fe7, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Xmmword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movss, 2, 0xf310, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
movss, 2, 0xf310, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movss, 2, 0xf30f10, None, 2, CpuSSE, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
movups, 2, 0x10, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movups, 2, 0xf10, None, 2, CpuSSE, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
mulps, 2, 0x59, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
mulps, 2, 0xf59, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
mulss, 2, 0xf359, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
mulss, 2, 0xf30f59, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
orps, 2, 0x56, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
orps, 2, 0xf56, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pavgb, 2, 0xfe0, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pavgb, 2, 0x66e0, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pavgb, 2, 0x660fe0, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pavgw, 2, 0xfe3, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pavgw, 2, 0x66e3, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pavgw, 2, 0x660fe3, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
pextrw, 3, 0x66c5, None, 1, CpuAVX, Load|Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg32|Reg64 }
pextrw, 3, 0x6615, None, 1, CpuAVX, RegMem|Vex|VexOpcode=2|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pextrw, 3, 0x6615, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Word|Unspecified|BaseIndex }
x86: allow {store} to select alternative {,}PEXTRW encoding The 0F C5 encoding is indeed a load type one (just that memory operands are not permitted), while the 0F 3A 15 encoding is obviously a store. Allow the pseudo prefixes to be used to select between them. Also move (without any change) the secondary AVX512BW templates next to the primary one.
2018-11-06 11:40:25 +01:00
pextrw, 3, 0x660fc5, None, 2, CpuSSE2, Load|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Imm8, RegXMM, Reg32|Reg64 }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
pextrw, 3, 0x660f3a15, None, 3, CpuSSE4_1, RegMem|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, RegXMM, Reg32|Reg64 }
x86: replace Reg8, Reg16, Reg32, and Reg64 Use a combination of a single new Reg bit and Byte, Word, Dword, or Qword instead. Besides shrinking the number of operand type bits this has the benefit of making register handling more similar to accumulator handling (a generic flag is being accompanied by a "size qualifier"). It requires, however, to split a few insn templates, as it is no longer correct to have combinations like Reg32|Reg64|Byte. This slight growth in size will hopefully be outweighed by this change paving the road for folding a presumably much larger number of templates later on.
2017-12-18 09:34:00 +01:00
pextrw, 3, 0x660f3a15, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Word|Unspecified|BaseIndex }
gas/ 2008-04-22 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Don't check SSE instructions if noavx is 0. opcodes/ 2008-04-22 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Add NoAVX. * i386-opc.h (NoAVX): New. (OldGcc): Updated. (i386_opcode_modifier): Add noavx. * i386-opc.tbl: Add NoAVX to SSE, SSE2, SSE3 and SSSE3 instructions which don't have AVX equivalent. * i386-tbl.h: Regenerated.
2008-04-23 00:27:13 +02:00
pextrw, 3, 0xfc5, None, 2, CpuSSE|Cpu3dnowA, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64|NoAVX, { Imm8, RegMMX, Reg32|Reg64 }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
pinsrw, 3, 0x66c4, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { Imm8, Reg32|Reg64, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pinsrw, 3, 0x66c4, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Word|Unspecified|BaseIndex, RegXMM }
x86: replace Reg8, Reg16, Reg32, and Reg64 Use a combination of a single new Reg bit and Byte, Word, Dword, or Qword instead. Besides shrinking the number of operand type bits this has the benefit of making register handling more similar to accumulator handling (a generic flag is being accompanied by a "size qualifier"). It requires, however, to split a few insn templates, as it is no longer correct to have combinations like Reg32|Reg64|Byte. This slight growth in size will hopefully be outweighed by this change paving the road for folding a presumably much larger number of templates later on.
2017-12-18 09:34:00 +01:00
pinsrw, 3, 0x660fc4, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { Imm8, Reg32|Reg64, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pinsrw, 3, 0x660fc4, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Word|Unspecified|BaseIndex, RegXMM }
x86: replace Reg8, Reg16, Reg32, and Reg64 Use a combination of a single new Reg bit and Byte, Word, Dword, or Qword instead. Besides shrinking the number of operand type bits this has the benefit of making register handling more similar to accumulator handling (a generic flag is being accompanied by a "size qualifier"). It requires, however, to split a few insn templates, as it is no longer correct to have combinations like Reg32|Reg64|Byte. This slight growth in size will hopefully be outweighed by this change paving the road for folding a presumably much larger number of templates later on.
2017-12-18 09:34:00 +01:00
pinsrw, 3, 0xfc4, None, 2, CpuSSE|Cpu3dnowA, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64|NoAVX, { Imm8, Reg32|Reg64, RegMMX }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pinsrw, 3, 0xfc4, None, 2, CpuSSE|Cpu3dnowA, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { Imm8, Word|Unspecified|BaseIndex, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmaxsw, 2, 0x66ee, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmaxsw, 2, 0x660fee, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmaxsw, 2, 0xfee, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmaxub, 2, 0x66de, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmaxub, 2, 0x660fde, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmaxub, 2, 0xfde, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pminsw, 2, 0x66ea, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pminsw, 2, 0x660fea, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pminsw, 2, 0xfea, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pminub, 2, 0x66da, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pminub, 2, 0x660fda, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pminub, 2, 0xfda, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pmovmskb, 2, 0x66d7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { RegXMM, Reg32|Reg64 }
gas/testsuite/ 2008-01-04 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/rexw.d: New. * gas/i386/rexw.s: Likewise. * gas/i386/x86-64-sse4_1-intel.d: Updated. * gas/i386/x86-64-sse4_1.d: Likewise. opcodes/ 2008-01-04 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Add NoRex64 to extractps, movmskpd, movmskps, pextrb, pextrw, pinsrb, pinsrw and pmovmskb. * i386-tbl.h: Regenerated.
2008-01-04 19:03:02 +01:00
pmovmskb, 2, 0x660fd7, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { RegXMM, Reg32|Reg64 }
gas/ 2008-04-22 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Don't check SSE instructions if noavx is 0. opcodes/ 2008-04-22 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Add NoAVX. * i386-opc.h (NoAVX): New. (OldGcc): Updated. (i386_opcode_modifier): Add noavx. * i386-opc.tbl: Add NoAVX to SSE, SSE2, SSE3 and SSSE3 instructions which don't have AVX equivalent. * i386-tbl.h: Regenerated.
2008-04-23 00:27:13 +02:00
pmovmskb, 2, 0xfd7, None, 2, CpuSSE|Cpu3dnowA, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64|NoAVX, { RegMMX, Reg32|Reg64 }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmulhuw, 2, 0x66e4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmulhuw, 2, 0x660fe4, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmulhuw, 2, 0xfe4, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
prefetchnta, 1, 0xf18, 0x0, 2, CpuSSE|Cpu3dnowA, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { BaseIndex }
prefetcht0, 1, 0xf18, 0x1, 2, CpuSSE|Cpu3dnowA, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { BaseIndex }
prefetcht1, 1, 0xf18, 0x2, 2, CpuSSE|Cpu3dnowA, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { BaseIndex }
prefetcht2, 1, 0xf18, 0x3, 2, CpuSSE|Cpu3dnowA, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psadbw, 2, 0xff6, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
psadbw, 2, 0x66f6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psadbw, 2, 0x660ff6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pshufw, 3, 0xf70, None, 2, CpuSSE|Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Imm8, Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
rcpps, 2, 0x53, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
rcpps, 2, 0xf53, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
rcpss, 2, 0xf353, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
rcpss, 2, 0xf30f53, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
rsqrtps, 2, 0x52, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
rsqrtps, 2, 0xf52, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
rsqrtss, 2, 0xf352, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
rsqrtss, 2, 0xf30f52, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
sfence, 0, 0xfaef8, None, 3, CpuSSE|Cpu3dnowA, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { 0 }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
shufps, 3, 0xc6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
shufps, 3, 0xfc6, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
sqrtps, 2, 0x51, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
sqrtps, 2, 0xf51, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
sqrtss, 2, 0xf351, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
sqrtss, 2, 0xf30f51, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
stmxcsr, 1, 0xae, 0x3, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex }
x86: avoid SSE check for LDMXCSR/STMXCSR Neither touches any XMM register, so the check is pointless. It is imo even questionable whether in SSE2AVX mode the two should be converted to their AVX counterparts.
2018-03-08 08:35:48 +01:00
stmxcsr, 1, 0xfae, 0x3, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { Dword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
subps, 2, 0x5c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
subps, 2, 0xf5c, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
subss, 2, 0xf35c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
subss, 2, 0xf30f5c, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
ucomiss, 2, 0x2e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
ucomiss, 2, 0xf2e, None, 2, CpuSSE, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
unpckhps, 2, 0x15, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpckhps, 2, 0xf15, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpcklps, 2, 0x14, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpcklps, 2, 0xf14, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
xorps, 2, 0x57, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE insns
2018-09-13 11:11:26 +02:00
xorps, 2, 0xf57, None, 2, CpuSSE, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSE2 instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
addpd, 2, 0x6658, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
addpd, 2, 0x660f58, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
addsd, 2, 0xf258, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
addsd, 2, 0xf20f58, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
andnpd, 2, 0x6655, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
andnpd, 2, 0x660f55, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
andpd, 2, 0x6654, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
andpd, 2, 0x660f54, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: use template for SSE floating point comparison insns These all follow an almost common pattern, with the exception of being commutative. This exception can be easily taken care of.
2020-03-09 10:13:04 +01:00
cmp<sse_frel>pd, 2, 0x66c2, <sse_frel:imm>, 1, CpuAVX, Modrm|<sse_frel:comm>|Vex|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>pd, 2, 0x660fc2, <sse_frel:imm>, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>sd, 2, 0xf2c2, <sse_frel:imm>, 1, CpuAVX, Modrm|<sse_frel:comm>|VexLIG|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
cmp<sse_frel>sd, 2, 0xf20fc2, <sse_frel:imm>, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|NoRex64, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
cmppd, 3, 0x66c2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
cmppd, 3, 0x660fc2, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Intel mode string compare.
x86/Intel: drop IgnoreSize from operand-less MOVSD/CMPSD again These were mistakenly added by d241b91073 ("x86/Intel: correct MOVSD and CMPSD handling"). This addresses part of PR/gas 25167.
2019-11-07 09:27:16 +01:00
cmpsd, 0, 0xa7, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
cmpsd, 2, 0xa7, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsStringEsOp0|RepPrefixOk, { Dword|Unspecified|BaseIndex, Dword|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cmpsd, 3, 0xf2c2, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cmpsd, 3, 0xf20fc2, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
comisd, 2, 0x662f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
comisd, 2, 0x660f2f, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cvtpi2pd, 2, 0x660f2a, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegXMM }
i386: Update VexW field for VEX instructions 1. Mark VEX.W0 VEX instructions with VexW=1. 2. Mark VEX.W1 VEX instructions with VexW=2. 3. Remove VexW=1 from WIG VEX instructions. * i386-opc.tbl: Add VexW=1 to VEX.W0 VEX movd, cvtsi2ss, cvtsi2sd, pextrd, pinsrd, vcvtsi2sd, vcvtsi2ss, vmovd, vpextrd and vpinsrd. Add VexW=2 to VEX.W1 VEX movd, movq, pextrq, pinsrq, vmod, vmovq, vpextrq and vpinsrq. Remove VexW=1 from WIG VEX movq and vmovq. * i386-tbl.h: Regenerated.
2018-09-13 00:12:59 +02:00
cvtsi2sd, 2, 0xf22a, None, 1, CpuAVX|CpuNo64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
cvtsi2sd, 2, 0xf22a, None, 1, CpuAVX|Cpu64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX|ATTSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
cvtsi2sd, 2, 0xf22a, None, 1, CpuAVX|Cpu64, Modrm|Vex=3|VexOpcode=0|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX|IntelSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvtsi2sd, 2, 0xf20f2a, None, 2, CpuSSE2|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
cvtsi2sd, 2, 0xf20f2a, None, 2, CpuSSE2|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
cvtsi2sd, 2, 0xf20f2a, None, 2, CpuSSE2|Cpu64, Modrm|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
divpd, 2, 0x665e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
divpd, 2, 0x660f5e, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
divsd, 2, 0xf25e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
divsd, 2, 0xf20f5e, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
maxpd, 2, 0x665f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
maxpd, 2, 0x660f5f, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
maxsd, 2, 0xf25f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
maxsd, 2, 0xf20f5f, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
minpd, 2, 0x665d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
minpd, 2, 0x660f5d, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
minsd, 2, 0xf25d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
minsd, 2, 0xf20f5d, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movapd, 2, 0x6628, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movapd, 2, 0x660f28, None, 2, CpuSSE2, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movhpd, 2, 0x6616, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex, RegXMM }
movhpd, 2, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Qword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movhpd, 2, 0x660f16, None, 2, CpuSSE2, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movlpd, 2, 0x6612, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex, RegXMM }
movlpd, 2, 0x6613, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Qword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movlpd, 2, 0x660f12, None, 2, CpuSSE2, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
movmskpd, 2, 0x6650, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { RegXMM, Reg32|Reg64 }
gas/testsuite/ 2008-01-04 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/rexw.d: New. * gas/i386/rexw.s: Likewise. * gas/i386/x86-64-sse4_1-intel.d: Updated. * gas/i386/x86-64-sse4_1.d: Likewise. opcodes/ 2008-01-04 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Add NoRex64 to extractps, movmskpd, movmskps, pextrb, pextrw, pinsrb, pinsrw and pmovmskb. * i386-tbl.h: Regenerated.
2008-01-04 19:03:02 +01:00
movmskpd, 2, 0x660f50, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|NoRex64, { RegXMM, Reg32|Reg64 }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
movntpd, 2, 0x662b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, Xmmword|Unspecified|BaseIndex }
movntpd, 2, 0x660f2b, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Xmmword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Intel mode string move.
x86/Intel: drop IgnoreSize from operand-less MOVSD/CMPSD again These were mistakenly added by d241b91073 ("x86/Intel: correct MOVSD and CMPSD handling"). This addresses part of PR/gas 25167.
2019-11-07 09:27:16 +01:00
movsd, 0, 0xa5, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
x86: fold EsSeg into IsString EsSeg (a per-operand bit) is used with IsString (a per-insn attribute) only. Extend the attribute to 2 bits, thus allowing to encode - not a string insn, - string insn with neither operand requiring use of %es:, - string insn with 1st operand requiring use of %es:, - string insn with 2nd operand requiring use of %es:, which covers all possible cases, allowing to drop EsSeg. The (transient) need to comment out the OTUnused #define did uncover an oversight in the earlier OTMax -> OTNum conversion, which is being taken care of here.
2019-11-12 09:09:31 +01:00
movsd, 2, 0xa5, None, 1, 0, Size32|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsStringEsOp1|RepPrefixOk, { Dword|Unspecified|BaseIndex, Dword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movsd, 2, 0xf210, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
movsd, 2, 0xf210, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movsd, 2, 0xf20f10, None, 2, CpuSSE2, D|Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
movupd, 2, 0x6610, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movupd, 2, 0x660f10, None, 2, CpuSSE2, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
mulpd, 2, 0x6659, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
mulpd, 2, 0x660f59, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
mulsd, 2, 0xf259, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
mulsd, 2, 0xf20f59, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
orpd, 2, 0x6656, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
orpd, 2, 0x660f56, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
shufpd, 3, 0x66c6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
shufpd, 3, 0x660fc6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
sqrtpd, 2, 0x6651, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
sqrtpd, 2, 0x660f51, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
sqrtsd, 2, 0xf251, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
sqrtsd, 2, 0xf20f51, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
subpd, 2, 0x665c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
subpd, 2, 0x660f5c, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
subsd, 2, 0xf25c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
subsd, 2, 0xf20f5c, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
ucomisd, 2, 0x662e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
ucomisd, 2, 0x660f2e, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
unpckhpd, 2, 0x6615, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpckhpd, 2, 0x660f15, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpcklpd, 2, 0x6614, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
unpcklpd, 2, 0x660f14, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
xorpd, 2, 0x6657, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
xorpd, 2, 0x660f57, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvtdq2pd, 2, 0xf3e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cvtdq2pd, 2, 0xf30fe6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
cvtpd2dq, 2, 0xf2e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtpd2dq, 2, 0xf20fe6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtdq2ps, 2, 0x5b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtdq2ps, 2, 0xf5b, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtpd2pi, 2, 0x660f2d, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegXMM|Unspecified|BaseIndex, RegMMX }
cvtpd2ps, 2, 0x665a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtpd2ps, 2, 0x660f5a, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvtps2pd, 2, 0x5a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cvtps2pd, 2, 0xf5a, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
cvtps2dq, 2, 0x665b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvtps2dq, 2, 0x660f5b, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvtsd2si, 2, 0xf22d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
cvtsd2si, 2, 0xf20f2d, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
cvtsd2ss, 2, 0xf25a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cvtsd2ss, 2, 0xf20f5a, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
cvtss2sd, 2, 0xf35a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvtss2sd, 2, 0xf30f5a, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
cvttpd2pi, 2, 0x660f2c, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegXMM|Unspecified|BaseIndex, RegMMX }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
cvttsd2si, 2, 0xf22c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
cvttsd2si, 2, 0xf20f2c, None, 2, CpuSSE2, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
cvttpd2dq, 2, 0x66e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvttpd2dq, 2, 0x660fe6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvttps2dq, 2, 0xf35b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
cvttps2dq, 2, 0xf30f5b, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
maskmovdqu, 2, 0x66f7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM, RegXMM }
maskmovdqu, 2, 0x660ff7, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
movdqa, 2, 0x666f, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movdqa, 2, 0x660f6f, None, 2, CpuSSE2, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
movdqu, 2, 0xf36f, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movdqu, 2, 0xf30f6f, None, 2, CpuSSE2, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
movdq2q, 2, 0xf20fd6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegXMM, RegMMX }
movq2dq, 2, 0xf30fd6, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { RegMMX, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
pmuludq, 2, 0x66f4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pmuludq, 2, 0x660ff4, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmuludq, 2, 0xff4, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSE2 insns
2018-09-13 11:11:55 +02:00
pshufd, 3, 0x6670, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pshufd, 3, 0x660f70, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pshufhw, 3, 0xf370, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pshufhw, 3, 0xf30f70, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pshuflw, 3, 0xf270, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pshuflw, 3, 0xf20f70, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pslldq, 2, 0x6673, 0x7, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
pslldq, 2, 0x660f73, 0x7, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
psrldq, 2, 0x6673, 0x3, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM }
psrldq, 2, 0x660f73, 0x3, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM }
punpckhqdq, 2, 0x666d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpckhqdq, 2, 0x660f6d, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpcklqdq, 2, 0x666c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
punpcklqdq, 2, 0x660f6c, None, 2, CpuSSE2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSE3 instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop bogus IgnoreSize from SSE3 insns
2018-09-13 11:12:23 +02:00
addsubpd, 2, 0x66d0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
addsubpd, 2, 0x660fd0, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
addsubps, 2, 0xf2d0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
addsubps, 2, 0xf20fd0, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
cmpxchg16b, 1, 0xfc7, 0x1, 2, CpuCX16|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64|IsLockable, { Oword|Unspecified|BaseIndex }
x86: drop bogus NoAVX These are meaningful only for insns with CpuSSE* (and a few other Cpu*) attribute.
2018-03-08 08:36:41 +01:00
fisttp, 1, 0xdf, 0x1, 1, CpuFISTTP, Modrm|FloatMF|No_bSuf|No_wSuf|No_qSuf|No_ldSuf, { Word|Dword|Unspecified|BaseIndex }
fisttp, 1, 0xdd, 0x1, 1, CpuFISTTP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: don't accept FI{LD,STP,STTP}LL in Intel syntax mode As of commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic") these have been accepted even with "qword ptr" operand size specifier, but in 64-bit mode they're now wrongly having a REX prefix (with REX.W set) emitted in this case. These aren't Intel syntax mnemonics, so rather than fixing code generation, let's simply reject them. As a result, the Qword attribute can then be dropped, too.
2020-03-06 08:53:56 +01:00
fisttpll, 1, 0xdd, 0x1, 1, CpuFISTTP, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from SSE3 insns
2018-09-13 11:12:23 +02:00
haddpd, 2, 0x667c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
haddpd, 2, 0x660f7c, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
haddps, 2, 0xf27c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
haddps, 2, 0xf20f7c, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
hsubpd, 2, 0x667d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
hsubpd, 2, 0x660f7d, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
hsubps, 2, 0xf27d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
hsubps, 2, 0xf20f7d, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
lddqu, 2, 0xf2f0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Xmmword|Unspecified|BaseIndex, RegXMM }
lddqu, 2, 0xf20ff0, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex, RegXMM }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
monitor, 0, 0xf01c8, None, 3, CpuSSE3, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { 0 }
x86: adjust monitor/mwait templates Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or 32-bit register outside of 64-bit mode, and a 64- or 32-bit register inside 64-bit mode. The other register operands, including all of them for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain the original 64-bit MONITOR{,X} templates for compatibility only, and fold the MWAIT{,X} ones.
2018-07-11 10:30:00 +02:00
// monitor is very special. CX and DX are always 32 bits. The
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// address size override prefix can be used to overrride the AX size in
// all modes.
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
monitor, 3, 0xf01c8, None, 3, CpuSSE3, AddrPrefixOpReg|NoAVX, { Acc|Word|Dword|Qword, RegC|Dword, RegD|Dword }
x86: adjust monitor/mwait templates Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or 32-bit register outside of 64-bit mode, and a 64- or 32-bit register inside 64-bit mode. The other register operands, including all of them for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain the original 64-bit MONITOR{,X} templates for compatibility only, and fold the MWAIT{,X} ones.
2018-07-11 10:30:00 +02:00
// The 64-bit form exists only for compatibility with older gas.
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
monitor, 3, 0xf01c8, None, 3, CpuSSE3|Cpu64, AddrPrefixOpReg|NoAVX, { Acc|Dword|Qword, RegC|Qword, RegD|Qword }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
movddup, 2, 0xf212, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
movddup, 2, 0xf20f12, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE3 insns
2018-09-13 11:12:23 +02:00
movshdup, 2, 0xf316, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movshdup, 2, 0xf30f16, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
movsldup, 2, 0xf312, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
movsldup, 2, 0xf30f12, None, 2, CpuSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
mwait, 0, 0xf01c9, None, 3, CpuSSE3, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { 0 }
x86: adjust monitor/mwait templates Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or 32-bit register outside of 64-bit mode, and a 64- or 32-bit register inside 64-bit mode. The other register operands, including all of them for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain the original 64-bit MONITOR{,X} templates for compatibility only, and fold the MWAIT{,X} ones.
2018-07-11 10:30:00 +02:00
// mwait is very special. AX and CX are always 32 bits.
// The 64-bit form exists only for compatibility with older gas.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
mwait, 2, 0xf01c9, None, 3, CpuSSE3, CheckRegSize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Acc|Dword|Qword, RegC|Dword|Qword }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// VMX instructions.
gas/testsuite/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run smx. * gas/i386/smx.d: New. * gas/i386/smx.s: Likewise. opcodes/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (dis386_twobyte): Add getsec. * i386-gen.c (cpu_flags): Add CpuSMX. * i386-opc.h (CpuSMX): New. (CpuSSSE3): Updated. (i386_cpu_flags): Add cpusmx. * i386-opc.tbl: Add getsec. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2007-10-05 21:04:06 +02:00
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmcall, 0, 0xf01c1, None, 3, CpuVMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmclear, 1, 0x660fc7, 0x6, 2, CpuVMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmlaunch, 0, 0xf01c2, None, 3, CpuVMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
vmresume, 0, 0xf01c3, None, 3, CpuVMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmptrld, 1, 0xfc7, 0x6, 2, CpuVMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
vmptrst, 1, 0xfc7, 0x7, 2, CpuVMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
vmread, 2, 0xf78, None, 2, CpuVMX|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, Reg32|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmread, 2, 0xf78, None, 2, CpuVMX|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg64, Reg64|Qword|Unspecified|BaseIndex }
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
vmwrite, 2, 0xf79, None, 2, CpuVMX|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, Reg32 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmwrite, 2, 0xf79, None, 2, CpuVMX|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoRex64, { Reg64|Qword|Unspecified|BaseIndex, Reg64 }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmxoff, 0, 0xf01c4, None, 3, CpuVMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmxon, 1, 0xf30fc7, 0x6, 2, CpuVMX, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
Add vmfunc gas/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (cpu_arch): Add ".vmfunc". * doc/c-i386.texi: Document vmfunc. gas/testsuite/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run vmfunc and x86-64-vmfunc. * gas/i386/vmfunc.d: New. * gas/i386/vmfunc.s: Likewise. * gas/i386/x86-64-vmfunc.d: Likewise. opcodes/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (mod_table): Add vmfunc. * i386-gen.c (cpu_flag_init): Add CPU_VMFUNC_FLAGS. (cpu_flags): CpuVMFUNC. * i386-opc.h (CpuVMFUNC): New. (i386_cpu_flags): Add cpuvmfunc. * i386-opc.tbl: Add vmfunc. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2012-01-13 23:19:32 +01:00
// VMFUNC instruction
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmfunc, 0, 0xf01d4, None, 3, CpuVMFUNC, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Add vmfunc gas/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (cpu_arch): Add ".vmfunc". * doc/c-i386.texi: Document vmfunc. gas/testsuite/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run vmfunc and x86-64-vmfunc. * gas/i386/vmfunc.d: New. * gas/i386/vmfunc.s: Likewise. * gas/i386/x86-64-vmfunc.d: Likewise. opcodes/ 2012-01-13 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (mod_table): Add vmfunc. * i386-gen.c (cpu_flag_init): Add CPU_VMFUNC_FLAGS. (cpu_flags): CpuVMFUNC. * i386-opc.h (CpuVMFUNC): New. (i386_cpu_flags): Add cpuvmfunc. * i386-opc.tbl: Add vmfunc. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2012-01-13 23:19:32 +01:00
gas/testsuite/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run smx. * gas/i386/smx.d: New. * gas/i386/smx.s: Likewise. opcodes/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (dis386_twobyte): Add getsec. * i386-gen.c (cpu_flags): Add CpuSMX. * i386-opc.h (CpuSMX): New. (CpuSSSE3): Updated. (i386_cpu_flags): Add cpusmx. * i386-opc.tbl: Add getsec. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2007-10-05 21:04:06 +02:00
// SMX instructions.
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
getsec, 0, 0xf37, None, 2, CpuSMX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
gas/testsuite/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run smx. * gas/i386/smx.d: New. * gas/i386/smx.s: Likewise. opcodes/ 2007-10-05 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (dis386_twobyte): Add getsec. * i386-gen.c (cpu_flags): Add CpuSMX. * i386-opc.h (CpuSMX): New. (CpuSSSE3): Updated. (i386_cpu_flags): Add cpusmx. * i386-opc.tbl: Add getsec. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2007-10-05 21:04:06 +02:00
gas/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention XSAVE, EPT and MOVBE. * config/tc-i386.c (cpu_arch): Add .movbe and .ept. (md_show_usage): Add .movbe and .ept. * doc/c-i386.texi: Add movbe and ept to -march=. Document .movbe and .ept. gas/testsuite/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run movbe, movbe-intel, inval-movbe, ept, ept-intel, inval-ept, x86-64-movbe, x86-64-movbe-intel, x86-64-inval-movbe. x86-64-ept, x86-64-ept-intel and x86-64-inval-ept. * gas/i386/arch-10.s: Add movbe and invept. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/ept.d: New file * gas/i386/ept-intel.d: Likewise. * gas/i386/ept.s: Likewise. * gas/i386/inval-ept.l: Likewise. * gas/i386/inval-ept.s: Likewise. * gas/i386/inval-movbe.l: Likewise. * gas/i386/inval-movbe.s: Likewise. * gas/i386/movbe.d: Likewise. * gas/i386/movbe-intel.d: Likewise. * gas/i386/movbe.s: Likewise. * gas/i386/x86-64-inval-ept.l: Likewise. * gas/i386/x86-64-inval-ept.s: Likewise. * gas/i386/x86-64-inval-movbe.l: Likewise. * gas/i386/x86-64-inval-movbe.s: Likewise. * gas/i386/x86-64-ept.d: Likewise. * gas/i386/x86-64-ept-intel.d: Likewise. * gas/i386/x86-64-ept.s: Likewise. * gas/i386/x86-64-movbe.d: Likewise. * gas/i386/x86-64-movbe-intel.d: Likewise. * gas/i386/x86-64-movbe.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (MOVBE_Fixup): New. (Mo): Likewise. (PREFIX_0F3880): Likewise. (PREFIX_0F3881): Likewise. (PREFIX_0F38F0): Updated. (prefix_table): Add PREFIX_0F3880 and PREFIX_0F3881. Update PREFIX_0F38F0 and PREFIX_0F38F1 for movbe. (three_byte_table): Use PREFIX_0F3880 and PREFIX_0F3881. * i386-gen.c (cpu_flag_init): Add CPU_MOVBE_FLAGS and CPU_EPT_FLAGS. (cpu_flags): Add CpuMovbe and CpuEPT. * i386-opc.h (CpuMovbe): New. (CpuEPT): Likewise. (CpuLM): Updated. (i386_cpu_flags): Add cpumovbe and cpuept. * i386-opc.tbl: Add entries for movbe and EPT instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-05-02 18:53:40 +02:00
// EPT instructions.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
invept, 2, 0x660f3880, None, 3, CpuEPT|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf, { Oword|Unspecified|BaseIndex, Reg32 }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
invept, 2, 0x660f3880, None, 3, CpuEPT|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf|NoRex64, { Oword|Unspecified|BaseIndex, Reg64 }
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
invvpid, 2, 0x660f3881, None, 3, CpuEPT|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf, { Oword|Unspecified|BaseIndex, Reg32 }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
invvpid, 2, 0x660f3881, None, 3, CpuEPT|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf|NoRex64, { Oword|Unspecified|BaseIndex, Reg64 }
gas/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention XSAVE, EPT and MOVBE. * config/tc-i386.c (cpu_arch): Add .movbe and .ept. (md_show_usage): Add .movbe and .ept. * doc/c-i386.texi: Add movbe and ept to -march=. Document .movbe and .ept. gas/testsuite/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run movbe, movbe-intel, inval-movbe, ept, ept-intel, inval-ept, x86-64-movbe, x86-64-movbe-intel, x86-64-inval-movbe. x86-64-ept, x86-64-ept-intel and x86-64-inval-ept. * gas/i386/arch-10.s: Add movbe and invept. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/ept.d: New file * gas/i386/ept-intel.d: Likewise. * gas/i386/ept.s: Likewise. * gas/i386/inval-ept.l: Likewise. * gas/i386/inval-ept.s: Likewise. * gas/i386/inval-movbe.l: Likewise. * gas/i386/inval-movbe.s: Likewise. * gas/i386/movbe.d: Likewise. * gas/i386/movbe-intel.d: Likewise. * gas/i386/movbe.s: Likewise. * gas/i386/x86-64-inval-ept.l: Likewise. * gas/i386/x86-64-inval-ept.s: Likewise. * gas/i386/x86-64-inval-movbe.l: Likewise. * gas/i386/x86-64-inval-movbe.s: Likewise. * gas/i386/x86-64-ept.d: Likewise. * gas/i386/x86-64-ept-intel.d: Likewise. * gas/i386/x86-64-ept.s: Likewise. * gas/i386/x86-64-movbe.d: Likewise. * gas/i386/x86-64-movbe-intel.d: Likewise. * gas/i386/x86-64-movbe.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2008-05-02 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (MOVBE_Fixup): New. (Mo): Likewise. (PREFIX_0F3880): Likewise. (PREFIX_0F3881): Likewise. (PREFIX_0F38F0): Updated. (prefix_table): Add PREFIX_0F3880 and PREFIX_0F3881. Update PREFIX_0F38F0 and PREFIX_0F38F1 for movbe. (three_byte_table): Use PREFIX_0F3880 and PREFIX_0F3881. * i386-gen.c (cpu_flag_init): Add CPU_MOVBE_FLAGS and CPU_EPT_FLAGS. (cpu_flags): Add CpuMovbe and CpuEPT. * i386-opc.h (CpuMovbe): New. (CpuEPT): Likewise. (CpuLM): Updated. (i386_cpu_flags): Add cpumovbe and cpuept. * i386-opc.tbl: Add entries for movbe and EPT instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-05-02 18:53:40 +02:00
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
// INVPCID instruction
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
invpcid, 2, 0x660f3882, None, 3, CpuINVPCID|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf, { Oword|Unspecified|BaseIndex, Reg32 }
x86/Intel: accept "oword ptr" for INVPCID The insn is no different in this reagrd from INVEPT and INVVPID.
2018-06-01 08:37:24 +02:00
invpcid, 2, 0x660f3882, None, 3, CpuINVPCID|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_qSuf|No_sSuf|No_ldSuf|NoRex64, { Oword|Unspecified|BaseIndex, Reg64 }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSSE3 instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phaddw, 2, 0x6601, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phaddw, 2, 0x660f3801, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phaddw, 2, 0xf3801, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phaddd, 2, 0x6602, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phaddd, 2, 0x660f3802, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phaddd, 2, 0xf3802, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phaddsw, 2, 0x6603, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phaddsw, 2, 0x660f3803, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phaddsw, 2, 0xf3803, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phsubw, 2, 0x6605, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phsubw, 2, 0x660f3805, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phsubw, 2, 0xf3805, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phsubd, 2, 0x6606, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phsubd, 2, 0x660f3806, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phsubd, 2, 0xf3806, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
phsubsw, 2, 0x6607, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phsubsw, 2, 0x660f3807, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
phsubsw, 2, 0xf3807, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pmaddubsw, 2, 0x6604, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaddubsw, 2, 0x660f3804, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmaddubsw, 2, 0xf3804, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pmulhrsw, 2, 0x660b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmulhrsw, 2, 0x660f380b, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmulhrsw, 2, 0xf380b, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pshufb, 2, 0x6600, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pshufb, 2, 0x660f3800, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pshufb, 2, 0xf3800, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
psignb, 2, 0x6608, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psignb, 2, 0x660f3808, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psignb, 2, 0xf3808, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
psignw, 2, 0x6609, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psignw, 2, 0x660f3809, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psignw, 2, 0xf3809, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
psignd, 2, 0x660a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
psignd, 2, 0x660f380a, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
psignd, 2, 0xf380a, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
palignr, 3, 0x660f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
palignr, 3, 0x660f3a0f, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
palignr, 3, 0xf3a0f, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Imm8, Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pabsb, 2, 0x661c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pabsb, 2, 0x660f381c, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pabsb, 2, 0xf381c, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pabsw, 2, 0x661d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pabsw, 2, 0x660f381d, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pabsw, 2, 0xf381d, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
x86: drop bogus IgnoreSize from SSSE3 insns
2018-09-13 11:12:51 +02:00
pabsd, 2, 0x661e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pabsd, 2, 0x660f381e, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pabsd, 2, 0xf381e, None, 3, CpuSSSE3, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|NoAVX, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSE4.1 instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop bogus IgnoreSize from SSE4.1 insns
2018-09-13 11:14:32 +02:00
blendpd, 3, 0x660d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
blendpd, 3, 0x660f3a0d, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
blendps, 3, 0x660c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
blendps, 3, 0x660f3a0c, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
blendvpd, 3, 0x664b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
blendvpd, 2, 0x664b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Implicit1stXmm0|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
blendvpd, 3, 0x660f3815, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
blendvpd, 2, 0x660f3815, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
blendvps, 3, 0x664a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
blendvps, 2, 0x664a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Implicit1stXmm0|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
blendvps, 3, 0x660f3814, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
blendvps, 2, 0x660f3814, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
dppd, 3, 0x6641, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
dppd, 3, 0x660f3a41, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
dpps, 3, 0x6640, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
dpps, 3, 0x660f3a40, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: add a few more missing VexWIG Alternatively it could also be VexW0 (to match other SSE2AVX), but the VexW attribute shouldn't be left unset.
2020-01-16 10:06:21 +01:00
extractps, 3, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
extractps, 3, 0x6617, None, 1, CpuAVX|Cpu64, RegMem|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg64 }
x86: replace Reg8, Reg16, Reg32, and Reg64 Use a combination of a single new Reg bit and Byte, Word, Dword, or Qword instead. Besides shrinking the number of operand type bits this has the benefit of making register handling more similar to accumulator handling (a generic flag is being accompanied by a "size qualifier"). It requires, however, to split a few insn templates, as it is no longer correct to have combinations like Reg32|Reg64|Byte. This slight growth in size will hopefully be outweighed by this change paving the road for folding a presumably much larger number of templates later on.
2017-12-18 09:34:00 +01:00
extractps, 3, 0x660f3a17, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
extractps, 3, 0x660f3a17, None, 3, CpuSSE4_1|Cpu64, RegMem|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, RegXMM, Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
insertps, 3, 0x6621, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
insertps, 3, 0x660f3a21, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.1 insns
2018-09-13 11:14:32 +02:00
movntdqa, 2, 0x662a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Xmmword|Unspecified|BaseIndex, RegXMM }
movntdqa, 2, 0x660f382a, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex, RegXMM }
mpsadbw, 3, 0x6642, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
mpsadbw, 3, 0x660f3a42, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
packusdw, 2, 0x662b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
packusdw, 2, 0x660f382b, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pblendvb, 3, 0x664c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
pblendvb, 2, 0x664c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Implicit1stXmm0|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pblendvb, 3, 0x660f3810, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
pblendvb, 2, 0x660f3810, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pblendw, 3, 0x660e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pblendw, 3, 0x660f3a0e, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpeqq, 2, 0x6629, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpeqq, 2, 0x660f3829, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
pextrb, 3, 0x6614, None, 1, CpuAVX, RegMem|Vex|VexOpcode=2|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pextrb, 3, 0x6614, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Byte|Unspecified|BaseIndex }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
pextrb, 3, 0x660f3a14, None, 3, CpuSSE4_1, RegMem|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pextrb, 3, 0x660f3a14, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Byte|Unspecified|BaseIndex }
i386: Update VexW field for VEX instructions 1. Mark VEX.W0 VEX instructions with VexW=1. 2. Mark VEX.W1 VEX instructions with VexW=2. 3. Remove VexW=1 from WIG VEX instructions. * i386-opc.tbl: Add VexW=1 to VEX.W0 VEX movd, cvtsi2ss, cvtsi2sd, pextrd, pinsrd, vcvtsi2sd, vcvtsi2ss, vmovd, vpextrd and vpinsrd. Add VexW=2 to VEX.W1 VEX movd, movq, pextrq, pinsrq, vmod, vmovq, vpextrq and vpinsrq. Remove VexW=1 from WIG VEX movq and vmovq. * i386-tbl.h: Regenerated.
2018-09-13 00:12:59 +02:00
pextrd, 3, 0x6616, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pextrd, 3, 0x660f3a16, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: VPEXTRQ/VPINSRQ are unavailable outside of 64-bit mode The AVX512DQ patterns lacking a Cpu64 attribute made the memory operand forms accepted even outside of 64-bit mode, and this even without any {evex} pseudo-prefix (otherwise one could argue that this is an attempt to follow one possible, albeit somewhat odd, interpretation of the SDM wording to this effect). For consistency between the various involved templates drop the * (now) unnecessary IgnoreSize attributes * unnecessary (due to VexW1) Size64 attributes from VEX encoded forms * redundant (with Reg64) Qword operand attributes uniformly.
2020-01-16 10:05:35 +01:00
pextrq, 3, 0x6616, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM, Reg64|Unspecified|BaseIndex }
pextrq, 3, 0x660f3a16, None, 3, CpuSSE4_1|Cpu64, Modrm|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg64|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from SSE4.1 insns
2018-09-13 11:14:32 +02:00
phminposuw, 2, 0x6641, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
phminposuw, 2, 0x660f3841, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
pinsrb, 3, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Reg32|Reg64, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pinsrb, 3, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Byte|Unspecified|BaseIndex, RegXMM }
x86: replace Reg8, Reg16, Reg32, and Reg64 Use a combination of a single new Reg bit and Byte, Word, Dword, or Qword instead. Besides shrinking the number of operand type bits this has the benefit of making register handling more similar to accumulator handling (a generic flag is being accompanied by a "size qualifier"). It requires, however, to split a few insn templates, as it is no longer correct to have combinations like Reg32|Reg64|Byte. This slight growth in size will hopefully be outweighed by this change paving the road for folding a presumably much larger number of templates later on.
2017-12-18 09:34:00 +01:00
pinsrb, 3, 0x660f3a20, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, Reg32|Reg64, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pinsrb, 3, 0x660f3a20, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Byte|Unspecified|BaseIndex, RegXMM }
i386: Update VexW field for VEX instructions 1. Mark VEX.W0 VEX instructions with VexW=1. 2. Mark VEX.W1 VEX instructions with VexW=2. 3. Remove VexW=1 from WIG VEX instructions. * i386-opc.tbl: Add VexW=1 to VEX.W0 VEX movd, cvtsi2ss, cvtsi2sd, pextrd, pinsrd, vcvtsi2sd, vcvtsi2ss, vmovd, vpextrd and vpinsrd. Add VexW=2 to VEX.W1 VEX movd, movq, pextrq, pinsrq, vmod, vmovq, vpextrq and vpinsrq. Remove VexW=1 from WIG VEX movq and vmovq. * i386-tbl.h: Regenerated.
2018-09-13 00:12:59 +02:00
pinsrd, 3, 0x6622, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pinsrd, 3, 0x660f3a22, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Dword|Unspecified|BaseIndex, RegXMM }
x86: VPEXTRQ/VPINSRQ are unavailable outside of 64-bit mode The AVX512DQ patterns lacking a Cpu64 attribute made the memory operand forms accepted even outside of 64-bit mode, and this even without any {evex} pseudo-prefix (otherwise one could argue that this is an attempt to follow one possible, albeit somewhat odd, interpretation of the SDM wording to this effect). For consistency between the various involved templates drop the * (now) unnecessary IgnoreSize attributes * unnecessary (due to VexW1) Size64 attributes from VEX encoded forms * redundant (with Reg64) Qword operand attributes uniformly.
2020-01-16 10:05:35 +01:00
pinsrq, 3, 0x6622, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Reg64|Unspecified|BaseIndex, RegXMM }
pinsrq, 3, 0x660f3a22, None, 3, CpuSSE4_1|Cpu64, Modrm|Size64|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg64|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from SSE4.1 insns
2018-09-13 11:14:32 +02:00
pmaxsb, 2, 0x663c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxsb, 2, 0x660f383c, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxsd, 2, 0x663d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxsd, 2, 0x660f383d, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxud, 2, 0x663f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxud, 2, 0x660f383f, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxuw, 2, 0x663e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmaxuw, 2, 0x660f383e, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminsb, 2, 0x6638, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminsb, 2, 0x660f3838, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminsd, 2, 0x6639, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminsd, 2, 0x660f3839, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminud, 2, 0x663b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminud, 2, 0x660f383b, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminuw, 2, 0x663a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pminuw, 2, 0x660f383a, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovsxbw, 2, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovsxbw, 2, 0x660f3820, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { QWord|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovsxbd, 2, 0x6621, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovsxbd, 2, 0x660f3821, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovsxbq, 2, 0x6622, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovsxbq, 2, 0x660f3822, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovsxwd, 2, 0x6623, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovsxwd, 2, 0x660f3823, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovsxwq, 2, 0x6624, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovsxwq, 2, 0x660f3824, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovsxdq, 2, 0x6625, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovsxdq, 2, 0x660f3825, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { QWord|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxbw, 2, 0x6630, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxbw, 2, 0x660f3830, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxbd, 2, 0x6631, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovzxbd, 2, 0x660f3831, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovzxbq, 2, 0x6632, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovzxbq, 2, 0x660f3832, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovzxwd, 2, 0x6633, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxwd, 2, 0x660f3833, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxwq, 2, 0x6634, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pmovzxwq, 2, 0x660f3834, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pmovzxdq, 2, 0x6635, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
pmovzxdq, 2, 0x660f3835, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { QWord|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.1 insns
2018-09-13 11:14:32 +02:00
pmuldq, 2, 0x6628, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmuldq, 2, 0x660f3828, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmulld, 2, 0x6640, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pmulld, 2, 0x660f3840, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
ptest, 2, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
ptest, 2, 0x660f3817, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
roundpd, 3, 0x6609, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
roundpd, 3, 0x660f3a09, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
roundps, 3, 0x6608, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
roundps, 3, 0x660f3a08, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
roundsd, 3, 0x660b, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=2|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
roundsd, 3, 0x660f3a0b, None, 3, CpuSSE4_1, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
roundss, 3, 0x660a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=2|VexVVVV=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
roundss, 3, 0x660f3a0a, None, 3, CpuSSE4_1, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
2007-10-03 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Update SSE comments.
2007-10-03 21:03:20 +02:00
// SSE4.2 instructions.
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: drop bogus IgnoreSize from SSE4.2 insns
2018-09-13 11:15:01 +02:00
pcmpgtq, 2, 0x6637, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpgtq, 2, 0x660f3837, None, 3, CpuSSE4_2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpestri, 3, 0x6661, None, 1, CpuAVX|CpuNo64, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pcmpestri, 3, 0x6661, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.2 insns
2018-09-13 11:15:01 +02:00
pcmpestri, 3, 0x660f3a61, None, 3, CpuSSE4_2|CpuNo64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pcmpestri, 3, 0x660f3a61, None, 3, CpuSSE4_2|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.2 insns
2018-09-13 11:15:01 +02:00
pcmpestrm, 3, 0x6660, None, 1, CpuAVX|CpuNo64, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pcmpestrm, 3, 0x6660, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SSE2AVX, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.2 insns
2018-09-13 11:15:01 +02:00
pcmpestrm, 3, 0x660f3a60, None, 3, CpuSSE4_2|CpuNo64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
pcmpestrm, 3, 0x660f3a60, None, 3, CpuSSE4_2|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from SSE4.2 insns
2018-09-13 11:15:01 +02:00
pcmpistri, 3, 0x6663, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpistri, 3, 0x660f3a63, None, 3, CpuSSE4_2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpistrm, 3, 0x6662, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pcmpistrm, 3, 0x660f3a62, None, 3, CpuSSE4_2, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
x86: fold CRC32 templates Just like other insns having byte and word forms, these can also make use of the W modifier, which at the same time allows simplifying some other code a little bit.
2018-09-14 11:21:15 +02:00
crc32, 2, 0xf20f38f0, None, 3, CpuSSE4_2, W|Modrm|No_sSuf|No_qSuf|No_ldSuf|NoAVX, { Reg8|Reg16|Reg32|Unspecified|BaseIndex, Reg32 }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
crc32, 2, 0xf20f38f0, None, 3, CpuSSE4_2|Cpu64, W|Modrm|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|NoAVX, { Reg8|Reg64|Unspecified|BaseIndex, Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
gas/testsuite/ 2002-02-11 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run xsave, xsave-intel, x86-64-xsave and x86-64-xsave-intel. * gas/i386/x86-64-xsave-intel.d: New file. * gas/i386/x86-64-xsave.d: Likewise. * gas/i386/x86-64-xsave.s: Likewise. * gas/i386/xsave-intel.d: Likewise. * gas/i386/xsave.d: Likewise. * gas/i386/xsave.s: Likewise. opcodes/ 2008-02-11 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (cpu_flags): Add CpuXsave. * i386-opc.h (CpuXsave): New. (Cpu64): Updated. (i386_cpu_flags): Add cpuxsave. * i386-dis.c (MOD_0FAE_REG_4): New. (RM_0F01_REG_2): Likewise. (MOD_0FAE_REG_5): Updated. (RM_0F01_REG_3): Likewise. (reg_table): Use MOD_0FAE_REG_4. (mod_table): Use RM_0F01_REG_2. Add MOD_0FAE_REG_4. Updated for xrstor. (rm_table): Add RM_0F01_REG_2. * i386-opc.tbl: Add xsave, xrstor, xgetbv and xsetbv. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-02-12 01:04:45 +01:00
// xsave/xrstor New Instructions.
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xsave, 1, 0xfae, 0x4, 2, CpuXsave, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xsave64, 1, 0xfae, 0x4, 2, CpuXsave|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xrstor, 1, 0xfae, 0x5, 2, CpuXsave, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xrstor64, 1, 0xfae, 0x5, 2, CpuXsave|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
xgetbv, 0, 0xf01d0, None, 3, CpuXsave, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
xsetbv, 0, 0xf01d1, None, 3, CpuXsave, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
gas/testsuite/ 2002-02-11 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run xsave, xsave-intel, x86-64-xsave and x86-64-xsave-intel. * gas/i386/x86-64-xsave-intel.d: New file. * gas/i386/x86-64-xsave.d: Likewise. * gas/i386/x86-64-xsave.s: Likewise. * gas/i386/xsave-intel.d: Likewise. * gas/i386/xsave.d: Likewise. * gas/i386/xsave.s: Likewise. opcodes/ 2008-02-11 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (cpu_flags): Add CpuXsave. * i386-opc.h (CpuXsave): New. (Cpu64): Updated. (i386_cpu_flags): Add cpuxsave. * i386-dis.c (MOD_0FAE_REG_4): New. (RM_0F01_REG_2): Likewise. (MOD_0FAE_REG_5): Updated. (RM_0F01_REG_3): Likewise. (reg_table): Use MOD_0FAE_REG_4. (mod_table): Use RM_0F01_REG_2. Add MOD_0FAE_REG_4. Updated for xrstor. (rm_table): Add RM_0F01_REG_2. * i386-opc.tbl: Add xsave, xrstor, xgetbv and xsetbv. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-02-12 01:04:45 +01:00
gas/ * config/tc-i386.c (parse_insn): Don't complain about REP prefix when the template has opcode_modifier.repprefixok set. * NEWS: Mention the change. gas/testsuite/ * gas/i386/rep-bsf.d: New file. * gas/i386/rep-bsf.s: New file. * gas/i386/i386.exp: Add the new test. opcodes/ * i386-opc.h (RepPrefixOk): New enum constant. (i386_opcode_modifier): New bitfield 'repprefixok'. * i386-gen.c (opcode_modifiers): Add RepPrefixOk. * i386-opc.tbl: Add RepPrefixOk to bsf, bsr, and to all instructions that have IsString. * i386-tbl.h: Regenerate.
2012-06-22 18:42:08 +02:00
// xsaveopt
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
xsaveopt, 1, 0xfae, 0x6, 2, CpuXsaveopt, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xsaveopt64, 1, 0xfae, 0x6, 2, CpuXsaveopt|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
Support AVX Programming Reference (June, 2010) gas/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * config/tc-i386.c (cpu_arch): Add .xsaveopt, .fsgsbase, .rdrnd and .f16c. * doc/c-i386.texi: Document xsaveopt, fsgsbase, rdrnd and f16c. gas/testsuite/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * gas/i386/arch-10.s: Add xsaveopt. * gas/i386/x86-64-arch-2.s: Likwise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/f16c-intel.d: New. * gas/i386/f16c.d: Likewise. * gas/i386/f16c.s: Likewise. * gas/i386/fsgs-intel.d: Likewise. * gas/i386/fsgs.d: Likewise. * gas/i386/fsgs.s: Likewise. * gas/i386/rdrnd-intel.d: Likewise. * gas/i386/rdrnd.d: Likewise. * gas/i386/rdrnd.s: Likewise. * gas/i386/x86-64-f16c-intel.d: Likewise. * gas/i386/x86-64-f16c.d: Likewise. * gas/i386/x86-64-f16c.s: Likewise. * gas/i386/x86-64-fsgs-intel.d: Likewise. * gas/i386/x86-64-fsgs.d: Likewise. * gas/i386/x86-64-fsgs.s: Likewise. * gas/i386/x86-64-rdrnd-intel.d: Likewise. * gas/i386/x86-64-rdrnd.d: Likewise. * gas/i386/x86-64-rdrnd.s: Likewise. * gas/i386/i386.exp: Run f16c, f16c-intel, fsgs, fsgs-intel, rdrnd, rdrnd-intel, x86-64-f16c, x86-64-f16c-intel, x86-64-fsgs, x86-64-fsgs-intel, x86-64-rdrnd, x86-64-rdrnd-intel. * gas/i386/x86-64-xsave.s: Add tests for xsaveopt64. * gas/i386/x86-64-xsave-intel.d: Updated. * gas/i386/x86-64-xsave.d: Likewise. opcodes/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * i386-dis.c (PREFIX_0FAE_REG_0): New. (PREFIX_0FAE_REG_1): Likewise. (PREFIX_0FAE_REG_2): Likewise. (PREFIX_0FAE_REG_3): Likewise. (PREFIX_VEX_3813): Likewise. (PREFIX_VEX_3A1D): Likewise. (prefix_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3, PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (vex_table): Add PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (mod_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3 xsaveopt and rdrnd. * i386-gen.c (cpu_flag_init): Add CPU_XSAVEOPT_FLAGS, CPU_FSGSBASE_FLAGS, CPU_RDRND_FLAGS and CPU_F16C_FLAGS. (cpu_flags): Add CpuXsaveopt, CpuFSGSBase, CpuRdRnd and CpuF16C. * i386-opc.h (CpuXsaveopt): New. (CpuFSGSBase):Likewise. (CpuRdRnd): Likewise. (CpuF16C): Likewise. (i386_cpu_flags): Add cpuxsaveopt, cpufsgsbase, cpurdrnd and cpuf16c. * i386-opc.tbl: Add xsaveopt, rdfsbase, rdgsbase, rdrnd, wrfsbase, wrgsbase, vcvtph2ps and vcvtps2ph.
2010-07-01 23:55:02 +02:00
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
// AES instructions.
x86: drop bogus IgnoreSize from AES/VAES insns
2018-09-13 11:15:38 +02:00
aesdec, 2, 0x66de, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesdec, 2, 0x660f38de, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesdeclast, 2, 0x66df, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesdeclast, 2, 0x660f38df, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesenc, 2, 0x66dc, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesenc, 2, 0x660f38dc, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesenclast, 2, 0x66dd, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesenclast, 2, 0x660f38dd, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesimc, 2, 0x66db, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
aesimc, 2, 0x660f38db, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
aeskeygenassist, 3, 0x66df, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
aeskeygenassist, 3, 0x660f3adf, None, 3, CpuAES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
Enable Intel VAES instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VAES. * doc/c-i386.texi: Document VAES. * testsuite/gas/i386/i386.exp: Run VAES tests. * testsuite/gas/i386/avx512f_vaes-intel.d: New test. * testsuite/gas/i386/avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vaes.d: Ditto. * testsuite/gas/i386/avx512f_vaes.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.s: Ditto. * testsuite/gas/i386/vaes-intel.d: Ditto. * testsuite/gas/i386/vaes.d: Ditto. * testsuite/gas/i386/vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.s: Ditto. * testsuite/gas/i386/x86-64-vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-vaes.d: Ditto. * testsuite/gas/i386/x86-64-vaes.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F38DC, PREFIX_EVEX_0F38DD, PREFIX_EVEX_0F38DE, PREFIX_EVEX_0F38DF. (enum): Remove VEX_LEN_0F38DC_P_2, VEX_LEN_0F38DD_P_2, VEX_LEN_0F38DE_P_2, VEX_LEN_0F38DF_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F38DC_P_2, VEX_W_0F38DD_P_2, VEX_W_0F38DE_P_2, VEX_W_0F38DF_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VAES. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. * i386-opc.tbl (vaes{enc,dec}{last,}): New instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:35:45 +02:00
// VAES
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vaesdec, 3, 0x66de, None, 1, CpuVAES, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegYMM, RegYMM }
vaesdeclast, 3, 0x66df, None, 1, CpuVAES, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegYMM, RegYMM }
vaesenc, 3, 0x66dc, None, 1, CpuVAES, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegYMM, RegYMM }
vaesenclast, 3, 0x66dd, None, 1, CpuVAES, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegYMM, RegYMM }
Enable Intel VAES instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VAES. * doc/c-i386.texi: Document VAES. * testsuite/gas/i386/i386.exp: Run VAES tests. * testsuite/gas/i386/avx512f_vaes-intel.d: New test. * testsuite/gas/i386/avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vaes.d: Ditto. * testsuite/gas/i386/avx512f_vaes.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.s: Ditto. * testsuite/gas/i386/vaes-intel.d: Ditto. * testsuite/gas/i386/vaes.d: Ditto. * testsuite/gas/i386/vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.s: Ditto. * testsuite/gas/i386/x86-64-vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-vaes.d: Ditto. * testsuite/gas/i386/x86-64-vaes.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F38DC, PREFIX_EVEX_0F38DD, PREFIX_EVEX_0F38DE, PREFIX_EVEX_0F38DF. (enum): Remove VEX_LEN_0F38DC_P_2, VEX_LEN_0F38DD_P_2, VEX_LEN_0F38DE_P_2, VEX_LEN_0F38DF_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F38DC_P_2, VEX_W_0F38DD_P_2, VEX_W_0F38DE_P_2, VEX_W_0F38DF_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VAES. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. * i386-opc.tbl (vaes{enc,dec}{last,}): New instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:35:45 +02:00
gas/ 2008-04-04 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention XSAVE. Change CLMUL to PCLMUL. * config/tc-i386.c (cpu_arch): Add .pclmul. (md_show_usage): Replace clmul with pclmul. * doc/c-i386.texi: Likewise. gas/testsuite/ 2008-04-04 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/arch-10-1.l: Replace CLMUL with PCLMUL. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/arch-10.d: Replace clmul with pclmul. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2008-04-04 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (cpu_flag_init): Replace CPU_CLMUL_FLAGS/CpuCLMUL with CPU_PCLMUL_FLAGS/CpuPCLMUL. (cpu_flags): Replace CpuCLMUL with CpuPCLMUL. * i386-opc.tbl: Likewise. * i386-opc.h (CpuCLMUL): Renamed to ... (CpuPCLMUL): This. (CpuFMA): Updated. (i386_cpu_flags): Replace cpuclmul with cpupclmul. * i386-init.h: Regenerated.
2008-04-04 18:34:23 +02:00
// PCLMUL
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
x86: drop bogus IgnoreSize from PCLMUL/VPCLMUL insns
2018-09-13 11:16:19 +02:00
pclmulqdq, 3, 0x6644, None, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pclmulqdq, 3, 0x660f3a44, None, 3, CpuPCLMUL, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
pclmullqlqdq, 2, 0x6644, 0x0, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmullqlqdq, 2, 0x660f3a44, 0x0, 3, CpuPCLMUL, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmulhqlqdq, 2, 0x6644, 0x1, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmulhqlqdq, 2, 0x660f3a44, 0x1, 3, CpuPCLMUL, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmullqhqdq, 2, 0x6644, 0x10, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmullqhqdq, 2, 0x660f3a44, 0x10, 3, CpuPCLMUL, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmulhqhqdq, 2, 0x6644, 0x11, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
pclmulhqhqdq, 2, 0x660f3a44, 0x11, 3, CpuPCLMUL, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM }
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
// GFNI
x86: drop bogus IgnoreSize from GNFI insns
2018-09-13 11:16:49 +02:00
gf2p8affineqb, 3, 0x66ce, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
gf2p8affineqb, 3, 0x660f3ace, None, 3, CpuGFNI, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
gf2p8affineinvqb, 3, 0x66cf, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
gf2p8affineinvqb, 3, 0x660f3acf, None, 3, CpuGFNI, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
gf2p8mulb, 2, 0x66cf, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SSE2AVX, { RegXMM|Unspecified|BaseIndex, RegXMM }
gf2p8mulb, 2, 0x660f38cf, None, 3, CpuGFNI, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
// AVX instructions.
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
<avx_frel:imm:comm, eq:00:C, eq_oq:00:C, lt:01:, lt_os:01:, le:02:, le_os:02:, \
unord:03:C, unord_q:03:C, neq:04:C, neq_uq:04:C, nlt:05:, nlt_us:05:, \
nle:06:, nle_us:06:, ord:07:C, ord_q:07:C, eq_uq:08:C, \
nge:09:, nge_us:09:, ngt:0a:, ngt_us:0a:, false:0b:C, false_oq:0b:C, \
neq_oq:0c:C, ge:0d:, ge_os:0d:, gt:0e:, gt_os:0e:, true:0f:C, \
true_uq:0f:C, eq_os:10:C, lt_oq:11:, le_oq:12:, \
unord_s:13:C, neq_us:14:C, nlt_uq:15:, nle_uq:16:, ord_s:17:C, eq_us:18:C, \
nge_uq:19:, ngt_uq:1a:, false_os:1b:C, neq_os:1c:C, ge_oq:1d:, gt_oq:1e:, \
true_us:1f:C>
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vaddpd, 3, 0x6658, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vaddps, 3, 0x58, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vaddsd, 3, 0xf258, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vaddss, 3, 0xf358, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vaddsubpd, 3, 0x66d0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vaddsubps, 3, 0xf2d0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: add more VexWIG Commits 6865c0435a ("x86: Support VEX/EVEX WIG encoding") and 6fa52824c3 ("x86: Replace VexW=3 with VexWIG") omitted quite a few templates, oddly enough in some cases despite testcases getting added (which then were recorded with wrong expected output). Also adjust VPMAXUB's attributes in the AVX512BW case to match ordering of that of neighboring templates. For the moment SSE2AVX templates are left alone, as it isn't clear whether they were intentionally left untouched by the original commits (the descriptions don't say either way). In this context I question the decision in commit 0375113302 ("x86: Add -mvexwig=[0|1] option to assembler") to move the logic to determine the value of the W bit ahead of the decision whether to use 2-byte VEX: While I can see this as one possible interpretation of -mvexwig=, the other alternative (setting the value of the bit only if it actually exists in the encoding) looks as reasonable to me, and perhaps even more in line with us generally trying to pick the shortest encoding.
2018-11-06 11:39:42 +01:00
vandnpd, 3, 0x6655, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vandnps, 3, 0x55, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vandpd, 3, 0x6654, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vandps, 3, 0x54, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vblendpd, 4, 0x660d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vblendps, 4, 0x660c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vblendvpd, 4, 0x664b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vblendvps, 4, 0x664a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop bogus IgnoreSize from AVX insns
2018-09-13 11:18:52 +02:00
vbroadcastf128, 2, 0x661a, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vbroadcastsd, 2, 0x6619, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegYMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vbroadcastss, 2, 0x6618, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM|RegYMM }
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
vcmp<avx_frel>pd, 3, 0x66c2, 0x<avx_frel:imm>, 1, CpuAVX, Modrm|<avx_frel:comm>|Vex|VexOpcode=0|VexVVVV|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
vcmp<avx_frel>ps, 3, 0xc2, 0x<avx_frel:imm>, 1, CpuAVX, Modrm|<avx_frel:comm>|Vex|VexOpcode=0|VexVVVV|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
vcmp<avx_frel>sd, 3, 0xf2c2, 0x<avx_frel:imm>, 1, CpuAVX, Modrm|<avx_frel:comm>|VexLIG|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vcmp<avx_frel>ss, 3, 0xf3c2, 0x<avx_frel:imm>, 1, CpuAVX, Modrm|<avx_frel:comm>|VexLIG|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vcmppd, 4, 0x66c2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vcmpps, 4, 0xc2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcmpsd, 4, 0xf2c2, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vcmpss, 4, 0xf3c2, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vcomisd, 2, 0x662f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vcomiss, 2, 0x2f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtdq2pd, 2, 0xf3e6, None, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtdq2pd, 2, 0xf3e6, None, 1, CpuAVX, Modrm|Vex256|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vcvtdq2ps, 2, 0x5b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtpd2dq, 2, 0xf2e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Unspecified|BaseIndex, RegXMM }
vcvtpd2dq, 2, 0xf2e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vcvtpd2dqx, 2, 0xf2e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex|RegXMM, RegXMM }
vcvtpd2dqy, 2, 0xf2e6, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex|RegYMM, RegXMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtpd2ps, 2, 0x665a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Unspecified|BaseIndex, RegXMM }
vcvtpd2ps, 2, 0x665a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vcvtpd2psx, 2, 0x665a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex|RegXMM, RegXMM }
vcvtpd2psy, 2, 0x665a, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Unspecified|BaseIndex|RegYMM, RegXMM }
vcvtps2dq, 2, 0x665b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtps2pd, 2, 0x5a, None, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtps2pd, 2, 0x5a, None, 1, CpuAVX, Modrm|Vex256|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vcvtsd2si, 2, 0xf22d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtsd2ss, 3, 0xf25a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
vcvtsi2sd, 3, 0xf22a, None, 1, CpuAVX, Modrm|VexLIG|VexOpcode=0|VexVVVV|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtsi2sd, 3, 0xf22a, None, 1, CpuAVX, Modrm|VexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtsi2ss, 3, 0xf32a, None, 1, CpuAVX, Modrm|VexLIG|VexOpcode=0|VexVVVV|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtsi2ss, 3, 0xf32a, None, 1, CpuAVX, Modrm|VexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtss2sd, 3, 0xf35a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vcvtss2si, 2, 0xf32d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvttpd2dq, 2, 0x66e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Unspecified|BaseIndex, RegXMM }
vcvttpd2dq, 2, 0x66e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vcvttpd2dqx, 2, 0x66e6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTsyntax, { Unspecified|BaseIndex|RegXMM, RegXMM }
vcvttpd2dqy, 2, 0x66e6, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTsyntax, { Unspecified|BaseIndex|RegYMM, RegXMM }
vcvttps2dq, 2, 0xf35b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vcvttsd2si, 2, 0xf22c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { Qword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvttss2si, 2, 0xf32c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { Dword|Unspecified|BaseIndex|RegXMM, Reg32|Reg64 }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vdivpd, 3, 0x665e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vdivps, 3, 0x5e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vdivsd, 3, 0xf25e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vdivss, 3, 0xf35e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vdppd, 4, 0x6641, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vdpps, 4, 0x6640, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop bogus IgnoreSize from AVX insns
2018-09-13 11:18:52 +02:00
vextractf128, 3, 0x6619, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, Unspecified|BaseIndex|RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vextractps, 3, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vextractps, 3, 0x6617, None, 1, CpuAVX|Cpu64, RegMem|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg64 }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vhaddpd, 3, 0x667c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vhaddps, 3, 0xf27c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vhsubpd, 3, 0x667d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vhsubps, 3, 0xf27d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop bogus IgnoreSize from AVX insns
2018-09-13 11:18:52 +02:00
vinsertf128, 4, 0x6618, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vinsertps, 4, 0x6621, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vlddqu, 2, 0xf2f0, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Ymmword|Unspecified|BaseIndex, RegXMM|RegYMM }
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
vldmxcsr, 1, 0xae, 0x2, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmaskmovdqu, 2, 0x66f7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vmaskmovpd, 3, 0x662f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vmaskmovpd, 3, 0x662d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Ymmword|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
vmaskmovps, 3, 0x662e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vmaskmovps, 3, 0x662c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Ymmword|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmaxpd, 3, 0x665f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vmaxps, 3, 0x5f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmaxsd, 3, 0xf25f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vmaxss, 3, 0xf35f, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vminpd, 3, 0x665d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vminps, 3, 0x5d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vminsd, 3, 0xf25d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vminss, 3, 0xf35d, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovapd, 2, 0x6628, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovaps, 2, 0x28, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
gas/testsuite/ 2008-05-30 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/x86-64-avx.s: Add tests for vmovd on 64bit operands. * gas/i386/x86-64-sse2avx.s: Add tests for movd on 64bit operands. * gas/testsuite/gas/i386/x86-64-avx.d: Updated. * gas/testsuite/gas/i386/x86-64-avx-intel.d: Likewise. * gas/testsuite/gas/i386/x86-64-sse2avx.d: Likewise. opcodes/ 2008-05-30 H.J. Lu <hongjiu.lu@intel.com> * i386-opc.tbl: Add vmovd with 64bit operand. * i386-tbl.h: Regenerated.
2008-05-30 21:49:18 +02:00
// vmovd really shouldn't allow for 64bit operand (vmovq is the right
// mnemonic for copying between Reg64/Mem64 and RegXMM, as is mandated
// by Intel AVX spec). To avoid extra template in gcc x86 backend and
// support assembler for AMD64, we accept 64bit operand on vmovd so
// that we can use one template for both SSE and AVX instructions.
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vmovd, 2, 0x666e, None, 1, CpuAVX, D|Modrm|Vex=1|VexOpcode=0|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vmovd, 2, 0x667e, None, 1, CpuAVX|Cpu64, D|RegMem|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { RegXMM, Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmovddup, 2, 0xf212, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovddup, 2, 0xf212, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM }
vmovdqa, 2, 0x666f, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovdqu, 2, 0xf36f, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovhlps, 3, 0x12, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
vmovhpd, 3, 0x6616, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovhpd, 2, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vmovhps, 3, 0x16, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovhps, 2, 0x17, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vmovlhps, 3, 0x16, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
vmovlpd, 3, 0x6612, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovlpd, 2, 0x6613, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vmovlps, 3, 0x12, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovlps, 2, 0x13, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vmovmskpd, 2, 0x6650, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { RegXMM|RegYMM, Reg32|Reg64 }
vmovmskps, 2, 0x50, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { RegXMM|RegYMM, Reg32|Reg64 }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovntdq, 2, 0x66e7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vmovntdqa, 2, 0x662a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex, RegXMM }
vmovntpd, 2, 0x662b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vmovntps, 2, 0x2b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmovq, 2, 0xf37e, None, 1, CpuAVX, Load|Modrm|Vex=1|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vmovq, 2, 0x66d6, None, 1, CpuAVX, Modrm|Vex=1|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM }
x86: Set Vex=1 on VEX.128 only vmovd and vmovq AVX "VMOVD xmm1, r32/m32", "VMOVD r32/m32, xmm2", "VMOVQ xmm1, r64/m64" and "VMOVD r64/m64, xmm2" can only be encoded with VEX.128. Set Vex=1 on VEX.128 only vmovd and vmovq. gas/ PR gas/23665 * testsuite/gas/i386/avx-scalar.s: Remove vmovq and vmovd tests. * testsuite/gas/i386/x86-64-avx-scalar.s: Likewise. * testsuite/gas/i386/avx-scalar-intel.d: Updated. * testsuite/gas/i386/avx-scalar.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar.d: Likewise. * testsuite/gas/i386/i386.exp: Run avx-scalar2 and x86-64-avx-scalar2. * testsuite/gas/i386/avx-scalar-2.d: New file. * testsuite/gas/i386/avx-scalar-2.s: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-2.d: Likewise. * testsuite/gas/i386/x86-64-avx-scalar-2.s: Likewise. opcodes/ PR gas/23665 * i386-dis.c (vex_len_table): Update VEX_LEN_0F6E_P_2 and VEX_LEN_0F7E_P_2 entries. * i386-opc.tbl: Set Vex=1 on VEX.128 only vmovd and vmovq. * i386-tbl.h: Regenerated.
2018-09-17 18:31:07 +02:00
vmovq, 2, 0x666e, None, 1, CpuAVX|Cpu64, D|Modrm|Vex=1|VexOpcode=0|VexW=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64|Unspecified|BaseIndex, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovsd, 2, 0xf210, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vmovsd, 3, 0xf210, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovshdup, 2, 0xf316, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovsldup, 2, 0xf312, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovss, 2, 0xf310, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vmovss, 3, 0xf310, None, 1, CpuAVX, D|Modrm|Vex=3|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vmovupd, 2, 0x6610, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmovups, 2, 0x10, None, 1, CpuAVX, D|Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vmpsadbw, 4, 0x6642, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vmulpd, 3, 0x6659, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vmulps, 3, 0x59, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmulsd, 3, 0xf259, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vmulss, 3, 0xf359, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vorpd, 3, 0x6656, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vorps, 3, 0x56, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpabsb, 2, 0x661c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM }
vpabsd, 2, 0x661e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM }
vpabsw, 2, 0x661d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM }
vpackssdw, 3, 0x666b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpacksswb, 3, 0x6663, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpackusdw, 3, 0x662b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpackuswb, 3, 0x6667, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpaddsb, 3, 0x66ec, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddsw, 3, 0x66ed, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddb, 3, 0x66fc, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddd, 3, 0x66fe, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddq, 3, 0x66d4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddw, 3, 0x66fd, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddusb, 3, 0x66dc, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpaddusw, 3, 0x66dd, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpalignr, 4, 0x660f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpand, 3, 0x66db, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpandn, 3, 0x66df, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpavgb, 3, 0x66e0, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpavgw, 3, 0x66e3, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX insns
2018-09-13 11:18:52 +02:00
vpblendvb, 4, 0x664c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpblendw, 4, 0x660e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpcmpeqb, 3, 0x6674, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpcmpeqd, 3, 0x6676, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpeqq, 3, 0x6629, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpcmpeqw, 3, 0x6675, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpcmpestri, 3, 0x6661, None, 1, CpuAVX|CpuNo64, Modrm|Vex|VexOpcode=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vpcmpestri, 3, 0x6661, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpcmpestrm, 3, 0x6660, None, 1, CpuAVX|CpuNo64, Modrm|Vex|VexOpcode=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vpcmpestrm, 3, 0x6660, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Xmmword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpgtb, 3, 0x6664, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpcmpgtd, 3, 0x6666, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpcmpgtq, 3, 0x6637, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpcmpgtw, 3, 0x6665, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpcmpistri, 3, 0x6663, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
vpcmpistrm, 3, 0x6662, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX insns
2018-09-13 11:18:52 +02:00
vperm2f128, 4, 0x6606, None, 1, CpuAVX, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vpermilpd, 3, 0x660d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpermilpd, 3, 0x6605, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vpermilps, 3, 0x660c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpermilps, 3, 0x6604, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vpextrb, 3, 0x6614, None, 1, CpuAVX, RegMem|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpextrb, 3, 0x6614, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Byte|Unspecified|BaseIndex }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpextrd, 3, 0x6616, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: VPEXTRQ/VPINSRQ are unavailable outside of 64-bit mode The AVX512DQ patterns lacking a Cpu64 attribute made the memory operand forms accepted even outside of 64-bit mode, and this even without any {evex} pseudo-prefix (otherwise one could argue that this is an attempt to follow one possible, albeit somewhat odd, interpretation of the SDM wording to this effect). For consistency between the various involved templates drop the * (now) unnecessary IgnoreSize attributes * unnecessary (due to VexW1) Size64 attributes from VEX encoded forms * redundant (with Reg64) Qword operand attributes uniformly.
2020-01-16 10:05:35 +01:00
vpextrq, 3, 0x6616, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg64|Unspecified|BaseIndex }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vpextrw, 3, 0x66c5, None, 1, CpuAVX, Load|Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
vpextrw, 3, 0x6615, None, 1, CpuAVX, RegMem|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpextrw, 3, 0x6615, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Word|Unspecified|BaseIndex }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vphaddd, 3, 0x6602, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vphaddsw, 3, 0x6603, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vphaddw, 3, 0x6601, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vphminposuw, 2, 0x6641, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM }
vphsubd, 3, 0x6606, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vphsubsw, 3, 0x6607, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vphsubw, 3, 0x6605, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vpinsrb, 4, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpinsrb, 4, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Byte|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpinsrd, 4, 0x6622, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexVVVV=1|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: VPEXTRQ/VPINSRQ are unavailable outside of 64-bit mode The AVX512DQ patterns lacking a Cpu64 attribute made the memory operand forms accepted even outside of 64-bit mode, and this even without any {evex} pseudo-prefix (otherwise one could argue that this is an attempt to follow one possible, albeit somewhat odd, interpretation of the SDM wording to this effect). For consistency between the various involved templates drop the * (now) unnecessary IgnoreSize attributes * unnecessary (due to VexW1) Size64 attributes from VEX encoded forms * redundant (with Reg64) Qword operand attributes uniformly.
2020-01-16 10:05:35 +01:00
vpinsrq, 4, 0x6622, None, 1, CpuAVX|Cpu64, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: replace NoRex64 on VEX-encoded insns When the template specifies any of the possible VexW settings, we can use this instead of a separate NoRex64 to suppress the setting of REX_W. Note that this ends up addressing an inconsistency between VEX- and EVEX-encoded VEXTRACTPS, VPEXTR{B,W}, and VPINSR{B,W} - while the former avoided setting VEX.W, the latter pointlessly set EVEX.W when there is a 64-bit GPR operand. Adjust the testcase to cover both cases. Convert VexW= to their respective VexW* on lines touched anyway.
2020-03-06 08:53:18 +01:00
vpinsrw, 4, 0x66c4, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpinsrw, 4, 0x66c4, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Word|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaddubsw, 3, 0x6604, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmaddwd, 3, 0x66f5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaxsb, 3, 0x663c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmaxsd, 3, 0x663d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmaxsw, 3, 0x66ee, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmaxub, 3, 0x66de, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaxud, 3, 0x663f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmaxuw, 3, 0x663e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpminsb, 3, 0x6638, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpminsd, 3, 0x6639, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpminsw, 3, 0x66ea, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpminub, 3, 0x66da, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpminud, 3, 0x663b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpminuw, 3, 0x663a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovmskb, 2, 0x66d7, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { RegXMM, Reg32|Reg64 }
vpmovsxbd, 2, 0x6621, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovsxbq, 2, 0x6622, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovsxbw, 2, 0x6620, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovsxdq, 2, 0x6625, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovsxwd, 2, 0x6623, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovsxwq, 2, 0x6624, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxbd, 2, 0x6631, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxbq, 2, 0x6632, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxbw, 2, 0x6630, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxdq, 2, 0x6635, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxwd, 2, 0x6633, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vpmovzxwq, 2, 0x6634, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmuldq, 3, 0x6628, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmulhrsw, 3, 0x660b, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmulhuw, 3, 0x66e4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmulhw, 3, 0x66e5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmulld, 3, 0x6640, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmullw, 3, 0x66d5, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpmuludq, 3, 0x66f4, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpor, 3, 0x66eb, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpsadbw, 3, 0x66f6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpshufb, 3, 0x6600, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpshufd, 3, 0x6670, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
vpshufhw, 3, 0xf370, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
vpshuflw, 3, 0xf270, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
vpsignb, 3, 0x6608, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsignd, 3, 0x660a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsignw, 3, 0x6609, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpslld, 3, 0x6672, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpslld, 3, 0x66f2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpslldq, 3, 0x6673, 0x7, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsllq, 3, 0x6673, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsllq, 3, 0x66f3, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsllw, 3, 0x6671, 0x6, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsllw, 3, 0x66f1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsrad, 3, 0x6672, 0x4, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsrad, 3, 0x66e2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsraw, 3, 0x6671, 0x4, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsraw, 3, 0x66e1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsrld, 3, 0x6672, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsrld, 3, 0x66d2, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsrldq, 3, 0x6673, 0x3, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsrlq, 3, 0x6673, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsrlq, 3, 0x66d3, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsrlw, 3, 0x6671, 0x2, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM }
vpsrlw, 3, 0x66d1, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubb, 3, 0x66f8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubd, 3, 0x66fa, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubq, 3, 0x66fb, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubsb, 3, 0x66e8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubsw, 3, 0x66e9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubusb, 3, 0x66d8, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubusw, 3, 0x66d9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpsubw, 3, 0x66f9, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vptest, 2, 0x6617, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vpunpckhbw, 3, 0x6668, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpckhdq, 3, 0x666a, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpckhqdq, 3, 0x666d, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpckhwd, 3, 0x6669, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpcklbw, 3, 0x6660, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpckldq, 3, 0x6662, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpcklqdq, 3, 0x666c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpunpcklwd, 3, 0x6661, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpxor, 3, 0x66ef, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vrcpps, 2, 0x53, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrcpss, 3, 0xf353, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vroundpd, 3, 0x6609, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vroundps, 3, 0x6608, None, 1, CpuAVX, Modrm|Vex|VexOpcode=2|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vroundsd, 4, 0x660b, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=2|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vroundss, 4, 0x660a, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=2|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vrsqrtps, 2, 0x52, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrsqrtss, 3, 0xf352, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vshufpd, 4, 0x66c6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vshufps, 4, 0xc6, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vsqrtpd, 2, 0x6651, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vsqrtps, 2, 0x51, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsqrtsd, 3, 0xf251, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vsqrtss, 3, 0xf351, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
vstmxcsr, 1, 0xae, 0x3, 1, CpuAVX, Modrm|Vex128|VexOpcode=0|VexWIG|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vsubpd, 3, 0x665c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vsubps, 3, 0x5c, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsubsd, 3, 0xf25c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vsubss, 3, 0xf35c, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexVVVV|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vtestpd, 2, 0x660f, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vtestps, 2, 0x660e, None, 1, CpuAVX, Modrm|Vex|VexOpcode=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vucomisd, 2, 0x662e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vucomiss, 2, 0x2e, None, 1, CpuAVX, Modrm|Vex=3|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vunpckhpd, 3, 0x6615, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vunpckhps, 3, 0x15, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vunpcklpd, 3, 0x6614, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vunpcklps, 3, 0x14, None, 1, CpuAVX, Modrm|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vxorpd, 3, 0x6657, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vxorps, 3, 0x57, None, 1, CpuAVX, Modrm|C|Vex|VexOpcode=0|VexVVVV=1|VexWIG|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vzeroall, 0, 0x77, None, 1, CpuAVX, Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
vzeroupper, 0, 0x77, None, 1, CpuAVX, Vex|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
// 256bit integer AVX2 instructions.
x86: add more VexWIG Commits 6865c0435a ("x86: Support VEX/EVEX WIG encoding") and 6fa52824c3 ("x86: Replace VexW=3 with VexWIG") omitted quite a few templates, oddly enough in some cases despite testcases getting added (which then were recorded with wrong expected output). Also adjust VPMAXUB's attributes in the AVX512BW case to match ordering of that of neighboring templates. For the moment SSE2AVX templates are left alone, as it isn't clear whether they were intentionally left untouched by the original commits (the descriptions don't say either way). In this context I question the decision in commit 0375113302 ("x86: Add -mvexwig=[0|1] option to assembler") to move the logic to determine the value of the W bit ahead of the decision whether to use 2-byte VEX: While I can see this as one possible interpretation of -mvexwig=, the other alternative (setting the value of the bit only if it actually exists in the encoding) looks as reasonable to me, and perhaps even more in line with us generally trying to pick the shortest encoding.
2018-11-06 11:39:42 +01:00
vmovntdqa, 2, 0x662a, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Ymmword|Unspecified|BaseIndex, RegYMM }
vmpsadbw, 4, 0x6642, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpabsb, 2, 0x661c, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM }
vpabsd, 2, 0x661e, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM }
vpabsw, 2, 0x661d, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM }
vpackssdw, 3, 0x666b, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpacksswb, 3, 0x6663, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpackusdw, 3, 0x662b, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpackuswb, 3, 0x6667, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpaddsb, 3, 0x66ec, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddsw, 3, 0x66ed, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddb, 3, 0x66fc, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddd, 3, 0x66fe, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddq, 3, 0x66d4, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddw, 3, 0x66fd, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddusb, 3, 0x66dc, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpaddusw, 3, 0x66dd, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpalignr, 4, 0x660f, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpand, 3, 0x66db, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpandn, 3, 0x66df, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpavgb, 3, 0x66e0, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpavgw, 3, 0x66e3, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: drop bogus IgnoreSize from AVX2 insns
2018-09-13 11:19:21 +02:00
vpblendvb, 4, 0x664c, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexSources=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpblendw, 4, 0x660e, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpcmpeqb, 3, 0x6674, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpcmpeqd, 3, 0x6676, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpeqq, 3, 0x6629, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpcmpeqw, 3, 0x6675, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpgtb, 3, 0x6664, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpcmpgtd, 3, 0x6666, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpcmpgtq, 3, 0x6637, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpcmpgtw, 3, 0x6665, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vphaddd, 3, 0x6602, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vphaddsw, 3, 0x6603, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vphaddw, 3, 0x6601, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: add more VexWIG Commits 6865c0435a ("x86: Support VEX/EVEX WIG encoding") and 6fa52824c3 ("x86: Replace VexW=3 with VexWIG") omitted quite a few templates, oddly enough in some cases despite testcases getting added (which then were recorded with wrong expected output). Also adjust VPMAXUB's attributes in the AVX512BW case to match ordering of that of neighboring templates. For the moment SSE2AVX templates are left alone, as it isn't clear whether they were intentionally left untouched by the original commits (the descriptions don't say either way). In this context I question the decision in commit 0375113302 ("x86: Add -mvexwig=[0|1] option to assembler") to move the logic to determine the value of the W bit ahead of the decision whether to use 2-byte VEX: While I can see this as one possible interpretation of -mvexwig=, the other alternative (setting the value of the bit only if it actually exists in the encoding) looks as reasonable to me, and perhaps even more in line with us generally trying to pick the shortest encoding.
2018-11-06 11:39:42 +01:00
vphsubd, 3, 0x6606, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vphsubsw, 3, 0x6607, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vphsubw, 3, 0x6605, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaddubsw, 3, 0x6604, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmaddwd, 3, 0x66f5, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaxsb, 3, 0x663c, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmaxsd, 3, 0x663d, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmaxsw, 3, 0x66ee, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmaxub, 3, 0x66de, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaxud, 3, 0x663f, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmaxuw, 3, 0x663e, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpminsb, 3, 0x6638, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpminsd, 3, 0x6639, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpminsw, 3, 0x66ea, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpminub, 3, 0x66da, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpminud, 3, 0x663b, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpminuw, 3, 0x663a, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovmskb, 2, 0x66d7, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { RegYMM, Reg32|Reg64 }
vpmovsxbd, 2, 0x6621, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovsxbq, 2, 0x6622, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxbw, 2, 0x6620, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovsxdq, 2, 0x6625, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovsxwd, 2, 0x6623, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxwq, 2, 0x6624, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovzxbd, 2, 0x6631, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovzxbq, 2, 0x6632, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovzxbw, 2, 0x6630, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovzxdq, 2, 0x6635, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
vpmovzxwd, 2, 0x6633, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovzxwq, 2, 0x6634, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmuldq, 3, 0x6628, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmulhrsw, 3, 0x660b, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmulhuw, 3, 0x66e4, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmulhw, 3, 0x66e5, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmulld, 3, 0x6640, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpmullw, 3, 0x66d5, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpmuludq, 3, 0x66f4, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpor, 3, 0x66eb, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpsadbw, 3, 0x66f6, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpshufb, 3, 0x6600, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpshufd, 3, 0x6670, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM }
vpshufhw, 3, 0xf370, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM }
vpshuflw, 3, 0xf270, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM }
vpsignb, 3, 0x6608, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsignd, 3, 0x660a, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsignw, 3, 0x6609, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpslld, 3, 0x6672, 0x6, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpslld, 3, 0x66f2, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpslldq, 3, 0x6673, 0x7, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsllq, 3, 0x6673, 0x6, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsllq, 3, 0x66f3, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsllw, 3, 0x6671, 0x6, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsllw, 3, 0x66f1, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsrad, 3, 0x6672, 0x4, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsrad, 3, 0x66e2, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsraw, 3, 0x6671, 0x4, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsraw, 3, 0x66e1, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsrld, 3, 0x6672, 0x2, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsrld, 3, 0x66d2, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsrldq, 3, 0x6673, 0x3, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsrlq, 3, 0x6673, 0x2, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsrlq, 3, 0x66d3, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsrlw, 3, 0x6671, 0x2, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegYMM }
vpsrlw, 3, 0x66d1, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
vpsubb, 3, 0x66f8, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubd, 3, 0x66fa, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubq, 3, 0x66fb, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubsb, 3, 0x66e8, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubsw, 3, 0x66e9, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubusb, 3, 0x66d8, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubusw, 3, 0x66d9, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpsubw, 3, 0x66f9, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpckhbw, 3, 0x6668, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpckhdq, 3, 0x666a, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpckhqdq, 3, 0x666d, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpckhwd, 3, 0x6669, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpcklbw, 3, 0x6660, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpckldq, 3, 0x6662, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpcklqdq, 3, 0x666c, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpunpcklwd, 3, 0x6661, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86-64: optimize certain commutative VEX-encoded insns When they're in the 0F opcode space, swapping their source operands may allow switching from 3-byte to 2-byte VEX prefix encoding. Note that NaN behavior precludes us doing so for many packed and scalar floating point insns; such an optimization would need to be done by the compiler instead in this case, when it knows that NaN-s have undefined behavior anyway. While for explicitly specified AVX/AVX2 insns the optimization (for now at least) gets done only for -O2 and -Os, it is utilized by default in SSE2AVX mode, as there we're re-writing the programmer's specified insns anyway. Rather than introducing a new attribute flag, the change re-uses one which so far was meaningful only for EVEX-encoded insns.
2019-07-01 08:33:56 +02:00
vpxor, 3, 0x66ef, None, 1, CpuAVX2, Modrm|C|Vex=2|VexOpcode=0|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
// New AVX2 instructions.
x86: drop bogus IgnoreSize from AVX2 insns
2018-09-13 11:19:21 +02:00
vbroadcasti128, 2, 0x665A, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex, RegYMM }
vbroadcastsd, 2, 0x6619, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegYMM }
vbroadcastss, 2, 0x6618, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|RegYMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vpblendd, 4, 0x6602, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpbroadcastb, 2, 0x6678, None, 1, CpuAVX2, Modrm|Vex=1|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Byte|Unspecified|BaseIndex|RegXMM, RegXMM|RegYMM }
vpbroadcastd, 2, 0x6658, None, 1, CpuAVX2, Modrm|Vex=1|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM|RegYMM }
vpbroadcastq, 2, 0x6659, None, 1, CpuAVX2, Modrm|Vex=1|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM|RegYMM }
vpbroadcastw, 2, 0x6679, None, 1, CpuAVX2, Modrm|Vex=1|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Word|Unspecified|BaseIndex|RegXMM, RegXMM|RegYMM }
x86: drop bogus IgnoreSize from AVX2 insns
2018-09-13 11:19:21 +02:00
vperm2i128, 4, 0x6646, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpermd, 3, 0x6636, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpermpd, 3, 0x6601, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM }
vpermps, 3, 0x6616, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpermq, 3, 0x6600, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM }
vextracti128, 3, 0x6639, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, Unspecified|BaseIndex|RegXMM }
vinserti128, 4, 0x6638, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegYMM, RegYMM }
x86: fold certain AVX and AVX2 templates Just like for instructions in GPRs, there's no need to have separate templates for otherwise identical insns acting on XMM or YMM registers (or memory of the same size).
2017-12-18 09:37:15 +01:00
vpmaskmovd, 3, 0x668e, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vpmaskmovd, 3, 0x668c, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Ymmword|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
vpmaskmovq, 3, 0x668e, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, RegXMM|RegYMM, Xmmword|Ymmword|Unspecified|BaseIndex }
vpmaskmovq, 3, 0x668c, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Ymmword|Unspecified|BaseIndex, RegXMM|RegYMM, RegXMM|RegYMM }
vpsllvd, 3, 0x6647, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpsllvq, 3, 0x6647, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpsravd, 3, 0x6646, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpsrlvd, 3, 0x6645, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpsrlvq, 3, 0x6645, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
// AVX gather instructions
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherdpd, 3, 0x6692, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Qword|Unspecified|BaseIndex, RegXMM }
vgatherdpd, 3, 0x6692, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegYMM, Qword|Unspecified|BaseIndex, RegYMM }
vgatherdps, 3, 0x6692, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vgatherdps, 3, 0x6692, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegYMM, Dword|Unspecified|BaseIndex, RegYMM }
vgatherqpd, 3, 0x6693, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Qword|Unspecified|BaseIndex, RegXMM }
vgatherqpd, 3, 0x6693, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegYMM, Qword|Unspecified|BaseIndex, RegYMM }
vgatherqps, 3, 0x6693, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vgatherqps, 3, 0x6693, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vpgatherdd, 3, 0x6690, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vpgatherdd, 3, 0x6690, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegYMM, Dword|Unspecified|BaseIndex, RegYMM }
vpgatherdq, 3, 0x6690, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Qword|Unspecified|BaseIndex, RegXMM }
vpgatherdq, 3, 0x6690, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegYMM, Qword|Unspecified|BaseIndex, RegYMM }
vpgatherqd, 3, 0x6691, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vpgatherqd, 3, 0x6691, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegXMM, Dword|Unspecified|BaseIndex, RegXMM }
vpgatherqq, 3, 0x6691, None, 1, CpuAVX2, Modrm|Vex|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=1, { RegXMM, Qword|Unspecified|BaseIndex, RegXMM }
vpgatherqq, 3, 0x6691, None, 1, CpuAVX2, Modrm|Vex=2|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VecSIB=2, { RegYMM, Qword|Unspecified|BaseIndex, RegYMM }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
gas/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * config/tc-i386.c (CPU_FLAGS_AES_MATCH): New. (CPU_FLAGS_AVX_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Updated. (cpu_flags_match): Likewise. gas/testsuite/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * gas/i386/avx.s: Add AES + AVX tests. * gas/i386/arch-10.s: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/i386.exp: Run arch-avx-1, arch-avx-1-1 and arch-avx-1-2. * gas/i386/arch-avx-1.d: New. * gas/i386/arch-avx-1.s: Likewise. * gas/i386/arch-avx-1-1.l: Likewise. * gas/i386/arch-avx-1-1.s: Likewise. * gas/i386/arch-avx-1-2.l: Likewise. * gas/i386/arch-avx-1-2.s: Likewise. opcodes/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * i386-dis.c (PREFIX_VEX_38DB): New. (PREFIX_VEX_38DC): Likewise. (PREFIX_VEX_38DD): Likewise. (PREFIX_VEX_38DE): Likewise. (PREFIX_VEX_38DF): Likewise. (PREFIX_VEX_3ADF): Likewise. (VEX_LEN_38DB_P_2): Likewise. (VEX_LEN_38DC_P_2): Likewise. (VEX_LEN_38DD_P_2): Likewise. (VEX_LEN_38DE_P_2): Likewise. (VEX_LEN_38DF_P_2): Likewise. (VEX_LEN_3ADF_P_2): Likewise. (PREFIX_VEX_3A04): Updated. (VEX_LEN_3A06_P_2): Likewise. (prefix_table): Add PREFIX_VEX_38DB, PREFIX_VEX_38DC, PREFIX_VEX_38DD, PREFIX_VEX_38DE and PREFIX_VEX_3ADF. (x86_64_table): Likewise. (vex_len_table): Add VEX_LEN_38DB_P_2, VEX_LEN_38DC_P_2, VEX_LEN_38DD_P_2, VEX_LEN_38DE_P_2, VEX_LEN_38DF_P_2 and VEX_LEN_3ADF_P_2. * i386-opc.tbl: Add AES + AVX instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-08-20 20:38:40 +02:00
// AES + AVX
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vaesdec, 3, 0x66de, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vaesdeclast, 3, 0x66df, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vaesenc, 3, 0x66dc, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vaesenclast, 3, 0x66dd, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vaesimc, 2, 0x66db, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegXMM }
vaeskeygenassist, 3, 0x66df, None, 1, CpuAVX|CpuAES, Modrm|Vex|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM }
gas/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * config/tc-i386.c (CPU_FLAGS_AES_MATCH): New. (CPU_FLAGS_AVX_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Updated. (cpu_flags_match): Likewise. gas/testsuite/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * gas/i386/avx.s: Add AES + AVX tests. * gas/i386/arch-10.s: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/i386.exp: Run arch-avx-1, arch-avx-1-1 and arch-avx-1-2. * gas/i386/arch-avx-1.d: New. * gas/i386/arch-avx-1.s: Likewise. * gas/i386/arch-avx-1-1.l: Likewise. * gas/i386/arch-avx-1-1.s: Likewise. * gas/i386/arch-avx-1-2.l: Likewise. * gas/i386/arch-avx-1-2.s: Likewise. opcodes/ 2008-08-20 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (August, 2008) * i386-dis.c (PREFIX_VEX_38DB): New. (PREFIX_VEX_38DC): Likewise. (PREFIX_VEX_38DD): Likewise. (PREFIX_VEX_38DE): Likewise. (PREFIX_VEX_38DF): Likewise. (PREFIX_VEX_3ADF): Likewise. (VEX_LEN_38DB_P_2): Likewise. (VEX_LEN_38DC_P_2): Likewise. (VEX_LEN_38DD_P_2): Likewise. (VEX_LEN_38DE_P_2): Likewise. (VEX_LEN_38DF_P_2): Likewise. (VEX_LEN_3ADF_P_2): Likewise. (PREFIX_VEX_3A04): Updated. (VEX_LEN_3A06_P_2): Likewise. (prefix_table): Add PREFIX_VEX_38DB, PREFIX_VEX_38DC, PREFIX_VEX_38DD, PREFIX_VEX_38DE and PREFIX_VEX_3ADF. (x86_64_table): Likewise. (vex_len_table): Add VEX_LEN_38DB_P_2, VEX_LEN_38DC_P_2, VEX_LEN_38DD_P_2, VEX_LEN_38DE_P_2, VEX_LEN_38DF_P_2 and VEX_LEN_3ADF_P_2. * i386-opc.tbl: Add AES + AVX instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-08-20 20:38:40 +02:00
gas/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * config/tc-i386.c (CPU_FLAGS_PCLMUL_MATCH): New. (CPU_FLAGS_AVX_MATCH): Updated. (CPU_FLAGS_32BIT_MATCH): Likewise. (cpu_flags_match): Likewise. gas/testsuite/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * gas/i386/arch-avx-1-3.l: New. * gas/i386/arch-avx-1-3.s: Likewise. * gas/i386/arch-avx-1-4.l: Likewise. * gas/i386/arch-avx-1-4.s: Likewise. * gas/i386/arch-avx-1-5.l: Likewise. * gas/i386/arch-avx-1-5.s: Likewise. * gas/i386/arch-avx-1-6.l: Likewise. * gas/i386/arch-avx-1-6.s: Likewise. * gas/i386/arch-10.s: Add vpclmul instructions. * gas/i386/arch-avx-1.s: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/sse2avx.s: Add pclmul instructions. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-avx-1.d: Likewise. * gas/i386/arch-avx-1-1.l: Likewise. * gas/i386/arch-avx-1-2.l: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/i386.exp: Run arch-avx-1-3, arch-avx-1-4, arch-avx-1-5 and arch-avx-1-6. opcodes/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * i386-dis.c (PREFIX_VEX_3A44): New. (VEX_LEN_3A44_P_2): Likewise. (PREFIX_VEX_3A48): Updated. (VEX_LEN_3A4C_P_2): Likewise. (prefix_table): Add PREFIX_VEX_3A44. (vex_table): Likewise. (vex_len_table): Add VEX_LEN_3A44_P_2. * i386-opc.tbl: Add PCLMUL + AVX instructions. * i386-tbl.h: Regenerated.
2009-02-04 17:03:31 +01:00
// PCLMUL + AVX
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpclmulqdq, 4, 0x6644, None, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpclmullqlqdq, 3, 0x6644, 0x0, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpclmulhqlqdq, 3, 0x6644, 0x1, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpclmullqhqdq, 3, 0x6644, 0x10, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vpclmulhqhqdq, 3, 0x6644, 0x11, 1, CpuAVX|CpuPCLMUL, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
gas/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * config/tc-i386.c (CPU_FLAGS_PCLMUL_MATCH): New. (CPU_FLAGS_AVX_MATCH): Updated. (CPU_FLAGS_32BIT_MATCH): Likewise. (cpu_flags_match): Likewise. gas/testsuite/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * gas/i386/arch-avx-1-3.l: New. * gas/i386/arch-avx-1-3.s: Likewise. * gas/i386/arch-avx-1-4.l: Likewise. * gas/i386/arch-avx-1-4.s: Likewise. * gas/i386/arch-avx-1-5.l: Likewise. * gas/i386/arch-avx-1-5.s: Likewise. * gas/i386/arch-avx-1-6.l: Likewise. * gas/i386/arch-avx-1-6.s: Likewise. * gas/i386/arch-10.s: Add vpclmul instructions. * gas/i386/arch-avx-1.s: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/sse2avx.s: Add pclmul instructions. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-avx-1.d: Likewise. * gas/i386/arch-avx-1-1.l: Likewise. * gas/i386/arch-avx-1-2.l: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/i386.exp: Run arch-avx-1-3, arch-avx-1-4, arch-avx-1-5 and arch-avx-1-6. opcodes/ 2009-02-04 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (January, 2009) * i386-dis.c (PREFIX_VEX_3A44): New. (VEX_LEN_3A44_P_2): Likewise. (PREFIX_VEX_3A48): Updated. (VEX_LEN_3A4C_P_2): Likewise. (prefix_table): Add PREFIX_VEX_3A44. (vex_table): Likewise. (vex_len_table): Add VEX_LEN_3A44_P_2. * i386-opc.tbl: Add PCLMUL + AVX instructions. * i386-tbl.h: Regenerated.
2009-02-04 17:03:31 +01:00
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
// GFNI + AVX
x86: fold VEX-encoded GFNI templates
2018-03-08 08:57:19 +01:00
vgf2p8affineinvqb, 4, 0x66cf, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vgf2p8affineqb, 4, 0x66ce, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vgf2p8mulb, 3, 0x66cf, None, 1, CpuAVX|CpuGFNI, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
Support AVX Programming Reference (June, 2010) gas/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * config/tc-i386.c (cpu_arch): Add .xsaveopt, .fsgsbase, .rdrnd and .f16c. * doc/c-i386.texi: Document xsaveopt, fsgsbase, rdrnd and f16c. gas/testsuite/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * gas/i386/arch-10.s: Add xsaveopt. * gas/i386/x86-64-arch-2.s: Likwise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/f16c-intel.d: New. * gas/i386/f16c.d: Likewise. * gas/i386/f16c.s: Likewise. * gas/i386/fsgs-intel.d: Likewise. * gas/i386/fsgs.d: Likewise. * gas/i386/fsgs.s: Likewise. * gas/i386/rdrnd-intel.d: Likewise. * gas/i386/rdrnd.d: Likewise. * gas/i386/rdrnd.s: Likewise. * gas/i386/x86-64-f16c-intel.d: Likewise. * gas/i386/x86-64-f16c.d: Likewise. * gas/i386/x86-64-f16c.s: Likewise. * gas/i386/x86-64-fsgs-intel.d: Likewise. * gas/i386/x86-64-fsgs.d: Likewise. * gas/i386/x86-64-fsgs.s: Likewise. * gas/i386/x86-64-rdrnd-intel.d: Likewise. * gas/i386/x86-64-rdrnd.d: Likewise. * gas/i386/x86-64-rdrnd.s: Likewise. * gas/i386/i386.exp: Run f16c, f16c-intel, fsgs, fsgs-intel, rdrnd, rdrnd-intel, x86-64-f16c, x86-64-f16c-intel, x86-64-fsgs, x86-64-fsgs-intel, x86-64-rdrnd, x86-64-rdrnd-intel. * gas/i386/x86-64-xsave.s: Add tests for xsaveopt64. * gas/i386/x86-64-xsave-intel.d: Updated. * gas/i386/x86-64-xsave.d: Likewise. opcodes/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * i386-dis.c (PREFIX_0FAE_REG_0): New. (PREFIX_0FAE_REG_1): Likewise. (PREFIX_0FAE_REG_2): Likewise. (PREFIX_0FAE_REG_3): Likewise. (PREFIX_VEX_3813): Likewise. (PREFIX_VEX_3A1D): Likewise. (prefix_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3, PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (vex_table): Add PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (mod_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3 xsaveopt and rdrnd. * i386-gen.c (cpu_flag_init): Add CPU_XSAVEOPT_FLAGS, CPU_FSGSBASE_FLAGS, CPU_RDRND_FLAGS and CPU_F16C_FLAGS. (cpu_flags): Add CpuXsaveopt, CpuFSGSBase, CpuRdRnd and CpuF16C. * i386-opc.h (CpuXsaveopt): New. (CpuFSGSBase):Likewise. (CpuRdRnd): Likewise. (CpuF16C): Likewise. (i386_cpu_flags): Add cpuxsaveopt, cpufsgsbase, cpurdrnd and cpuf16c. * i386-opc.tbl: Add xsaveopt, rdfsbase, rdgsbase, rdrnd, wrfsbase, wrgsbase, vcvtph2ps and vcvtps2ph.
2010-07-01 23:55:02 +02:00
// FSGSBASE, RDRND and F16C
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
rdfsbase, 1, 0xf30fae, 0x0, 2, CpuFSGSBase, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Reg64 }
rdgsbase, 1, 0xf30fae, 0x1, 2, CpuFSGSBase, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Reg64 }
Replace rdrnd with rdrand. gas/testsuite/ 2010-07-05 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * gas/i386/rdrnd.s: Replace rdrnd with rdrand. * gas/i386/rdrnd-intel.d: Likewise. * gas/i386/rdrnd.d: Likewise. * gas/i386/x86-64-rdrnd-intel.d: Likewise. * gas/i386/x86-64-rdrnd.d: Likewise. * gas/i386/x86-64-rdrnd.s: Likewise. opcodes/ 2010-07-05 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * i386-dis.c (mod_table): Replace rdrnd with rdrand. * i386-opc.tbl: Likewise. * i386-tbl.h: Regenerated.
2010-07-05 19:14:22 +02:00
rdrand, 1, 0xfc7, 0x6, 2, CpuRdRnd, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32|Reg64 }
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
wrfsbase, 1, 0xf30fae, 0x2, 2, CpuFSGSBase, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Reg64 }
wrgsbase, 1, 0xf30fae, 0x3, 2, CpuFSGSBase, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtph2ps, 2, 0x6613, None, 1, CpuF16C, Modrm|Vex|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
vcvtph2ps, 2, 0x6613, None, 1, CpuF16C, Modrm|Vex=2|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtps2ph, 3, 0x661d, None, 1, CpuF16C, Modrm|Vex|VexOpcode=2|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Qword|Unspecified|BaseIndex|RegXMM }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
vcvtps2ph, 3, 0x661d, None, 1, CpuF16C, Modrm|Vex=2|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, Unspecified|BaseIndex|RegXMM }
Support AVX Programming Reference (June, 2010) gas/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * config/tc-i386.c (cpu_arch): Add .xsaveopt, .fsgsbase, .rdrnd and .f16c. * doc/c-i386.texi: Document xsaveopt, fsgsbase, rdrnd and f16c. gas/testsuite/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * gas/i386/arch-10.s: Add xsaveopt. * gas/i386/x86-64-arch-2.s: Likwise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/f16c-intel.d: New. * gas/i386/f16c.d: Likewise. * gas/i386/f16c.s: Likewise. * gas/i386/fsgs-intel.d: Likewise. * gas/i386/fsgs.d: Likewise. * gas/i386/fsgs.s: Likewise. * gas/i386/rdrnd-intel.d: Likewise. * gas/i386/rdrnd.d: Likewise. * gas/i386/rdrnd.s: Likewise. * gas/i386/x86-64-f16c-intel.d: Likewise. * gas/i386/x86-64-f16c.d: Likewise. * gas/i386/x86-64-f16c.s: Likewise. * gas/i386/x86-64-fsgs-intel.d: Likewise. * gas/i386/x86-64-fsgs.d: Likewise. * gas/i386/x86-64-fsgs.s: Likewise. * gas/i386/x86-64-rdrnd-intel.d: Likewise. * gas/i386/x86-64-rdrnd.d: Likewise. * gas/i386/x86-64-rdrnd.s: Likewise. * gas/i386/i386.exp: Run f16c, f16c-intel, fsgs, fsgs-intel, rdrnd, rdrnd-intel, x86-64-f16c, x86-64-f16c-intel, x86-64-fsgs, x86-64-fsgs-intel, x86-64-rdrnd, x86-64-rdrnd-intel. * gas/i386/x86-64-xsave.s: Add tests for xsaveopt64. * gas/i386/x86-64-xsave-intel.d: Updated. * gas/i386/x86-64-xsave.d: Likewise. opcodes/ 2010-07-01 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2010) * i386-dis.c (PREFIX_0FAE_REG_0): New. (PREFIX_0FAE_REG_1): Likewise. (PREFIX_0FAE_REG_2): Likewise. (PREFIX_0FAE_REG_3): Likewise. (PREFIX_VEX_3813): Likewise. (PREFIX_VEX_3A1D): Likewise. (prefix_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3, PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (vex_table): Add PREFIX_VEX_3813 and PREFIX_VEX_3A1D. (mod_table): Add PREFIX_0FAE_REG_0, PREFIX_0FAE_REG_1, PREFIX_0FAE_REG_2, PREFIX_0FAE_REG_3 xsaveopt and rdrnd. * i386-gen.c (cpu_flag_init): Add CPU_XSAVEOPT_FLAGS, CPU_FSGSBASE_FLAGS, CPU_RDRND_FLAGS and CPU_F16C_FLAGS. (cpu_flags): Add CpuXsaveopt, CpuFSGSBase, CpuRdRnd and CpuF16C. * i386-opc.h (CpuXsaveopt): New. (CpuFSGSBase):Likewise. (CpuRdRnd): Likewise. (CpuF16C): Likewise. (i386_cpu_flags): Add cpuxsaveopt, cpufsgsbase, cpurdrnd and cpuf16c. * i386-opc.tbl: Add xsaveopt, rdfsbase, rdgsbase, rdrnd, wrfsbase, wrgsbase, vcvtph2ps and vcvtps2ph.
2010-07-01 23:55:02 +02:00
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
// FMA instructions
x86: fold FMA and FMA4 templates
2018-03-08 08:55:37 +01:00
vfmadd132pd, 3, 0x6698, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmadd132ps, 3, 0x6698, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmadd213pd, 3, 0x66a8, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmadd213ps, 3, 0x66a8, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmadd231pd, 3, 0x66b8, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmadd231ps, 3, 0x66b8, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd132sd, 3, 0x6699, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmadd132ss, 3, 0x6699, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmadd213sd, 3, 0x66a9, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmadd213ss, 3, 0x66a9, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmadd231sd, 3, 0x66b9, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmadd231ss, 3, 0x66b9, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: fold FMA and FMA4 templates
2018-03-08 08:55:37 +01:00
vfmaddsub132pd, 3, 0x6696, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsub132ps, 3, 0x6696, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsub213pd, 3, 0x66a6, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsub213ps, 3, 0x66a6, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsub231pd, 3, 0x66b6, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsub231ps, 3, 0x66b6, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd132pd, 3, 0x6697, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd132ps, 3, 0x6697, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd213pd, 3, 0x66a7, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd213ps, 3, 0x66a7, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd231pd, 3, 0x66b7, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubadd231ps, 3, 0x66b7, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub132pd, 3, 0x669a, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub132ps, 3, 0x669a, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub213pd, 3, 0x66aa, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub213ps, 3, 0x66aa, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub231pd, 3, 0x66ba, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsub231ps, 3, 0x66ba, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub132sd, 3, 0x669b, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsub132ss, 3, 0x669b, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsub213sd, 3, 0x66ab, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsub213ss, 3, 0x66ab, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsub231sd, 3, 0x66bb, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsub231ss, 3, 0x66bb, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: fold FMA and FMA4 templates
2018-03-08 08:55:37 +01:00
vfnmadd132pd, 3, 0x669c, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmadd132ps, 3, 0x669c, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmadd213pd, 3, 0x66ac, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmadd213ps, 3, 0x66ac, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmadd231pd, 3, 0x66bc, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmadd231ps, 3, 0x66bc, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd132sd, 3, 0x669d, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmadd132ss, 3, 0x669d, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmadd213sd, 3, 0x66ad, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmadd213ss, 3, 0x66ad, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmadd231sd, 3, 0x66bd, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmadd231ss, 3, 0x66bd, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: fold FMA and FMA4 templates
2018-03-08 08:55:37 +01:00
vfnmsub132pd, 3, 0x669e, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsub132ps, 3, 0x669e, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsub213pd, 3, 0x66ae, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsub213ps, 3, 0x66ae, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsub231pd, 3, 0x66be, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsub231ps, 3, 0x66be, None, 1, CpuFMA, Modrm|Vex|VexOpcode=1|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub132sd, 3, 0x669f, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsub132ss, 3, 0x669f, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsub213sd, 3, 0x66af, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsub213ss, 3, 0x66af, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsub231sd, 3, 0x66bf, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsub231ss, 3, 0x66bf, None, 1, CpuFMA, Modrm|Vex=3|VexOpcode=1|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
binutils/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * dwarf.c (dwarf_regnames_i386): Add AVX registers. (dwarf_regnames_x86_64): Likewise. gas/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * NEWS: Mention AES, CLMUL, AVX/FMA and -msse2avx. * doc/c-i386.texi: Add avx, aes, clmul and fma to -march=. Document -msse2avx, .avx, .aes, .clmul and .fma. * config/tc-i386.c (YMMWORD_MNEM_SUFFIX): New. (vex_prefix): Likewise. (sse2avx): Likewise. (CPU_FLAGS_ARCH_MATCH): Likewise. (CPU_FLAGS_64BIT_MATCH): Likewise. (CPU_FLAGS_32BIT_MATCH): Likewise. (CPU_FLAGS_PERFECT_MATCH): Likewise. (regymm): Likewise. (vex_imm4): Likewise. (fits_in_imm4): Likewise. (build_vex_prefix): Likewise. (VEX_check_operands): Likewise. (bad_implicit_operand): Likewise. (OPTION_MSSE2AVX): Likewise. (T_YMMWORD): Likewise. (_i386_insn): Add vex. (cpu_arch): Add .avx, .aes, .clmul and .fma. (cpu_flags_match): Changed to take a pointer to const template. Enable encoding SSE instructions with VEX prefix for -msse2avx. (match_mem_size): Also check ymmword. (operand_type_match): Clear ymmword. (md_begin): Allow '_' in mnemonic. (type_names): Add OPERAND_TYPE_VEX_IMM4. (process_immext): Update assert. (md_assemble): Don't call process_immext if sse2avx and immext are true. Call build_vex_prefix if vex is true. (parse_insn): Updated for cpu_flags_match. (swap_operands): Handle 5 operands. (match_template): Handle 5 operands. Updated for cpu_flags_match. Check regymm. Call VEX_check_operands. Handle YMMWORD_MNEM_SUFFIX. (process_suffix): Handle YMMWORD_MNEM_SUFFIX. (check_byte_reg): Check regymm. (process_operands): Duplicate the destination register for -msse2avx if needed. (build_modrm_byte): Updated for instructions with VEX encoding. (output_insn): Output VEX prefix if needed. (md_longopts): Add msse2avx. (md_parse_option): Handle OPTION_MSSE2AVX. (md_show_usage): Add avx, aes, clmul, fma and -msse2avx. (intel_e09): Support YMMWORD. (intel_e11): Likewise. (intel_get_token): Likewise. gas/testsuite/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/i386.exp: Run aes, aes-intel, x86-64-aes, x86-64-aes-intel, avx, avx-intel, inval-avx, x86-64-avx, x86-64-avx-intel and x86-64-inval-avx. * gas/cfi/cfi-i386.s: Add tests for AVX register maps. * gas/cfi/cfi-x86_64.s: Likewise. * gas/i386/aes.d: New. * gas/i386/aes.s: Likewise. * gas/i386/aes-intel.d: Likewise. * gas/i386/avx.d: Likewise. * gas/i386/avx.s: Likewise. * gas/i386/avx-intel.d: Likewise. * gas/i386/clmul.d: Likewise. * gas/i386/clmul-intel.d: Likewise. * gas/i386/clmul.s: Likewise. * gas/i386/i386.exp: Likewise. * gas/i386/inval-avx.l: Likewise. * gas/i386/inval-avx.s: Likewise. * gas/i386/sse2avx.d: Likewise. * gas/i386/sse2avx.s: Likewise. * gas/i386/x86-64-aes.d: Likewise. * gas/i386/x86-64-aes.s: Likewise. * gas/i386/x86-64-aes-intel.d: Likewise. * gas/i386/x86-64-avx.d: Likewise. * gas/i386/x86-64-avx.s: Likewise. * gas/i386/x86-64-avx-intel.d: Likewise. * gas/i386/x86-64-clmul.d: Likewise. * gas/i386/x86-64-clmul-intel.d: Likewise. * gas/i386/x86-64-clmul.s: Likewise. * gas/i386/x86-64-inval-avx.l: Likewise. * gas/i386/x86-64-inval-avx.s: Likewise. * gas/i386/x86-64-sse2avx.d: Likewise. * gas/i386/x86-64-sse2avx.s: Likewise. * gas/i386/arch-10.s: Add tests for AVX, AES, CLMUL and FMA. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/rexw.s: Add AVX tests. * gas/i386/x86-64-opcode-inval.s: Remove lds/les test. * gas/cfi/cfi-i386.d: Updated. * gas/cfi/cfi-x86_64.d: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/rexw.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-opcode-inval.d: Likewise. * gas/i386/x86-64-opcode-inval-intel.d: Likewise. include/opcode/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386.h (MAX_OPERANDS): Set to 5. (MAX_MNEM_SIZE): Changed to 20. opcodes/ 2008-04-03 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (OP_E_register): New. (OP_E_memory): Likewise. (OP_VEX): Likewise. (OP_EX_Vex): Likewise. (OP_EX_VexW): Likewise. (OP_XMM_Vex): Likewise. (OP_XMM_VexW): Likewise. (OP_REG_VexI4): Likewise. (PCLMUL_Fixup): Likewise. (VEXI4_Fixup): Likewise. (VZERO_Fixup): Likewise. (VCMP_Fixup): Likewise. (VPERMIL2_Fixup): Likewise. (rex_original): Likewise. (rex_ignored): Likewise. (Mxmm): Likewise. (XMM): Likewise. (EXxmm): Likewise. (EXxmmq): Likewise. (EXymmq): Likewise. (Vex): Likewise. (Vex128): Likewise. (Vex256): Likewise. (VexI4): Likewise. (EXdVex): Likewise. (EXqVex): Likewise. (EXVexW): Likewise. (EXdVexW): Likewise. (EXqVexW): Likewise. (XMVex): Likewise. (XMVexW): Likewise. (XMVexI4): Likewise. (PCLMUL): Likewise. (VZERO): Likewise. (VCMP): Likewise. (VPERMIL2): Likewise. (xmm_mode): Likewise. (xmmq_mode): Likewise. (ymmq_mode): Likewise. (vex_mode): Likewise. (vex128_mode): Likewise. (vex256_mode): Likewise. (USE_VEX_C4_TABLE): Likewise. (USE_VEX_C5_TABLE): Likewise. (USE_VEX_LEN_TABLE): Likewise. (VEX_C4_TABLE): Likewise. (VEX_C5_TABLE): Likewise. (VEX_LEN_TABLE): Likewise. (REG_VEX_XX): Likewise. (MOD_VEX_XXX): Likewise. (PREFIX_0F38DB..PREFIX_0F38DF): Likewise. (PREFIX_0F3A44): Likewise. (PREFIX_0F3ADF): Likewise. (PREFIX_VEX_XXX): Likewise. (VEX_OF): Likewise. (VEX_OF38): Likewise. (VEX_OF3A): Likewise. (VEX_LEN_XXX): Likewise. (vex): Likewise. (need_vex): Likewise. (need_vex_reg): Likewise. (vex_i4_done): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (OP_REG_VexI4): Likewise. (vex_cmp_op): Likewise. (pclmul_op): Likewise. (vpermil2_op): Likewise. (m_mode): Updated. (es_reg): Likewise. (PREFIX_0F38F0): Likewise. (PREFIX_0F3A60): Likewise. (reg_table): Add REG_VEX_71...REG_VEX_73 and REG_VEX_AE. (prefix_table): Add PREFIX_0F38DB..PREFIX_0F38DF, PREFIX_0F3ADF and PREFIX_VEX_XXX entries. (x86_64_table): Use VEX_C4_TABLE and VEX_C5_TABLE. (three_byte_table): Use PREFIX_0F38DB..PREFIX_0F38DF and PREFIX_0F3ADF. (mod_table): Use VEX_C4_TABLE, VEX_C5_TABLE and VEX_LEN_TABLE. Add MOD_VEX_XXX entries. (ckprefix): Initialize rex_original and rex_ignored. Store the REX byte in rex_original. (get_valid_dis386): Handle the implicit prefix in VEX prefix bytes and USE_VEX_LEN_TABLE/USE_VEX_C4_TABLE/USE_VEX_C5_TABLE. (print_insn): Set need_vex/need_vex_reg/vex_i4_done to 0 before calling get_valid_dis386. Use rex_original and rex_ignored when printing out REX. (putop): Handle "XY". (intel_operand_size): Handle VEX, xmm_mode, xmmq_mode and ymmq_mode. (OP_E_extended): Updated to use OP_E_register and OP_E_memory. (OP_XMM): Handle VEX. (OP_EX): Likewise. (XMM_Fixup): Likewise. (CMP_Fixup): Use ARRAY_SIZE. * i386-gen.c (cpu_flag_init): Add CpuAES, CPU_CLMUL_FLAGS, CPU_FMA_FLAGS and CPU_AVX_FLAGS. (operand_type_init): Add OPERAND_TYPE_REGYMM and OPERAND_TYPE_VEX_IMM4. (cpu_flags): Add CpuAVX, CpuAES, CpuCLMUL and CpuFMA. (opcode_modifiers): Add Implicit1stXmm0, Vex, Vex256, VexNDD, VexNDS, VexW0, VexW1, Vex0F, Vex0F38, Vex0F3A, Vex3Sources, VexImmExt and SSE2AVX. (operand_types): Add RegYMM, Ymmword and Vex_Imm4. * i386-opc.h (CpuAVX): New. (CpuAES): Likewise. (CpuCLMUL): Likewise. (CpuFMA): Likewise. (Vex): Likewise. (Vex256): Likewise. (VexNDS): Likewise. (VexNDD): Likewise. (VexW0): Likewise. (VexW1): Likewise. (Vex0F): Likewise. (Vex0F38): Likewise. (Vex0F3A): Likewise. (Vex3Sources): Likewise. (VexImmExt): Likewise. (SSE2AVX): Likewise. (RegYMM): Likewise. (Ymmword): Likewise. (Vex_Imm4): Likewise. (Implicit1stXmm0): Likewise. (CpuXsave): Updated. (CpuLM): Likewise. (ByteOkIntel): Likewise. (OldGcc): Likewise. (Control): Likewise. (Unspecified): Likewise. (OTMax): Likewise. (i386_cpu_flags): Add cpuavx, cpuaes, cpuclmul and cpufma. (i386_opcode_modifier): Add implicit1stxmm0, vex, vex256, vexnds, vexndd, vexw0, vexw1, vex0f, vex0f38, vex0f3a, vex3sources, veximmext and sse2avx. (i386_operand_type): Add regymm, ymmword and vex_imm4. * i386-opc.tbl: Add AES, CLMUL, AVX and FMA new instructions. * i386-reg.tbl: Add AVX registers, ymm0..ymm15. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2008-04-03 16:03:21 +02:00
Implement Intel Transactional Synchronization Extensions gas/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (HLE_PREFIX): New. (check_hle): Likewise. (_i386_insn): Add have_hle. (cpu_arch): Add .hle and .rtm. (md_assemble): Call check_hle if i.have_hle isn't zero. (parse_insn): Set i.have_hle to 1 for HLE prefix. (output_jump): Support up to 2 byte opcode. * doc/c-i386.texi: Document hle/.hle and rtm/.rtm. gas/testsuite/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/hle-intel.d: New. * gas/i386/hle.d: Likewise. * gas/i386/hle.s: Likewise. * gas/i386/hlebad.l: Likewise. * gas/i386/hlebad.s: Likewise. * gas/i386/rtm-intel.d: Likewise. * gas/i386/rtm.d: Likewise. * gas/i386/rtm.s: Likewise. * gas/i386/x86-64-hle-intel.d: Likewise. * gas/i386/x86-64-hle.d: Likewise. * gas/i386/x86-64-hle.s: Likewise. * gas/i386/x86-64-hlebad.l: Likewise. * gas/i386/x86-64-hlebad.s: Likewise. * gas/i386/x86-64-rtm-intel.d: Likewise. * gas/i386/x86-64-rtm.d: Likewise. * gas/i386/x86-64-rtm.s: Likewise. * gas/i386/i386.exp: Run hle, hle-intel, hlebad x86-64-hle, rtm, rtm-intel, x86-64-hle-intel, x86-64-hlebad, x86-64-rtm and x86-64-rtm-intel. include/opcode/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * i386.h (XACQUIRE_PREFIX_OPCODE): New. (XRELEASE_PREFIX_OPCODE): Likewise. opcodes/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (HLE_Fixup1): New. (HLE_Fixup2): Likewise. (HLE_Fixup3): Likewise. (Ebh1): Likewise. (Evh1): Likewise. (Ebh2): Likewise. (Evh2): Likewise. (Ebh3): Likewise. (Evh3): Likewise. (MOD_C6_REG_7): Likewise. (MOD_C7_REG_7): Likewise. (RM_C6_REG_7): Likewise. (RM_C7_REG_7): Likewise. (XACQUIRE_PREFIX): Likewise. (XRELEASE_PREFIX): Likewise. (dis386): Use Ebh1/Evh1 on add, adc, and, btc, btr, bts, cmpxchg, dec, inc, neg, not, or, sbb, sub, xor and xadd. Use Ebh2/Evh2 on xchg. Use Ebh3/Evh3 on mov. (reg_table): Use Ebh1/Evh1 on add, adc, and, dec, inc, neg, not, or, sbb, sub and xor. Use Ebh3/Evh3 on mov. Use MOD_C6_REG_7 and MOD_C7_REG_7. (mod_table): Add MOD_C6_REG_7 and MOD_C7_REG_7. (rm_table): Add RM_C6_REG_7 and RM_C7_REG_7. Add xend and xtest. (prefix_name): Handle XACQUIRE_PREFIX and XRELEASE_PREFIX. (CMPXCHG8B_Fixup): Handle HLE prefix on cmpxchg8b. * i386-gen.c (cpu_flag_init): Add CPU_HLE_FLAGS and CPU_RTM_FLAGS. (cpu_flags): Add CpuHLE and CpuRTM. (opcode_modifiers): Add HLEPrefixOk. * i386-opc.h (CpuHLE): New. (CpuRTM): Likewise. (HLEPrefixOk): Likewise. (i386_cpu_flags): Add cpuhle and cpurtm. (i386_opcode_modifier): Add hleprefixok. * i386-opc.tbl: Add HLEPrefixOk=3 to mov. Add HLEPrefixOk to add, adc, and, btc, btr, bts, cmpxchg, dec, inc, neg, not, or, sbb, sub, xor and xadd. Add HLEPrefixOk=2 to xchg with memory operand. Add xacquire, xrelease, xabort, xbegin, xend and xtest. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2012-02-08 19:20:41 +01:00
// HLE prefixes
xacquire, 0, 0xf2, None, 1, CpuHLE, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
xrelease, 0, 0xf3, None, 1, CpuHLE, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
// RTM instructions
xabort, 1, 0xc6f8, None, 2, CpuRTM, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8 }
x86: consolidate Disp<NN> handling a little In memory operand addressing, which forms of displacement are permitted besides Disp8 is pretty clearly limited - outside of 64-bit mode, Disp16 or Disp32 only, depending on address size (MPX being special in not allowing Disp16), - in 64-bit mode, Disp32s or Disp64 without address size override, and solely Disp32 with one. Adjust assembler and i386-gen to match this, observing that templates already get adjusted before trying to match them against input depending on the presence of an address size prefix. This adjustment logic gets extended to all cases, as certain DispNN values should also be dropped when there's no such prefix. In fact behavior of the assembler, perhaps besides the exact diagnostics wording, should not differ between there being templates applicable to 64-bit and non-64-bit at the same time, or there being fully separate sets of templates, with their DispNN settings already reduced accordingly. This adjustment logic further gets guarded such that there wouldn't be and Disp<N> conversion based on address size prefix when this prefix doesn't control the width of the displacement (on branches other than absolute ones). These adjustments then also allow folding two MOV templates, which had been split between 64-bit and non-64-bits variants so far. Once in this area also - drop the bogus DispNN from JumpByte templates, leaving just the correct Disp8 there (compensated by i386_finalize_displacement() now setting Disp8 on their operands), - add the missing Disp32S to XBEGIN. Note that the changes make it necessary to temporarily mark a test as XFAIL; this will get taken care of by a subsequent patch. The failing parts are entirely bogus and will get replaced.
2019-12-27 09:22:03 +01:00
xbegin, 1, 0xc7f8, None, 2, CpuRTM, JumpDword|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Disp16|Disp32|Disp32S }
Implement Intel Transactional Synchronization Extensions gas/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (HLE_PREFIX): New. (check_hle): Likewise. (_i386_insn): Add have_hle. (cpu_arch): Add .hle and .rtm. (md_assemble): Call check_hle if i.have_hle isn't zero. (parse_insn): Set i.have_hle to 1 for HLE prefix. (output_jump): Support up to 2 byte opcode. * doc/c-i386.texi: Document hle/.hle and rtm/.rtm. gas/testsuite/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/hle-intel.d: New. * gas/i386/hle.d: Likewise. * gas/i386/hle.s: Likewise. * gas/i386/hlebad.l: Likewise. * gas/i386/hlebad.s: Likewise. * gas/i386/rtm-intel.d: Likewise. * gas/i386/rtm.d: Likewise. * gas/i386/rtm.s: Likewise. * gas/i386/x86-64-hle-intel.d: Likewise. * gas/i386/x86-64-hle.d: Likewise. * gas/i386/x86-64-hle.s: Likewise. * gas/i386/x86-64-hlebad.l: Likewise. * gas/i386/x86-64-hlebad.s: Likewise. * gas/i386/x86-64-rtm-intel.d: Likewise. * gas/i386/x86-64-rtm.d: Likewise. * gas/i386/x86-64-rtm.s: Likewise. * gas/i386/i386.exp: Run hle, hle-intel, hlebad x86-64-hle, rtm, rtm-intel, x86-64-hle-intel, x86-64-hlebad, x86-64-rtm and x86-64-rtm-intel. include/opcode/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * i386.h (XACQUIRE_PREFIX_OPCODE): New. (XRELEASE_PREFIX_OPCODE): Likewise. opcodes/ 2012-02-08 H.J. Lu <hongjiu.lu@intel.com> * i386-dis.c (HLE_Fixup1): New. (HLE_Fixup2): Likewise. (HLE_Fixup3): Likewise. (Ebh1): Likewise. (Evh1): Likewise. (Ebh2): Likewise. (Evh2): Likewise. (Ebh3): Likewise. (Evh3): Likewise. (MOD_C6_REG_7): Likewise. (MOD_C7_REG_7): Likewise. (RM_C6_REG_7): Likewise. (RM_C7_REG_7): Likewise. (XACQUIRE_PREFIX): Likewise. (XRELEASE_PREFIX): Likewise. (dis386): Use Ebh1/Evh1 on add, adc, and, btc, btr, bts, cmpxchg, dec, inc, neg, not, or, sbb, sub, xor and xadd. Use Ebh2/Evh2 on xchg. Use Ebh3/Evh3 on mov. (reg_table): Use Ebh1/Evh1 on add, adc, and, dec, inc, neg, not, or, sbb, sub and xor. Use Ebh3/Evh3 on mov. Use MOD_C6_REG_7 and MOD_C7_REG_7. (mod_table): Add MOD_C6_REG_7 and MOD_C7_REG_7. (rm_table): Add RM_C6_REG_7 and RM_C7_REG_7. Add xend and xtest. (prefix_name): Handle XACQUIRE_PREFIX and XRELEASE_PREFIX. (CMPXCHG8B_Fixup): Handle HLE prefix on cmpxchg8b. * i386-gen.c (cpu_flag_init): Add CPU_HLE_FLAGS and CPU_RTM_FLAGS. (cpu_flags): Add CpuHLE and CpuRTM. (opcode_modifiers): Add HLEPrefixOk. * i386-opc.h (CpuHLE): New. (CpuRTM): Likewise. (HLEPrefixOk): Likewise. (i386_cpu_flags): Add cpuhle and cpurtm. (i386_opcode_modifier): Add hleprefixok. * i386-opc.tbl: Add HLEPrefixOk=3 to mov. Add HLEPrefixOk to add, adc, and, btc, btr, bts, cmpxchg, dec, inc, neg, not, or, sbb, sub, xor and xadd. Add HLEPrefixOk=2 to xchg with memory operand. Add xacquire, xrelease, xabort, xbegin, xend and xtest. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2012-02-08 19:20:41 +01:00
xend, 0, 0xf01d5, None, 3, CpuRTM, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
xtest, 0, 0xf01d6, None, 3, CpuHLE|CpuRTM, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
// BMI2 instructions.
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
bzhi, 3, 0xf5, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
mulx, 3, 0xf2f6, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64, Reg32|Reg64 }
pdep, 3, 0xf2f5, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64, Reg32|Reg64 }
pext, 3, 0xf3f5, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64, Reg32|Reg64 }
rorx, 3, 0xf2f0, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=2|VexVVVV=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm8, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
sarx, 3, 0xf3f7, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
shlx, 3, 0x66f7, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
shrx, 3, 0xf2f7, None, 1, CpuBMI2, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
Support AVX Programming Reference (June, 2011). gas/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * config/tc-i386.c (i386_error): Add invalid_vsib_address and unsupported_vector_index_register. (cpu_arch): Add .avx2, .bmi2, .lzcnt and .invpcid. (check_VecOperands): New. (match_template): Call check_VecOperands. Handle invalid_vsib_address and unsupported_vector_index_register. (build_modrm_byte): Support VecSIB. Check register-only source operand when two source operands are swapped. (i386_index_check): Allow Xmm/Ymm index registers. * doc/c-i386.texi: Document avx2/.avx2, bmi2/.bmi2, lzcnt/.lzcnt and invpcid./invpcid. gas/testsuite/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * gas/i386/arch-10-1.l: Updated. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.s: Add LZCNT to comments. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10-lzcnt.d: New. * gas/i386/avx-gather-intel.d: Likewise. * gas/i386/avx-gather.d: Likewise. * gas/i386/avx-gather.s: Likewise. * gas/i386/avx2-intel.d: Likewise. * gas/i386/avx2.d: Likewise. * gas/i386/avx2.s: Likewise * gas/i386/avx256int-intel.d: Likewise. * gas/i386/avx256int.d: Likewise. * gas/i386/avx256int.s: Likewise. * gas/i386/bmi2-intel.d: Likewise. * gas/i386/bmi2.d: Likewise. * gas/i386/bmi2.s: Likewise. * gas/i386/inval-invpcid.l:Likewise. * gas/i386/inval-invpcid.s: Likewise. * gas/i386/invpcid-intel.d: Likewise. * gas/i386/invpcid.d: Likewise. * gas/i386/invpcid.s: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/x86-64-avx-gather-intel.d: Likewise. * gas/i386/x86-64-avx-gather.d: Likewise. * gas/i386/x86-64-avx-gather.s: Likewise. * gas/i386/x86-64-avx2-intel.d: Likewise. * gas/i386/x86-64-avx2.d: Likewise. * gas/i386/x86-64-avx2.s: Likewise. * gas/i386/x86-64-avx256int-intel.d: Likewise. * gas/i386/x86-64-avx256int.d: Likewise. * gas/i386/x86-64-avx256int.s: Likewise. * gas/i386/x86-64-bmi2-intel.d: Likewise. * gas/i386/x86-64-bmi2.d: Likewise. * gas/i386/x86-64-bmi2.s: Likewise. * gas/i386/x86-64-inval-invpcid.l: Likewise. * gas/i386/x86-64-inval-invpcid.s: Likewise. * gas/i386/x86-64-invpcid-intel.d: Likewise. * gas/i386/x86-64-invpcid.d: Likewise. * gas/i386/x86-64-invpcid.s: Likewise. opcodes/ 2011-06-10 H.J. Lu <hongjiu.lu@intel.com> AVX Programming Reference (June, 2011) * i386-dis.c (XMGatherQ): New. * i386-dis.c (EXxmm_mb): New. (EXxmm_mb): Likewise. (EXxmm_mw): Likewise. (EXxmm_md): Likewise. (EXxmm_mq): Likewise. (EXxmmdw): Likewise. (EXxmmqd): Likewise. (VexGatherQ): Likewise. (MVexVSIBDWpX): Likewise. (MVexVSIBQWpX): Likewise. (xmm_mb_mode): Likewise. (xmm_mw_mode): Likewise. (xmm_md_mode): Likewise. (xmm_mq_mode): Likewise. (xmmdw_mode): Likewise. (xmmqd_mode): Likewise. (ymmxmm_mode): Likewise. (vex_vsib_d_w_dq_mode): Likewise. (vex_vsib_q_w_dq_mode): Likewise. (MOD_VEX_0F385A_PREFIX_2): Likewise. (MOD_VEX_0F388C_PREFIX_2): Likewise. (MOD_VEX_0F388E_PREFIX_2): Likewise. (PREFIX_0F3882): Likewise. (PREFIX_VEX_0F3816): Likewise. (PREFIX_VEX_0F3836): Likewise. (PREFIX_VEX_0F3845): Likewise. (PREFIX_VEX_0F3846): Likewise. (PREFIX_VEX_0F3847): Likewise. (PREFIX_VEX_0F3858): Likewise. (PREFIX_VEX_0F3859): Likewise. (PREFIX_VEX_0F385A): Likewise. (PREFIX_VEX_0F3878): Likewise. (PREFIX_VEX_0F3879): Likewise. (PREFIX_VEX_0F388C): Likewise. (PREFIX_VEX_0F388E): Likewise. (PREFIX_VEX_0F3890..PREFIX_VEX_0F3893): Likewise. (PREFIX_VEX_0F38F5): Likewise. (PREFIX_VEX_0F38F6): Likewise. (PREFIX_VEX_0F3A00): Likewise. (PREFIX_VEX_0F3A01): Likewise. (PREFIX_VEX_0F3A02): Likewise. (PREFIX_VEX_0F3A38): Likewise. (PREFIX_VEX_0F3A39): Likewise. (PREFIX_VEX_0F3A46): Likewise. (PREFIX_VEX_0F3AF0): Likewise. (VEX_LEN_0F3816_P_2): Likewise. (VEX_LEN_0F3819_P_2): Likewise. (VEX_LEN_0F3836_P_2): Likewise. (VEX_LEN_0F385A_P_2_M_0): Likewise. (VEX_LEN_0F38F5_P_0): Likewise. (VEX_LEN_0F38F5_P_1): Likewise. (VEX_LEN_0F38F5_P_3): Likewise. (VEX_LEN_0F38F6_P_3): Likewise. (VEX_LEN_0F38F7_P_1): Likewise. (VEX_LEN_0F38F7_P_2): Likewise. (VEX_LEN_0F38F7_P_3): Likewise. (VEX_LEN_0F3A00_P_2): Likewise. (VEX_LEN_0F3A01_P_2): Likewise. (VEX_LEN_0F3A38_P_2): Likewise. (VEX_LEN_0F3A39_P_2): Likewise. (VEX_LEN_0F3A46_P_2): Likewise. (VEX_LEN_0F3AF0_P_3): Likewise. (VEX_W_0F3816_P_2): Likewise. (VEX_W_0F3818_P_2): Likewise. (VEX_W_0F3819_P_2): Likewise. (VEX_W_0F3836_P_2): Likewise. (VEX_W_0F3846_P_2): Likewise. (VEX_W_0F3858_P_2): Likewise. (VEX_W_0F3859_P_2): Likewise. (VEX_W_0F385A_P_2_M_0): Likewise. (VEX_W_0F3878_P_2): Likewise. (VEX_W_0F3879_P_2): Likewise. (VEX_W_0F3A00_P_2): Likewise. (VEX_W_0F3A01_P_2): Likewise. (VEX_W_0F3A02_P_2): Likewise. (VEX_W_0F3A38_P_2): Likewise. (VEX_W_0F3A39_P_2): Likewise. (VEX_W_0F3A46_P_2): Likewise. (MOD_VEX_0F3818_PREFIX_2): Removed. (MOD_VEX_0F3819_PREFIX_2): Likewise. (VEX_LEN_0F60_P_2..VEX_LEN_0F6D_P_2): Likewise. (VEX_LEN_0F70_P_1..VEX_LEN_0F76_P_2): Likewise. (VEX_LEN_0FD1_P_2..VEX_LEN_0FD5_P_2): Likewise. (VEX_LEN_0FD7_P_2_M_1..VEX_LEN_0F3819_P_2_M_0): Likewise. (VEX_LEN_0F381C_P_2..VEX_LEN_0F3840_P_2): Likewise. (VEX_LEN_0F3A0E_P_2): Likewise. (VEX_LEN_0F3A0F_P_2): Likewise. (VEX_LEN_0F3A42_P_2): Likewise. (VEX_LEN_0F3A4C_P_2): Likewise. (VEX_W_0F3818_P_2_M_0): Likewise. (VEX_W_0F3819_P_2_M_0): Likewise. (prefix_table): Updated. (three_byte_table): Likewise. (vex_table): Likewise. (vex_len_table): Likewise. (vex_w_table): Likewise. (mod_table): Likewise. (putop): Handle "LW". (intel_operand_size): Handle xmm_mb_mode, xmm_mw_mode, xmm_md_mode, xmm_mq_mode, xmmdw_mode, xmmqd_mode, ymmxmm_mode, vex_vsib_d_w_dq_mode, vex_vsib_q_w_dq_mode. (OP_EX): Likewise. (OP_E_memory): Handle vex_vsib_d_w_dq_mode and vex_vsib_q_w_dq_mode. (OP_XMM): Handle vex_vsib_q_w_dq_mode. (OP_VEX): Likewise. * i386-gen.c (cpu_flag_init): Add CpuAVX2 to CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_BMI2_FLAGS, CPU_LZCNT_FLAGS, CPU_INVPCID_FLAGS and CPU_AVX2_FLAGS. (cpu_flags): Add CpuAVX2, CpuBMI2, CpuLZCNT and CpuINVPCID. (opcode_modifiers): Add VecSIB. * i386-opc.h (CpuAVX2): New. (CpuBMI2): Likewise. (CpuLZCNT): Likewise. (CpuINVPCID): Likewise. (VecSIB128): Likewise. (VecSIB256): Likewise. (VecSIB): Likewise. (i386_cpu_flags): Add cpuavx2, cpubmi2, cpulzcnt and cpuinvpcid. (i386_opcode_modifier): Add vecsib. * i386-opc.tbl: Add invpcid, AVX2 and BMI2 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2011-06-10 23:27:40 +02:00
<gas changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * config/tc-i386.c (cpu_arch): Add .fma4 and CPU_FMA4_FLAGS. (build_modrm_byte): Add support to handle FMA4 instructions. (md_show_usage): Add fma4. <gas/testsuite changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * gas/i386/i386.exp: Add FMA4 tests. * gas/i386/x86-64-fma4.d: Ditto. * gas/i386/fma4.d: Ditto. * gas/i386/x86-64-fma4.s: Ditto. * gas/i386/fma4.s: Ditto. <opcodes changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * i386-opc.h (CpuFMA4): Add CpuFMA4. (i386_cpu_flags): New. * i386-gen.c: Add CPU_FMA4_FLAGS. * i386-opc.tbl: Add FMA4 instructions. * i386-tbl.h: Regenerate. * i386-init.h: Regenerate. * i386-dis.c (OP_VEX_FMA): New. Handle FMA4. (OP_XMM_VexW): Ditto. (OP_EX_VexW): Ditto. (VEXI4_Fixup): Ditto. (VexI4, VexFMA, Vex128FMA, EXVexW, EXdVexW, XMVexW): New Macros. (PREFIX_VEX_3A5C, PREFIX_VEX_3A5D, PREFIX_VEX_3A5E): New. (PREFIX_VEX_3A5F, PREFIX_VEX_3A60): New. (PREFIX_VEX_3A68, PREFIX_VEX_3A69, PREFIX_VEX_3A6A): New. (PREFIX_VEX_3A6B, PREFIX_VEX_3A6C, PREFIX_VEX_3A6D): New. (PREFIX_VEX_3A6E, PREFIX_VEX_3A6F, PREFIX_VEX_3A7A): New. (PREFIX_VEX_3A7B, PREFIX_VEX_3A7C, PREFIX_VEX_3A7D): New. (PREFIX_VEX_3A7E, PREFIX_VEX_3A7F): New. (VEX_LEN_3A6A_P_2,VEX_LEN_3A6B_P_2, VEX_LEN_3A6E_P_2): New. (VEX_LEN_3A6F_P_2,VEX_LEN_3A7A_P_2, VEX_LEN_3A7B_P_2): New. (VEX_LEN_3A7E_P_2,VEX_LEN_3A7F_P_2): New. (get_vex_imm8): New. handle FMA4. (OP_EX_VexReg): Ditto.
2009-07-06 21:34:30 +02:00
// FMA4 instructions
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vfmaddpd, 4, 0x6669, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddpd, 4, 0x6669, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddps, 4, 0x6668, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddps, 4, 0x6668, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmaddsd, 4, 0x666b, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfmaddsd, 4, 0x666b, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmaddss, 4, 0x666a, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfmaddss, 4, 0x666a, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vfmaddsubpd, 4, 0x665d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsubpd, 4, 0x665d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsubps, 4, 0x665c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmaddsubps, 4, 0x665c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubaddpd, 4, 0x665f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubaddpd, 4, 0x665f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubaddps, 4, 0x665e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubaddps, 4, 0x665e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubpd, 4, 0x666d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubpd, 4, 0x666d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubps, 4, 0x666c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfmsubps, 4, 0x666c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsubsd, 4, 0x666f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfmsubsd, 4, 0x666f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfmsubss, 4, 0x666e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfmsubss, 4, 0x666e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vfnmaddpd, 4, 0x6679, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmaddpd, 4, 0x6679, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmaddps, 4, 0x6678, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmaddps, 4, 0x6678, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmaddsd, 4, 0x667b, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfnmaddsd, 4, 0x667b, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmaddss, 4, 0x667a, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfnmaddss, 4, 0x667a, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vfnmsubpd, 4, 0x667d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsubpd, 4, 0x667d, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsubps, 4, 0x667c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=2|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vfnmsubps, 4, 0x667c, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV=1|VexW=1|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, {RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsubsd, 4, 0x667f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfnmsubsd, 4, 0x667f, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
vfnmsubss, 4, 0x667e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW1|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM, RegXMM }
vfnmsubss, 4, 0x667e, None, 1, CpuFMA4, Modrm|Vex|VexOpcode=2|VexVVVV|VexW0|VexSources=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex|RegXMM, RegXMM, RegXMM }
<gas changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * config/tc-i386.c (cpu_arch): Add .fma4 and CPU_FMA4_FLAGS. (build_modrm_byte): Add support to handle FMA4 instructions. (md_show_usage): Add fma4. <gas/testsuite changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * gas/i386/i386.exp: Add FMA4 tests. * gas/i386/x86-64-fma4.d: Ditto. * gas/i386/fma4.d: Ditto. * gas/i386/x86-64-fma4.s: Ditto. * gas/i386/fma4.s: Ditto. <opcodes changes> 2009-07-06 Dwarakanath Rajagopal <dwarak.rajagopal@amd.com> * i386-opc.h (CpuFMA4): Add CpuFMA4. (i386_cpu_flags): New. * i386-gen.c: Add CPU_FMA4_FLAGS. * i386-opc.tbl: Add FMA4 instructions. * i386-tbl.h: Regenerate. * i386-init.h: Regenerate. * i386-dis.c (OP_VEX_FMA): New. Handle FMA4. (OP_XMM_VexW): Ditto. (OP_EX_VexW): Ditto. (VEXI4_Fixup): Ditto. (VexI4, VexFMA, Vex128FMA, EXVexW, EXdVexW, XMVexW): New Macros. (PREFIX_VEX_3A5C, PREFIX_VEX_3A5D, PREFIX_VEX_3A5E): New. (PREFIX_VEX_3A5F, PREFIX_VEX_3A60): New. (PREFIX_VEX_3A68, PREFIX_VEX_3A69, PREFIX_VEX_3A6A): New. (PREFIX_VEX_3A6B, PREFIX_VEX_3A6C, PREFIX_VEX_3A6D): New. (PREFIX_VEX_3A6E, PREFIX_VEX_3A6F, PREFIX_VEX_3A7A): New. (PREFIX_VEX_3A7B, PREFIX_VEX_3A7C, PREFIX_VEX_3A7D): New. (PREFIX_VEX_3A7E, PREFIX_VEX_3A7F): New. (VEX_LEN_3A6A_P_2,VEX_LEN_3A6B_P_2, VEX_LEN_3A6E_P_2): New. (VEX_LEN_3A6F_P_2,VEX_LEN_3A7A_P_2, VEX_LEN_3A7B_P_2): New. (VEX_LEN_3A7E_P_2,VEX_LEN_3A7F_P_2): New. (get_vex_imm8): New. handle FMA4. (OP_EX_VexReg): Ditto.
2009-07-06 21:34:30 +02:00
2009-11-17 Sebastian Pop <sebastian.pop@amd.com> Quentin Neill <quentin.neill@amd.com> gas/ * config/tc-i386.c (cpu_arch): Added .xop and .cvt16. (build_vex_prefix): Handle xop08. (md_assemble): Don't special case the constant 3 for insns using MODRM. (build_modrm_byte): Handle vex2sources. (md_show_usage): Add xop and cvt16. * doc/c-i386.texi: Document fma4, xop, and cvt16. gas/testsuite/ * gas/i386/i386.exp: Run xop and cvt16 in 32-bit mode. Run x86-64-xop and x86-64-cvt16 in 64-bit mode. * gas/i386/lwp.d: Update name of the testcase. * gas/i386/x86-64-xop.d: New. * gas/i386/x86-64-xop.s: New. * gas/i386/xop.d: New. * gas/i386/xop.s: New. * gas/i386/cvt16.d: New. * gas/i386/cvt16.s: New. opcodes/ * i386-dis.c (OP_Vex_2src_1): New. (OP_Vex_2src_2): New. (Vex_2src_1): New. (Vex_2src_2): New. (XOP_08): Added. (VEX_LEN_XOP_08_A0): Added. (VEX_LEN_XOP_08_A1): Added. (VEX_LEN_XOP_09_80): Added. (VEX_LEN_XOP_09_81): Added. (xop_table): Added an entry for XOP_08. Handle xop instructions. (vex_len_table): Added entries for VEX_LEN_XOP_08_A0, VEX_LEN_XOP_08_A1, VEX_LEN_XOP_09_80, VEX_LEN_XOP_09_81. (get_valid_dis386): Handle XOP_08. (OP_Vex_2src): New. * i386-gen.c (cpu_flag_init): Add CPU_XOP_FLAGS and CPU_CVT16_FLAGS. (cpu_flags): Add CpuXOP and CpuCVT16. (opcode_modifiers): Add XOP08, Vex2Sources. * i386-opc.h (CpuXOP): Added. (CpuCVT16): Added. (i386_cpu_flags): Add cpuxop and cpucvt16. (XOP08): Added. (Vex2Sources): Added. (i386_opcode_modifier): Add xop08, vex2sources. * i386-opc.tbl: Add entries for XOP and CVT16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated.
2009-11-18 05:04:17 +01:00
// XOP instructions
x86: use template for XOP integer comparison, shift, and rotate insns These all follow common patterns.
2020-03-09 10:14:17 +01:00
<xop_elem:opc, b:0, w:1, d:2, q:3>
<xop_irel:imm, lt:0, le:1, gt:2, ge:3, eq:4, neq:5, false:6, true:7>
<xop_sign:opc, :00, u:20>
x86: fold a few XOP templates Also add a new test case verifying that mixed operands of SIMD insns with a size-less memory operand in the middle are properly rejected.
2018-03-22 08:29:45 +01:00
vfrczpd, 2, 0x81, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
vfrczps, 2, 0x80, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfrczsd, 2, 0x83, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Qword|RegXMM|Unspecified|BaseIndex, RegXMM }
vfrczss, 2, 0x82, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Dword|RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop VexImmExt It's only used in assertions, and hence not really needed for correct code generation.
2018-04-26 08:30:06 +02:00
vpcmov, 4, 0xa2, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpcmov, 4, 0xa2, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: use template for XOP integer comparison, shift, and rotate insns These all follow common patterns.
2020-03-09 10:14:17 +01:00
vpcom<xop_sign><xop_elem>, 4, 0xcc | 0x<xop_sign:opc> | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=3|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpcom<xop_irel><xop_sign><xop_elem>, 3, 0xcc | 0x<xop_sign:opc> | <xop_elem:opc>, <xop_irel:imm>, 1, CpuXOP, Modrm|Vex128|VexOpcode=3|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop Vec_Imm4 It is pretty wasteful to have a per-operand flag which is used in exactly 4 cases. It can be relatively easily replaced, and by doing so I've actually found some dead code to remove at the same time (there's no case of ImmExt set at the same time as Vec_Imm4).
2019-07-01 08:38:50 +02:00
vpermil2pd, 5, 0x6649, None, 1, CpuXOP, Modrm|VexOpcode=2|VexVVVV=1|VexW=1|Vex|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpermil2pd, 5, 0x6649, None, 1, CpuXOP, Modrm|VexOpcode=2|VexVVVV=1|VexW=2|Vex|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpermil2ps, 5, 0x6648, None, 1, CpuXOP, Modrm|VexOpcode=2|VexVVVV=1|VexW=1|Vex|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
vpermil2ps, 5, 0x6648, None, 1, CpuXOP, Modrm|VexOpcode=2|VexVVVV=1|VexW=2|Vex|VexSources=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM, RegXMM|RegYMM }
x86: XOP VPHADD* / VPHSUB* are VEX.W0 Also avoid introducing further uses of VexW=1, by introducing and using VexW0 at this occasion. Move the marker past all #define-s.
2018-11-06 11:38:47 +01:00
vphaddbd, 2, 0xc2, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddbq, 2, 0xc3, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddbw, 2, 0xc1, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphadddq, 2, 0xcb, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddubd, 2, 0xd2, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddubq, 2, 0xd3, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddubw, 2, 0xd1, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddudq, 2, 0xdb, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphadduwd, 2, 0xd6, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphadduwq, 2, 0xd7, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddwd, 2, 0xc6, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphaddwq, 2, 0xc7, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphsubbw, 2, 0xe1, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphsubdq, 2, 0xe3, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
vphsubwd, 2, 0xe2, None, 1, CpuXOP, Modrm|VexOpcode=4|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from XOP and SSE4a insns
2018-09-13 11:21:36 +02:00
vpmacsdd, 4, 0x9e, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacsdqh, 4, 0x9f, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacsdql, 4, 0x97, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacssdd, 4, 0x8e, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacssdqh, 4, 0x8f, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacssdql, 4, 0x87, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacsswd, 4, 0x86, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacssww, 4, 0x85, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacswd, 4, 0x96, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmacsww, 4, 0x95, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmadcsswd, 4, 0xa6, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpmadcswd, 4, 0xb6, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpperm, 4, 0xa3, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpperm, 4, 0xa3, None, 1, CpuXOP, Modrm|VexOpcode=3|VexSources=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM, RegXMM }
x86: use template for XOP integer comparison, shift, and rotate insns These all follow common patterns.
2020-03-09 10:14:17 +01:00
vprot<xop_elem>, 3, 0x90 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW0|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM }
vprot<xop_elem>, 3, 0x90 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW1|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vprot<xop_elem>, 3, 0xc0 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=3|VexW0|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
vpsha<xop_elem>, 3, 0x98 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW0|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM }
vpsha<xop_elem>, 3, 0x98 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW1|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpshl<xop_elem>, 3, 0x94 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW0|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM, RegXMM|Unspecified|BaseIndex, RegXMM }
vpshl<xop_elem>, 3, 0x94 | <xop_elem:opc>, None, 1, CpuXOP, Modrm|Vex128|VexOpcode=4|VexW1|VexSources=1|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|No_qSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
2009-11-17 Sebastian Pop <sebastian.pop@amd.com> Quentin Neill <quentin.neill@amd.com> gas/ * config/tc-i386.c (cpu_arch): Added .xop and .cvt16. (build_vex_prefix): Handle xop08. (md_assemble): Don't special case the constant 3 for insns using MODRM. (build_modrm_byte): Handle vex2sources. (md_show_usage): Add xop and cvt16. * doc/c-i386.texi: Document fma4, xop, and cvt16. gas/testsuite/ * gas/i386/i386.exp: Run xop and cvt16 in 32-bit mode. Run x86-64-xop and x86-64-cvt16 in 64-bit mode. * gas/i386/lwp.d: Update name of the testcase. * gas/i386/x86-64-xop.d: New. * gas/i386/x86-64-xop.s: New. * gas/i386/xop.d: New. * gas/i386/xop.s: New. * gas/i386/cvt16.d: New. * gas/i386/cvt16.s: New. opcodes/ * i386-dis.c (OP_Vex_2src_1): New. (OP_Vex_2src_2): New. (Vex_2src_1): New. (Vex_2src_2): New. (XOP_08): Added. (VEX_LEN_XOP_08_A0): Added. (VEX_LEN_XOP_08_A1): Added. (VEX_LEN_XOP_09_80): Added. (VEX_LEN_XOP_09_81): Added. (xop_table): Added an entry for XOP_08. Handle xop instructions. (vex_len_table): Added entries for VEX_LEN_XOP_08_A0, VEX_LEN_XOP_08_A1, VEX_LEN_XOP_09_80, VEX_LEN_XOP_09_81. (get_valid_dis386): Handle XOP_08. (OP_Vex_2src): New. * i386-gen.c (cpu_flag_init): Add CPU_XOP_FLAGS and CPU_CVT16_FLAGS. (cpu_flags): Add CpuXOP and CpuCVT16. (opcode_modifiers): Add XOP08, Vex2Sources. * i386-opc.h (CpuXOP): Added. (CpuCVT16): Added. (i386_cpu_flags): Add cpuxop and cpucvt16. (XOP08): Added. (Vex2Sources): Added. (i386_opcode_modifier): Add xop08, vex2sources. * i386-opc.tbl: Add entries for XOP and CVT16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated.
2009-11-18 05:04:17 +01:00
2009-11-05 Sebastian Pop <sebastian.pop@amd.com> Quentin Neill <quentin.neill@amd.com> * gas/config/tc-i386.c (cpu_arch): Add CPU_LWP_FLAGS. (build_vex_prefix): Handle xop09 and xop0a. (build_modrm_byte): Handle vexlwp. (md_show_usage): Add lwp. * gas/doc/c-i386.texi (i386-LWP): New section. * gas/testsuite/gas/i386/i386.exp: Run x86-64-lwp in 64-bit mode, run lwp in 32-bit mode. * gas/testsuite/gas/i386/x86-64-lwp.d: New. * gas/testsuite/gas/i386/x86-64-lwp.s: New. * gas/testsuite/gas/i386/lwp.d: New. * gas/testsuite/gas/i386/lwp.s: New. * opcodes/i386-dis.c (OP_LWPCB_E): New. (OP_LWP_E): New. (OP_LWP_I): New. (USE_XOP_8F_TABLE): New. (XOP_8F_TABLE): New. (REG_XOP_LWPCB): New. (REG_XOP_LWP): New. (XOP_09): New. (XOP_0A): New. (reg_table): Redirect REG_8F to XOP_8F_TABLE. Add entries for REG_XOP_LWPCB and REG_XOP_LWP. (xop_table): New. (get_valid_dis386): Handle USE_XOP_8F_TABLE. Use the offsets VEX_0F, VEX_0F38, and VEX_0F3A instead of their values to access to the vex_table. (OP_LWPCB_E): New. (OP_LWP_E): New. (OP_LWP_I): New. * opcodes/i386-gen.c (cpu_flag_init): Add CPU_LWP_FLAGS, CpuLWP. (cpu_flags): Add CpuLWP. (opcode_modifiers): Add VexLWP, XOP09, and XOP0A. * opcodes/i386-opc.h (CpuLWP): New. (i386_cpu_flags): Add bit cpulwp. (VexLWP): New. (XOP09): New. (XOP0A): New. (i386_opcode_modifier): Add vexlwp, xop09, and xop0a. * opcodes/i386-opc.tbl (llwpcb): Added. (lwpval): Added. (lwpins): Added.
2009-11-06 00:40:05 +01:00
// LWP instructions
x86: fold LWP templates Also drop the no longer necessary explicit Disp<N> from them.
2018-03-08 08:56:08 +01:00
llwpcb, 1, 0x12, 0x0, 1, CpuLWP, Modrm|VexOpcode=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Reg32|Reg64 }
slwpcb, 1, 0x12, 0x1, 1, CpuLWP, Modrm|VexOpcode=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Vex, { Reg32|Reg64 }
lwpval, 3, 0x12, 0x1, 1, CpuLWP, Modrm|VexOpcode=5|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VexVVVV=3|Vex, { Imm32|Imm32S, Reg32|Unspecified|BaseIndex, Reg32|Reg64 }
lwpins, 3, 0x12, 0x0, 1, CpuLWP, Modrm|VexOpcode=5|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|VexVVVV=3|Vex, { Imm32|Imm32S, Reg32|Unspecified|BaseIndex, Reg32|Reg64 }
2009-11-05 Sebastian Pop <sebastian.pop@amd.com> Quentin Neill <quentin.neill@amd.com> * gas/config/tc-i386.c (cpu_arch): Add CPU_LWP_FLAGS. (build_vex_prefix): Handle xop09 and xop0a. (build_modrm_byte): Handle vexlwp. (md_show_usage): Add lwp. * gas/doc/c-i386.texi (i386-LWP): New section. * gas/testsuite/gas/i386/i386.exp: Run x86-64-lwp in 64-bit mode, run lwp in 32-bit mode. * gas/testsuite/gas/i386/x86-64-lwp.d: New. * gas/testsuite/gas/i386/x86-64-lwp.s: New. * gas/testsuite/gas/i386/lwp.d: New. * gas/testsuite/gas/i386/lwp.s: New. * opcodes/i386-dis.c (OP_LWPCB_E): New. (OP_LWP_E): New. (OP_LWP_I): New. (USE_XOP_8F_TABLE): New. (XOP_8F_TABLE): New. (REG_XOP_LWPCB): New. (REG_XOP_LWP): New. (XOP_09): New. (XOP_0A): New. (reg_table): Redirect REG_8F to XOP_8F_TABLE. Add entries for REG_XOP_LWPCB and REG_XOP_LWP. (xop_table): New. (get_valid_dis386): Handle USE_XOP_8F_TABLE. Use the offsets VEX_0F, VEX_0F38, and VEX_0F3A instead of their values to access to the vex_table. (OP_LWPCB_E): New. (OP_LWP_E): New. (OP_LWP_I): New. * opcodes/i386-gen.c (cpu_flag_init): Add CPU_LWP_FLAGS, CpuLWP. (cpu_flags): Add CpuLWP. (opcode_modifiers): Add VexLWP, XOP09, and XOP0A. * opcodes/i386-opc.h (CpuLWP): New. (i386_cpu_flags): Add bit cpulwp. (VexLWP): New. (XOP09): New. (XOP0A): New. (i386_opcode_modifier): Add vexlwp, xop09, and xop0a. * opcodes/i386-opc.tbl (llwpcb): Added. (lwpval): Added. (lwpins): Added.
2009-11-06 00:40:05 +01:00
Implement BMI instructions.
2011-01-05 01:16:57 +01:00
// BMI instructions
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
andn, 3, 0xf2, None, 1, CpuBMI, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64, Reg32|Reg64 }
bextr, 3, 0xf7, None, 1, CpuBMI, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=1|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blsi, 2, 0xf3, 0x3, 1, CpuBMI, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blsmsk, 2, 0xf3, 0x2, 1, CpuBMI, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blsr, 2, 0xf3, 0x1, 1, CpuBMI, Modrm|CheckRegSize|Vex128|VexOpcode=1|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
tzcnt, 2, 0xf30fbc, None, 2, CpuBMI, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
Implement BMI instructions.
2011-01-05 01:16:57 +01:00
Add support for TBM instructions. gas/ 2011-01-17 Quentin Neill <quentin.neill@amd.com> * config/tc-i386.c (cpu_arch): Add CPU_TBM_FLAGS. * doc/c-i386.texi (i386-TBM): New section. opcodes/ 2011-01-17 Quentin Neill <quentin.neill@amd.com> * i386-dis.c (REG_XOP_TBM_01): New. (REG_XOP_TBM_02): New. (reg_table): Add REG_XOP_TBM_01 and REG_XOP_TBM_02 tables. (xop_table): Redirect to REG_XOP_TBM_01 and REG_XOP_TBM_02 entries, and add bextr instruction. * i386-gen.c (cpu_flag_init): Add CPU_TBM_FLAGS, CpuTBM. (cpu_flags): Add CpuTBM. * i386-opc.h (CpuTBM) New. (i386_cpu_flags): Add bit cputbm. * i386-opc.tbl: Add bextr, blcfill, blci, blcic, blcmsk, blcs, blsfill, blsic, t1mskc, and tzmsk. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated gas/testsuite 2011-01-17 Quentin Neill <quentin.neill@amd.com> * gas/i386/tbm.s: New. * gas/i386/tbm.d: New. * gas/i386/tbm-intel.d: New. * gas/i386/x86-64-tbm.s: New. * gas/i386/x86-64-tbm.d: New. * gas/i386/x86-64-tbm-intel.d: New. * gas/i386/arch-10.d: Add tbm flag and TBM instruction pattern. * gas/i386/arch-10.s: Add a TBM instruction. * gas/i386/arch-10-1.l: Add TBM instruction pattern. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-arch-2.d: Likewise.
2011-01-17 19:40:36 +01:00
// TBM instructions
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
bextr, 3, 0x10, None, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=5|VexVVVV=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Imm32|Imm32S, Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blcfill, 2, 0x01, 0x1, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blci, 2, 0x02, 0x6, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blcic, 2, 0x01, 0x5, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blcmsk, 2, 0x02, 0x1, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blcs, 2, 0x01, 0x3, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blsfill, 2, 0x01, 0x2, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
blsic, 2, 0x01, 0x6, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
t1mskc, 2, 0x01, 0x7, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
tzmsk, 2, 0x01, 0x4, 1, CpuTBM, Modrm|CheckRegSize|Vex128|VexOpcode=4|VexVVVV=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Dword|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
Add support for TBM instructions. gas/ 2011-01-17 Quentin Neill <quentin.neill@amd.com> * config/tc-i386.c (cpu_arch): Add CPU_TBM_FLAGS. * doc/c-i386.texi (i386-TBM): New section. opcodes/ 2011-01-17 Quentin Neill <quentin.neill@amd.com> * i386-dis.c (REG_XOP_TBM_01): New. (REG_XOP_TBM_02): New. (reg_table): Add REG_XOP_TBM_01 and REG_XOP_TBM_02 tables. (xop_table): Redirect to REG_XOP_TBM_01 and REG_XOP_TBM_02 entries, and add bextr instruction. * i386-gen.c (cpu_flag_init): Add CPU_TBM_FLAGS, CpuTBM. (cpu_flags): Add CpuTBM. * i386-opc.h (CpuTBM) New. (i386_cpu_flags): Add bit cputbm. * i386-opc.tbl: Add bextr, blcfill, blci, blcic, blcmsk, blcs, blsfill, blsic, t1mskc, and tzmsk. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated gas/testsuite 2011-01-17 Quentin Neill <quentin.neill@amd.com> * gas/i386/tbm.s: New. * gas/i386/tbm.d: New. * gas/i386/tbm-intel.d: New. * gas/i386/x86-64-tbm.s: New. * gas/i386/x86-64-tbm.d: New. * gas/i386/x86-64-tbm-intel.d: New. * gas/i386/arch-10.d: Add tbm flag and TBM instruction pattern. * gas/i386/arch-10.s: Add a TBM instruction. * gas/i386/arch-10-1.l: Add TBM instruction pattern. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/x86-64-arch-2.d: Likewise.
2011-01-17 19:40:36 +01:00
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// AMD 3DNow! instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
prefetch, 1, 0xf0d, 0x0, 2, Cpu3dnow|CpuPRFCHW, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
prefetchw, 1, 0xf0d, 0x1, 2, Cpu3dnow|CpuPRFCHW, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
gas/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * config/tc-i386.c (md_assemble): Replace no_xsuf with no_ldsuf. (match_template): Likewise. opcodes/ 2007-11-01 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (opcode_modifiers): Replace No_xSuf with No_ldSuf. * i386-opc.tbl: Likewise. * i386-opc.h (No_xSuf): Renamed to ... (No_ldSuf): This. (FWait): Updated.
2007-11-01 20:06:54 +01:00
femms, 0, 0xf0e, None, 2, Cpu3dnow, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
pavgusb, 2, 0xf0f, 0xbf, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pf2id, 2, 0xf0f, 0x1d, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pf2iw, 2, 0xf0f, 0x1c, 2, Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfacc, 2, 0xf0f, 0xae, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfadd, 2, 0xf0f, 0x9e, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfcmpeq, 2, 0xf0f, 0xb0, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfcmpge, 2, 0xf0f, 0x90, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfcmpgt, 2, 0xf0f, 0xa0, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfmax, 2, 0xf0f, 0xa4, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfmin, 2, 0xf0f, 0x94, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfmul, 2, 0xf0f, 0xb4, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfnacc, 2, 0xf0f, 0x8a, 2, Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfpnacc, 2, 0xf0f, 0x8e, 2, Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfrcp, 2, 0xf0f, 0x96, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfrcpit1, 2, 0xf0f, 0xa6, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfrcpit2, 2, 0xf0f, 0xb6, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfrsqit1, 2, 0xf0f, 0xa7, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfrsqrt, 2, 0xf0f, 0x97, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfsub, 2, 0xf0f, 0x9a, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pfsubr, 2, 0xf0f, 0xaa, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pi2fd, 2, 0xf0f, 0xd, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pi2fw, 2, 0xf0f, 0xc, 2, Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pmulhrw, 2, 0xf0f, 0xb7, 2, Cpu3dnow, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
pswapd, 2, 0xf0f, 0xbb, 2, Cpu3dnowA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64|ImmExt, { Qword|Unspecified|BaseIndex|RegMMX, RegMMX }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// AMD extensions.
gas/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * gas/config/tc-i386.c (cpu_arch): Add corei7, .clflush and .syscall. (i386_align_code): Handle PROCESSOR_COREI7. (md_show_usage): Add corei7, clflush and syscall. (i386_target_format): Replace cpup4 with cpuclflush. * gas/config/tc-i386.h (processor_type): Add PROCESSOR_COREI7. * doc/c-i386.texi: Document corei7, clflush and syscall. gas/testsuite/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * gas/i386/arch-10.s: Add clflush and syscall. * gas/i386/x86-64-arch-2.s: Likewise. * gas/i386/arch-10.d: Updated. * gas/i386/arch-10-1.l: Likewise. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. opcodes/ 2009-01-10 H.J. Lu <hongjiu.lu@intel.com> * i386-gen.c (cpu_flag_init): Replace CpuP4 and CpuK6 with CpuClflush and CpuSYSCALL, respectively. Remove CpuK8. Add CPU_COREI7_FLAGS, CPU_CLFLUSH_FLAGS and CPU_SYSCALL_FLAGS. (cpu_flags): Remove CpuP4, CpuK6 and CpuK8. Add CpuClflush and CpuSYSCALL. (lineno): Removed. (set_bitfield): Take an argument, lineno. Don't report lineno on error if it is -1. (process_i386_cpu_flag): Take an argument, lineno. (process_i386_opcode_modifier): Likewise. (process_i386_operand_type): Likewise. (output_i386_opcode): Likewise. (opcode_hash_entry): Add lineno. (process_i386_opcodes): Updated. (process_i386_registers): Likewise. (process_i386_initializers): Likewise. * i386-opc.h (CpuP4): Removed. (CpuK6): Likewise. (CpuK8): Likewise. (CpuClflush): New. (CpuSYSCALL): Likewise. (CpuMMX): Updated. (i386_cpu_flags): Remove cpup4, cpuk6 and cpuk8. Add cpuclflush and cpusyscall. * i386-opc.tbl: Update movnti, clflush, lfence, mfence, pause, syscall and sysret. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2009-01-10 18:25:52 +01:00
syscall, 0, 0xf05, None, 2, CpuSYSCALL, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: improve handling of insns with ambiguous operand sizes Commit b76bc5d54e ("x86: don't default variable shift count insns to 8-bit operand size") pointed out a very bad case, but the underlying problem is, as mentioned on various occasions, much larger: Silently selecting a (nowhere documented afaict) certain default operand size when there's no "sizing" suffix and no suitable register operand(s) is simply dangerous (for the programmer to make mistakes). While in Intel syntax mode such mistakes already lead to an error (which is going to remain that way), AT&T syntax mode now gains warnings in such cases by default, which can be suppressed or promoted to an error if so desired by the programmer. Furthermore at least general purpose insns now consistently have a default applied (alongside the warning emission), rather than accepting some and refusing others. No warnings are (as before) to be generated for "DefaultSize" insns as well as ones acting on selector and other fixed-width values. For SYSRET, however, the DefaultSize needs to be dropped - it had been wrongly put there in the first place, as it's unrelated to .code16gcc (no stack accesses involved). As set forth as a prereq when I first mentioned this intended change a few years back, Linux as well as gcc have meanwhile been patched to avoid (emission of) ambiguous operands (and hence triggering of the new warning). Note that I think that in 64-bit mode IRET and far RET would better get a diagnostic too, as it's reasonably likely that a suffix-less instance really is meant to be a 64-bit one. But I guess I better make this a separate follow-on patch. Note further that floating point operations with integer operands are an exception for now: They continue to use short (16-bit) operands by default even in 32- and 64-bit modes. Finally note that while {,V}PCMPESTR{I,M} would, strictly speaking, also need to be diagnosed, with their 64-bit forms not being very useful I think it is better to continue to avoid warning about them (by way of them carrying IgnoreSize attributes).
2020-01-21 08:28:25 +01:00
sysret, 0, 0xf07, None, 2, CpuSYSCALL, No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { 0 }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
swapgs, 0, 0xf01f8, None, 3, Cpu64, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
rdtscp, 0, 0xf01f9, None, 3, CpuRdtscp, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// AMD Pacifica additions.
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
clgi, 0, 0xf01dd, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
invlpga, 0, 0xf01df, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
invlpga, 2, 0xf01df, None, 3, CpuSVME, AddrPrefixOpReg, { Acc|Word|Dword|Qword, RegC|Dword }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
skinit, 0, 0xf01de, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: eliminate ImmExt abuse Drop the remaining instances left in place by commit c3949f432f ("x86: limit ImmExt abuse), now that we have a way to specify specific GPRs. Take the opportunity and also introduce proper 16-bit forms of applicable SVME insns as well as 1-operand forms of CLZERO.
2019-11-12 09:08:32 +01:00
skinit, 1, 0xf01de, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Dword }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
stgi, 0, 0xf01dc, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: support VMGEXIT It has been publicly documented for quite some time, albeit not in the "General-Purpose and System Instructions" volume: https://www.amd.com/system/files/TechDocs/24593.pdf.
2020-03-04 08:58:13 +01:00
vmgexit, 0, 0xf30f01d9, None, 3, CpuSEV_ES, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmload, 0, 0xf01da, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
vmload, 1, 0xf01da, None, 3, CpuSVME, AddrPrefixOpReg, { Acc|Word|Dword|Qword }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmmcall, 0, 0xf01d9, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
vmrun, 0, 0xf01d8, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
vmrun, 1, 0xf01d8, None, 3, CpuSVME, AddrPrefixOpReg, { Acc|Word|Dword|Qword }
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
vmsave, 0, 0xf01db, None, 3, CpuSVME, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
vmsave, 1, 0xf01db, None, 3, CpuSVME, AddrPrefixOpReg, { Acc|Word|Dword|Qword }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// SSE4a instructions
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
movntsd, 2, 0xf20f2b, None, 2, CpuSSE4a, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
movntss, 2, 0xf30f2b, None, 2, CpuSSE4a, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from XOP and SSE4a insns
2018-09-13 11:21:36 +02:00
extrq, 3, 0x660f78, 0x0, 2, CpuSSE4a, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Imm8, RegXMM }
extrq, 2, 0x660f79, None, 2, CpuSSE4a, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
insertq, 2, 0xf20f79, None, 2, CpuSSE4a, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM }
insertq, 4, 0xf20f78, None, 2, CpuSSE4a, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Imm8, RegXMM, RegXMM }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
x86: Remove CpuABM and add CpuPOPCNT AMD ABM has 2 instructions: popcnt and lzcnt. ABM CPUID feature bit has been reused for lzcnt and a POPCNT CPUID feature bit is added for popcnt which used to be the part of SSE4.2. This patch removes CpuABM and adds CpuPOPCNT. It changes ABM to enable both lzcnt and popcnt, changes SSE4.2 to also enable popcnt. gas/ * config/tc-i386.c (cpu_arch): Add .popcnt. * doc/c-i386.texi: Remove abm and .abm. Add popcnt and .popcnt. Add a tab before @samp{.sse4a}. opcodes/ * i386-gen.c (cpu_flag_init): Replace CpuABM with CpuLZCNT|CpuPOPCNT. Add CpuPOPCNT to CPU_SSE4_2_FLAGS. Add CPU_POPCNT_FLAGS. (cpu_flags): Remove CpuABM. Add CpuPOPCNT. * i386-opc.h (CpuABM): Removed. (CpuPOPCNT): New. (i386_cpu_flags): Remove cpuabm. Add cpupopcnt. * i386-opc.tbl: Replace CpuABM|CpuSSE4_2 with CpuPOPCNT on popcnt. Remove CpuABM from lzcnt. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2020-02-17 16:12:10 +01:00
// LZCNT instruction
lzcnt, 2, 0xf30fbd, None, 2, CpuLZCNT, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
// POPCNT instruction
popcnt, 2, 0xf30fb8, None, 2, CpuPOPCNT, Modrm|CheckRegSize|No_bSuf|No_sSuf|No_ldSuf|NoAVX, { Reg16|Reg32|Reg64|Word|Dword|Qword|Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// VIA PadLock extensions.
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
xstore-rng, 0, 0xfa7c0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcrypt-ecb, 0, 0xf30fa7c8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcrypt-cbc, 0, 0xf30fa7d0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcrypt-ctr, 0, 0xf30fa7d8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcrypt-cfb, 0, 0xf30fa7e0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcrypt-ofb, 0, 0xf30fa7e8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
montmul, 0, 0xf30fa6c0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xsha1, 0, 0xf30fa6c8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xsha256, 0, 0xf30fa6d0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Aliases without hyphens.
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
xstorerng, 0, 0xfa7c0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcryptecb, 0, 0xf30fa7c8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcryptcbc, 0, 0xf30fa7d0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcryptctr, 0, 0xf30fa7d8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcryptcfb, 0, 0xf30fa7e0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
xcryptofb, 0, 0xf30fa7e8, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
2007-06-28 H.J. Lu <hongjiu.lu@intel.com> * Makefile.am (HFILES): Add i386-opc.h and i386-tbl.h. (CFILES): Add i386-gen.c. (i386-gen): New rule. (i386-gen.o): Likewise. (i386-tbl.h): Likewise. Run "make dep-am". * Makefile.in: Regenerated. * i386-gen.c: New file. * i386-opc.tbl: Likewise. * i386-reg.tbl: Likewise. * i386-tbl.h: Likewise. * i386-opc.c: Include "i386-tbl.h". (i386_optab): Removed. (i386_regtab): Likewise. (i386_regtab_size): Likewise.
2007-06-28 16:29:56 +02:00
// Alias for xstore-rng.
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
xstore, 0, 0xfa7c0, None, 3, CpuPadLock, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsString|RepPrefixOk, { 0 }
Implement RDRSEED, ADX and PRFCHW instructions gas/ * config/tc-i386.c: Add ADX, RDSEED and PRFCHW asm directives. * doc/c-i386.texi: Document the new directives. gas/testsuite/ * gas/i386/i386.exp: Run adx, rdseed and prefetchw tests. * gas/i386/x86-64-arch-2.s: Use prefetchw as 3dnow and Prfchw tests. * gas/i386/arch-10.s: Likewise. * gas/i386/arch-10-1.l: Changed correspondingly. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-lzcnt.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/ilp32/x86-64-arch-2.d: Likewise. * gas/i386/arch-10-prefetchw.d: New file. * gas/i386/x86-64-arch-2-prefetchw.d: Likewise. * gas/i386/rdseed.s: Likewise. * gas/i386/rdseed.d: Likewise. * gas/i386/rdseed-intel.d: Likewise. * gas/i386/adx.s: Likewise. * gas/i386/adx.d: Likewise. * gas/i386/adx-intel.d: Likewise. * gas/i386/x86-64-rdseed.s: Likewise. * gas/i386/x86-64-rdseed.d: Likewise. * gas/i386/x86-64-rdseed-intel.d: Likewise. * gas/i386/x86-64-adx.s: Likewise. * gas/i386/x86-64-adx.d: Likewise. * gas/i386/x86-64-adx-intel.d: Likewise. opcodes/ * i386-dis.c (PREFIX_0F38F6): New. (prefix_table): Add adcx, adox instructions. (three_byte_table): Use PREFIX_0F38F6. (mod_table): Add rdseed instruction. * i386-gen.c (cpu_flag_init): Add CpuADX, CpuRDSEED, CpuPRFCHW. (cpu_flags): Likewise. * i386-opc.h: Add CpuADX, CpuRDSEED, CpuPRFCHW. (i386_cpu_flags): Add fields cpurdseed, cpuadx, cpuprfchw. * i386-opc.tbl: Add instrcutions adcx, adox, rdseed. Extend prefetchw. * i386-tbl.h: Regenerate. * i386-init.h: Likewise.
2012-07-16 14:58:29 +02:00
// Multy-precision Add Carry, rdseed instructions.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
adcx, 2, 0x660f38f6, None, 3, CpuADX, Modrm|CheckRegSize|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Unspecified|BaseIndex, Reg32|Reg64 }
adox, 2, 0xf30f38f6, None, 3, CpuADX, Modrm|CheckRegSize|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Unspecified|BaseIndex, Reg32|Reg64 }
Implement RDRSEED, ADX and PRFCHW instructions gas/ * config/tc-i386.c: Add ADX, RDSEED and PRFCHW asm directives. * doc/c-i386.texi: Document the new directives. gas/testsuite/ * gas/i386/i386.exp: Run adx, rdseed and prefetchw tests. * gas/i386/x86-64-arch-2.s: Use prefetchw as 3dnow and Prfchw tests. * gas/i386/arch-10.s: Likewise. * gas/i386/arch-10-1.l: Changed correspondingly. * gas/i386/arch-10-2.l: Likewise. * gas/i386/arch-10-3.l: Likewise. * gas/i386/arch-10-4.l: Likewise. * gas/i386/arch-10.d: Likewise. * gas/i386/arch-10-lzcnt.d: Likewise. * gas/i386/x86-64-arch-2.d: Likewise. * gas/i386/x86-64-arch-2-lzcnt.d: Likewise. * gas/i386/ilp32/x86-64-arch-2.d: Likewise. * gas/i386/arch-10-prefetchw.d: New file. * gas/i386/x86-64-arch-2-prefetchw.d: Likewise. * gas/i386/rdseed.s: Likewise. * gas/i386/rdseed.d: Likewise. * gas/i386/rdseed-intel.d: Likewise. * gas/i386/adx.s: Likewise. * gas/i386/adx.d: Likewise. * gas/i386/adx-intel.d: Likewise. * gas/i386/x86-64-rdseed.s: Likewise. * gas/i386/x86-64-rdseed.d: Likewise. * gas/i386/x86-64-rdseed-intel.d: Likewise. * gas/i386/x86-64-adx.s: Likewise. * gas/i386/x86-64-adx.d: Likewise. * gas/i386/x86-64-adx-intel.d: Likewise. opcodes/ * i386-dis.c (PREFIX_0F38F6): New. (prefix_table): Add adcx, adox instructions. (three_byte_table): Use PREFIX_0F38F6. (mod_table): Add rdseed instruction. * i386-gen.c (cpu_flag_init): Add CpuADX, CpuRDSEED, CpuPRFCHW. (cpu_flags): Likewise. * i386-opc.h: Add CpuADX, CpuRDSEED, CpuPRFCHW. (i386_cpu_flags): Add fields cpurdseed, cpuadx, cpuprfchw. * i386-opc.tbl: Add instrcutions adcx, adox, rdseed. Extend prefetchw. * i386-tbl.h: Regenerate. * i386-init.h: Likewise.
2012-07-16 14:58:29 +02:00
rdseed, 1, 0xfc7, 0x7, 2, CpuRdSeed, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg16|Reg32|Reg64 }
Implement Intel SMAP instructions gas/ PR gas/15159 * config/tc-i386.c (cpu_arch): Add ".smap". * doc/c-i386.texi: Document smap. gas/testsuite/ PR gas/15159 * gas/i386/i386.exp: Run smap and x86-64-smap. * gas/i386/smap.d: New file. * gas/i386/smap.s: likewise. * gas/i386/x86-64-smap.d: likewise. opcodes/ PR gas/15159 * i386-dis.c (rm_table): Add clac and stac to RM_0F01_REG_1. * i386-gen.c (cpu_flag_init): Add CPU_SMAP_FLAGS. (cpu_flags): Add CpuSMAP. * i386-opc.h (CpuSMAP): New. (i386_cpu_flags): Add cpusmap. * i386-opc.tbl: Add clac and stac. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2013-02-19 20:10:31 +01:00
// SMAP instructions.
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
clac, 0, 0xf01ca, None, 3, CpuSMAP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
stac, 0, 0xf01cb, None, 3, CpuSMAP, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Support Intel MPX gas/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386.c (BND_PREFIX): New. (struct _i386_insn): Add new field bnd_prefix. (add_bnd_prefix): New. (cpu_arch): Add MPX. (i386_operand_type): Add regbnd. (md_assemble): Handle BND prefixes. (parse_insn): Likewise. (output_branch): Likewise. (output_jump): Likewise. (build_modrm_byte): Handle regbnd. (OPTION_MADD_BND_PREFIX): New. (md_longopts): Add entry for 'madd-bnd-prefix'. (md_parse_option): Handle madd-bnd-prefix option. (md_show_usage): Add description for madd-bnd-prefix option. * doc/c-i386.texi: Document mpx/.mpx and -madd-bnd-prefix. gas/testsuite/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/i386/mpx-add-bnd-prefix.s: New. * gas/i386/mpx-add-bnd-prefix.d: New. * gas/i386/mpx-inval-1.l: New. * gas/i386/mpx-inval-1.s: New. * gas/i386/mpx.d: New. * gas/i386/mpx.s: New. * gas/i386/x86-64-mpx-add-bnd-prefix.d: New. * gas/i386/x86-64-mpx-add-bnd-prefix.s: New. * gas/i386/x86-64-mpx-addr32.d: New. * gas/i386/x86-64-mpx-addr32.s: New. * gas/i386/x86-64-mpx-inval-1.l: New. * gas/i386/x86-64-mpx-inval-1.s: New. * gas/i386/x86-64-mpx-inval-2.l: New. * gas/i386/x86-64-mpx-inval-2.s: New. * gas/i386/x86-64-mpx.d: New. * gas/i386/x86-64-mpx.s: New. * gas/i386/nops.d: Adjust to MPX changes. * gas/i386/nops.s: Likewise. * gas/i386/x86-64-nops.d: Likewise. * gas/i386/x86-64-nops.s: Likewise. * gas/i386/ilp32/x86-64-nops.d: Likewise. * gas/i386/i386.exp: Run new MPX tests. include/opcode/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386.h (BND_PREFIX_OPCODE): New. opcodes/ 2013-07-24 Anna Tikhonova <anna.tikhonova@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis.c (BND_Fixup): New. (Ebnd): New. (Ev_bnd): New. (Gbnd): New. (BND): New. (v_bnd_mode): New. (bnd_mode): New. (MOD enum): Add new entries. (PREFIX enum): Likewise. (dis tables): Replace XX with BND for near branch and call instructions. (prefix_table): Add new entries. (mod_table): Likewise. (names_bnd): New. (intel_names_bnd): New. (att_names_bnd): New. (BND_PREFIX): New. (prefix_name): Handle BND_PREFIX. (print_insn): Initialize names_bnd. (intel_operand_size): Handle new modes. (OP_E_register): Likewise. (OP_E_memory): Likewise. (OP_G): Likewise. * i386-gen.c (cpu_flag_init): Add CpuMPX. (cpu_flags): Add CpuMPX. (operand_type_init): Add RegBND. (opcode_modifiers): Add BNDPrefixOk. (operand_types): Add RegBND. * i386-init.h: Regenerate. * i386-opc.h (CpuMPX): New. (CpuUnused): Comment out. (i386_cpu_flags): Add cpumpx. (BNDPrefixOk): New. (i386_opcode_modifier): Add bndprefixok. (RegBND): New. (i386_operand_type): Add regbnd. * i386-opc.tbl: Add BNDPrefixOk to near jumps, calls and rets. Add MPX instructions and bnd prefix. * i386-reg.tbl: Add bnd0-bnd3 registers. * i386-tbl.h: Regenerate.
2013-07-24 17:47:25 +02:00
// BND prefix
bnd, 0, 0xf2, None, 1, CpuMPX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
// MPX instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
bndmk, 2, 0xf30f1b, None, 2, CpuMPX, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex, RegBND }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
bndmov, 2, 0x660f1a, None, 2, CpuMPX, D|Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Xmmword|Unspecified|BaseIndex|RegBND, RegBND }
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
bndcl, 2, 0xf30f1a, None, 2, CpuMPX|CpuNo64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|BaseIndex, RegBND }
bndcl, 2, 0xf30f1a, None, 2, CpuMPX|Cpu64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Reg64|BaseIndex, RegBND }
bndcu, 2, 0xf20f1a, None, 2, CpuMPX|CpuNo64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|BaseIndex, RegBND }
bndcu, 2, 0xf20f1a, None, 2, CpuMPX|Cpu64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Reg64|BaseIndex, RegBND }
bndcn, 2, 0xf20f1b, None, 2, CpuMPX|CpuNo64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|BaseIndex, RegBND }
bndcn, 2, 0xf20f1b, None, 2, CpuMPX|Cpu64, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Reg64|BaseIndex, RegBND }
bndstx, 2, 0x0f1b, None, 2, CpuMPX, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegBND, BaseIndex }
bndldx, 2, 0x0f1a, None, 2, CpuMPX, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex, RegBND }
Support Intel SHA gas/ 2013-07-25 Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386.c (cpu_arch): Add .sha. * doc/c-i386.texi: Document sha/.sha. gas/testsuite/ 2013-07-25 Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/i386/sha.d: New. * gas/i386/sha.s: New. * gas/i386/x86-64-sha.d: New. * gas/i386/x86-64-sha.s: New. * gas/i386/i386.exp: Run new SHA tests. opcodes/ 2013-07-25 Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis.c (PREFIX enum): Add PREFIX_0F38C8, PREFIX_0F38C9, PREFIX_0F38CA, PREFIX_0F38CB, PREFIX_0F38CC, PREFIX_0F38CD, PREFIX_0F3ACC. (prefix_table): Updated. (three_byte_table): Likewise. * i386-gen.c (cpu_flag_init): Add CPU_SHA_FLAGS. (cpu_flags): Add CpuSHA. (i386_cpu_flags): Add cpusha. * i386-init.h: Regenerate. * i386-opc.h (CpuSHA): New. (CpuUnused): Restored. (i386_cpu_flags): Add cpusha. * i386-opc.tbl: Add SHA instructions. * i386-tbl.h: Regenerate.
2013-07-25 18:16:35 +02:00
// SHA instructions.
x86: drop bogus IgnoreSize from SHA insns
2018-09-13 11:22:03 +02:00
sha1rnds4, 3, 0xf3acc, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegXMM }
sha1nexte, 2, 0xf38c8, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
sha1msg1, 2, 0xf38c9, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
sha1msg2, 2, 0xf38ca, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
sha256rnds2, 3, 0xf38cb, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Acc|Xmmword, RegXMM|Unspecified|BaseIndex, RegXMM }
sha256rnds2, 2, 0xf38cb, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
sha256msg1, 2, 0xf38cc, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
sha256msg2, 2, 0xf38cd, None, 3, CpuSHA, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
Enable Intel VPCLMULQDQ instruction. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VPCLMULQDQ. * doc/c-i386.texi: Document VPCLMULQDQ. * testsuite/gas/i386/i386.exp: Run VPCLMULQDQ tests. * testsuite/gas/i386/avx512f_vpclmulqdq-intel.d: New test. * testsuite/gas/i386/avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/vpclmulqdq.d: Ditto. * testsuite/gas/i386/vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3A44. (enum): Remove VEX_LEN_0F3A44_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F3A44_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VPCLMULQDQ. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. (CpuUnused): Comment out to avoid zero-width field problem. * i386-opc.tbl (vpclmulqdq): New instruction. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:42:40 +02:00
// VPCLMULQDQ instructions
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpclmulqdq, 4, 0x6644, None, 1, CpuVPCLMULQDQ, Modrm|Vex=2|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
x86: add missing pseudo ops for VPCLMULQDQ ISA extension While the ISA extensions doc suggests them to be made available just like the SDM does for the PCLMULQDQ ISA extension, these weren't added when supposrt for the new extension was introduced. Also make sure the 64-bit non-AVX512 test actually tests VEX encodings, not EVEX ones.
2019-07-01 08:28:58 +02:00
vpclmullqlqdq, 3, 0x6644, 0x00, 1, CpuVPCLMULQDQ, Modrm|Vex=2|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpclmulhqlqdq, 3, 0x6644, 0x01, 1, CpuVPCLMULQDQ, Modrm|Vex=2|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpclmullqhqdq, 3, 0x6644, 0x10, 1, CpuVPCLMULQDQ, Modrm|Vex=2|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
vpclmulhqhqdq, 3, 0x6644, 0x11, 1, CpuVPCLMULQDQ, Modrm|Vex=2|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { Unspecified|BaseIndex|RegYMM, RegYMM, RegYMM }
Enable Intel VPCLMULQDQ instruction. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VPCLMULQDQ. * doc/c-i386.texi: Document VPCLMULQDQ. * testsuite/gas/i386/i386.exp: Run VPCLMULQDQ tests. * testsuite/gas/i386/avx512f_vpclmulqdq-intel.d: New test. * testsuite/gas/i386/avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/vpclmulqdq.d: Ditto. * testsuite/gas/i386/vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3A44. (enum): Remove VEX_LEN_0F3A44_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F3A44_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VPCLMULQDQ. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. (CpuUnused): Comment out to avoid zero-width field problem. * i386-opc.tbl (vpclmulqdq): New instruction. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:42:40 +02:00
// VPCLMULQDQ instructions end
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
// AVX512F instructions.
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
#define Disp8ShiftVL Disp8MemShift=DISP8_SHIFT_VL
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
#define MaskingMorZ Masking=DYNAMIC_MASKING
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
kandnw, 3, 0x42, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandw, 3, 0x41, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
korw, 3, 0x45, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxnorw, 3, 0x46, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxorw, 3, 0x47, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
kmovw, 2, 0x90, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask|Word|Unspecified|BaseIndex, RegMask }
kmovw, 2, 0x91, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, Word|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
kmovw, 2, 0x92, None, 1, CpuAVX512F, D|Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
knotw, 2, 0x44, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
kortestw, 2, 0x98, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
kshiftlw, 3, 0x6632, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
kshiftrw, 3, 0x6630, None, 1, CpuAVX512F, Modrm|Vex=1|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
kunpckbw, 3, 0x664B, None, 1, CpuAVX512F, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vaddpd, 3, 0x6658, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vaddpd, 4, 0x6658, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vdivpd, 3, 0x665E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vdivpd, 4, 0x665E, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmulpd, 3, 0x6659, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmulpd, 4, 0x6659, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vsubpd, 3, 0x665C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsubpd, 4, 0x665C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vaddps, 3, 0x58, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vaddps, 4, 0x58, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vdivps, 3, 0x5E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vdivps, 4, 0x5E, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmulps, 3, 0x59, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmulps, 4, 0x59, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vsubps, 3, 0x5C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsubps, 4, 0x5C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vaddsd, 3, 0xF258, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vaddsd, 4, 0xF258, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vdivsd, 3, 0xF25E, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vdivsd, 4, 0xF25E, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmulsd, 3, 0xF259, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmulsd, 4, 0xF259, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsqrtsd, 3, 0xF251, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsqrtsd, 4, 0xF251, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsubsd, 3, 0xF25C, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsubsd, 4, 0xF25C, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vaddss, 3, 0xF358, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vaddss, 4, 0xF358, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vdivss, 3, 0xF35E, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vdivss, 4, 0xF35E, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmulss, 3, 0xF359, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmulss, 4, 0xF359, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsqrtss, 3, 0xF351, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsqrtss, 4, 0xF351, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vsubss, 3, 0xF35C, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsubss, 4, 0xF35C, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
valignd, 4, 0x6603, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpternlogd, 4, 0x6625, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
valignq, 4, 0x6603, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpternlogq, 4, 0x6625, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vblendmpd, 3, 0x6665, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpblendmq, 3, 0x6664, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2pd, 3, 0x6677, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2q, 3, 0x6676, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2pd, 3, 0x667F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2q, 3, 0x667E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxsq, 3, 0x663D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxuq, 3, 0x663F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminsq, 3, 0x6639, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminuq, 3, 0x663B, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmuldq, 3, 0x6628, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vprolvq, 3, 0x6615, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vprorvq, 3, 0x6614, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsllvq, 3, 0x6647, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsravq, 3, 0x6646, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsrlvq, 3, 0x6645, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vblendmps, 3, 0x6665, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpblendmd, 3, 0x6664, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2d, 3, 0x6676, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2ps, 3, 0x6677, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2d, 3, 0x667E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2ps, 3, 0x667F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxsd, 3, 0x663D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxud, 3, 0x663F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminsd, 3, 0x6639, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminud, 3, 0x663B, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmulld, 3, 0x6640, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vprolvd, 3, 0x6615, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vprorvd, 3, 0x6614, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsllvd, 3, 0x6647, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsravd, 3, 0x6646, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsrlvd, 3, 0x6645, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vbroadcastf32x4, 2, 0x661A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { XMMword|Unspecified|BaseIndex, RegYMM|RegZMM }
vbroadcasti32x4, 2, 0x665A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { XMMword|Unspecified|BaseIndex, RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vbroadcastf64x4, 2, 0x661B, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { YMMword|Unspecified|BaseIndex, RegZMM }
vbroadcasti64x4, 2, 0x665B, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { YMMword|Unspecified|BaseIndex, RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vbroadcastss, 2, 0x6618, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vbroadcastsd, 2, 0x6619, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpbroadcastd, 2, 0x6658, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpbroadcastd, 2, 0x667C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
vcmp<avx_frel>pd, 3, 0x66C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV|VexW1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vcmp<avx_frel>pd, 4, 0x66C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVex512|Masking=2|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SAE, { Imm8, RegZMM, RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcmppd, 4, 0x66C2, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcmppd, 5, 0x66C2, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
vcmp<avx_frel>ps, 3, 0xC2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vcmp<avx_frel>ps, 4, 0xC2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVex512|Masking=2|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SAE, { Imm8, RegZMM, RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcmpps, 4, 0xC2, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcmpps, 5, 0xC2, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
vcmp<avx_frel>sd, 3, 0xF2C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVexLIG|Masking=2|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegMask }
vcmp<avx_frel>sd, 4, 0xF2C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVexLIG|Masking=2|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SAE, { Imm8, RegXMM, RegXMM, RegMask }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcmpsd, 4, 0xF2C2, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=2|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegMask }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcmpsd, 5, 0xF2C2, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use template for AVX/AVX512 floating point comparison insns These all follow an almost common pattern, again with the exception of being commutative, which can be easily taken care of. Note that, as an intended side effect (and in fact one of the reason to introduce templates), AVX long-form pseudo-ops get introduced alongside the already existing AVX512 ones.
2020-03-09 10:13:43 +01:00
vcmp<avx_frel>ss, 3, 0xF3C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVexLIG|Masking=2|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegMask }
vcmp<avx_frel>ss, 4, 0xF3C2, 0x<avx_frel:imm>, 1, CpuAVX512F, Modrm|EVexLIG|Masking=2|VexOpcode=0|VexVVVV|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt|SAE, { Imm8, RegXMM, RegXMM, RegMask }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcmpss, 4, 0xF3C2, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=2|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegMask }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcmpss, 5, 0xF3C2, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegMask }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcomisd, 2, 0x662F, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcomisd, 3, 0x662F, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vucomisd, 2, 0x662E, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vucomisd, 3, 0x662E, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcomiss, 2, 0x2F, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcomiss, 3, 0x2F, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vucomiss, 2, 0x2E, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vucomiss, 3, 0x2E, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vcompresspd, 2, 0x668A, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=1|VexW=2|Disp8MemShift=3|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
vcompressps, 2, 0x668A, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
vpcompressq, 2, 0x668B, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=1|VexW=2|Disp8MemShift=3|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
vpcompressd, 2, 0x668B, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpscatterdq, 2, 0x66A0, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
vpscatterqq, 2, 0x66A1, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
vscatterdpd, 2, 0x66A2, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
vscatterqpd, 2, 0x66A3, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpscatterdd, 2, 0x66A0, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
vscatterdps, 2, 0x66A2, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, Dword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtdq2pd, 2, 0xF3E6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvtudq2pd, 2, 0xF37A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvtdq2ps, 2, 0x5B, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtdq2ps, 3, 0x5B, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvtps2udq, 2, 0x79, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtps2udq, 3, 0x79, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vsqrtps, 2, 0x51, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsqrtps, 3, 0x51, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtpd2dq, 2, 0xF2E6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegYMM }
vcvtpd2dq, 3, 0xF2E6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtpd2ps, 2, 0x665A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegYMM }
vcvtpd2ps, 3, 0x665A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtpd2udq, 2, 0x79, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegYMM }
vcvtpd2udq, 3, 0x79, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtph2ps, 2, 0x6613, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
vcvtph2ps, 3, 0x6613, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegYMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvtps2dq, 2, 0x665B, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtps2dq, 3, 0x665B, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtps2pd, 2, 0x5A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvtps2pd, 3, 0x5A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegYMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vcvtps2ph, 3, 0x661D, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=2|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM, RegYMM|Unspecified|BaseIndex }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vcvtps2ph, 4, 0x661D, None, 1, CpuAVX512F, RegMem|EVex=1|Masking=3|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold to-scalar-int conversion insns
2018-03-28 14:22:56 +02:00
vcvtsd2si, 2, 0xF22D, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { RegXMM|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtsd2si, 3, 0xF22D, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, Reg32|Reg64 }
x86: fold to-scalar-int conversion insns
2018-03-28 14:22:56 +02:00
vcvtsd2usi, 2, 0xF279, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ToDword, { RegXMM|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtsd2usi, 3, 0xF279, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, Reg32|Reg64 }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtsd2ss, 3, 0xF25A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtsd2ss, 4, 0xF25A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
vcvtsi2sd, 3, 0xF22A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtsi2sd, 3, 0xF22A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: fold certain VCVT{,U}SI2S{S,D} templates There don't really need to be separate Cpu64 and CpuNo64 templates for these. One small issue with this is that slightly strange code .intel_syntax noprefix .code16 .arch i286 .arch .avx vcvtsi2sd xmm0, xmm0, dword ptr [bx] vcvtsi2sd xmm0, xmm0, qword ptr [bx] vcvtsi2sd xmm0, xmm0, ebx vcvtsi2sd xmm0, xmm0, rbx now will match in behavior with the AVX512 counterparts in that not only the 2nd vcvtsi2sd won't assemble, but also the first. The last two, otoh, will continue to assemble fine (due to the lack of any memory operand size specifier). As a result, another way to make things behave more consistently would be to avoid the folding and add IgnoreSize to the CpuNo64 AVX512 variants. A 3rd way to do so would be to add Cpu386 to any such insn template. While doing this also make the usual cosmetic adjustments for the insns touched anyway. Additionally drop the redundant Cpu64 from the SAE forms of VCVT{,U}SI2SD - they won't assemble outside of 64-bit mode due to there not being anything to match the Reg64 operand.
2020-02-17 08:59:52 +01:00
vcvtsi2sd, 4, 0xF22A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Reg64, Imm8, RegXMM, RegXMM }
vcvtsi2sd, 4, 0xF22A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|StaticRounding|SAE|IntelSyntax, { Imm8, Reg64, RegXMM, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
vcvtusi2sd, 3, 0xF27B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtusi2sd, 3, 0xF27B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: fold certain VCVT{,U}SI2S{S,D} templates There don't really need to be separate Cpu64 and CpuNo64 templates for these. One small issue with this is that slightly strange code .intel_syntax noprefix .code16 .arch i286 .arch .avx vcvtsi2sd xmm0, xmm0, dword ptr [bx] vcvtsi2sd xmm0, xmm0, qword ptr [bx] vcvtsi2sd xmm0, xmm0, ebx vcvtsi2sd xmm0, xmm0, rbx now will match in behavior with the AVX512 counterparts in that not only the 2nd vcvtsi2sd won't assemble, but also the first. The last two, otoh, will continue to assemble fine (due to the lack of any memory operand size specifier). As a result, another way to make things behave more consistently would be to avoid the folding and add IgnoreSize to the CpuNo64 AVX512 variants. A 3rd way to do so would be to add Cpu386 to any such insn template. While doing this also make the usual cosmetic adjustments for the insns touched anyway. Additionally drop the redundant Cpu64 from the SAE forms of VCVT{,U}SI2SD - they won't assemble outside of 64-bit mode due to there not being anything to match the Reg64 operand.
2020-02-17 08:59:52 +01:00
vcvtusi2sd, 4, 0xF27B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Reg64, Imm8, RegXMM, RegXMM }
vcvtusi2sd, 4, 0xF27B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_ldSuf|StaticRounding|SAE|IntelSyntax, { Imm8, Reg64, RegXMM, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
vcvtsi2ss, 3, 0xF32A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtsi2ss, 3, 0xF32A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: fold certain VCVT{,U}SI2S{S,D} templates There don't really need to be separate Cpu64 and CpuNo64 templates for these. One small issue with this is that slightly strange code .intel_syntax noprefix .code16 .arch i286 .arch .avx vcvtsi2sd xmm0, xmm0, dword ptr [bx] vcvtsi2sd xmm0, xmm0, qword ptr [bx] vcvtsi2sd xmm0, xmm0, ebx vcvtsi2sd xmm0, xmm0, rbx now will match in behavior with the AVX512 counterparts in that not only the 2nd vcvtsi2sd won't assemble, but also the first. The last two, otoh, will continue to assemble fine (due to the lack of any memory operand size specifier). As a result, another way to make things behave more consistently would be to avoid the folding and add IgnoreSize to the CpuNo64 AVX512 variants. A 3rd way to do so would be to add Cpu386 to any such insn template. While doing this also make the usual cosmetic adjustments for the insns touched anyway. Additionally drop the redundant Cpu64 from the SAE forms of VCVT{,U}SI2SD - they won't assemble outside of 64-bit mode due to there not being anything to match the Reg64 operand.
2020-02-17 08:59:52 +01:00
vcvtsi2ss, 4, 0xF32A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Reg32|Reg64, Imm8, RegXMM, RegXMM }
vcvtsi2ss, 4, 0xF32A, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE|IntelSyntax, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
x86: Allow integer conversion without suffix in AT&T syntax According to gas manual, suffix in instruction mnemonics isn't always required: When there is no sizing suffix and no (suitable) register operands to deduce the size of memory operands, with a few exceptions and where long operand size is possible in the first place, operand size will default to long in 32- and 64-bit modes. This includes cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss. Since they are used in GCC 8 and older GCC releases, they must be allowed without suffix in AT&T syntax. gas/ PR gas/25622 * testsuite/gas/i386/i386.exp: Run x86-64-default-suffix and x86-64-default-suffix-avx. * testsuite/gas/i386/noreg64.s: Remove cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss entries. * testsuite/gas/i386/noreg64.d: Updated. * testsuite/gas/i386/noreg64.l: Likewise. * testsuite/gas/i386/x86-64-default-suffix-avx.d: New file. * testsuite/gas/i386/x86-64-default-suffix.d: Likewise. * testsuite/gas/i386/x86-64-default-suffix.s: Likewise. opcodes/ PR gas/25622 * i386-opc.tbl: Add IgnoreSize to cvtsi2sd, cvtsi2ss, vcvtsi2sd, vcvtsi2ss, vcvtusi2sd and vcvtusi2ss for AT&T syntax. * i386-tbl.h: Regenerated.
2020-03-03 16:39:18 +01:00
vcvtusi2ss, 3, 0xF37B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ATTSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
vcvtusi2ss, 3, 0xF37B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|Disp8ShiftVL|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|IntelSyntax, { Reg32|Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: fold certain VCVT{,U}SI2S{S,D} templates There don't really need to be separate Cpu64 and CpuNo64 templates for these. One small issue with this is that slightly strange code .intel_syntax noprefix .code16 .arch i286 .arch .avx vcvtsi2sd xmm0, xmm0, dword ptr [bx] vcvtsi2sd xmm0, xmm0, qword ptr [bx] vcvtsi2sd xmm0, xmm0, ebx vcvtsi2sd xmm0, xmm0, rbx now will match in behavior with the AVX512 counterparts in that not only the 2nd vcvtsi2sd won't assemble, but also the first. The last two, otoh, will continue to assemble fine (due to the lack of any memory operand size specifier). As a result, another way to make things behave more consistently would be to avoid the folding and add IgnoreSize to the CpuNo64 AVX512 variants. A 3rd way to do so would be to add Cpu386 to any such insn template. While doing this also make the usual cosmetic adjustments for the insns touched anyway. Additionally drop the redundant Cpu64 from the SAE forms of VCVT{,U}SI2SD - they won't assemble outside of 64-bit mode due to there not being anything to match the Reg64 operand.
2020-02-17 08:59:52 +01:00
vcvtusi2ss, 4, 0xF37B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Reg32|Reg64, Imm8, RegXMM, RegXMM }
vcvtusi2ss, 4, 0xF37B, None, 1, CpuAVX512F, Modrm|EVexLIG|VexOpcode=0|VexVVVV|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE|IntelSyntax, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtss2sd, 3, 0xF35A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtss2sd, 4, 0xF35A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtss2si, 2, 0xF32D, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { RegXMM|Dword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtss2si, 3, 0xF32D, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtss2usi, 2, 0xF379, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ToQword, { RegXMM|Dword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtss2usi, 3, 0xF379, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, Reg32|Reg64 }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttpd2dq, 2, 0x66E6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|ZMMword|Unspecified|BaseIndex, RegYMM }
vcvttpd2dq, 3, 0x66E6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegYMM }
vcvttpd2udq, 2, 0x78, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|ZMMword|Unspecified|BaseIndex, RegYMM }
vcvttpd2udq, 3, 0x78, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvttps2dq, 2, 0xF35B, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttps2dq, 3, 0xF35B, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvttps2udq, 2, 0x78, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttps2udq, 3, 0x78, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold to-scalar-int conversion insns
2018-03-28 14:22:56 +02:00
vcvttsd2si, 2, 0xF22C, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToDword, { RegXMM|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttsd2si, 3, 0xF22C, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SAE, { Imm8, RegXMM, Reg32|Reg64 }
x86: fold to-scalar-int conversion insns
2018-03-28 14:22:56 +02:00
vcvttsd2usi, 2, 0xF278, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ToDword, { RegXMM|Qword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttsd2usi, 3, 0xF278, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, Reg32|Reg64 }
x86: fold to-scalar-int conversion insns
2018-03-28 14:22:56 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvttss2si, 2, 0xF32C, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=2|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|ToQword, { RegXMM|Dword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttss2si, 3, 0xF32C, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_sSuf|No_ldSuf|SAE, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvttss2usi, 2, 0xF378, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ToQword, { RegXMM|Dword|Unspecified|BaseIndex, Reg32|Reg64 }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvttss2usi, 3, 0xF378, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, Reg32|Reg64 }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vcvtudq2ps, 2, 0xF27A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vcvtudq2ps, 3, 0xF27A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vexpandpd, 2, 0x6688, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=3|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpexpandq, 2, 0x6689, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=3|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vexpandps, 2, 0x6688, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpexpandd, 2, 0x6689, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=2|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vextractf32x4, 3, 0x6619, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=2|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM, RegXMM|Unspecified|BaseIndex }
vextracti32x4, 3, 0x6639, None, 1, CpuAVX512F, Modrm|MaskingMorZ|VexOpcode=2|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM, RegXMM|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vextractf64x4, 3, 0x661B, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=2|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM, RegYMM|Unspecified|BaseIndex }
vextracti64x4, 3, 0x663B, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=2|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM, RegYMM|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
vextractps, 3, 0x6617, None, 1, CpuAVX512F, Modrm|EVex128|VexOpcode=2|VexWIG|Disp8MemShift=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vextractps, 3, 0x6617, None, 1, CpuAVX512F|Cpu64, RegMem|EVex128|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg64 }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfixupimmpd, 4, 0x6654, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfixupimmpd, 5, 0x6654, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfixupimmps, 4, 0x6654, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfixupimmps, 5, 0x6654, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfixupimmsd, 4, 0x6655, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfixupimmsd, 5, 0x6655, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgetmantsd, 4, 0x6627, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetmantsd, 5, 0x6627, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrndscalesd, 4, 0x660B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vrndscalesd, 5, 0x660B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfixupimmss, 4, 0x6655, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfixupimmss, 5, 0x6655, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgetmantss, 4, 0x6627, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetmantss, 5, 0x6627, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrndscaless, 4, 0x660A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vrndscaless, 5, 0x660A, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd132pd, 3, 0x6698, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd132pd, 4, 0x6698, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd213pd, 3, 0x66A8, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd213pd, 4, 0x66A8, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd231pd, 3, 0x66B8, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd231pd, 4, 0x66B8, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub132pd, 3, 0x6696, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub132pd, 4, 0x6696, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub213pd, 3, 0x66A6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub213pd, 4, 0x66A6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub231pd, 3, 0x66B6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub231pd, 4, 0x66B6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub132pd, 3, 0x669A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub132pd, 4, 0x669A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub213pd, 3, 0x66AA, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub213pd, 4, 0x66AA, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub231pd, 3, 0x66BA, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub231pd, 4, 0x66BA, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd132pd, 3, 0x6697, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd132pd, 4, 0x6697, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd213pd, 3, 0x66A7, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd213pd, 4, 0x66A7, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd231pd, 3, 0x66B7, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd231pd, 4, 0x66B7, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd132pd, 3, 0x669C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd132pd, 4, 0x669C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd213pd, 3, 0x66AC, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd213pd, 4, 0x66AC, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd231pd, 3, 0x66BC, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd231pd, 4, 0x66BC, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub132pd, 3, 0x669E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub132pd, 4, 0x669E, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub213pd, 3, 0x66AE, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub213pd, 4, 0x66AE, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub231pd, 3, 0x66BE, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub231pd, 4, 0x66BE, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vscalefpd, 3, 0x662C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vscalefpd, 4, 0x662C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd132ps, 3, 0x6698, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd132ps, 4, 0x6698, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd213ps, 3, 0x66A8, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd213ps, 4, 0x66A8, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmadd231ps, 3, 0x66B8, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd231ps, 4, 0x66B8, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub132ps, 3, 0x6696, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub132ps, 4, 0x6696, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub213ps, 3, 0x66A6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub213ps, 4, 0x66A6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmaddsub231ps, 3, 0x66B6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmaddsub231ps, 4, 0x66B6, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub132ps, 3, 0x669A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub132ps, 4, 0x669A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub213ps, 3, 0x66AA, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub213ps, 4, 0x66AA, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsub231ps, 3, 0x66BA, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub231ps, 4, 0x66BA, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd132ps, 3, 0x6697, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd132ps, 4, 0x6697, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd213ps, 3, 0x66A7, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd213ps, 4, 0x66A7, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfmsubadd231ps, 3, 0x66B7, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsubadd231ps, 4, 0x66B7, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd132ps, 3, 0x669C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd132ps, 4, 0x669C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd213ps, 3, 0x66AC, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd213ps, 4, 0x66AC, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmadd231ps, 3, 0x66BC, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd231ps, 4, 0x66BC, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub132ps, 3, 0x669E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub132ps, 4, 0x669E, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub213ps, 3, 0x66AE, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub213ps, 4, 0x66AE, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vfnmsub231ps, 3, 0x66BE, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub231ps, 4, 0x66BE, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vscalefps, 3, 0x662C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vscalefps, 4, 0x662C, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd132sd, 3, 0x6699, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd132sd, 4, 0x6699, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd213sd, 3, 0x66A9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd213sd, 4, 0x66A9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd231sd, 3, 0x66B9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd231sd, 4, 0x66B9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub132sd, 3, 0x669B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub132sd, 4, 0x669B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub213sd, 3, 0x66AB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub213sd, 4, 0x66AB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub231sd, 3, 0x66BB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub231sd, 4, 0x66BB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd132sd, 3, 0x669D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd132sd, 4, 0x669D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd213sd, 3, 0x66AD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd213sd, 4, 0x66AD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd231sd, 3, 0x66BD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd231sd, 4, 0x66BD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub132sd, 3, 0x669F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub132sd, 4, 0x669F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub213sd, 3, 0x66AF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub213sd, 4, 0x66AF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub231sd, 3, 0x66BF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub231sd, 4, 0x66BF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vscalefsd, 3, 0x662D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vscalefsd, 4, 0x662D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd132ss, 3, 0x6699, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd132ss, 4, 0x6699, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd213ss, 3, 0x66A9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd213ss, 4, 0x66A9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmadd231ss, 3, 0x66B9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmadd231ss, 4, 0x66B9, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub132ss, 3, 0x669B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub132ss, 4, 0x669B, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub213ss, 3, 0x66AB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub213ss, 4, 0x66AB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfmsub231ss, 3, 0x66BB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfmsub231ss, 4, 0x66BB, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd132ss, 3, 0x669D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd132ss, 4, 0x669D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd213ss, 3, 0x66AD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd213ss, 4, 0x66AD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmadd231ss, 3, 0x66BD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmadd231ss, 4, 0x66BD, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub132ss, 3, 0x669F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub132ss, 4, 0x669F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub213ss, 3, 0x66AF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub213ss, 4, 0x66AF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfnmsub231ss, 3, 0x66BF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vfnmsub231ss, 4, 0x66BF, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vscalefss, 3, 0x662D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vscalefss, 4, 0x662D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherdpd, 2, 0x6692, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegZMM }
vgatherqpd, 2, 0x6693, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegZMM }
vpgatherdq, 2, 0x6690, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegZMM }
vpgatherqq, 2, 0x6691, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherdps, 2, 0x6692, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovntdqa, 2, 0x662A, None, 1, CpuAVX512F, Modrm|VexOpcode=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { XMMword|YMMword|ZMMword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpgatherdd, 2, 0x6690, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherqps, 2, 0x6693, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegYMM }
vpgatherqd, 2, 0x6691, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegYMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vgetexppd, 2, 0x6642, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetexppd, 3, 0x6642, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vgetexpps, 2, 0x6642, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetexpps, 3, 0x6642, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgetexpsd, 3, 0x6643, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetexpsd, 4, 0x6643, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgetexpss, 3, 0x6643, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetexpss, 4, 0x6643, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vgetmantpd, 3, 0x6626, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetmantpd, 4, 0x6626, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vrndscalepd, 3, 0x6609, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vrndscalepd, 4, 0x6609, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vgetmantps, 3, 0x6626, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vgetmantps, 4, 0x6626, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vrndscaleps, 3, 0x6608, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vrndscaleps, 4, 0x6608, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vinsertf32x4, 4, 0x6618, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|XMMword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vinserti32x4, 4, 0x6638, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|XMMword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vinsertf64x4, 4, 0x661A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Unspecified|BaseIndex, RegZMM, RegZMM }
vinserti64x4, 4, 0x663A, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Unspecified|BaseIndex, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vinsertps, 4, 0x6621, None, 1, CpuAVX512F, Modrm|EVex128|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmaxpd, 3, 0x665F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmaxpd, 4, 0x665F, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vminpd, 3, 0x665D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vminpd, 4, 0x665D, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmaxps, 3, 0x5F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmaxps, 4, 0x5F, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vminps, 3, 0x5D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vminps, 4, 0x5D, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmaxsd, 3, 0xF25F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmaxsd, 4, 0xF25F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vminsd, 3, 0xF25D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vminsd, 4, 0xF25D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmaxss, 3, 0xF35F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmaxss, 4, 0xF35F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vminss, 3, 0xF35D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vminss, 4, 0xF35D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovapd, 2, 0x6628, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovntpd, 2, 0x662B, None, 1, CpuAVX512F, Modrm|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, XMMword|YMMword|ZMMword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovupd, 2, 0x6610, None, 1, CpuAVX512F, D|Modrm|Load|MaskingMorZ|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovaps, 2, 0x28, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovntps, 2, 0x2B, None, 1, CpuAVX512F, Modrm|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, XMMword|YMMword|ZMMword|Unspecified|BaseIndex }
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovups, 2, 0x10, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vmovd, 2, 0x666E, None, 1, CpuAVX512F, D|Modrm|EVex=2|VexOpcode=0|Disp8MemShift=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovddup, 2, 0xF212, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|RegZMM|Unspecified|BaseIndex, RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Optimize EVEX vector load/store instructions When there is no write mask, we can encode lower 16 128-bit/256-bit EVEX vector register load and store instructions as VEX vector register load and store instructions with -O1. gas/ PR gas/24348 * config/tc-i386.c (optimize_encoding): Encode 128-bit and 256-bit EVEX vector register load/store instructions as VEX vector register load/store instructions for -O1. * doc/c-i386.texi: Update -O1 documentation. * testsuite/gas/i386/i386.exp: Run PR gas/24348 tests. * testsuite/gas/i386/optimize-1.s: Add tests for EVEX vector load/store instructions. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. * testsuite/gas/i386/x86-64-optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-6.s: Likewise. * testsuite/gas/i386/optimize-1.d: Updated. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-4.d: Likewise. * testsuite/gas/i386/optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-6.d: Likewise. * testsuite/gas/i386/optimize-7.d: New file. * testsuite/gas/i386/optimize-7.s: Likewise. * testsuite/gas/i386/x86-64-optimize-8.d: Likewise. * testsuite/gas/i386/x86-64-optimize-8.s: Likewise. opcodes/ PR gas/24348 * i386-opc.tbl: Add Optimize to vmovdqa32, vmovdqa64, vmovdqu8, vmovdqu16, vmovdqu32 and vmovdqu64. * i386-tbl.h: Regenerated.
2019-03-18 01:56:10 +01:00
vmovdqa64, 2, 0x666F, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vmovdqa32, 2, 0x666F, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovntdq, 2, 0x66E7, None, 1, CpuAVX512F, Modrm|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, XMMword|YMMword|ZMMword|Unspecified|BaseIndex }
x86: Optimize EVEX vector load/store instructions When there is no write mask, we can encode lower 16 128-bit/256-bit EVEX vector register load and store instructions as VEX vector register load and store instructions with -O1. gas/ PR gas/24348 * config/tc-i386.c (optimize_encoding): Encode 128-bit and 256-bit EVEX vector register load/store instructions as VEX vector register load/store instructions for -O1. * doc/c-i386.texi: Update -O1 documentation. * testsuite/gas/i386/i386.exp: Run PR gas/24348 tests. * testsuite/gas/i386/optimize-1.s: Add tests for EVEX vector load/store instructions. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. * testsuite/gas/i386/x86-64-optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-6.s: Likewise. * testsuite/gas/i386/optimize-1.d: Updated. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-4.d: Likewise. * testsuite/gas/i386/optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-6.d: Likewise. * testsuite/gas/i386/optimize-7.d: New file. * testsuite/gas/i386/optimize-7.s: Likewise. * testsuite/gas/i386/x86-64-optimize-8.d: Likewise. * testsuite/gas/i386/x86-64-optimize-8.s: Likewise. opcodes/ PR gas/24348 * i386-opc.tbl: Add Optimize to vmovdqa32, vmovdqa64, vmovdqu8, vmovdqu16, vmovdqu32 and vmovdqu64. * i386-tbl.h: Regenerated.
2019-03-18 01:56:10 +01:00
vmovdqu32, 2, 0xF36F, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vmovdqu64, 2, 0xF36F, None, 1, CpuAVX512F, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vmovhlps, 3, 0x12, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
vmovlhps, 3, 0x16, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop pointless VecESize The attribute is meaningful only in templates allowing embedded broadcast. Drop them everywhere else.
2018-03-22 08:46:25 +01:00
vmovhpd, 3, 0x6616, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovhpd, 2, 0x6617, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vmovlpd, 3, 0x6612, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovlpd, 2, 0x6613, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
vmovhps, 3, 0x16, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovhps, 2, 0x17, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=1|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vmovlps, 3, 0x12, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vmovlps, 2, 0x13, None, 1, CpuAVX512F, Modrm|EVex=4|VexOpcode=0|VexW=1|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Set EVex=2 on EVEX.128 only vmovd and vmovq EVEX "VMOVD xmm1, r32/m32", "VMOVD r32/m32, xmm2", "VMOVQ xmm1, r64/m64", "VMOVD r64/m64, xmm2", "VMOVQ xmm1, xmm2/m64" and "VMOVQ xmm1/m64, xmm2" can only be encoded with EVEX.128. Set EVex=2 on EVEX.128 only vmovd and vmovq. gas/ PR gas/23670 * testsuite/gas/i386/evex-lig-2.d: New file. * testsuite/gas/i386/evex-lig-2.s: Likewise. * testsuite/gas/i386/x86-64-evex-lig-2.d: Likewise. * testsuite/gas/i386/x86-64-evex-lig-2.s: Likewise. * testsuite/gas/i386/i386.exp: Run evex-lig-2 and x86-64-evex-lig-2. opcodes/ PR gas/23670 * i386-dis-evex.h (evex_table): Use EVEX_LEN_0F6E_P_2, EVEX_LEN_0F7E_P_1, EVEX_LEN_0F7E_P_2 and EVEX_LEN_0FD6_P_2. (EVEX_LEN_0F6E_P_2): New EVEX_LEN_TABLE entry. (EVEX_LEN_0F7E_P_1): Likewise. (EVEX_LEN_0F7E_P_2): Likewise. (EVEX_LEN_0FD6_P_2): Likewise. * i386-dis.c (USE_EVEX_LEN_TABLE): New. (EVEX_LEN_TABLE): Likewise. (EVEX_LEN_0F6E_P_2): New enum. (EVEX_LEN_0F7E_P_1): Likewise. (EVEX_LEN_0F7E_P_2): Likewise. (EVEX_LEN_0FD6_P_2): Likewise. (evex_len_table): New. (get_valid_dis386): Handle USE_EVEX_LEN_TABLE. * i386-opc.tbl: Set EVex=2 on EVEX.128 only vmovd and vmovq. * i386-tbl.h: Regenerated.
2018-09-17 18:33:20 +02:00
vmovq, 2, 0x666E, None, 1, CpuAVX512F|Cpu64, D|Modrm|EVex=2|VexOpcode=0|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg64|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmovq, 2, 0xF37E, None, 1, CpuAVX512F, Load|Modrm|EVex=2|VexOpcode=0|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex|RegXMM, RegXMM }
vmovq, 2, 0x66D6, None, 1, CpuAVX512F, Modrm|EVex=2|VexOpcode=0|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex|RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovsd, 2, 0xF210, None, 1, CpuAVX512F, D|Modrm|EVex=4|MaskingMorZ|VexOpcode=0|VexW=2|Disp8MemShift=3|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vmovsd, 3, 0xF210, None, 1, CpuAVX512F, D|Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vmovshdup, 2, 0xF316, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vmovsldup, 2, 0xF312, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use D attribute also for SIMD templates Various moves come in load and store forms, and just like on the GPR and FPU sides there would better be only one pattern. In some cases this is not feasible because the opcodes are too different, but quite a few cases follow a similar standard scheme. Introduce Opcode_SIMD_FloatD and Opcode_SIMD_IntD, generalize handling in operand_size_match() (reverse operand handling there simply needs to match "straight" operand one), and fix a long standing, but so far only latent bug with when to zap found_reverse_match. Also once again drop IgnoreSize where pointlessly applied to templates touched anyway as well as *word when redundant with Reg*.
2018-09-13 11:07:07 +02:00
vmovss, 2, 0xF310, None, 1, CpuAVX512F, D|Modrm|EVex=4|MaskingMorZ|VexOpcode=0|VexW=1|Disp8MemShift=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vmovss, 3, 0xF310, None, 1, CpuAVX512F, D|Modrm|EVex=4|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpabsd, 2, 0x661E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vrcp14ps, 2, 0x664C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vrsqrt14ps, 2, 0x664E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpabsq, 2, 0x661F, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vrcp14pd, 2, 0x664C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vrsqrt14pd, 2, 0x664E, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpaddd, 3, 0x66FE, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: optimize EVEX packed integer logical instructions As long as there's no write mask as well as no broadcast, and as long as the scaled Disp8 wouldn't result in a shorter EVEX encoding, encode VPAND{D,Q}, VPANDN{D,Q}, VPOR{D,Q}, and VPXOR{D,Q} acting on only the lower 16 XMM/YMM registers using their VEX equivalents with -O1. Also take the opportunity and avoid looping twice over all operands when dealing with memory-with-displacement ones.
2019-07-01 08:31:14 +02:00
vpandd, 3, 0x66DB, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpandnd, 3, 0x66DF, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: optimize EVEX packed integer logical instructions As long as there's no write mask as well as no broadcast, and as long as the scaled Disp8 wouldn't result in a shorter EVEX encoding, encode VPAND{D,Q}, VPANDN{D,Q}, VPOR{D,Q}, and VPXOR{D,Q} acting on only the lower 16 XMM/YMM registers using their VEX equivalents with -O1. Also take the opportunity and avoid looping twice over all operands when dealing with memory-with-displacement ones.
2019-07-01 08:31:14 +02:00
vpord, 3, 0x66EB, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsubd, 3, 0x66FA, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpckhdq, 3, 0x666A, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpckldq, 3, 0x6662, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpxord, 3, 0x66EF, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddq, 3, 0x66D4, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpandnq, 3, 0x66DF, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: optimize EVEX packed integer logical instructions As long as there's no write mask as well as no broadcast, and as long as the scaled Disp8 wouldn't result in a shorter EVEX encoding, encode VPAND{D,Q}, VPANDN{D,Q}, VPOR{D,Q}, and VPXOR{D,Q} acting on only the lower 16 XMM/YMM registers using their VEX equivalents with -O1. Also take the opportunity and avoid looping twice over all operands when dealing with memory-with-displacement ones.
2019-07-01 08:31:14 +02:00
vpandq, 3, 0x66DB, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpmuludq, 3, 0x66F4, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: optimize EVEX packed integer logical instructions As long as there's no write mask as well as no broadcast, and as long as the scaled Disp8 wouldn't result in a shorter EVEX encoding, encode VPAND{D,Q}, VPANDN{D,Q}, VPOR{D,Q}, and VPXOR{D,Q} acting on only the lower 16 XMM/YMM registers using their VEX equivalents with -O1. Also take the opportunity and avoid looping twice over all operands when dealing with memory-with-displacement ones.
2019-07-01 08:31:14 +02:00
vporq, 3, 0x66EB, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsubq, 3, 0x66FB, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpckhqdq, 3, 0x666D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpcklqdq, 3, 0x666C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpxorq, 3, 0x66EF, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vunpckhpd, 3, 0x6615, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vunpcklpd, 3, 0x6614, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpbroadcastq, 2, 0x6659, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpbroadcastq, 2, 0x667C, None, 1, CpuAVX512F|Cpu64, Modrm|Masking=3|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg64, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
<avx_irel:imm, eq:0, lt:1, le:2, neq:4, nlt:5, nle:6>
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpcmpeqd, 3, 0x6676, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpgtd, 3, 0x6666, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpd, 4, 0x661F, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>d, 3, 0x661F, <avx_irel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpcmpud, 4, 0x661E, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>ud, 3, 0x661E, <avx_irel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpcmpeqq, 3, 0x6629, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpgtq, 3, 0x6637, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpq, 4, 0x661F, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>q, 3, 0x661F, <avx_irel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpcmpuq, 4, 0x661E, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>uq, 3, 0x661E, <avx_irel:imm>, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vptestmd, 3, 0x6627, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vptestnmd, 3, 0xF327, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vptestmq, 3, 0x6627, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vptestnmq, 3, 0xF327, None, 1, CpuAVX512F, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpermd, 3, 0x6636, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpermps, 3, 0x6616, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpermilpd, 3, 0x6605, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpermilpd, 3, 0x660D, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermilps, 3, 0x6604, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpermilps, 3, 0x660C, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermpd, 3, 0x6601, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM }
vpermpd, 3, 0x6616, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpermq, 3, 0x6600, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM }
vpermq, 3, 0x6636, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovdb, 2, 0xF331, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
vpmovsdb, 2, 0xF321, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
vpmovusdb, 2, 0xF311, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
vpmovdw, 2, 0xF333, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
vpmovsdw, 2, 0xF323, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
vpmovusdw, 2, 0xF313, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovqb, 2, 0xF332, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovsqb, 2, 0xF322, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovusqb, 2, 0xF312, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovqd, 2, 0xF335, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
vpmovsqd, 2, 0xF325, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
vpmovusqd, 2, 0xF315, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
vpmovqw, 2, 0xF334, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
vpmovsqw, 2, 0xF324, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
vpmovusqw, 2, 0xF314, None, 1, CpuAVX512F, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegXMM|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxbd, 2, 0x6621, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegZMM }
vpmovzxbd, 2, 0x6631, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxbq, 2, 0x6622, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegZMM }
vpmovzxbq, 2, 0x6632, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vpmovsxdq, 2, 0x6625, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
vpmovzxdq, 2, 0x6635, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxwd, 2, 0x6623, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
vpmovzxwd, 2, 0x6633, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxwq, 2, 0x6624, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegZMM }
vpmovzxwq, 2, 0x6634, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vprold, 3, 0x6672, 1, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vprord, 3, 0x6672, 0, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vprolq, 3, 0x6672, 1, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vprorq, 3, 0x6672, 0, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpscatterqd, 2, 0x66A1, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Dword|Unspecified|BaseIndex }
vscatterqps, 2, 0x66A3, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Dword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpshufd, 3, 0x6670, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpslld, 3, 0x66F2, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpslld, 3, 0x6672, 6, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsrad, 3, 0x66E2, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsrad, 3, 0x6672, 4, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsrld, 3, 0x66D2, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsrld, 3, 0x6672, 2, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsllq, 3, 0x66F3, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsllq, 3, 0x6673, 6, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsraq, 3, 0x66E2, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsraq, 3, 0x6672, 4, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsrlq, 3, 0x66D3, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vpsrlq, 3, 0x6673, 2, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrcp14sd, 3, 0x664D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
vrsqrt14sd, 3, 0x664F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrcp14ss, 3, 0x664D, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
vrsqrt14ss, 3, 0x664F, None, 1, CpuAVX512F, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vshuff32x4, 4, 0x6623, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vshufi32x4, 4, 0x6643, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vshuff64x2, 4, 0x6623, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vshufi64x2, 4, 0x6643, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vshufpd, 4, 0x66C6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vshufps, 4, 0xC6, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vsqrtpd, 2, 0x6651, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512F insns
2018-09-13 11:22:49 +02:00
vsqrtpd, 3, 0x6651, None, 1, CpuAVX512F, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
vunpckhps, 3, 0x15, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vunpcklps, 3, 0x14, None, 1, CpuAVX512F, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Change cpu for vptestnmd and vptestnmq instructions. In latest release of AVX512 spec http://download-software.intel.com/sites/default/files/managed/50/1a/319433-018.pdf Vptestnmq and vptestnmq instructions have CPUID AVX512F, not AVX512CD. This patch fixes it. opcodes/ * i386-opc.tbl: Change CPU of vptestnmq, vptestnmd from CpuAVX512CD, to CpuAVX512F. * i386-tbl.h: Regenerate. gas/testsuite/ * gas/i386/avx512cd-intel.d: Remove vptestnmq, vptestnmd. * gas/i386/avx512cd.s: Ditto. * gas/i386/avx512cd.d: Ditto. * gas/i386/x86-64-avx512cd-intel.d: Ditto. * gas/i386/x86-64-avx512cd.s: Ditto. * gas/i386/x86-64-avx512cd.d: Ditto. * gas/i386/avx512f-intel.d: Add vptestnmq, vptestnmd. * gas/i386/avx512f.s: Ditto. * gas/i386/avx512f.d: Ditto. * gas/i386/x86-64-avx512f-intel.d: Ditto. * gas/i386/x86-64-avx512f.s: Ditto. * gas/i386/x86-64-avx512f.d: Ditto.
2014-02-20 15:31:11 +01:00
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
// AVX512F instructions end.
// AVX512CD instructions.
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpbroadcastmb2q, 2, 0xF32A, None, 1, CpuAVX512CD, Modrm|VexOpcode=1|EVex=5|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
vpbroadcastmw2d, 2, 0xF33A, None, 1, CpuAVX512CD, Modrm|VexOpcode=1|EVex=5|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512CD templates
2018-07-19 08:31:24 +02:00
vpconflictd, 2, 0x66C4, None, 1, CpuAVX512CD, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpconflictq, 2, 0x66C4, None, 1, CpuAVX512CD, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: fold various AVX512CD templates
2018-07-19 08:31:24 +02:00
vplzcntd, 2, 0x6644, None, 1, CpuAVX512CD, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vplzcntq, 2, 0x6644, None, 1, CpuAVX512CD, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
// AVX512CD instructions end.
// AVX512ER instructions.
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vexp2pd, 2, 0x66C8, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegZMM }
vexp2pd, 3, 0x66C8, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vexp2ps, 2, 0x66C8, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Dword|Unspecified|BaseIndex, RegZMM }
vexp2ps, 3, 0x66C8, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrcp28pd, 2, 0x66CA, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegZMM }
vrcp28pd, 3, 0x66CA, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
vrsqrt28pd, 2, 0x66CC, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegZMM }
vrsqrt28pd, 3, 0x66CC, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrcp28ps, 2, 0x66CA, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Dword|Unspecified|BaseIndex, RegZMM }
vrcp28ps, 3, 0x66CA, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
vrsqrt28ps, 2, 0x66CC, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Dword|Unspecified|BaseIndex, RegZMM }
vrsqrt28ps, 3, 0x66CC, None, 1, CpuAVX512ER, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrcp28sd, 3, 0x66CB, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrcp28sd, 4, 0x66CB, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrsqrt28sd, 3, 0x66CD, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrsqrt28sd, 4, 0x66CD, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrcp28ss, 3, 0x66CB, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrcp28ss, 4, 0x66CB, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrsqrt28ss, 3, 0x66CD, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512ER insns
2018-09-13 11:23:17 +02:00
vrsqrt28ss, 4, 0x66CD, None, 1, CpuAVX512ER, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegXMM, RegXMM, RegXMM }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
// AVX512ER instructions end.
// AVX512PF instructions.
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherpf0dpd, 1, 0x66C6, 1, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vgatherpf0qpd, 1, 0x66C7, 1, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vgatherpf1dpd, 1, 0x66C6, 2, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vgatherpf1qpd, 1, 0x66C7, 2, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vscatterpf0dpd, 1, 0x66C6, 5, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vscatterpf0qpd, 1, 0x66C7, 5, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vscatterpf1dpd, 1, 0x66C6, 6, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vscatterpf1qpd, 1, 0x66C7, 6, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex }
vgatherpf0dps, 1, 0x66C6, 1, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vgatherpf0qps, 1, 0x66C7, 1, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vgatherpf1dps, 1, 0x66C6, 2, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vgatherpf1qps, 1, 0x66C7, 2, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vscatterpf0dps, 1, 0x66C6, 5, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vscatterpf0qps, 1, 0x66C7, 5, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vscatterpf1dps, 1, 0x66C6, 6, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
vscatterpf1qps, 1, 0x66C7, 6, 1, CpuAVX512PF, Modrm|EVex=1|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex }
Add Intel AVX-512 support binutils/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * dwarf.c (dwarf_regnames_i386): Add k0-k7 registers and numeration in comments. (dwarf_regnames_x86_64): Add xmm16-31 and k0-k7 registers to dwarf table. gas/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * config/tc-i386-intel.c (O_zmmword_ptr): New. (i386_types): Add zmmword. (i386_intel_simplify_register): Allow regzmm. (i386_intel_simplify): Handle zmmwords. (i386_intel_operand): Handle RC/SAE, vector operations and zmmwords. * config/tc-i386.c (ZMMWORD_MNEM_SUFFIX): New. (struct RC_Operation): New. (struct Mask_Operation): New. (struct Broadcast_Operation): New. (vex_prefix): Size of bytes increased to 4 to support EVEX encoding. (enum i386_error): Add new error codes: unsupported_broadcast, broadcast_not_on_src_operand, broadcast_needed, unsupported_masking, mask_not_on_destination, no_default_mask, unsupported_rc_sae, rc_sae_operand_not_last_imm, invalid_register_operand, try_vector_disp8. (struct _i386_insn): Add new fields vrex, need_vrex, mask, rounding, broadcast, memshift. (struct RC_name): New. (RC_NamesTable): New. (evexlig): New. (evexwig): New. (extra_symbol_chars): Add '{'. (cpu_arch): Add AVX512F, AVX512CD, AVX512ER and AVX512PF. (i386_operand_type): Add regzmm, regmask and vec_disp8. (match_mem_size): Handle zmmwords. (operand_type_match): Handle zmm-registers. (mode_from_disp_size): Handle vec_disp8. (fits_in_vec_disp8): New. (md_begin): Handle {} properly. (type_names): Add "rZMM", "Mask reg" and "Vector d8". (build_vex_prefix): Handle vrex. (build_evex_prefix): New. (process_immext): Adjust to properly handle EVEX. (md_assemble): Add EVEX encoding support. (swap_2_operands): Correctly handle operands with masking, broadcasting or RC/SAE. (check_VecOperands): Support EVEX features. (VEX_check_operands): Properly handle 16 upper [xyz]mm registers. (match_template): Support regzmm and handle new error codes. (process_suffix): Handle zmmwords and zmm-registers. (check_byte_reg): Extend to zmm-registers. (process_operands): Extend to zmm-registers. (build_modrm_byte): Handle EVEX. (output_insn): Adjust to properly handle EVEX case. (disp_size): Handle vec_disp8. (output_disp): Support compressed disp8*N evex feature. (output_imm): Handle RC/SAE immediates properly. (check_VecOperations): New. (i386_immediate): Handle EVEX features. (i386_index_check): Handle zmmwords and zmm-registers. (RC_SAE_immediate): New. (i386_att_operand): Handle EVEX features. (parse_real_register): Add a check for ZMM/Mask registers. (OPTION_MEVEXLIG): New. (OPTION_MEVEXWIG): New. (md_longopts): Add mevexlig and mevexwig. (md_parse_option): Handle mevexlig and mevexwig options. (md_show_usage): Add description for mevexlig and mevexwig. * doc/c-i386.texi: Document avx512f/.avx512f, avx512cd/.avx512cd, avx512er/.avx512er, avx512pf/.avx512pf, mevexlig and mevexwig. gas/testsuite/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * gas/cfi/cfi-i386.s: Add tests for k0-k7. * gas/cfi/cfi-i386.d: Change to reflect above mentioned changes. * gas/cfi/cfi-x86_64.s: Add tests for xmm16-31, k0-7. * gas/cfi/cfi-x86_64.d: Change to reflect above mentioned changes. * gas/i386/ilp32/cfi/cfi-x86_64.d: Ditto. * gas/i386/intel-regs.s: Add tests for zmm0 and xmm16 registers. * gas/i386/intel-regs.d: Change correspondingly. * gas/i386/prefetch-intel.d: Reflect implementation of prefetchwt1. * gas/i386/prefetch.d: Ditto. * gas/i386/x86-64-prefetch-intel.d: Ditto. * gas/i386/x86-64-prefetch.d: Ditto. * gas/i386/avx512f-intel.d: New. * gas/i386/avx512f-nondef.d: New. * gas/i386/avx512f-nondef.s: New. * gas/i386/avx512f-opts-intel.d: New. * gas/i386/avx512f-opts.d: New. * gas/i386/avx512f-opts.s: New. * gas/i386/avx512f.d: New. * gas/i386/avx512f.s: New. * gas/i386/avx512cd-intel.d: New. * gas/i386/avx512cd.d: New. * gas/i386/avx512cd.s: New. * gas/i386/avx512er-intel.d: New. * gas/i386/avx512er.d: New. * gas/i386/avx512er.s: New. * gas/i386/avx512pf-intel.d: New. * gas/i386/avx512pf.d: New. * gas/i386/avx512pf.s: New. * gas/i386/evex-lig.s: New. * gas/i386/evex-lig256-intel.d: New. * gas/i386/evex-lig256.d: New. * gas/i386/evex-lig512-intel.d: New. * gas/i386/evex-lig512.d: New. * gas/i386/evex-wig.s: New. * gas/i386/evex-wig1-intel.d: New. * gas/i386/evex-wig1.d: New. * gas/i386/inval-avx512f.l: New. * gas/i386/inval-avx512f.s: New. * gas/i386/x86-64-avx512f-intel.d: New. * gas/i386/x86-64-avx512f-nondef.d: New. * gas/i386/x86-64-avx512f-nondef.s: New. * gas/i386/x86-64-avx512f-opts-intel.d: New. * gas/i386/x86-64-avx512f-opts.d: New. * gas/i386/x86-64-avx512f-opts.s: New. * gas/i386/x86-64-avx512f.d: New. * gas/i386/x86-64-avx512f.s: New. * gas/i386/x86-64-avx512cd-intel.d: New. * gas/i386/x86-64-avx512cd.d: New. * gas/i386/x86-64-avx512cd.s: New. * gas/i386/x86-64-avx512er-intel.d: New. * gas/i386/x86-64-avx512er.d: New. * gas/i386/x86-64-avx512er.s: New. * gas/i386/x86-64-avx512pf-intel.d: New. * gas/i386/x86-64-avx512pf.d: New. * gas/i386/x86-64-avx512pf.s: New. * gas/i386/x86-64-evex-lig.s: New. * gas/i386/x86-64-evex-lig256-intel.d: New. * gas/i386/x86-64-evex-lig256.d: New. * gas/i386/x86-64-evex-lig512-intel.d: New. * gas/i386/x86-64-evex-lig512.d: New. * gas/i386/x86-64-evex-wig.s: New. * gas/i386/x86-64-evex-wig1-intel.d: New. * gas/i386/x86-64-evex-wig1.d: New. * gas/i386/x86-64-inval-avx512f.l: New. * gas/i386/x86-64-inval-avx512f.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. opcodes/ 2013-07-26 Sergey Guriev <sergey.s.guriev@intel.com> Alexander Ivchenko <alexander.ivchenko@intel.com> Maxim Kuznetsov <maxim.kuznetsov@intel.com> Sergey Lega <sergey.s.lega@intel.com> Anna Tikhonova <anna.tikhonova@intel.com> Ilya Tocar <ilya.tocar@intel.com> Andrey Turetskiy <andrey.turetskiy@intel.com> Ilya Verbin <ilya.verbin@intel.com> Kirill Yukhin <kirill.yukhin@intel.com> Michael Zolotukhin <michael.v.zolotukhin@intel.com> * i386-dis-evex.h: New. * i386-dis.c (OP_Rounding): New. (VPCMP_Fixup): New. (OP_Mask): New. (Rdq): New. (XMxmmq): New. (EXdScalarS): New. (EXymm): New. (EXEvexHalfBcstXmmq): New. (EXxmm_mdq): New. (EXEvexXGscat): New. (EXEvexXNoBcst): New. (VPCMP): New. (EXxEVexR): New. (EXxEVexS): New. (XMask): New. (MaskG): New. (MaskE): New. (MaskR): New. (MaskVex): New. (modes enum): Add evex_x_gscat_mode, evex_x_nobcst_mode, evex_half_bcst_xmmq_mode, xmm_mdq_mode, ymm_mode, evex_rounding_mode, evex_sae_mode, mask_mode. (USE_EVEX_TABLE): New. (EVEX_TABLE): New. (EVEX enum): New. (REG enum): Add REG_EVEX_0F72, REG_EVEX_0F73, REG_EVEX_0F38C6, REG_EVEX_0F38C7. (MOD enum): Add MOD_EVEX_0F10_PREFIX_1, MOD_EVEX_0F10_PREFIX_3, MOD_EVEX_0F11_PREFIX_1, MOD_EVEX_0F11_PREFIX_3, MOD_EVEX_0F12_PREFIX_0, MOD_EVEX_0F16_PREFIX_0, MOD_EVEX_0F38C6_REG_1, MOD_EVEX_0F38C6_REG_2, MOD_EVEX_0F38C6_REG_5, MOD_EVEX_0F38C6_REG_6, MOD_EVEX_0F38C7_REG_1, MOD_EVEX_0F38C7_REG_2, MOD_EVEX_0F38C7_REG_5, MOD_EVEX_0F38C7_REG_6. (PREFIX enum): Add PREFIX_VEX_0F41, PREFIX_VEX_0F42, PREFIX_VEX_0F44, PREFIX_VEX_0F45, PREFIX_VEX_0F46, PREFIX_VEX_0F47, PREFIX_VEX_0F4B, PREFIX_VEX_0F90, PREFIX_VEX_0F91, PREFIX_VEX_0F92, PREFIX_VEX_0F93, PREFIX_VEX_0F98, PREFIX_VEX_0F3A30, PREFIX_VEX_0F3A32, PREFIX_VEX_0F3AF0, PREFIX_EVEX_0F10, PREFIX_EVEX_0F11, PREFIX_EVEX_0F12, PREFIX_EVEX_0F13, PREFIX_EVEX_0F14, PREFIX_EVEX_0F15, PREFIX_EVEX_0F16, PREFIX_EVEX_0F17, PREFIX_EVEX_0F28, PREFIX_EVEX_0F29, PREFIX_EVEX_0F2A, PREFIX_EVEX_0F2B, PREFIX_EVEX_0F2C, PREFIX_EVEX_0F2D, PREFIX_EVEX_0F2E, PREFIX_EVEX_0F2F, PREFIX_EVEX_0F51, PREFIX_EVEX_0F58, PREFIX_EVEX_0F59, PREFIX_EVEX_0F5A, PREFIX_EVEX_0F5B, PREFIX_EVEX_0F5C, PREFIX_EVEX_0F5D, PREFIX_EVEX_0F5E, PREFIX_EVEX_0F5F, PREFIX_EVEX_0F62, PREFIX_EVEX_0F66, PREFIX_EVEX_0F6A, PREFIX_EVEX_0F6C, PREFIX_EVEX_0F6D, PREFIX_EVEX_0F6E, PREFIX_EVEX_0F6F, PREFIX_EVEX_0F70, PREFIX_EVEX_0F72_REG_0, PREFIX_EVEX_0F72_REG_1, PREFIX_EVEX_0F72_REG_2, PREFIX_EVEX_0F72_REG_4, PREFIX_EVEX_0F72_REG_6, PREFIX_EVEX_0F73_REG_2, PREFIX_EVEX_0F73_REG_6, PREFIX_EVEX_0F76, PREFIX_EVEX_0F78, PREFIX_EVEX_0F79, PREFIX_EVEX_0F7A, PREFIX_EVEX_0F7B, PREFIX_EVEX_0F7E, PREFIX_EVEX_0F7F, PREFIX_EVEX_0FC2, PREFIX_EVEX_0FC6, PREFIX_EVEX_0FD2, PREFIX_EVEX_0FD3, PREFIX_EVEX_0FD4, PREFIX_EVEX_0FD6, PREFIX_EVEX_0FDB, PREFIX_EVEX_0FDF, PREFIX_EVEX_0FE2, PREFIX_EVEX_0FE6 PREFIX_EVEX_0FE7, PREFIX_EVEX_0FEB, PREFIX_EVEX_0FEF, PREFIX_EVEX_0FF2, PREFIX_EVEX_0FF3, PREFIX_EVEX_0FF4, PREFIX_EVEX_0FFA, PREFIX_EVEX_0FFB, PREFIX_EVEX_0FFE, PREFIX_EVEX_0F380C, PREFIX_EVEX_0F380D, PREFIX_EVEX_0F3811, PREFIX_EVEX_0F3812, PREFIX_EVEX_0F3813, PREFIX_EVEX_0F3814, PREFIX_EVEX_0F3815, PREFIX_EVEX_0F3816, PREFIX_EVEX_0F3818, PREFIX_EVEX_0F3819, PREFIX_EVEX_0F381A, PREFIX_EVEX_0F381B, PREFIX_EVEX_0F381E, PREFIX_EVEX_0F381F, PREFIX_EVEX_0F3821, PREFIX_EVEX_0F3822, PREFIX_EVEX_0F3823, PREFIX_EVEX_0F3824, PREFIX_EVEX_0F3825, PREFIX_EVEX_0F3827, PREFIX_EVEX_0F3828, PREFIX_EVEX_0F3829, PREFIX_EVEX_0F382A, PREFIX_EVEX_0F382C, PREFIX_EVEX_0F382D, PREFIX_EVEX_0F3831, PREFIX_EVEX_0F3832, PREFIX_EVEX_0F3833, PREFIX_EVEX_0F3834, PREFIX_EVEX_0F3835, PREFIX_EVEX_0F3836, PREFIX_EVEX_0F3837, PREFIX_EVEX_0F3839, PREFIX_EVEX_0F383A, PREFIX_EVEX_0F383B, PREFIX_EVEX_0F383D, PREFIX_EVEX_0F383F, PREFIX_EVEX_0F3840, PREFIX_EVEX_0F3842, PREFIX_EVEX_0F3843, PREFIX_EVEX_0F3844, PREFIX_EVEX_0F3845, PREFIX_EVEX_0F3846, PREFIX_EVEX_0F3847, PREFIX_EVEX_0F384C, PREFIX_EVEX_0F384D, PREFIX_EVEX_0F384E, PREFIX_EVEX_0F384F, PREFIX_EVEX_0F3858, PREFIX_EVEX_0F3859, PREFIX_EVEX_0F385A, PREFIX_EVEX_0F385B, PREFIX_EVEX_0F3864, PREFIX_EVEX_0F3865, PREFIX_EVEX_0F3876, PREFIX_EVEX_0F3877, PREFIX_EVEX_0F387C, PREFIX_EVEX_0F387E, PREFIX_EVEX_0F387F, PREFIX_EVEX_0F3888, PREFIX_EVEX_0F3889, PREFIX_EVEX_0F388A, PREFIX_EVEX_0F388B, PREFIX_EVEX_0F3890, PREFIX_EVEX_0F3891, PREFIX_EVEX_0F3892, PREFIX_EVEX_0F3893, PREFIX_EVEX_0F3896, PREFIX_EVEX_0F3897, PREFIX_EVEX_0F3898, PREFIX_EVEX_0F3899, PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F389C, PREFIX_EVEX_0F389D, PREFIX_EVEX_0F389E, PREFIX_EVEX_0F389F, PREFIX_EVEX_0F38A0, PREFIX_EVEX_0F38A1, PREFIX_EVEX_0F38A2, PREFIX_EVEX_0F38A3, PREFIX_EVEX_0F38A6, PREFIX_EVEX_0F38A7, PREFIX_EVEX_0F38A8, PREFIX_EVEX_0F38A9, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB, PREFIX_EVEX_0F38AC, PREFIX_EVEX_0F38AD, PREFIX_EVEX_0F38AE, PREFIX_EVEX_0F38AF, PREFIX_EVEX_0F38B6, PREFIX_EVEX_0F38B7, PREFIX_EVEX_0F38B8, PREFIX_EVEX_0F38B9, PREFIX_EVEX_0F38BA, PREFIX_EVEX_0F38BB, PREFIX_EVEX_0F38BC, PREFIX_EVEX_0F38BD, PREFIX_EVEX_0F38BE, PREFIX_EVEX_0F38BF, PREFIX_EVEX_0F38C4, PREFIX_EVEX_0F38C6_REG_1, PREFIX_EVEX_0F38C6_REG_2, PREFIX_EVEX_0F38C6_REG_5, PREFIX_EVEX_0F38C6_REG_6, PREFIX_EVEX_0F38C7_REG_1, PREFIX_EVEX_0F38C7_REG_2, PREFIX_EVEX_0F38C7_REG_5, PREFIX_EVEX_0F38C7_REG_6, PREFIX_EVEX_0F38C8, PREFIX_EVEX_0F38CA, PREFIX_EVEX_0F38CB, PREFIX_EVEX_0F38CC, PREFIX_EVEX_0F38CD, PREFIX_EVEX_0F3A00, PREFIX_EVEX_0F3A01, PREFIX_EVEX_0F3A03, PREFIX_EVEX_0F3A04, PREFIX_EVEX_0F3A05, PREFIX_EVEX_0F3A08, PREFIX_EVEX_0F3A09, PREFIX_EVEX_0F3A0A, PREFIX_EVEX_0F3A0B, PREFIX_EVEX_0F3A17, PREFIX_EVEX_0F3A18, PREFIX_EVEX_0F3A19, PREFIX_EVEX_0F3A1A, PREFIX_EVEX_0F3A1B, PREFIX_EVEX_0F3A1D, PREFIX_EVEX_0F3A1E, PREFIX_EVEX_0F3A1F, PREFIX_EVEX_0F3A21, PREFIX_EVEX_0F3A23, PREFIX_EVEX_0F3A25, PREFIX_EVEX_0F3A26, PREFIX_EVEX_0F3A27, PREFIX_EVEX_0F3A38, PREFIX_EVEX_0F3A39, PREFIX_EVEX_0F3A3A, PREFIX_EVEX_0F3A3B, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A43, PREFIX_EVEX_0F3A54, PREFIX_EVEX_0F3A55. (VEX_LEN enum): Add VEX_LEN_0F41_P_0, VEX_LEN_0F42_P_0, VEX_LEN_0F44_P_0, VEX_LEN_0F45_P_0, VEX_LEN_0F46_P_0, VEX_LEN_0F47_P_0, VEX_LEN_0F4B_P_2, VEX_LEN_0F90_P_0, VEX_LEN_0F91_P_0, VEX_LEN_0F92_P_0, VEX_LEN_0F93_P_0, VEX_LEN_0F98_P_0, VEX_LEN_0F3A30_P_2, VEX_LEN_0F3A32_P_2, VEX_W_0F41_P_0_LEN_1, VEX_W_0F42_P_0_LEN_1, VEX_W_0F44_P_0_LEN_0, VEX_W_0F45_P_0_LEN_1, VEX_W_0F46_P_0_LEN_1, VEX_W_0F47_P_0_LEN_1, VEX_W_0F4B_P_2_LEN_1, VEX_W_0F90_P_0_LEN_0, VEX_W_0F91_P_0_LEN_0, VEX_W_0F92_P_0_LEN_0, VEX_W_0F93_P_0_LEN_0, VEX_W_0F98_P_0_LEN_0, VEX_W_0F3A30_P_2_LEN_0, VEX_W_0F3A32_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F10_P_0, EVEX_W_0F10_P_1_M_0, EVEX_W_0F10_P_1_M_1, EVEX_W_0F10_P_2, EVEX_W_0F10_P_3_M_0, EVEX_W_0F10_P_3_M_1, EVEX_W_0F11_P_0, EVEX_W_0F11_P_1_M_0, EVEX_W_0F11_P_1_M_1, EVEX_W_0F11_P_2, EVEX_W_0F11_P_3_M_0, EVEX_W_0F11_P_3_M_1, EVEX_W_0F12_P_0_M_0, EVEX_W_0F12_P_0_M_1, EVEX_W_0F12_P_1, EVEX_W_0F12_P_2, EVEX_W_0F12_P_3, EVEX_W_0F13_P_0, EVEX_W_0F13_P_2, EVEX_W_0F14_P_0, EVEX_W_0F14_P_2, EVEX_W_0F15_P_0, EVEX_W_0F15_P_2, EVEX_W_0F16_P_0_M_0, EVEX_W_0F16_P_0_M_1, EVEX_W_0F16_P_1, EVEX_W_0F16_P_2, EVEX_W_0F17_P_0, EVEX_W_0F17_P_2, EVEX_W_0F28_P_0, EVEX_W_0F28_P_2, EVEX_W_0F29_P_0, EVEX_W_0F29_P_2, EVEX_W_0F2A_P_1, EVEX_W_0F2A_P_3, EVEX_W_0F2B_P_0, EVEX_W_0F2B_P_2, EVEX_W_0F2E_P_0, EVEX_W_0F2E_P_2, EVEX_W_0F2F_P_0, EVEX_W_0F2F_P_2, EVEX_W_0F51_P_0, EVEX_W_0F51_P_1, EVEX_W_0F51_P_2, EVEX_W_0F51_P_3, EVEX_W_0F58_P_0, EVEX_W_0F58_P_1, EVEX_W_0F58_P_2, EVEX_W_0F58_P_3, EVEX_W_0F59_P_0, EVEX_W_0F59_P_1, EVEX_W_0F59_P_2, EVEX_W_0F59_P_3, EVEX_W_0F5A_P_0, EVEX_W_0F5A_P_1, EVEX_W_0F5A_P_2, EVEX_W_0F5A_P_3, EVEX_W_0F5B_P_0, EVEX_W_0F5B_P_1, EVEX_W_0F5B_P_2, EVEX_W_0F5C_P_0, EVEX_W_0F5C_P_1, EVEX_W_0F5C_P_2, EVEX_W_0F5C_P_3, EVEX_W_0F5D_P_0, EVEX_W_0F5D_P_1, EVEX_W_0F5D_P_2, EVEX_W_0F5D_P_3, EVEX_W_0F5E_P_0, EVEX_W_0F5E_P_1, EVEX_W_0F5E_P_2, EVEX_W_0F5E_P_3, EVEX_W_0F5F_P_0, EVEX_W_0F5F_P_1, EVEX_W_0F5F_P_2, EVEX_W_0F5F_P_3, EVEX_W_0F62_P_2, EVEX_W_0F66_P_2, EVEX_W_0F6A_P_2, EVEX_W_0F6C_P_2, EVEX_W_0F6D_P_2, EVEX_W_0F6E_P_2, EVEX_W_0F6F_P_1, EVEX_W_0F6F_P_2, EVEX_W_0F70_P_2, EVEX_W_0F72_R_2_P_2, EVEX_W_0F72_R_6_P_2, EVEX_W_0F73_R_2_P_2, EVEX_W_0F73_R_6_P_2, EVEX_W_0F76_P_2, EVEX_W_0F78_P_0, EVEX_W_0F79_P_0, EVEX_W_0F7A_P_1, EVEX_W_0F7A_P_3, EVEX_W_0F7B_P_1, EVEX_W_0F7B_P_3, EVEX_W_0F7E_P_1, EVEX_W_0F7E_P_2, EVEX_W_0F7F_P_1, EVEX_W_0F7F_P_2, EVEX_W_0FC2_P_0, EVEX_W_0FC2_P_1, EVEX_W_0FC2_P_2, EVEX_W_0FC2_P_3, EVEX_W_0FC6_P_0, EVEX_W_0FC6_P_2, EVEX_W_0FD2_P_2, EVEX_W_0FD3_P_2, EVEX_W_0FD4_P_2, EVEX_W_0FD6_P_2, EVEX_W_0FE6_P_1, EVEX_W_0FE6_P_2, EVEX_W_0FE6_P_3, EVEX_W_0FE7_P_2, EVEX_W_0FF2_P_2, EVEX_W_0FF3_P_2, EVEX_W_0FF4_P_2, EVEX_W_0FFA_P_2, EVEX_W_0FFB_P_2, EVEX_W_0FFE_P_2, EVEX_W_0F380C_P_2, EVEX_W_0F380D_P_2, EVEX_W_0F3811_P_1, EVEX_W_0F3812_P_1, EVEX_W_0F3813_P_1, EVEX_W_0F3813_P_2, EVEX_W_0F3814_P_1, EVEX_W_0F3815_P_1, EVEX_W_0F3818_P_2, EVEX_W_0F3819_P_2, EVEX_W_0F381A_P_2, EVEX_W_0F381B_P_2, EVEX_W_0F381E_P_2, EVEX_W_0F381F_P_2, EVEX_W_0F3821_P_1, EVEX_W_0F3822_P_1, EVEX_W_0F3823_P_1, EVEX_W_0F3824_P_1, EVEX_W_0F3825_P_1, EVEX_W_0F3825_P_2, EVEX_W_0F3828_P_2, EVEX_W_0F3829_P_2, EVEX_W_0F382A_P_1, EVEX_W_0F382A_P_2, EVEX_W_0F3831_P_1, EVEX_W_0F3832_P_1, EVEX_W_0F3833_P_1, EVEX_W_0F3834_P_1, EVEX_W_0F3835_P_1, EVEX_W_0F3835_P_2, EVEX_W_0F3837_P_2, EVEX_W_0F383A_P_1, EVEX_W_0F3840_P_2, EVEX_W_0F3858_P_2, EVEX_W_0F3859_P_2, EVEX_W_0F385A_P_2, EVEX_W_0F385B_P_2, EVEX_W_0F3891_P_2, EVEX_W_0F3893_P_2, EVEX_W_0F38A1_P_2, EVEX_W_0F38A3_P_2, EVEX_W_0F38C7_R_1_P_2, EVEX_W_0F38C7_R_2_P_2, EVEX_W_0F38C7_R_5_P_2, EVEX_W_0F38C7_R_6_P_2, EVEX_W_0F3A00_P_2, EVEX_W_0F3A01_P_2, EVEX_W_0F3A04_P_2, EVEX_W_0F3A05_P_2, EVEX_W_0F3A08_P_2, EVEX_W_0F3A09_P_2, EVEX_W_0F3A0A_P_2, EVEX_W_0F3A0B_P_2, EVEX_W_0F3A18_P_2, EVEX_W_0F3A19_P_2, EVEX_W_0F3A1A_P_2, EVEX_W_0F3A1B_P_2, EVEX_W_0F3A1D_P_2, EVEX_W_0F3A21_P_2, EVEX_W_0F3A23_P_2, EVEX_W_0F3A38_P_2, EVEX_W_0F3A39_P_2, EVEX_W_0F3A3A_P_2, EVEX_W_0F3A3B_P_2, EVEX_W_0F3A43_P_2. (struct vex): Add fields evex, r, v, mask_register_specifier, zeroing, ll, b. (intel_names_xmm): Add upper 16 registers. (att_names_xmm): Ditto. (intel_names_ymm): Ditto. (att_names_ymm): Ditto. (names_zmm): New. (intel_names_zmm): Ditto. (att_names_zmm): Ditto. (names_mask): Ditto. (intel_names_mask): Ditto. (att_names_mask): Ditto. (names_rounding): Ditto. (names_broadcast): Ditto. (x86_64_table): Add escape to evex-table. (reg_table): Include reg_table evex-entries from i386-dis-evex.h. Fix prefetchwt1 instruction. (prefix_table): Add entries for new instructions. (vex_table): Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (mod_table): Ditto. (get_valid_dis386): Properly handle new instructions. (print_insn): Handle zmm and mask registers, print mask operand. (intel_operand_size): Support EVEX, new modes and sizes. (OP_E_register): Handle new modes. (OP_E_memory): Ditto. (OP_G): Ditto. (OP_XMM): Ditto. (OP_EX): Ditto. (OP_VEX): Ditto. * i386-gen.c (cpu_flag_init): Update CPU_ANY_SSE_FLAGS and CPU_ANY_AVX_FLAGS. Add CPU_AVX512F_FLAGS, CPU_AVX512CD_FLAGS, CPU_AVX512ER_FLAGS and CPU_AVX512PF_FLAGS. (cpu_flags): Add CpuAVX512F, CpuAVX512CD, CpuAVX512ER, CpuAVX512PF and CpuVREX. (operand_type_init): Add OPERAND_TYPE_REGZMM, OPERAND_TYPE_REGMASK and OPERAND_TYPE_VEC_DISP8. (opcode_modifiers): Add EVex, Masking, VecESize, Broadcast, StaticRounding, SAE, Disp8MemShift, NoDefMask. (operand_types): Add RegZMM, RegMask, Vec_Disp8, Zmmword. * i386-init.h: Regenerate. * i386-opc.h (CpuAVX512F): New. (CpuAVX512CD): New. (CpuAVX512ER): New. (CpuAVX512PF): New. (CpuVREX): New. (i386_cpu_flags): Add cpuavx512f, cpuavx512cd, cpuavx512er, cpuavx512pf and cpuvrex fields. (VecSIB): Add VecSIB512. (EVex): New. (Masking): New. (VecESize): New. (Broadcast): New. (StaticRounding): New. (SAE): New. (Disp8MemShift): New. (NoDefMask): New. (i386_opcode_modifier): Add evex, masking, vecesize, broadcast, staticrounding, sae, disp8memshift and nodefmask. (RegZMM): New. (Zmmword): Ditto. (Vec_Disp8): Ditto. (i386_operand_type): Add regzmm, regmask, zmmword and vec_disp8 fields. (RegVRex): New. * i386-opc.tbl: Add AVX512 instructions. * i386-reg.tbl: Add 16 upper XMM and YMM registers, 32 new ZMM registers, mask registers. * i386-tbl.h: Regenerate.
2013-07-26 19:20:25 +02:00
// AVX512PF instructions end.
Add clflushopt, xsaves, xsavec, xrstors gas/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * config/tc-i386.c (cpu_arch): Add .clflushopt, .xsavec, .xsaves. * doc/c-i386.texi: Document .xsavec/xsavec/.xsaves/xsaves/ clflushopt/.clfushopt. gas/testsuite/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * gas/i386/clflushopt-intel.d: New. * gas/i386/clflushopt.d: Ditto. * gas/i386/clflushopt.s: Ditto. * gas/i386/i386.exp: Run new tests. * gas/i386/x86-64-clflushopt-intel.d: New. * gas/i386/x86-64-clflushopt.d: Ditto. * gas/i386/x86-64-clflushopt.s: Ditto. * gas/i386/x86-64-xsavec-intel.d: Ditto. * gas/i386/x86-64-xsavec.d: Ditto. * gas/i386/x86-64-xsavec.s: Ditto. * gas/i386/x86-64-xsaves-intel.d: Ditto. * gas/i386/x86-64-xsaves.d: Ditto. * gas/i386/x86-64-xsaves.s: Ditto. * gas/i386/xsavec-intel.d: Ditto. * gas/i386/xsavec.d: Ditto. * gas/i386/xsavec.s: Ditto. * gas/i386/xsaves-intel.d: Ditto. * gas/i386/xsaves.d: Ditto. * gas/i386/xsaves.s: Ditto. opcodes/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * i386-dis.c (MOD enum): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4, MOD_0FC7_REG_5. (PREFIX enum): Add PREFIX_0FAE_REG_7. (reg_table): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4 MOD_0FC7_REG_5. (prefix_table): Add clflusopt. (mod_table): Add xrstors, xsavec, xsaves. * i386-gen.c (cpu_flag_init): Add CPU_CLFLUSHOPT_FLAGS, CPU_XSAVES_FLAGS, CPU_XSAVEC_FLAGS. (cpu_flags): Add CpuClflushOpt, CpuXSAVES, CpuXSAVEC. * i386-init.h: Regenerate. * i386-opc.tbl: Add clflushopt, xrstors, xrstors64, xsaves, xsaves64, xsavec, xsavec64. * i386-tbl.h: Regenerate.
2013-11-25 13:35:42 +01:00
Add support for CPUID PREFETCHWT1 Latest AVX512 spec http://download-software.intel.com/sites/default/files/managed/50/1a/319433-018.pdf Has CPUID PREFETCHWT1 for prefetchwt1 instruction, which we list as AVX512PF. This patch introduces CPUID PREFETCHWT1. gas/ * config/tc-i386.c (cpu_arch): Add .prefetchwt1. * doc/c-i386.texi: Document .prefetchwt1/prefetchwt1. opcodes/ * i386-gen.c (cpu_flag_init): Add CPU_PREFETCHWT1_FLAGS/ (cpu_flags): Add CpuPREFETCHWT1. * i386-init.h: Regenerate. * i386-opc.h (CpuPREFETCHWT1): New. (i386_cpu_flags): Add cpuprefetchwt1. * i386-opc.tbl: Cahnge CPU of prefetchwt1 from CpuAVX512PF to CpuPREFETCHWT1. * i386-tbl.h: Regenerate. gas/testsuite * gas/i386/avx512pf-intel.d: Remove prefetchwt1. * gas/i386/avx512pf.s: Ditto. * gas/i386/avx512pf.d: Ditto. * gas/i386/x86-64-avx512pf-intel.d: Ditto. * gas/i386/x86-64-avx512pf.s: Ditto. * gas/i386/x86-64-avx512pf.d: Ditto. * gas/i386/prefetchwt1-intel.d: New file. * gas/i386/prefetchwt1.s: Ditto. * gas/i386/prefetchwt1.d: Ditto. * gas/i386/x86-64-prefetchwt1-intel.d: Ditto. * gas/i386/x86-64-prefetchwt1.s: Ditto. * gas/i386/x86-64-prefetchwt1.d: Ditto.
2014-02-20 15:57:31 +01:00
// CpuPREFETCHWT1 instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
prefetchwt1, 1, 0x0F0D, 2, 2, CpuPREFETCHWT1, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
Add support for CPUID PREFETCHWT1 Latest AVX512 spec http://download-software.intel.com/sites/default/files/managed/50/1a/319433-018.pdf Has CPUID PREFETCHWT1 for prefetchwt1 instruction, which we list as AVX512PF. This patch introduces CPUID PREFETCHWT1. gas/ * config/tc-i386.c (cpu_arch): Add .prefetchwt1. * doc/c-i386.texi: Document .prefetchwt1/prefetchwt1. opcodes/ * i386-gen.c (cpu_flag_init): Add CPU_PREFETCHWT1_FLAGS/ (cpu_flags): Add CpuPREFETCHWT1. * i386-init.h: Regenerate. * i386-opc.h (CpuPREFETCHWT1): New. (i386_cpu_flags): Add cpuprefetchwt1. * i386-opc.tbl: Cahnge CPU of prefetchwt1 from CpuAVX512PF to CpuPREFETCHWT1. * i386-tbl.h: Regenerate. gas/testsuite * gas/i386/avx512pf-intel.d: Remove prefetchwt1. * gas/i386/avx512pf.s: Ditto. * gas/i386/avx512pf.d: Ditto. * gas/i386/x86-64-avx512pf-intel.d: Ditto. * gas/i386/x86-64-avx512pf.s: Ditto. * gas/i386/x86-64-avx512pf.d: Ditto. * gas/i386/prefetchwt1-intel.d: New file. * gas/i386/prefetchwt1.s: Ditto. * gas/i386/prefetchwt1.d: Ditto. * gas/i386/x86-64-prefetchwt1-intel.d: Ditto. * gas/i386/x86-64-prefetchwt1.s: Ditto. * gas/i386/x86-64-prefetchwt1.d: Ditto.
2014-02-20 15:57:31 +01:00
// CpuPREFETCHWT1 instructions end.
Add clflushopt, xsaves, xsavec, xrstors gas/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * config/tc-i386.c (cpu_arch): Add .clflushopt, .xsavec, .xsaves. * doc/c-i386.texi: Document .xsavec/xsavec/.xsaves/xsaves/ clflushopt/.clfushopt. gas/testsuite/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * gas/i386/clflushopt-intel.d: New. * gas/i386/clflushopt.d: Ditto. * gas/i386/clflushopt.s: Ditto. * gas/i386/i386.exp: Run new tests. * gas/i386/x86-64-clflushopt-intel.d: New. * gas/i386/x86-64-clflushopt.d: Ditto. * gas/i386/x86-64-clflushopt.s: Ditto. * gas/i386/x86-64-xsavec-intel.d: Ditto. * gas/i386/x86-64-xsavec.d: Ditto. * gas/i386/x86-64-xsavec.s: Ditto. * gas/i386/x86-64-xsaves-intel.d: Ditto. * gas/i386/x86-64-xsaves.d: Ditto. * gas/i386/x86-64-xsaves.s: Ditto. * gas/i386/xsavec-intel.d: Ditto. * gas/i386/xsavec.d: Ditto. * gas/i386/xsavec.s: Ditto. * gas/i386/xsaves-intel.d: Ditto. * gas/i386/xsaves.d: Ditto. * gas/i386/xsaves.s: Ditto. opcodes/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * i386-dis.c (MOD enum): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4, MOD_0FC7_REG_5. (PREFIX enum): Add PREFIX_0FAE_REG_7. (reg_table): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4 MOD_0FC7_REG_5. (prefix_table): Add clflusopt. (mod_table): Add xrstors, xsavec, xsaves. * i386-gen.c (cpu_flag_init): Add CPU_CLFLUSHOPT_FLAGS, CPU_XSAVES_FLAGS, CPU_XSAVEC_FLAGS. (cpu_flags): Add CpuClflushOpt, CpuXSAVES, CpuXSAVEC. * i386-init.h: Regenerate. * i386-opc.tbl: Add clflushopt, xrstors, xrstors64, xsaves, xsaves64, xsavec, xsavec64. * i386-tbl.h: Regenerate.
2013-11-25 13:35:42 +01:00
// CLFLUSHOPT instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
clflushopt, 1, 0x660fae, 0x7, 2, CpuClflushOpt, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
Add clflushopt, xsaves, xsavec, xrstors gas/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * config/tc-i386.c (cpu_arch): Add .clflushopt, .xsavec, .xsaves. * doc/c-i386.texi: Document .xsavec/xsavec/.xsaves/xsaves/ clflushopt/.clfushopt. gas/testsuite/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * gas/i386/clflushopt-intel.d: New. * gas/i386/clflushopt.d: Ditto. * gas/i386/clflushopt.s: Ditto. * gas/i386/i386.exp: Run new tests. * gas/i386/x86-64-clflushopt-intel.d: New. * gas/i386/x86-64-clflushopt.d: Ditto. * gas/i386/x86-64-clflushopt.s: Ditto. * gas/i386/x86-64-xsavec-intel.d: Ditto. * gas/i386/x86-64-xsavec.d: Ditto. * gas/i386/x86-64-xsavec.s: Ditto. * gas/i386/x86-64-xsaves-intel.d: Ditto. * gas/i386/x86-64-xsaves.d: Ditto. * gas/i386/x86-64-xsaves.s: Ditto. * gas/i386/xsavec-intel.d: Ditto. * gas/i386/xsavec.d: Ditto. * gas/i386/xsavec.s: Ditto. * gas/i386/xsaves-intel.d: Ditto. * gas/i386/xsaves.d: Ditto. * gas/i386/xsaves.s: Ditto. opcodes/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * i386-dis.c (MOD enum): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4, MOD_0FC7_REG_5. (PREFIX enum): Add PREFIX_0FAE_REG_7. (reg_table): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4 MOD_0FC7_REG_5. (prefix_table): Add clflusopt. (mod_table): Add xrstors, xsavec, xsaves. * i386-gen.c (cpu_flag_init): Add CPU_CLFLUSHOPT_FLAGS, CPU_XSAVES_FLAGS, CPU_XSAVEC_FLAGS. (cpu_flags): Add CpuClflushOpt, CpuXSAVES, CpuXSAVEC. * i386-init.h: Regenerate. * i386-opc.tbl: Add clflushopt, xrstors, xrstors64, xsaves, xsaves64, xsavec, xsavec64. * i386-tbl.h: Regenerate.
2013-11-25 13:35:42 +01:00
// CLFLUSHOPT instructions end.
// XSAVES/XRSTORS instructions.
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
xrstors, 1, 0xfc7, 0x3, 2, CpuXSAVES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xrstors64, 1, 0xfc7, 0x3, 2, CpuXSAVES|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
xsaves, 1, 0xfc7, 0x5, 2, CpuXSAVES, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xsaves64, 1, 0xfc7, 0x5, 2, CpuXSAVES|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
Add clflushopt, xsaves, xsavec, xrstors gas/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * config/tc-i386.c (cpu_arch): Add .clflushopt, .xsavec, .xsaves. * doc/c-i386.texi: Document .xsavec/xsavec/.xsaves/xsaves/ clflushopt/.clfushopt. gas/testsuite/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * gas/i386/clflushopt-intel.d: New. * gas/i386/clflushopt.d: Ditto. * gas/i386/clflushopt.s: Ditto. * gas/i386/i386.exp: Run new tests. * gas/i386/x86-64-clflushopt-intel.d: New. * gas/i386/x86-64-clflushopt.d: Ditto. * gas/i386/x86-64-clflushopt.s: Ditto. * gas/i386/x86-64-xsavec-intel.d: Ditto. * gas/i386/x86-64-xsavec.d: Ditto. * gas/i386/x86-64-xsavec.s: Ditto. * gas/i386/x86-64-xsaves-intel.d: Ditto. * gas/i386/x86-64-xsaves.d: Ditto. * gas/i386/x86-64-xsaves.s: Ditto. * gas/i386/xsavec-intel.d: Ditto. * gas/i386/xsavec.d: Ditto. * gas/i386/xsavec.s: Ditto. * gas/i386/xsaves-intel.d: Ditto. * gas/i386/xsaves.d: Ditto. * gas/i386/xsaves.s: Ditto. opcodes/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * i386-dis.c (MOD enum): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4, MOD_0FC7_REG_5. (PREFIX enum): Add PREFIX_0FAE_REG_7. (reg_table): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4 MOD_0FC7_REG_5. (prefix_table): Add clflusopt. (mod_table): Add xrstors, xsavec, xsaves. * i386-gen.c (cpu_flag_init): Add CPU_CLFLUSHOPT_FLAGS, CPU_XSAVES_FLAGS, CPU_XSAVEC_FLAGS. (cpu_flags): Add CpuClflushOpt, CpuXSAVES, CpuXSAVEC. * i386-init.h: Regenerate. * i386-opc.tbl: Add clflushopt, xrstors, xrstors64, xsaves, xsaves64, xsavec, xsavec64. * i386-tbl.h: Regenerate.
2013-11-25 13:35:42 +01:00
// XSAVES instructions end.
// XSAVEC instructions.
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
xsavec, 1, 0xfc7, 0x4, 2, CpuXSAVEC, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
xsavec64, 1, 0xfc7, 0x4, 2, CpuXSAVEC|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Unspecified|BaseIndex }
Add clflushopt, xsaves, xsavec, xrstors gas/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * config/tc-i386.c (cpu_arch): Add .clflushopt, .xsavec, .xsaves. * doc/c-i386.texi: Document .xsavec/xsavec/.xsaves/xsaves/ clflushopt/.clfushopt. gas/testsuite/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * gas/i386/clflushopt-intel.d: New. * gas/i386/clflushopt.d: Ditto. * gas/i386/clflushopt.s: Ditto. * gas/i386/i386.exp: Run new tests. * gas/i386/x86-64-clflushopt-intel.d: New. * gas/i386/x86-64-clflushopt.d: Ditto. * gas/i386/x86-64-clflushopt.s: Ditto. * gas/i386/x86-64-xsavec-intel.d: Ditto. * gas/i386/x86-64-xsavec.d: Ditto. * gas/i386/x86-64-xsavec.s: Ditto. * gas/i386/x86-64-xsaves-intel.d: Ditto. * gas/i386/x86-64-xsaves.d: Ditto. * gas/i386/x86-64-xsaves.s: Ditto. * gas/i386/xsavec-intel.d: Ditto. * gas/i386/xsavec.d: Ditto. * gas/i386/xsavec.s: Ditto. * gas/i386/xsaves-intel.d: Ditto. * gas/i386/xsaves.d: Ditto. * gas/i386/xsaves.s: Ditto. opcodes/ 2014-02-12 Ilya Tocar <ilya.tocar@intel.com> * i386-dis.c (MOD enum): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4, MOD_0FC7_REG_5. (PREFIX enum): Add PREFIX_0FAE_REG_7. (reg_table): Add MOD_0FC7_REG_3, MOD_0FC7_REG_4 MOD_0FC7_REG_5. (prefix_table): Add clflusopt. (mod_table): Add xrstors, xsavec, xsaves. * i386-gen.c (cpu_flag_init): Add CPU_CLFLUSHOPT_FLAGS, CPU_XSAVES_FLAGS, CPU_XSAVEC_FLAGS. (cpu_flags): Add CpuClflushOpt, CpuXSAVES, CpuXSAVEC. * i386-init.h: Regenerate. * i386-opc.tbl: Add clflushopt, xrstors, xrstors64, xsaves, xsaves64, xsavec, xsavec64. * i386-tbl.h: Regenerate.
2013-11-25 13:35:42 +01:00
// XSAVEC instructions end.
Add support for Intel SGX instructions Add Intel SGX instructions support to assembler and disassembler. gas/ * config/tc-i386.c (cpu_arch): Add .se1. * doc/c-i386.texi: Document .se1/se1. gas/testsuite/ * gas/i386/i386.exp: Run SE1 tests. * gas/i386/se1.d: New file. * gas/i386/se1.s: Ditto. * gas/i386/x86-64-se1.d: Ditto. * gas/i386/x86-64-se1.s: Ditto. opcodes/ * i386-dis.c (rm_table): Add encls, enclu. * i386-gen.c (cpu_flag_init): Add CPU_SE1_FLAGS, (cpu_flags): Add CpuSE1. * i386-opc.h (enum): Add CpuSE1. (i386_cpu_flags): Add cpuse1. * i386-opc.tbl: Add encls, enclu. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2014-04-03 14:40:04 +02:00
// SGX instructions.
encls, 0, 0xf01cf, None, 3, CpuSE1, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
enclu, 0, 0xf01d7, None, 3, CpuSE1, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: Add Intel ENCLV to assembler and disassembler gas/ * testsuite/gas/i386/se1.s: Add enclv. * testsuite/gas/i386/x86-64-se1.s: Likewise. * testsuite/gas/i386/se1.d: Updated. * testsuite/gas/i386/x86-64-se1.d: Likewise. opcodes/ * i386-dis.c (rm_table): Add enclv. * i386-opc.tbl: Add enclv. * i386-tbl.h: Regenerated.
2018-10-05 20:56:42 +02:00
enclv, 0, 0xf01c0, None, 3, CpuSE1, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Add support for Intel SGX instructions Add Intel SGX instructions support to assembler and disassembler. gas/ * config/tc-i386.c (cpu_arch): Add .se1. * doc/c-i386.texi: Document .se1/se1. gas/testsuite/ * gas/i386/i386.exp: Run SE1 tests. * gas/i386/se1.d: New file. * gas/i386/se1.s: Ditto. * gas/i386/x86-64-se1.d: Ditto. * gas/i386/x86-64-se1.s: Ditto. opcodes/ * i386-dis.c (rm_table): Add encls, enclu. * i386-gen.c (cpu_flag_init): Add CPU_SE1_FLAGS, (cpu_flags): Add CpuSE1. * i386-opc.h (enum): Add CpuSE1. (i386_cpu_flags): Add cpuse1. * i386-opc.tbl: Add encls, enclu. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2014-04-03 14:40:04 +02:00
// SGX instructions end.
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
// AVX512VL instructions.
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vgatherdpd, 2, 0x6692, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM|RegYMM }
vgatherqpd, 2, 0x6693, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
vgatherqpd, 2, 0x6693, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegYMM }
vpgatherdq, 2, 0x6690, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM|RegYMM }
vpgatherqq, 2, 0x6691, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegXMM }
vpgatherqq, 2, 0x6691, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Qword|Unspecified|BaseIndex, RegYMM }
vpscatterdq, 2, 0x66A0, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Qword|Unspecified|BaseIndex }
vpscatterqq, 2, 0x66A1, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vpscatterqq, 2, 0x66A1, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Qword|Unspecified|BaseIndex }
vscatterdpd, 2, 0x66A2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM, Qword|Unspecified|BaseIndex }
vscatterqpd, 2, 0x66A3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Qword|Unspecified|BaseIndex }
vscatterqpd, 2, 0x66A3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW1|Disp8MemShift=3|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Qword|Unspecified|BaseIndex }
vgatherdps, 2, 0x6692, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vgatherdps, 2, 0x6692, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegYMM }
vgatherqps, 2, 0x6693, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vgatherqps, 2, 0x6693, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vpgatherdd, 2, 0x6690, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vpgatherdd, 2, 0x6690, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegYMM }
vpgatherqd, 2, 0x6691, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vpgatherqd, 2, 0x6691, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Dword|Unspecified|BaseIndex, RegXMM }
vpscatterdd, 2, 0x66A0, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
vpscatterdd, 2, 0x66A0, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Dword|Unspecified|BaseIndex }
vpscatterqd, 2, 0x66A1, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
vpscatterqd, 2, 0x66A1, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
vscatterdps, 2, 0x66A2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
vscatterdps, 2, 0x66A2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, Dword|Unspecified|BaseIndex }
vscatterqps, 2, 0x66A3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
vscatterqps, 2, 0x66A3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=2|NoDefMask|VexOpcode=1|VexW0|Disp8MemShift=2|VecSIB=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, Dword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtdq2pd, 2, 0xF3E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvtdq2pd, 2, 0xF3E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
vcvtudq2pd, 2, 0xF37A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvtudq2pd, 2, 0xF37A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
x86: convert broadcast insn attribute to boolean The (only) valid broadcast type for an insn can be inferred from other information.
2018-03-28 14:24:05 +02:00
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtpd2dq, 2, 0xF2E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2dq, 2, 0xF2E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtpd2dqx, 2, 0xF2E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2dqy, 2, 0xF2E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtpd2ps, 2, 0x665A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2ps, 2, 0x665A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtpd2psx, 2, 0x665A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2psy, 2, 0x665A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtpd2udq, 2, 0x79, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2udq, 2, 0x79, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtpd2udqx, 2, 0x79, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtpd2udqy, 2, 0x79, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vcvtph2ps, 2, 0x6613, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512VL insns
2018-09-13 11:23:50 +02:00
vcvtph2ps, 2, 0x6613, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtps2pd, 2, 0x5A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvtps2pd, 2, 0x5A, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtps2ph, 3, 0x661D, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex128|MaskingMorZ|VexOpcode=2|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
vcvtps2ph, 3, 0x661D, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex256|MaskingMorZ|VexOpcode=2|VexW0|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM, RegXMM|Unspecified|BaseIndex }
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvttpd2dq, 2, 0x66E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Unspecified|Qword|BaseIndex, RegXMM }
vcvttpd2dq, 2, 0x66E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvttpd2dqx, 2, 0x66E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvttpd2dqy, 2, 0x66E6, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvttpd2udq, 2, 0x78, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvttpd2udq, 2, 0x78, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvttpd2udqx, 2, 0x78, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvttpd2udqy, 2, 0x78, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vmovddup, 2, 0xF212, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpslld, 3, 0x66F2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsrad, 3, 0x66E2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsrld, 3, 0x66D2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: fold various AVX512VL templates into their AVX512F counterparts
2018-07-19 08:29:35 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpsllq, 3, 0x66F3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsraq, 3, 0x66E2, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsrlq, 3, 0x66D3, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovdb, 2, 0xF331, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovdb, 2, 0xF331, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovsdb, 2, 0xF321, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovsdb, 2, 0xF321, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovusdb, 2, 0xF311, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovusdb, 2, 0xF311, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovdw, 2, 0xF333, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovdw, 2, 0xF333, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsdw, 2, 0xF323, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovsdw, 2, 0xF323, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovusdw, 2, 0xF313, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovusdw, 2, 0xF313, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovqb, 2, 0xF332, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Word|Unspecified|BaseIndex }
vpmovqb, 2, 0xF332, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovsqb, 2, 0xF322, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Word|Unspecified|BaseIndex }
vpmovsqb, 2, 0xF322, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovusqb, 2, 0xF312, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Word|Unspecified|BaseIndex }
vpmovusqb, 2, 0xF312, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Dword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovqd, 2, 0xF335, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovqd, 2, 0xF335, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsqd, 2, 0xF325, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovsqd, 2, 0xF325, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovusqd, 2, 0xF315, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovusqd, 2, 0xF315, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovqw, 2, 0xF334, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovqw, 2, 0xF334, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovsqw, 2, 0xF324, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovsqw, 2, 0xF324, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
vpmovusqw, 2, 0xF314, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Dword|Unspecified|BaseIndex }
vpmovusqw, 2, 0xF314, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxbd, 2, 0x6621, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
vpmovsxbd, 2, 0x6621, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM }
vpmovzxbd, 2, 0x6631, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
vpmovzxbd, 2, 0x6631, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxbq, 2, 0x6622, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Word|Unspecified|BaseIndex, RegXMM }
vpmovsxbq, 2, 0x6622, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
vpmovzxbq, 2, 0x6632, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Word|Unspecified|BaseIndex, RegXMM }
vpmovzxbq, 2, 0x6632, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxdq, 2, 0x6625, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512VL insns
2018-09-13 11:23:50 +02:00
vpmovsxdq, 2, 0x6625, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovzxdq, 2, 0x6635, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: drop bogus IgnoreSize from AVX512VL insns
2018-09-13 11:23:50 +02:00
vpmovzxdq, 2, 0x6635, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxwd, 2, 0x6623, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxwd, 2, 0x6623, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovzxwd, 2, 0x6633, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovzxwd, 2, 0x6633, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxwq, 2, 0x6624, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
vpmovsxwq, 2, 0x6624, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM }
vpmovzxwq, 2, 0x6634, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
vpmovzxwq, 2, 0x6634, None, 1, CpuAVX512F|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
Add support for AVX512VL. Add AVX512VL versions of AVX512F instructions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512vl, CPU_AVX512VL_FLAGS. (build_vex_prefix): Don't abort on VEX.W. (check_VecOperands): Support BROADCAST_1TO4 and BROADCAST_1TO2. (check_VecOperations): Ditto. * doc/c-i386.texi: Document avx512vl/.avx512vl. gas/testsuite/ * gas/i386/avx512f_vl-intel.d: New. * gas/i386/avx512f_vl-opts-intel.d: New. * gas/i386/avx512f_vl-opts.d: New. * gas/i386/avx512f_vl-opts.s: New. * gas/i386/avx512f_vl-wig.s: New. * gas/i386/avx512f_vl-wig1-intel.d: New. * gas/i386/avx512f_vl-wig1.d: New. * gas/i386/avx512f_vl.d: New. * gas/i386/avx512f_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512f_vl-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts-intel.d: New. * gas/i386/x86-64-avx512f_vl-opts.d: New. * gas/i386/x86-64-avx512f_vl-opts.s: New. * gas/i386/x86-64-avx512f_vl-wig.s: New. * gas/i386/x86-64-avx512f_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512f_vl-wig1.d: New. * gas/i386/x86-64-avx512f_vl.d: New. * gas/i386/x86-64-avx512f_vl.s: New. opcodes/ * i386-dis.c (intel_operand_size): Support 128/256 length in vex_vsib_q_w_dq_mode. (OP_E_memory): Add ymmq_mode handling, handle new broadcast. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VL_FLAGS. (cpu_flags): Add CpuAVX512VL. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512VL): New. (i386_cpu_flags): Add cpuavx512vl. (BROADCAST_1TO4, BROADCAST_1TO2): Define. * i386-opc.tbl: Add AVX512VL instructions. * i386-tbl.h: Regenerate.
2014-07-18 12:59:54 +02:00
// AVX512VL instructions end.
Add support for AVX512VL versions of AVX512CD instructions. gas/testsuite/ * gas/i386/avx512cd_vl-intel.d: New. * gas/i386/avx512cd_vl.d: New. * gas/i386/avx512cd_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512cd_vl-intel.d: New. * gas/i386/x86-64-avx512cd_vl.d: New. * gas/i386/x86-64-avx512cd_vl.s: New. opcodes/ * i386-opc.tbl: Add AVX512VL and AVX512CD instructions. * i386-tbl.h: Regenerate.
2014-07-10 12:21:48 +02:00
x86: fold several AVX512VL templates The differences between some of the register and memory forms of the same insn often don't really require the templates to be separate. For example, Disp8MemShift is simply irrelevant to register forms. Fold these as far as possible, and also fold register-only forms. Further folding is possible, but needs other prereq work done first. A note regarding EVEXDYN: This is intended to be used only when no other properties of the template would make is_evex_encoding() return true. In all "normal" cases I think it is preferable to omit this indicator, to keep the table half way readable.
2018-03-08 08:58:55 +01:00
// AVX512BW instructions.
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: drop bogus IgnoreSize from AVX512BW insns
2018-09-13 11:24:23 +02:00
kaddd, 3, 0x664A, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandd, 3, 0x6641, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandnd, 3, 0x6642, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegMask, RegMask, RegMask }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
kmovd, 2, 0x6690, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask|Dword|Unspecified|BaseIndex, RegMask }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
kmovd, 2, 0x6691, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, Dword|Unspecified|BaseIndex }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
kmovd, 2, 0xF292, None, 1, CpuAVX512BW, D|Modrm|Vex=1|VexOpcode=0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegMask }
x86: drop bogus IgnoreSize from AVX512BW insns
2018-09-13 11:24:23 +02:00
knotd, 2, 0x6644, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
kord, 3, 0x6645, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kortestd, 2, 0x6698, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
ktestd, 2, 0x6699, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
kxnord, 3, 0x6646, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxord, 3, 0x6647, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegMask, RegMask, RegMask }
kaddq, 3, 0x4A, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandnq, 3, 0x42, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegMask, RegMask, RegMask }
kandq, 3, 0x41, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
kmovq, 2, 0x90, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask|Qword|Unspecified|BaseIndex, RegMask }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
kmovq, 2, 0x91, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, Qword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from AVX512BW insns
2018-09-13 11:24:23 +02:00
kmovq, 2, 0xF292, None, 1, CpuAVX512BW, D|Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg64, RegMask }
knotq, 2, 0x44, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
korq, 3, 0x45, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kortestq, 2, 0x98, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
ktestq, 2, 0x99, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
kunpckdq, 3, 0x4B, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kunpckwd, 3, 0x4B, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxnorq, 3, 0x46, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxorq, 3, 0x47, None, 1, CpuAVX512BW, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegMask, RegMask, RegMask }
kshiftld, 3, 0x6633, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
kshiftlq, 3, 0x6633, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
kshiftrd, 3, 0x6631, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
kshiftrq, 3, 0x6631, None, 1, CpuAVX512BW, Modrm|Vex=1|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vdbpsadbw, 4, 0x6642, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Optimize EVEX vector load/store instructions When there is no write mask, we can encode lower 16 128-bit/256-bit EVEX vector register load and store instructions as VEX vector register load and store instructions with -O1. gas/ PR gas/24348 * config/tc-i386.c (optimize_encoding): Encode 128-bit and 256-bit EVEX vector register load/store instructions as VEX vector register load/store instructions for -O1. * doc/c-i386.texi: Update -O1 documentation. * testsuite/gas/i386/i386.exp: Run PR gas/24348 tests. * testsuite/gas/i386/optimize-1.s: Add tests for EVEX vector load/store instructions. * testsuite/gas/i386/optimize-2.s: Likewise. * testsuite/gas/i386/optimize-3.s: Likewise. * testsuite/gas/i386/optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-2.s: Likewise. * testsuite/gas/i386/x86-64-optimize-3.s: Likewise. * testsuite/gas/i386/x86-64-optimize-4.s: Likewise. * testsuite/gas/i386/x86-64-optimize-5.s: Likewise. * testsuite/gas/i386/x86-64-optimize-6.s: Likewise. * testsuite/gas/i386/optimize-1.d: Updated. * testsuite/gas/i386/optimize-2.d: Likewise. * testsuite/gas/i386/optimize-3.d: Likewise. * testsuite/gas/i386/optimize-4.d: Likewise. * testsuite/gas/i386/optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-2.d: Likewise. * testsuite/gas/i386/x86-64-optimize-3.d: Likewise. * testsuite/gas/i386/x86-64-optimize-4.d: Likewise. * testsuite/gas/i386/x86-64-optimize-5.d: Likewise. * testsuite/gas/i386/x86-64-optimize-6.d: Likewise. * testsuite/gas/i386/optimize-7.d: New file. * testsuite/gas/i386/optimize-7.s: Likewise. * testsuite/gas/i386/x86-64-optimize-8.d: Likewise. * testsuite/gas/i386/x86-64-optimize-8.s: Likewise. opcodes/ PR gas/24348 * i386-opc.tbl: Add Optimize to vmovdqa32, vmovdqa64, vmovdqu8, vmovdqu16, vmovdqu32 and vmovdqu64. * i386-tbl.h: Regenerated.
2019-03-18 01:56:10 +01:00
vmovdqu8, 2, 0xF26F, None, 1, CpuAVX512BW, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vmovdqu16, 2, 0xF26F, None, 1, CpuAVX512BW, D|Modrm|MaskingMorZ|VexOpcode=0|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpabsb, 2, 0x661C, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpmaxsb, 3, 0x663C, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminsb, 3, 0x6638, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshufb, 3, 0x6600, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpabsw, 2, 0x661D, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpmaddubsw, 3, 0x6604, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxuw, 3, 0x663E, None, 1, CpuAVX512BW, Modrm|Masking=3|VexWIG|VexOpcode=1|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminuw, 3, 0x663A, None, 1, CpuAVX512BW, Modrm|Masking=3|VexWIG|VexOpcode=1|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmulhrsw, 3, 0x660B, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpackssdw, 3, 0x666B, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpacksswb, 3, 0x6663, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpackuswb, 3, 0x6667, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpackusdw, 3, 0x662B, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpaddb, 3, 0x66FC, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddsb, 3, 0x66EC, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddusb, 3, 0x66DC, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpavgb, 3, 0x66E0, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: add more VexWIG Commits 6865c0435a ("x86: Support VEX/EVEX WIG encoding") and 6fa52824c3 ("x86: Replace VexW=3 with VexWIG") omitted quite a few templates, oddly enough in some cases despite testcases getting added (which then were recorded with wrong expected output). Also adjust VPMAXUB's attributes in the AVX512BW case to match ordering of that of neighboring templates. For the moment SSE2AVX templates are left alone, as it isn't clear whether they were intentionally left untouched by the original commits (the descriptions don't say either way). In this context I question the decision in commit 0375113302 ("x86: Add -mvexwig=[0|1] option to assembler") to move the logic to determine the value of the W bit ahead of the decision whether to use 2-byte VEX: While I can see this as one possible interpretation of -mvexwig=, the other alternative (setting the value of the bit only if it actually exists in the encoding) looks as reasonable to me, and perhaps even more in line with us generally trying to pick the shortest encoding.
2018-11-06 11:39:42 +01:00
vpmaxub, 3, 0x66DE, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpminub, 3, 0x66DA, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsubb, 3, 0x66F8, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsubsb, 3, 0x66E8, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsubusb, 3, 0x66D8, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpckhbw, 3, 0x6668, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpcklbw, 3, 0x6660, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddsw, 3, 0x66ED, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddusw, 3, 0x66DD, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpaddw, 3, 0x66FD, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpavgw, 3, 0x66E3, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmaxsw, 3, 0x66EE, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpminsw, 3, 0x66EA, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmulhuw, 3, 0x66E4, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmulhw, 3, 0x66E5, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmullw, 3, 0x66D5, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsllw, 3, 0x6671, 6, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpsllw, 3, 0x66F1, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexVVVV=1|VexWIG|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsllw, 3, 0x66F1, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexWIG|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpsraw, 3, 0x6671, 4, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpsraw, 3, 0x66E1, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsraw, 3, 0x66E1, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpsrlw, 3, 0x6671, 2, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpsrlw, 3, 0x66D1, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegXMM, RegXMM }
vpsrlw, 3, 0x66D1, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vpsubsw, 3, 0x66E9, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsubusw, 3, 0x66D9, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsubw, 3, 0x66F9, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpckhwd, 3, 0x6669, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpunpcklwd, 3, 0x6661, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpalignr, 4, 0x660F, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpblendmb, 3, 0x6666, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpbroadcastb, 2, 0x6678, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Byte|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpbroadcastb, 2, 0x667A, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpblendmw, 3, 0x6666, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2w, 3, 0x6675, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2w, 3, 0x667D, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermw, 3, 0x668D, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsllvw, 3, 0x6612, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsravw, 3, 0x6611, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpsrlvw, 3, 0x6610, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpbroadcastw, 2, 0x6679, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Word|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop unnecessary {,No}Rex64
2018-09-13 11:08:37 +02:00
vpbroadcastw, 2, 0x667B, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpeqb, 3, 0x6674, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpgtb, 3, 0x6664, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpcmpb, 4, 0x663F, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>b, 3, 0x663F, <avx_irel:imm>, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW0|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpcmpub, 4, 0x663E, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>ub, 3, 0x663E, <avx_irel:imm>, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW0|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpcmpeqw, 3, 0x6675, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vpcmpgtw, 3, 0x6665, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=0|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpcmpw, 4, 0x663F, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>w, 3, 0x663F, <avx_irel:imm>, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vpcmpuw, 4, 0x663E, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: use template for AVX512 integer comparison insns These all follow a common pattern.
2020-03-09 10:14:55 +01:00
vpcmp<avx_irel>uw, 3, 0x663E, <avx_irel:imm>, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=2|VexVVVV|VexW1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpslldq, 3, 0x6673, 7, 1, CpuAVX512BW, Modrm|VexOpcode=0|VexWIG|VexVVVV=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpsrldq, 3, 0x6673, 3, 1, CpuAVX512BW, Modrm|VexOpcode=0|VexWIG|VexVVVV=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
vpextrw, 3, 0x66C5, None, 1, CpuAVX512BW, Load|Modrm|EVex128|VexOpcode=0|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vpextrw, 3, 0x6615, None, 1, CpuAVX512BW, RegMem|EVex128|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpextrw, 3, 0x6615, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=2|VexWIG|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Word|Unspecified|BaseIndex }
x86: drop bogus Disp8MemShift attributes In commit dc821c5f9a ("x86: replace Reg8, Reg16, Reg32, and Reg64") I apparently blindly copied the original register/memory templates into separate ones, in particular without removing the Disp8MemShift which are applicable to templates with memory operands only.
2019-07-01 08:27:38 +02:00
vpinsrw, 4, 0x66C4, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=0|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpinsrw, 4, 0x66C4, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=0|VexWIG|VexVVVV|Disp8MemShift=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Word|Unspecified|BaseIndex, RegXMM, RegXMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: make RegMem an opcode modifier ... instead of an operand type bit: It's an insn property, not an operand one. There's just one actual change to be made to the templates: Most are now required to have the (unswapped) destination go into ModR/M.rm, so VMOVD template needs its opcode adjusted accordingly and its operands swapped. {,V}MOVS{S,D}, otoh, are left alone in this regard, as otherwise generated code would differ from what we've been producing so far (which I don't think is wanted). Take the opportunity and add a missing IgnoreSize to pextrb (leading to an error in 16-bit mode), and take the liberty to once again drop stray IgnoreSize attributes from lines changed and neighboring related ones.
2019-07-16 09:31:36 +02:00
vpextrb, 3, 0x6614, None, 1, CpuAVX512BW, RegMem|EVex128|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Reg64 }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpextrb, 3, 0x6614, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=2|VexWIG|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Byte|Unspecified|BaseIndex }
x86: fix various non-LIG templates Quite a few templates were marked LIG while really the insns aren't. Introduce descriptive shorthands once again, instead of continuing to use the less legible original forms.
2018-11-06 11:42:08 +01:00
vpinsrb, 4, 0x6620, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=2|VexWIG|VexVVVV=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Reg64, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpinsrb, 4, 0x6620, None, 1, CpuAVX512BW, Modrm|EVex128|VexOpcode=2|VexWIG|VexVVVV|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Byte|Unspecified|BaseIndex, RegXMM, RegXMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmaddwd, 3, 0x66F5, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpmovb2m, 2, 0xF329, None, 1, CpuAVX512BW, Modrm|EVex=5|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegMask }
vpmovw2m, 2, 0xF329, None, 1, CpuAVX512BW, Modrm|EVex=5|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegMask }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpmovm2b, 2, 0xF328, None, 1, CpuAVX512BW, Modrm|EVex=5|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
vpmovm2w, 2, 0xF328, None, 1, CpuAVX512BW, Modrm|EVex=5|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovswb, 2, 0xF320, None, 1, CpuAVX512BW, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovswb, 2, 0xF320, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovswb, 2, 0xF320, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
vpmovuswb, 2, 0xF310, None, 1, CpuAVX512BW, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovuswb, 2, 0xF310, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovuswb, 2, 0xF310, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
vpmovwb, 2, 0xF330, None, 1, CpuAVX512BW, Modrm|EVex=1|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM, RegYMM|Unspecified|BaseIndex }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovwb, 2, 0xF330, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|MaskingMorZ|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM, RegXMM|Qword|Unspecified|BaseIndex }
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpmovwb, 2, 0xF330, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=3|MaskingMorZ|VexOpcode=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM, RegXMM|Unspecified|BaseIndex }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxbw, 2, 0x6620, None, 1, CpuAVX512BW, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovsxbw, 2, 0x6620, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexWIG|VexOpcode=1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovsxbw, 2, 0x6620, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexWIG|VexOpcode=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
vpmovzxbw, 2, 0x6630, None, 1, CpuAVX512BW, Modrm|EVex=1|Masking=3|VexOpcode=1|VexWIG|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Unspecified|BaseIndex, RegZMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vpmovzxbw, 2, 0x6630, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexWIG|VexOpcode=1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpmovzxbw, 2, 0x6630, None, 1, CpuAVX512BW|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexWIG|VexOpcode=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Unspecified|BaseIndex, RegYMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpsadbw, 3, 0x66F6, None, 1, CpuAVX512BW, Modrm|VexOpcode=0|VexVVVV=1|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpshufhw, 3, 0xF370, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpshuflw, 3, 0xF270, None, 1, CpuAVX512BW, Modrm|Masking=3|VexOpcode=0|VexWIG|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vptestmb, 3, 0x6626, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vptestmw, 3, 0x6626, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
x86: fold various AVX512BW templates
2018-07-19 08:32:17 +02:00
vptestnmb, 3, 0xF326, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vptestnmw, 3, 0xF326, None, 1, CpuAVX512BW, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
Add support for AVX512BW instructions and their AVX512VL versions. gas/ * config/tc-i386.c (cpu_arch): Add .avx512bw, CPU_AVX512BW_FLAGS. * doc/c-i386.texi: Document avx512bw/.avx512bw. gas/testsuite/ * gas/i386/avx512bw-intel.d: New. * gas/i386/avx512bw-opts-intel.d: New. * gas/i386/avx512bw-opts.d: New. * gas/i386/avx512bw-opts.s: New. * gas/i386/avx512bw-wig.s: New. * gas/i386/avx512bw-wig1-intel.d: New. * gas/i386/avx512bw-wig1.d: New. * gas/i386/avx512bw.d: New. * gas/i386/avx512bw.s: New. * gas/i386/avx512bw_vl-intel.d: New. * gas/i386/avx512bw_vl-opts-intel.d: New. * gas/i386/avx512bw_vl-opts.d: New. * gas/i386/avx512bw_vl-opts.s: New. * gas/i386/avx512bw_vl-wig.s: New. * gas/i386/avx512bw_vl-wig1-intel.d: New. * gas/i386/avx512bw_vl-wig1.d: New. * gas/i386/avx512bw_vl.d: New. * gas/i386/avx512bw_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512bw-intel.d: New. * gas/i386/x86-64-avx512bw-opts-intel.d: New. * gas/i386/x86-64-avx512bw-opts.d: New. * gas/i386/x86-64-avx512bw-opts.s: New. * gas/i386/x86-64-avx512bw-wig.s: New. * gas/i386/x86-64-avx512bw-wig1-intel.d: New. * gas/i386/x86-64-avx512bw-wig1.d: New. * gas/i386/x86-64-avx512bw.d: New. * gas/i386/x86-64-avx512bw.s: New. * gas/i386/x86-64-avx512bw_vl-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts-intel.d: New. * gas/i386/x86-64-avx512bw_vl-opts.d: New. * gas/i386/x86-64-avx512bw_vl-opts.s: New. * gas/i386/x86-64-avx512bw_vl-wig.s: New. * gas/i386/x86-64-avx512bw_vl-wig1-intel.d: New. * gas/i386/x86-64-avx512bw_vl-wig1.d: New. * gas/i386/x86-64-avx512bw_vl.d: New. * gas/i386/x86-64-avx512bw_vl.s: New. opcodes/ * i386-dis-evex.h: Add new instructions (prefixes bellow). * i386-dis.c (fetch_data): Add EdqwS, Edb, Edw, MaskBDE. (enum): Add dqw_swap_mode, db_mode, dw_mode, mask_bd_mode, REG_EVEX_0F71. (PREFIX enum): Add PREFIX_VEX_0F4A, PREFIX_VEX_0F99, PREFIX_VEX_0F3A31, PREFIX_VEX_0F3A33, PREFIX_EVEX_0F60, PREFIX_EVEX_0F61, PREFIX_EVEX_0F63, PREFIX_EVEX_0F64, PREFIX_EVEX_0F65, PREFIX_EVEX_0F67, PREFIX_EVEX_0F68, PREFIX_EVEX_0F69, PREFIX_EVEX_0F6B, PREFIX_EVEX_0F71_REG_2, PREFIX_EVEX_0F71_REG_4, PREFIX_EVEX_0F71_REG_6, PREFIX_EVEX_0F73_REG_3, PREFIX_EVEX_0F73_REG_7, PREFIX_EVEX_0F74, PREFIX_EVEX_0F75, PREFIX_EVEX_0FC4, PREFIX_EVEX_0FC5, PREFIX_EVEX_0FD1, PREFIX_EVEX_0FD5, PREFIX_EVEX_0FD8, PREFIX_EVEX_0FD9, PREFIX_EVEX_0FDA, PREFIX_EVEX_0FDC, PREFIX_EVEX_0FDD, PREFIX_EVEX_0FDE, PREFIX_EVEX_0FE0, PREFIX_EVEX_0FE1, PREFIX_EVEX_0FE3, PREFIX_EVEX_0FE4, PREFIX_EVEX_0FE5, PREFIX_EVEX_0FE8, PREFIX_EVEX_0FE9, PREFIX_EVEX_0FEA, PREFIX_EVEX_0FEC, PREFIX_EVEX_0FED, PREFIX_EVEX_0FEE, PREFIX_EVEX_0FF1, PREFIX_EVEX_0FF5, PREFIX_EVEX_0FF6, PREFIX_EVEX_0FF8, PREFIX_EVEX_0FF9, PREFIX_EVEX_0FFC, PREFIX_EVEX_0FFD, PREFIX_EVEX_0F3800, PREFIX_EVEX_0F3804, PREFIX_EVEX_0F380B, PREFIX_EVEX_0F3810, PREFIX_EVEX_0F381C, PREFIX_EVEX_0F381D, PREFIX_EVEX_0F3820, PREFIX_EVEX_0F3826, PREFIX_EVEX_0F382B, PREFIX_EVEX_0F3830, PREFIX_EVEX_0F3838, PREFIX_EVEX_0F383C, PREFIX_EVEX_0F383E, PREFIX_EVEX_0F3866, PREFIX_EVEX_0F3875, PREFIX_EVEX_0F3878, PREFIX_EVEX_0F3879, PREFIX_EVEX_0F387A, PREFIX_EVEX_0F387B, PREFIX_EVEX_0F387D, PREFIX_EVEX_0F388D, PREFIX_EVEX_0F3A0F, PREFIX_EVEX_0F3A14, PREFIX_EVEX_0F3A15, PREFIX_EVEX_0F3A20, PREFIX_EVEX_0F3A3E, PREFIX_EVEX_0F3A3F, PREFIX_EVEX_0F3A42. (VEX_LEN enum): Add VEX_LEN_0F41_P_2, VEX_LEN_0F42_P_2, VEX_LEN_0F44_P_2, VEX_LEN_0F45_P_2, VEX_LEN_0F46_P_2, VEX_LEN_0F47_P_2, VEX_LEN_0F4A_P_0, VEX_LEN_0F4A_P_2, VEX_LEN_0F4B_P_0, VEX_LEN_0F90_P_2, VEX_LEN_0F91_P_2, VEX_LEN_0F92_P_3, VEX_LEN_0F93_P_3, VEX_LEN_0F98_P_2, VEX_LEN_0F99_P_0, VEX_LEN_0F99_P_2, VEX_LEN_0F3A31_P_2, VEX_LEN_0F3A33_P_2, VEX_W_0F41_P_2_LEN_1, VEX_W_0F42_P_2_LEN_1, VEX_W_0F44_P_2_LEN_0, VEX_W_0F45_P_2_LEN_1, VEX_W_0F46_P_2_LEN_1, VEX_W_0F47_P_2_LEN_1, VEX_W_0F4A_P_0_LEN_1, VEX_W_0F4A_P_2_LEN_1, VEX_W_0F4B_P_0_LEN_1, VEX_W_0F90_P_2_LEN_0, VEX_W_0F91_P_2_LEN_0, VEX_W_0F92_P_3_LEN_0, VEX_W_0F93_P_3_LEN_0, VEX_W_0F98_P_2_LEN_0, VEX_W_0F99_P_0_LEN_0, VEX_W_0F99_P_2_LEN_0, VEX_W_0F3A31_P_2_LEN_0, VEX_W_0F3A33_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F6B_P_2, EVEX_W_0F6F_P_3, EVEX_W_0F7F_P_3, EVEX_W_0F3810_P_1, EVEX_W_0F3810_P_2, EVEX_W_0F3811_P_2, EVEX_W_0F3812_P_2, EVEX_W_0F3820_P_1, EVEX_W_0F3826_P_1, EVEX_W_0F3826_P_2, EVEX_W_0F3828_P_1, EVEX_W_0F3829_P_1, EVEX_W_0F382B_P_2, EVEX_W_0F3830_P_1, EVEX_W_0F3866_P_2, EVEX_W_0F3875_P_2, EVEX_W_0F3878_P_2, EVEX_W_0F3879_P_2, EVEX_W_0F387A_P_2, EVEX_W_0F387B_P_2, EVEX_W_0F387D_P_2, EVEX_W_0F388D_P_2, EVEX_W_0F3A3E_P_2, EVEX_W_0F3A3F_P_2, EVEX_W_0F3A42_P_2. (prefix_table): Add entries for new instructions. (vex_table) : Ditto. (vex_len_table): Ditto. (vex_w_table): Ditto. (intel_operand_size): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_E_register): Add dqw_swap_mode, dw_mode, db_mode, mask_bd_mode handling. (OP_E_memory): Add dqw_mode, dw_mode, dqw_swap_mode, dqb_mode, db_mode handling. (OP_G): Add db_mode, dw_mode, dqw_swap_mode, mask_bd_mode handling. (OP_EX): Add dqw_swap_mode handling. (OP_VEX): Add mask_bd_mode handling. (OP_Mask): Add mask_bd_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512BW_FLAGS. (cpu_flags): Add CpuAVX512BW. * i386-init.h: Regenerated. * i386-opc.h (CpuAVX512BW): New. (i386_cpu_flags): Add cpuavx512bw. * i386-opc.tbl: Add AVX512BW instructions. * i386-tbl.h: Regenerate.
2014-07-11 14:32:29 +02:00
// AVX512BW instructions end.
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
// AVX512DQ instructions.
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
kaddb, 3, 0x664A, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandb, 3, 0x6641, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kandnb, 3, 0x6642, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
kmovb, 2, 0x6690, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask|Byte|Unspecified|BaseIndex, RegMask }
kmovb, 2, 0x6691, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW0|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, Byte|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
kmovb, 2, 0x6692, None, 1, CpuAVX512DQ, D|Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegMask }
knotb, 2, 0x6644, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
korb, 3, 0x6645, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kortestb, 2, 0x6698, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
ktestb, 2, 0x6699, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
kxnorb, 3, 0x6646, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
kxorb, 3, 0x6647, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
kaddw, 3, 0x4A, None, 1, CpuAVX512DQ, Modrm|Vex=2|VexOpcode=0|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask, RegMask }
ktestw, 2, 0x99, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegMask }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
kshiftlb, 3, 0x6632, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
kshiftrb, 3, 0x6630, None, 1, CpuAVX512DQ, Modrm|Vex=1|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegMask, RegMask }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vandnpd, 3, 0x6655, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vandpd, 3, 0x6654, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vorpd, 3, 0x6656, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vxorpd, 3, 0x6657, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vandnps, 3, 0x55, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vandps, 3, 0x54, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vorps, 3, 0x56, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vxorps, 3, 0x57, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Optimize, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vbroadcastf32x2, 2, 0x6619, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vbroadcastf32x8, 2, 0x661B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { YMMword|Unspecified|BaseIndex, RegZMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vbroadcasti32x2, 2, 0x6659, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=1|VexW0|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vbroadcasti32x8, 2, 0x665B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { YMMword|Unspecified|BaseIndex, RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vbroadcastf64x2, 2, 0x661A, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { XMMword|Unspecified|BaseIndex, RegYMM|RegZMM }
vbroadcasti64x2, 2, 0x665A, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { XMMword|Unspecified|BaseIndex, RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvtpd2qq, 2, 0x667B, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtpd2qq, 3, 0x667B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvtpd2uqq, 2, 0x6679, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtpd2uqq, 3, 0x6679, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtps2qq, 2, 0x667B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvtps2qq, 3, 0x667B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegYMM, RegZMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtps2qq, 2, 0x667B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvtps2qq, 2, 0x667B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtps2uqq, 2, 0x6679, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvtps2uqq, 3, 0x6679, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegYMM, RegZMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvtps2uqq, 2, 0x6679, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvtps2uqq, 2, 0x6679, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvtqq2pd, 2, 0xF3E6, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtqq2pd, 3, 0xF3E6, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvtuqq2pd, 2, 0xF37A, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvtuqq2pd, 3, 0xF37A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtqq2ps, 2, 0x5B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegYMM }
vcvtqq2ps, 3, 0x5B, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegYMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtqq2ps, 2, 0x5B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtqq2ps, 2, 0x5B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtqq2psx, 2, 0x5B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtqq2psy, 2, 0x5B, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvttpd2qq, 2, 0x667A, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvttpd2qq, 3, 0x667A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vcvttpd2uqq, 2, 0x6678, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvttpd2uqq, 3, 0x6678, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegZMM, RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvttps2qq, 2, 0x667A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvttps2qq, 3, 0x667A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegYMM, RegZMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvttps2qq, 2, 0x667A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvttps2qq, 2, 0x667A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vcvttps2uqq, 2, 0x6678, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegYMM|Dword|Unspecified|BaseIndex, RegZMM }
vcvttps2uqq, 3, 0x6678, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, RegYMM, RegZMM }
x86: reduce amount of various VCVT* templates Presumably as a result of various changes over the last several months, and - for some of them - with a generalization of logic in match_mem_size() plus mirroring of this generalization into the broadcast handling logic of check_VecOperands(), various register-only templates can be foled into their respective memory forms. This in particular then also allows dropping a few more instances of IgnoreSize.
2020-03-06 08:56:47 +01:00
vcvttps2uqq, 2, 0x6678, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex128|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Qword|Unspecified|BaseIndex, RegXMM }
vcvttps2uqq, 2, 0x6678, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex256|Masking=3|VexOpcode=0|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|Dword|Unspecified|BaseIndex, RegYMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtuqq2ps, 2, 0xF27A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Qword|Unspecified|BaseIndex, RegYMM }
vcvtuqq2ps, 3, 0xF27A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=0|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|StaticRounding|SAE, { Imm8, RegZMM, RegYMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtuqq2ps, 2, 0xF27A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtuqq2ps, 2, 0xF27A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Qword|BaseIndex, RegXMM }
x86: fold narrowing VCVT* templates When multiple (here: two) forms of an insn take different width inputs but produce identical size outputs (here: RegXMM), the templates can be combined. Also drop IgnoreSize (and the now redundant size specifiers) wherever applicable.
2018-07-19 08:36:19 +02:00
vcvtuqq2psx, 2, 0xF27A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=2|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Qword|Unspecified|BaseIndex, RegXMM }
vcvtuqq2psy, 2, 0xF27A, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=3|Masking=3|VexOpcode=0|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Qword|Unspecified|BaseIndex, RegXMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vextractf32x8, 3, 0x661B, None, 1, CpuAVX512DQ, Modrm|EVex=1|MaskingMorZ|VexOpcode=2|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM, RegYMM|Unspecified|BaseIndex }
vextracti32x8, 3, 0x663B, None, 1, CpuAVX512DQ, Modrm|EVex=1|MaskingMorZ|VexOpcode=2|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM, RegYMM|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vinsertf32x8, 4, 0x661A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Unspecified|BaseIndex, RegZMM, RegZMM }
vinserti32x8, 4, 0x663A, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Unspecified|BaseIndex, RegZMM, RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfpclassss, 3, 0x6667, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=2|VexOpcode=2|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegMask }
x86: adjust {,E}VEX.W handling outside of 64-bit mode Many VEX-/EVEX-encoded instructions accessing GPRs become WIG outside of 64-bit mode. The respective templates should specify neither VexWIG nor VexW0, but instead the setting of the bit should be determined from - REX.W in 64-bit mode, - the setting established through -mvexwig= / -mevexwig= otherwise. This implies that the evex-wig2 testcase needs to go away, as being wrong altogether. A few test additions desirable here will only happen in later patches, as the disassembler needs adjustments first. Once again SSE2AVX templates are left alone, for it being unclear what the behavior there should be.
2018-11-06 11:42:54 +01:00
vpextrd, 3, 0x6616, None, 1, CpuAVX512DQ, Modrm|EVex128|VexOpcode=2|Disp8MemShift=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg32|Dword|Unspecified|BaseIndex }
vpinsrd, 4, 0x6622, None, 1, CpuAVX512DQ, Modrm|EVex128|VexOpcode=2|VexVVVV=1|Disp8MemShift=2|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg32|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vfpclasssd, 3, 0x6667, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=2|VexOpcode=2|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegMask }
x86: VPEXTRQ/VPINSRQ are unavailable outside of 64-bit mode The AVX512DQ patterns lacking a Cpu64 attribute made the memory operand forms accepted even outside of 64-bit mode, and this even without any {evex} pseudo-prefix (otherwise one could argue that this is an attempt to follow one possible, albeit somewhat odd, interpretation of the SDM wording to this effect). For consistency between the various involved templates drop the * (now) unnecessary IgnoreSize attributes * unnecessary (due to VexW1) Size64 attributes from VEX encoded forms * redundant (with Reg64) Qword operand attributes uniformly.
2020-01-16 10:05:35 +01:00
vpextrq, 3, 0x6616, None, 1, CpuAVX512DQ|Cpu64, Modrm|EVex128|VexOpcode=2|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM, Reg64|Unspecified|BaseIndex }
vpinsrq, 4, 0x6622, None, 1, CpuAVX512DQ|Cpu64, Modrm|EVex128|VexOpcode=2|VexVVVV=1|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, Reg64|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop pointless VecESize The attribute is meaningful only in templates allowing embedded broadcast. Drop them everywhere else.
2018-03-22 08:46:25 +01:00
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vextractf64x2, 3, 0x6619, None, 1, CpuAVX512DQ, Modrm|MaskingMorZ|VexOpcode=2|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM, RegXMM|Unspecified|BaseIndex }
vextracti64x2, 3, 0x6639, None, 1, CpuAVX512DQ, Modrm|MaskingMorZ|VexOpcode=2|VexW=2|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|RegZMM, RegXMM|Unspecified|BaseIndex }
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vinsertf64x2, 4, 0x6618, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
vinserti64x2, 4, 0x6638, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8MemShift=4|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Unspecified|BaseIndex, RegYMM|RegZMM, RegYMM|RegZMM }
x86: derive DispN from BaseIndex BaseIndex implies - with the exception of string instructions the optional presence of a displacement. This is almost completely uniform for all instructions (the sole exception being MPX ones, which don't allow 16-bit addressing and hence Disp16), so there's no point in explicitly stating this in the main opcode table. Drop those explict specifications in favor of adding logic to i386-gen, shrinking the table size quite a bit and hence making it more readable. The opcodes/i386-tbl.h changes are due to a few cases where pointless Disp* still hadn't been removed from their insns.
2017-11-30 11:48:13 +01:00
x86: correct VFPCLASSP{S,D} operand size handling With AVX512VL disabled (e.g. when writing code for the Knights family of processors) these insns aren't ambiguous when used with a memory source, and hence should be accepted without suffix or operand size specifier. When AVX512VL is enabled, to be consistent with this as well as other ambiguous operand size handling it would seem better to just warn about the ambiguity in AT&T mode, and still default to 512-bit operands (on the assumption that the code may have been written without AVX512VL in mind yet), but it was requested to leave AT&T syntax mode alone here.
2020-02-12 16:20:56 +01:00
vfpclasspd, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|Masking=2|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Imm8, RegXMM|RegYMM|RegZMM|Qword|BaseIndex, RegMask }
vfpclasspd, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|Masking=2|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegMask }
x86: fold VFPCLASSP{D,S} templates These are special because they may not have a register operand to derive the vector length from, which requires to also deal with the braodcast case when determining vector length in build_evex_prefix(). Also drop IgnoreSize (and the now redundant size specifiers) from their suffixed counterparts.
2018-07-19 08:35:38 +02:00
vfpclasspdz, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=2|VexOpcode=2|VexW=2|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM|Qword|Unspecified|BaseIndex, RegMask }
vfpclasspdx, 3, 0x6666, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=2|Masking=2|VexOpcode=2|VexW=2|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegMask }
vfpclasspdy, 3, 0x6666, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=3|Masking=2|VexOpcode=2|VexW=2|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Qword|Unspecified|BaseIndex, RegMask }
x86: correct VFPCLASSP{S,D} operand size handling With AVX512VL disabled (e.g. when writing code for the Knights family of processors) these insns aren't ambiguous when used with a memory source, and hence should be accepted without suffix or operand size specifier. When AVX512VL is enabled, to be consistent with this as well as other ambiguous operand size handling it would seem better to just warn about the ambiguity in AT&T mode, and still default to 512-bit operands (on the assumption that the code may have been written without AVX512VL in mind yet), but it was requested to leave AT&T syntax mode alone here.
2020-02-12 16:20:56 +01:00
vfpclassps, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|Masking=2|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { Imm8, RegXMM|RegYMM|RegZMM|Dword|BaseIndex, RegMask }
vfpclassps, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|Masking=2|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegMask }
x86: fold VFPCLASSP{D,S} templates These are special because they may not have a register operand to derive the vector length from, which requires to also deal with the braodcast case when determining vector length in build_evex_prefix(). Also drop IgnoreSize (and the now redundant size specifiers) from their suffixed counterparts.
2018-07-19 08:35:38 +02:00
vfpclasspsz, 3, 0x6666, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=2|VexOpcode=2|VexW=1|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegZMM|Dword|Unspecified|BaseIndex, RegMask }
vfpclasspsx, 3, 0x6666, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=2|Masking=2|VexOpcode=2|VexW=1|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegMask }
vfpclasspsy, 3, 0x6666, None, 1, CpuAVX512DQ|CpuAVX512VL, Modrm|EVex=3|Masking=2|VexOpcode=2|VexW=1|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegYMM|Dword|Unspecified|BaseIndex, RegMask }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpmovd2m, 2, 0xF339, None, 1, CpuAVX512DQ, Modrm|EVex=5|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegMask }
vpmovq2m, 2, 0xF339, None, 1, CpuAVX512DQ, Modrm|EVex=5|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegMask }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpmovm2d, 2, 0xF338, None, 1, CpuAVX512DQ, Modrm|EVex=5|VexOpcode=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
vpmovm2q, 2, 0xF338, None, 1, CpuAVX512DQ, Modrm|EVex=5|VexOpcode=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegMask, RegXMM|RegYMM|RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vpmullq, 3, 0x6640, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vrangepd, 4, 0x6650, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vrangepd, 5, 0x6650, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vreducepd, 3, 0x6656, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vreducepd, 4, 0x6656, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vrangeps, 4, 0x6650, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vrangeps, 5, 0x6650, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM, RegZMM }
x86: fold various AVX512DQ templates
2018-07-19 08:34:01 +02:00
vreduceps, 3, 0x6656, None, 1, CpuAVX512DQ, Modrm|Masking=3|VexOpcode=2|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vreduceps, 4, 0x6656, None, 1, CpuAVX512DQ, Modrm|EVex=1|Masking=3|VexOpcode=2|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegZMM, RegZMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrangesd, 4, 0x6651, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vrangesd, 5, 0x6651, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vreducesd, 4, 0x6657, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW1|Disp8MemShift=3|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Qword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vreducesd, 5, 0x6657, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vrangess, 4, 0x6651, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vrangess, 5, 0x6651, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
x86: drop/replace IgnoreSize Even after commit dc2be329b950 ("i386: Only check suffix in instruction mnemonic"), by which many of its uses have become unnecessary (some were unnecessary even before), IgnoreSize is still used for various slightly different purposes: - to suppress emission of an operand size prefix, - in Intel syntax mode to zap "derived" suffixes in certain cases and to skip certain checks of remaining "derived" suffixes, - to suppress ambiguous operand size / missing suffix diagnostics, - for prefixes to suppress the "stand-alone ... prefix" warning. Drop entirely unnecessary ones and where possible also replace instances by the more focused (because of having just a single purpose) NoRex64. To further restrict when IgnoreSize is needed, also generalize the logic when to skip a template because of a present or derived L or Q suffix, by skipping immediate operands. Additionally consider mask registers and VecSIB there. Note that for the time being the attribute needs to be kept in place on MMX/SSE/etc insns (but not on VEX/EVEX encoded ones unless an operand template of them allows for only non-SIMD-register actuals) allowing for Dword operands - the logic when to emit a data size prefix would need further adjustment first. Note also that the memory forms of {,v}pinsrw get their permission for an L or Q suffix dropped. I can only assume that it being this way was a cut-and-paste mistake from the register forms, as the latter specifically have NoRex64 set, and the {,v}pextrw counterparts don't allow these suffixes either. Convert VexW= again to their respective VexW* on lines touched anyway.
2020-03-06 08:55:52 +01:00
vreducess, 4, 0x6657, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV|VexW0|Disp8MemShift=2|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|Dword|Unspecified|BaseIndex, RegXMM, RegXMM }
x86: drop bogus IgnoreSize from AVX512DQ insns
2018-09-13 11:24:53 +02:00
vreducess, 5, 0x6657, None, 1, CpuAVX512DQ, Modrm|EVex=4|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|SAE, { Imm8, Imm8, RegXMM, RegXMM, RegXMM }
Add AVX512DQ instructions and their AVX512VL variants. gas/ * config/tc-i386.c (cpu_arch): Add .avx512dq, CPU_AVX512DQ_FLAGS. * doc/c-i386.texi: Document avx512dq/.avx512dq. gas/testsuite/ * gas/i386/avx512dq-intel.d: New. * gas/i386/avx512dq.d: New. * gas/i386/avx512dq.s: New. * gas/i386/avx512dq_vl-intel.d: New. * gas/i386/avx512dq_vl.d: New. * gas/i386/avx512dq_vl.s: New. * gas/i386/i386.exp: Run new AVX-512 tests. * gas/i386/x86-64-avx512dq-intel.d: New. * gas/i386/x86-64-avx512dq.d: New. * gas/i386/x86-64-avx512dq.s: New. * gas/i386/x86-64-avx512dq_vl-intel.d: New. * gas/i386/x86-64-avx512dq_vl.d: New. * gas/i386/x86-64-avx512dq_vl.s: New. opcodes/ * i386-dis-evex.h: Updated. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F54, PREFIX_EVEX_0F55, PREFIX_EVEX_0F56, PREFIX_EVEX_0F57, PREFIX_EVEX_0F3A16, PREFIX_EVEX_0F3A22, PREFIX_EVEX_0F3A50, PREFIX_EVEX_0F3A51, PREFIX_EVEX_0F3A56, PREFIX_EVEX_0F3A57, PREFIX_EVEX_0F3A66, PREFIX_EVEX_0F3A67. (VEX_LEN enum): Add VEX_LEN_0F92_P_2, VEX_LEN_0F93_P_2, VEX_W_0F92_P_2_LEN_0, VEX_W_0F93_P_2_LEN_0. (VEX_W enum): Add EVEX_W_0F54_P_0, EVEX_W_0F54_P_2, EVEX_W_0F55_P_0, EVEX_W_0F55_P_2, EVEX_W_0F56_P_0, EVEX_W_0F56_P_2, EVEX_W_0F57_P_0, EVEX_W_0F57_P_2, EVEX_W_0F78_P_2, EVEX_W_0F79_P_2, EVEX_W_0F7A_P_2, EVEX_W_0F7B_P_2, EVEX_W_0F3838_P_1, EVEX_W_0F3839_P_1, EVEX_W_0F3A16_P_2, EVEX_W_0F3A22_P_2, EVEX_W_0F3A50_P_2, EVEX_W_0F3A51_P_2, EVEX_W_0F3A56_P_2, EVEX_W_0F3A57_P_2, EVEX_W_0F3A66_P_2, EVEX_W_0F3A67_P_2. (prefix_table): Add entries for new instructions. (vex_len_table): Ditto. (vex_w_table): Ditto. (OP_E_memory): Update xmmq_mode handling. * i386-gen.c (cpu_flag_init): Add CPU_AVX512DQ_FLAGS. (cpu_flags): Add CpuAVX512DQ. * i386-init.h: Regenerared. * i386-opc.h (CpuAVX512DQ): New. (i386_cpu_flags): Add cpuavx512dq. * i386-opc.tbl: Add AVX512DQ instructions. * i386-tbl.h: Regenerate.
2014-07-15 11:33:39 +02:00
// AVX512DQ instructions end.
Add clwb instruction gas/ * config/tc-i386.c (cpu_arch): Add .clwb. * doc/c-i386.texi: Document it. opcodes/ * i386-dis.c (PREFIX enum): Add PREFIX_0FAE_REG_6. (prefix_table): Add clwb. * i386-gen.c (cpu_flag_init): Add CPU_CLWB_FLAGS. (cpu_flags): Add CpuCLWB. * i386-opc.h (enum): Add CpuCLWB. (i386_cpu_flags): Add cpuclwb. * i386-opc.tbl: Add clwb. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/clwb-intel.d: New file. * gas/i386/clwb.d: Likewise. * gas/i386/clwb.s: Likewise. * gas/i386/x86-64-clwb-intel.d: Likewise. * gas/i386/x86-64-clwb.d: Likewise. * gas/i386/x86-64-clwb.s: Likewise.
2014-11-17 12:19:41 +01:00
// CLWB instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
clwb, 1, 0x660fae, 0x6, 2, CpuCLWB, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
Add clwb instruction gas/ * config/tc-i386.c (cpu_arch): Add .clwb. * doc/c-i386.texi: Document it. opcodes/ * i386-dis.c (PREFIX enum): Add PREFIX_0FAE_REG_6. (prefix_table): Add clwb. * i386-gen.c (cpu_flag_init): Add CPU_CLWB_FLAGS. (cpu_flags): Add CpuCLWB. * i386-opc.h (enum): Add CpuCLWB. (i386_cpu_flags): Add cpuclwb. * i386-opc.tbl: Add clwb. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/clwb-intel.d: New file. * gas/i386/clwb.d: Likewise. * gas/i386/clwb.s: Likewise. * gas/i386/x86-64-clwb-intel.d: Likewise. * gas/i386/x86-64-clwb.d: Likewise. * gas/i386/x86-64-clwb.s: Likewise.
2014-11-17 12:19:41 +01:00
// CLWB instructions end.
Add pcommit instruction gas/ * config/tc-i386.c (cpu_arch): Add .pcommit. * doc/c-i386.texi: Document it. /opcodes * i386-dis.c (PREFIX enum): Add PREFIX_RM_0_0FAE_REG_7. (prefix_table): Add pcommit. * i386-gen.c (cpu_flag_init): Add CPU_PCOMMIT_FLAGS. (cpu_flags): Add CpuPCOMMIT. * i386-opc.h (enum): Add CpuPCOMMIT. (i386_cpu_flags): Add cpupcommit. * i386-opc.tbl: Add pcommit. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. /gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/pcommit-intel.d: New file. * gas/i386/pcommit.d: Likewise. * gas/i386/pcommit.s: Likewise. * gas/i386/x86-64-pcommit-intel.d: Likewise. * gas/i386/x86-64-pcommit.d: Likewise. * gas/i386/x86-64-pcommit.s: Likewise.
2014-11-17 12:49:47 +01:00
Add AVX512IFMA instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512ifma. * doc/c-i386.texi: Document it. opcodes/ * i386-dis-evex.c (evex_table): Add vpmadd52luq, vpmadd52huq. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F38B4, PREFIX_EVEX_0F38B5. * i386-gen.c (cpu_flag_init): Add CPU_AVX512IFMA_FLAGS. (cpu_flags): Add CpuAVX512IFMA. * i386-opc.h (enum): Add CpuAVX512IFMA. (i386_cpu_flags): Add cpuavx512ifma. * i386-opc.tbl: Add vpmadd52huq, vpmadd52luq. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. /gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/avx512ifma-intel.d: New file. * gas/i386/avx512ifma.d: Likewise. * gas/i386/avx512ifma.s: Likewise. * gas/i386/avx512ifma_vl-intel.d: Likewise. * gas/i386/avx512ifma_vl.d: Likewise. * gas/i386/avx512ifma_vl.s: Likewise. * gas/i386/x86-64-avx512ifma-intel.d: Likewise. * gas/i386/x86-64-avx512ifma.d: Likewise. * gas/i386/x86-64-avx512ifma.s: Likewise. * gas/i386/x86-64-avx512ifma_vl-intel.d: Likewise. * gas/i386/x86-64-avx512ifma_vl.d: Likewise. * gas/i386/x86-64-avx512ifma_vl.s: Likewise.
2014-11-17 12:57:33 +01:00
// AVX512IFMA instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpmadd52huq, 3, 0x66B5, None, 1, CpuAVX512IFMA, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpmadd52luq, 3, 0x66B4, None, 1, CpuAVX512IFMA, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add AVX512IFMA instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512ifma. * doc/c-i386.texi: Document it. opcodes/ * i386-dis-evex.c (evex_table): Add vpmadd52luq, vpmadd52huq. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F38B4, PREFIX_EVEX_0F38B5. * i386-gen.c (cpu_flag_init): Add CPU_AVX512IFMA_FLAGS. (cpu_flags): Add CpuAVX512IFMA. * i386-opc.h (enum): Add CpuAVX512IFMA. (i386_cpu_flags): Add cpuavx512ifma. * i386-opc.tbl: Add vpmadd52huq, vpmadd52luq. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. /gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/avx512ifma-intel.d: New file. * gas/i386/avx512ifma.d: Likewise. * gas/i386/avx512ifma.s: Likewise. * gas/i386/avx512ifma_vl-intel.d: Likewise. * gas/i386/avx512ifma_vl.d: Likewise. * gas/i386/avx512ifma_vl.s: Likewise. * gas/i386/x86-64-avx512ifma-intel.d: Likewise. * gas/i386/x86-64-avx512ifma.d: Likewise. * gas/i386/x86-64-avx512ifma.s: Likewise. * gas/i386/x86-64-avx512ifma_vl-intel.d: Likewise. * gas/i386/x86-64-avx512ifma_vl.d: Likewise. * gas/i386/x86-64-avx512ifma_vl.s: Likewise.
2014-11-17 12:57:33 +01:00
// AVX512IFMA instructions end
Add AVX512VBMI instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512vbmi. * doc/c-i386.texi: Document it. opcodes/ * i386-dis-evex.c (evex_table): Add vpermi2b, vpermt2b, vpermb, vpmultishiftqb. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F3883, EVEX_W_0F3883_P_2. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VBMI_FLAGS. (cpu_flags): Add CpuAVX512VBMI. * i386-opc.h (enum): Add CpuAVX512VBMI. (i386_cpu_flags): Add cpuavx512vbmi. * i386-opc.tbl: Add vpmadd52luq, vpmultishiftqb, vpermb, vpermi2b, vpermt2b. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. /gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/avx512vbmi-intel.d: New file. * gas/i386/avx512vbmi.d: Likewise. * gas/i386/avx512vbmi.s: Likewise. * gas/i386/avx512vbmi_vl-intel.d: Likewise. * gas/i386/avx512vbmi_vl.d: Likewise. * gas/i386/avx512vbmi_vl.s: Likewise. * gas/i386/x86-64-avx512vbmi-intel.d: Likewise. * gas/i386/x86-64-avx512vbmi.d: Likewise. * gas/i386/x86-64-avx512vbmi.s: Likewise. * gas/i386/x86-64-avx512vbmi_vl-intel.d: Likewise. * gas/i386/x86-64-avx512vbmi_vl.d: Likewise. * gas/i386/x86-64-avx512vbmi_vl.s: Likewise.
2014-11-17 13:41:32 +01:00
// AVX512VBMI instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpmultishiftqb, 3, 0x6683, None, 1, CpuAVX512VBMI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermb, 3, 0x668D, None, 1, CpuAVX512VBMI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermi2b, 3, 0x6675, None, 1, CpuAVX512VBMI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpermt2b, 3, 0x667D, None, 1, CpuAVX512VBMI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Add AVX512VBMI instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512vbmi. * doc/c-i386.texi: Document it. opcodes/ * i386-dis-evex.c (evex_table): Add vpermi2b, vpermt2b, vpermb, vpmultishiftqb. * i386-dis.c (PREFIX enum): Add PREFIX_EVEX_0F3883, EVEX_W_0F3883_P_2. * i386-gen.c (cpu_flag_init): Add CPU_AVX512VBMI_FLAGS. (cpu_flags): Add CpuAVX512VBMI. * i386-opc.h (enum): Add CpuAVX512VBMI. (i386_cpu_flags): Add cpuavx512vbmi. * i386-opc.tbl: Add vpmadd52luq, vpmultishiftqb, vpermb, vpermi2b, vpermt2b. * i386-init.h: Regenerated. * i386-tbl.h: Likewise. /gas/testsuite/ * gas/i386/i386.exp: Run new tests. * gas/i386/avx512vbmi-intel.d: New file. * gas/i386/avx512vbmi.d: Likewise. * gas/i386/avx512vbmi.s: Likewise. * gas/i386/avx512vbmi_vl-intel.d: Likewise. * gas/i386/avx512vbmi_vl.d: Likewise. * gas/i386/avx512vbmi_vl.s: Likewise. * gas/i386/x86-64-avx512vbmi-intel.d: Likewise. * gas/i386/x86-64-avx512vbmi.d: Likewise. * gas/i386/x86-64-avx512vbmi.s: Likewise. * gas/i386/x86-64-avx512vbmi_vl-intel.d: Likewise. * gas/i386/x86-64-avx512vbmi_vl.d: Likewise. * gas/i386/x86-64-avx512vbmi_vl.s: Likewise.
2014-11-17 13:41:32 +01:00
// AVX512VBMI instructions end
Enable Intel AVX512_4FMAPS instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512_4fmaps. (cpu_noarch): Add noavx512_4fmaps. (process_operands): Handle implicit quad group. * doc/c-i386.texi: Document avx512_4fmaps, noavx512_4fmaps. * testsuite/gas/i386/i386.exp: Add AVX512_4FMAPS tests. * testsuite/gas/i386/avx512_4fmaps_vl-intel.d: New test. * testsuite/gas/i386/avx512_4fmaps_vl.d: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl.s: Ditto. * testsuite/gas/i386/avx512_4fmaps-intel.d: Ditto. * testsuite/gas/i386/avx512_4fmaps.d: Ditto. * testsuite/gas/i386/avx512_4fmaps.s: Ditto. * testsuite/gas/i386/avx512_4fmaps-warn.l: Ditto. * testsuite/gas/i386/avx512_4fmaps-warn.s: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl-warn.l: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl-warn.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-warn.l: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-warn.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-warn.l: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-warn.s: Ditto. opcodes/ * i386-dis.c. (enum): Add PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_4FMAPS_FLAGS, CPU_ANY_AVX512_4FMAPS_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_4FMAPS. (opcode_modifiers): Add ImplicitQuadGroup modifier. * i386-opc.h (AVX512_4FMAP): New. (i386_cpu_flags): Add cpuavx512_4fmaps. (ImplicitQuadGroup): New. (i386_opcode_modifier): Add implicitquadgroup. * i386-opc.tbl: Add Intel AVX512_4FMAPS instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2016-11-02 20:24:39 +01:00
// AVX512_4FMAPS instructions
x86: drop bogus IgnoreSize from AVX512_4* insns
2018-09-13 11:25:30 +02:00
v4fmaddps, 3, 0xf29a, None, 1, CpuAVX512_4FMAPS, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegZMM, RegZMM }
v4fnmaddps, 3, 0xf2aa, None, 1, CpuAVX512_4FMAPS, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegZMM, RegZMM }
v4fmaddss, 3, 0xf29b, None, 1, CpuAVX512_4FMAPS, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegXMM, RegXMM }
v4fnmaddss, 3, 0xf2ab, None, 1, CpuAVX512_4FMAPS, Modrm|EVex=4|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegXMM, RegXMM }
Enable Intel AVX512_4FMAPS instructions gas/ * config/tc-i386.c (cpu_arch): Add .avx512_4fmaps. (cpu_noarch): Add noavx512_4fmaps. (process_operands): Handle implicit quad group. * doc/c-i386.texi: Document avx512_4fmaps, noavx512_4fmaps. * testsuite/gas/i386/i386.exp: Add AVX512_4FMAPS tests. * testsuite/gas/i386/avx512_4fmaps_vl-intel.d: New test. * testsuite/gas/i386/avx512_4fmaps_vl.d: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl.s: Ditto. * testsuite/gas/i386/avx512_4fmaps-intel.d: Ditto. * testsuite/gas/i386/avx512_4fmaps.d: Ditto. * testsuite/gas/i386/avx512_4fmaps.s: Ditto. * testsuite/gas/i386/avx512_4fmaps-warn.l: Ditto. * testsuite/gas/i386/avx512_4fmaps-warn.s: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl-warn.l: Ditto. * testsuite/gas/i386/avx512_4fmaps_vl-warn.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-warn.l: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps-warn.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-warn.l: Ditto. * testsuite/gas/i386/x86-64-avx512_4fmaps_vl-warn.s: Ditto. opcodes/ * i386-dis.c. (enum): Add PREFIX_EVEX_0F389A, PREFIX_EVEX_0F389B, PREFIX_EVEX_0F38AA, PREFIX_EVEX_0F38AB. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_4FMAPS_FLAGS, CPU_ANY_AVX512_4FMAPS_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_4FMAPS. (opcode_modifiers): Add ImplicitQuadGroup modifier. * i386-opc.h (AVX512_4FMAP): New. (i386_cpu_flags): Add cpuavx512_4fmaps. (ImplicitQuadGroup): New. (i386_opcode_modifier): Add implicitquadgroup. * i386-opc.tbl: Add Intel AVX512_4FMAPS instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2016-11-02 20:24:39 +01:00
// AVX512_4FMAPS instructions end
Add znver1 processor
2015-03-17 17:19:15 +01:00
Enable Intel AVX512_4VNNIW instructions gas/ * config/tc-i386.c: (cpu_arch) Add .avx512_4vnniw. (cpu_noarch): Add noavx512_4vnniw. * doc/c-i386.texi: Document avx512_4vnniw, noavx512_4vnniw. * testsuite/gas/i386/i386.exp: Run AVX512_4VNNIW tests. * testsuite/gas/i386/avx512_4vnniwd_vl-intel.d: New test. * testsuite/gas/i386/avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3852, PREFIX_EVEX_0F3853. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_4VNNIW_FLAGS, CPU_ANY_AVX512_4VNNIW_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_4VNNIW. * i386-opc.h (enum): (AVX512_4VNNIW): New. (i386_cpu_flags): Add cpuavx512_4vnniw. * i386-opc.tbl: Add Intel AVX512_4VNNIW instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2016-11-02 20:31:25 +01:00
// AVX512_4VNNIW instructions
x86: drop bogus IgnoreSize from AVX512_4* insns
2018-09-13 11:25:30 +02:00
vp4dpwssd, 3, 0xf252, None, 1, CpuAVX512_4VNNIW, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegZMM, RegZMM }
vp4dpwssds, 3, 0xf253, None, 1, CpuAVX512_4VNNIW, Modrm|EVex=1|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImplicitQuadGroup, { XMMword|Unspecified|BaseIndex, RegZMM, RegZMM }
Enable Intel AVX512_4VNNIW instructions gas/ * config/tc-i386.c: (cpu_arch) Add .avx512_4vnniw. (cpu_noarch): Add noavx512_4vnniw. * doc/c-i386.texi: Document avx512_4vnniw, noavx512_4vnniw. * testsuite/gas/i386/i386.exp: Run AVX512_4VNNIW tests. * testsuite/gas/i386/avx512_4vnniwd_vl-intel.d: New test. * testsuite/gas/i386/avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3852, PREFIX_EVEX_0F3853. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_4VNNIW_FLAGS, CPU_ANY_AVX512_4VNNIW_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_4VNNIW. * i386-opc.h (enum): (AVX512_4VNNIW): New. (i386_cpu_flags): Add cpuavx512_4vnniw. * i386-opc.tbl: Add Intel AVX512_4VNNIW instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2016-11-02 20:31:25 +01:00
Enable Intel AVX512_VPOPCNTDQ instructions gas/ 2017-01-12 Igor Tsimbalist <igor.v.tsimbalist@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_vpopcntdq. (cpu_noarch): Add noavx512_vpopcntdq. * doc/c-i386.texi: Document avx512_vpopcntdq, noavx512_vpopcntdq. * testsuite/gas/i386/i386.exp: Run AVX512_VPOPCNTDQ tests. * testsuite/gas/i386/avx512_vpopcntdqd-intel.d: New file. * testsuite/gas/i386/avx512_vpopcntdqd.d: Ditto. * testsuite/gas/i386/avx512_vpopcntdqd.s: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd.d: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd.s: Ditto. opcodes/ 2017-01-12 Igor Tsimbalist <igor.v.tsimbalist@intel.com> * i386-dis.c (enum): Add PREFIX_EVEX_0F3855, EVEX_W_0F3855_P_2. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VPOPCNTDQ_FLAGS, CPU_ANY_AVX512_VPOPCNTDQ_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VPOPCNTDQ. * i386-opc.h (enum): (AVX512_VPOPCNTDQ): New. (i386_cpu_flags): Add cpuavx512_vpopcntdq. * i386-opc.tbl: Add Intel AVX512_VPOPCNTDQ instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-01-12 17:42:17 +01:00
// AVX512_4VNNIW instructions end
// AVX512_VPOPCNTDQ instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpopcntd, 2, 0x6655, None, 1, CpuAVX512_VPOPCNTDQ, Modrm|Masking=3|VexOpcode=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpopcntq, 2, 0x6655, None, 1, CpuAVX512_VPOPCNTDQ, Modrm|Masking=3|VexOpcode=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VPOPCNTDQ instructions gas/ 2017-01-12 Igor Tsimbalist <igor.v.tsimbalist@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_vpopcntdq. (cpu_noarch): Add noavx512_vpopcntdq. * doc/c-i386.texi: Document avx512_vpopcntdq, noavx512_vpopcntdq. * testsuite/gas/i386/i386.exp: Run AVX512_VPOPCNTDQ tests. * testsuite/gas/i386/avx512_vpopcntdqd-intel.d: New file. * testsuite/gas/i386/avx512_vpopcntdqd.d: Ditto. * testsuite/gas/i386/avx512_vpopcntdqd.s: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd.d: Ditto. * testsuite/gas/i386/x86-64-avx512_vpopcntdqd.s: Ditto. opcodes/ 2017-01-12 Igor Tsimbalist <igor.v.tsimbalist@intel.com> * i386-dis.c (enum): Add PREFIX_EVEX_0F3855, EVEX_W_0F3855_P_2. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VPOPCNTDQ_FLAGS, CPU_ANY_AVX512_VPOPCNTDQ_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VPOPCNTDQ. * i386-opc.h (enum): (AVX512_VPOPCNTDQ): New. (i386_cpu_flags): Add cpuavx512_vpopcntdq. * i386-opc.tbl: Add Intel AVX512_VPOPCNTDQ instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-01-12 17:42:17 +01:00
// AVX512_VPOPCNTDQ instructions end
Enable Intel AVX512_4VNNIW instructions gas/ * config/tc-i386.c: (cpu_arch) Add .avx512_4vnniw. (cpu_noarch): Add noavx512_4vnniw. * doc/c-i386.texi: Document avx512_4vnniw, noavx512_4vnniw. * testsuite/gas/i386/i386.exp: Run AVX512_4VNNIW tests. * testsuite/gas/i386/avx512_4vnniwd_vl-intel.d: New test. * testsuite/gas/i386/avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/avx512_4vnniwd.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd_vl.s: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.d: Ditto. * testsuite/gas/i386/x86-64-avx512_4vnniwd.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3852, PREFIX_EVEX_0F3853. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_4VNNIW_FLAGS, CPU_ANY_AVX512_4VNNIW_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_4VNNIW. * i386-opc.h (enum): (AVX512_4VNNIW): New. (i386_cpu_flags): Add cpuavx512_4vnniw. * i386-opc.tbl: Add Intel AVX512_4VNNIW instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2016-11-02 20:31:25 +01:00
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
// AVX512_VBMI2 instructions
x86: fold various AVX512 templates with so far differing Masking attributes There's no insn allowing ZEROING_MASKING alone. Re-purpose its value for handling the not uncommon case of insns allowing either form of masking with register operands, but only merging masking with a memory operand.
2018-07-31 10:57:09 +02:00
vpcompressb, 2, 0x6663, None, 1, CpuAVX512_VBMI2, Modrm|MaskingMorZ|VexOpcode=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
vpcompressw, 2, 0x6663, None, 1, CpuAVX512_VBMI2, Modrm|MaskingMorZ|VexOpcode=1|VexW=2|Disp8MemShift=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various non-memory operand AVX512VL templates There's little point carrying up to three templates per insn flavor when the sole difference is operand size and the dependency on AVX512VL being enabled. Instead the need for AVX512VL can be derived from an operand allowing for ZMMword as well as one or both or XMMword and YMMword (irrespective of whether this is a register or memory operand). Without further abstraction to deal with the different Disp8MemShift values between the templates, only a limited set (mostly ones only allowing for non-memory operands) can be folded, which is being done here. Also drop IgnoreSize wherever possible from anything that's being touched anyway.
2018-04-26 08:55:02 +02:00
vpexpandb, 2, 0x6662, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexW=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpexpandw, 2, 0x6662, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8MemShift=1|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldvd, 3, 0x6671, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdvd, 3, 0x6673, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldvq, 3, 0x6671, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdvq, 3, 0x6673, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldvw, 3, 0x6670, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdvw, 3, 0x6672, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldd, 4, 0x6671, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdd, 4, 0x6673, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldq, 4, 0x6671, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdq, 4, 0x6673, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshldw, 4, 0x6670, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpshrdw, 4, 0x6672, None, 1, CpuAVX512_VBMI2, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VBMI2 instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vbmi2. (cpu_noarch): noavx512_vbmi2. * doc/c-i386.texi: Document .avx512_vbmi2, noavx512_vbmi2. * testsuite/gas/i386/i386.exp: Add AVX512_VBMI2 tests. * testsuite/gas/i386/avx512vbmi2-intel.d: New test. * testsuite/gas/i386/avx512vbmi2.d: Likewise. * testsuite/gas/i386/avx512vbmi2.s: Likewise. * testsuite/gas/i386/avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/avx512vbmi2_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2.s: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vbmi2_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add b_scalar_mode, w_scalar_mode. Define EXbScalar and EXwScalar for OP_EX. (enum): Add PREFIX_EVEX_0F3862, PREFIX_EVEX_0F3863, PREFIX_EVEX_0F3870, PREFIX_EVEX_0F3871, PREFIX_EVEX_0F3872, PREFIX_EVEX_0F3873, PREFIX_EVEX_0F3A70, PREFIX_EVEX_0F3A71, PREFIX_EVEX_0F3A72, PREFIX_EVEX_0F3A73. (enum): Add EVEX_W_0F3862_P_2, EVEX_W_0F3863_P_2, EVEX_W_0F3870_P_2, EVEX_W_0F3871_P_2, EVEX_W_0F3872_P_2, EVEX_W_0F3873_P_2, EVEX_W_0F3A70_P_2, EVEX_W_0F3A71_P_2, EVEX_W_0F3A72_P_2, EVEX_W_0F3A73_P_2. (intel_operand_size): Handle b_scalar_mode and w_scalar_mode. (OP_E_memory): Likewise. * i386-dis-evex.h: Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VBMI2, CPU_ANY_AVX512_VBMI2_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VBMI2. * i386-opc.h (enum): Add CpuAVX512_VBMI2. (i386_cpu_flags): Add cpuavx512_vbmi2. * i386-opc.tbl: Add Intel AVX512_VBMI2 instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 21:52:59 +02:00
// AVX512_VBMI2 instructions end
Enable Intel AVX512_VNNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vnni. (cpu_noarch): Add noavx512_vnni. * doc/c-i386.texi: Document .avx512_vnni. * testsuite/gas/i386/i386.exp: Add AVX512_VNNI tests. * testsuite/gas/i386/avx512vnni-intel.d: New test. * testsuite/gas/i386/avx512vnni.d: Likewise. * testsuite/gas/i386/avx512vnni.s: Likewise. * testsuite/gas/i386/avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3850, PREFIX_EVEX_0F3851. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VNNI, CPU_ANY_AVX512_VNNI_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VNNI. * i386-opc.h (enum): Add CpuAVX512_VNNI. (i386_cpu_flags): Add cpuavx512_vnni. * i386-opc.tbl Add Intel AVX512_VNNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:52:52 +02:00
// AVX512_VNNI instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpdpbusd, 3, 0x6650, None, 1, CpuAVX512_VNNI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpdpwssd, 3, 0x6652, None, 1, CpuAVX512_VNNI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VNNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vnni. (cpu_noarch): Add noavx512_vnni. * doc/c-i386.texi: Document .avx512_vnni. * testsuite/gas/i386/i386.exp: Add AVX512_VNNI tests. * testsuite/gas/i386/avx512vnni-intel.d: New test. * testsuite/gas/i386/avx512vnni.d: Likewise. * testsuite/gas/i386/avx512vnni.s: Likewise. * testsuite/gas/i386/avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3850, PREFIX_EVEX_0F3851. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VNNI, CPU_ANY_AVX512_VNNI_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VNNI. * i386-opc.h (enum): Add CpuAVX512_VNNI. (i386_cpu_flags): Add cpuavx512_vnni. * i386-opc.tbl Add Intel AVX512_VNNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:52:52 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpdpbusds, 3, 0x6651, None, 1, CpuAVX512_VNNI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpdpwssds, 3, 0x6653, None, 1, CpuAVX512_VNNI, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_VNNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_vnni. (cpu_noarch): Add noavx512_vnni. * doc/c-i386.texi: Document .avx512_vnni. * testsuite/gas/i386/i386.exp: Add AVX512_VNNI tests. * testsuite/gas/i386/avx512vnni-intel.d: New test. * testsuite/gas/i386/avx512vnni.d: Likewise. * testsuite/gas/i386/avx512vnni.s: Likewise. * testsuite/gas/i386/avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.d: Likewise. * testsuite/gas/i386/avx512vnni_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512vnni_vl.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3850, PREFIX_EVEX_0F3851. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VNNI, CPU_ANY_AVX512_VNNI_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_VNNI. * i386-opc.h (enum): Add CpuAVX512_VNNI. (i386_cpu_flags): Add cpuavx512_vnni. * i386-opc.tbl Add Intel AVX512_VNNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:52:52 +02:00
// AVX512_VNNI instructions end
Enable Intel AVX512_BITALG instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_bitalg. (cpu_noarch): noavx512_bitalg. * doc/c-i386.texi: Document .avx512_bitalg, noavx512_bitalg. * testsuite/gas/i386/i386.exp: Add AVX512_BITALG tests. * testsuite/gas/i386/avx512f_bitalg-intel.d: New test. * testsuite/gas/i386/avx512f_bitalg.d: Likewise. * testsuite/gas/i386/avx512f_bitalg.s: Likewise. * testsuite/gas/i386/avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3854, PREFIX_EVEX_0F388F. (enum): Add EVEX_W_0F3854_P_2. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BITALG, CPU_ANY_AVX512_BITALG_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_BITALG. * i386-opc.h (enum): Add CpuAVX512_BITALG. (i386_cpu_flags): Add cpuavx512_bitalg.. * i386-opc.tbl: Add Intel AVX512_BITALG instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:56:30 +02:00
// AVX512_BITALG instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpopcntb, 2, 0x6654, None, 1, CpuAVX512_BITALG, Modrm|Masking=3|VexOpcode=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
vpopcntw, 2, 0x6654, None, 1, CpuAVX512_BITALG, Modrm|Masking=3|VexOpcode=1|VexW=2|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM }
Enable Intel AVX512_BITALG instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_bitalg. (cpu_noarch): noavx512_bitalg. * doc/c-i386.texi: Document .avx512_bitalg, noavx512_bitalg. * testsuite/gas/i386/i386.exp: Add AVX512_BITALG tests. * testsuite/gas/i386/avx512f_bitalg-intel.d: New test. * testsuite/gas/i386/avx512f_bitalg.d: Likewise. * testsuite/gas/i386/avx512f_bitalg.s: Likewise. * testsuite/gas/i386/avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3854, PREFIX_EVEX_0F388F. (enum): Add EVEX_W_0F3854_P_2. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BITALG, CPU_ANY_AVX512_BITALG_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_BITALG. * i386-opc.h (enum): Add CpuAVX512_BITALG. (i386_cpu_flags): Add cpuavx512_bitalg.. * i386-opc.tbl: Add Intel AVX512_BITALG instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:56:30 +02:00
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vpshufbitqmb, 3, 0x668f, None, 1, CpuAVX512_BITALG, Modrm|Masking=2|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
Enable Intel AVX512_BITALG instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .avx512_bitalg. (cpu_noarch): noavx512_bitalg. * doc/c-i386.texi: Document .avx512_bitalg, noavx512_bitalg. * testsuite/gas/i386/i386.exp: Add AVX512_BITALG tests. * testsuite/gas/i386/avx512f_bitalg-intel.d: New test. * testsuite/gas/i386/avx512f_bitalg.d: Likewise. * testsuite/gas/i386/avx512f_bitalg.s: Likewise. * testsuite/gas/i386/avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/avx512vl_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_bitalg.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_bitalg.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3854, PREFIX_EVEX_0F388F. (enum): Add EVEX_W_0F3854_P_2. * i386-dis-evex.h (evex_table): Updated. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BITALG, CPU_ANY_AVX512_BITALG_FLAGS. Update CPU_ANY_AVX512F_FLAGS. (cpu_flags): Add CpuAVX512_BITALG. * i386-opc.h (enum): Add CpuAVX512_BITALG. (i386_cpu_flags): Add cpuavx512_bitalg.. * i386-opc.tbl: Add Intel AVX512_BITALG instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:56:30 +02:00
// AVX512_BITALG instructions end
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
// AVX512 + GFNI instructions
x86: fold various AVX512* templates
2018-07-19 08:34:45 +02:00
vgf2p8affineinvqb, 4, 0x66cf, None, 1, CpuGFNI|CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vgf2p8affineqb, 4, 0x66ce, None, 1, CpuGFNI|CpuAVX512F, Modrm|Masking=3|VexOpcode=2|VexVVVV=1|VexW=2|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vgf2p8mulb, 3, 0x66cf, None, 1, CpuGFNI|CpuAVX512F, Modrm|Masking=3|VexOpcode=1|VexVVVV=1|VexW=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel GFNI instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add .gfni. * doc/c-i386.texi: Document .gfni. * testsuite/gas/i386/i386.exp: Add GFNI tests. * testsuite/gas/i386/avx.s: New GFNI test. * testsuite/gas/i386/x86-64-avx.s: Likewise. * testsuite/gas/i386/avx.d: Adjust. * testsuite/gas/i386/avx-intel.d: Likewise * testsuite/gas/i386/ilp32/x86-64-avx-intel.d: Likewise. * testsuite/gas/i386/ilp32/x86-64-avx.d: Likewise. * testsuite/gas/i386/avx512f_gfni-intel.d: New test. * testsuite/gas/i386/avx512f_gfni.d: Likewise. * testsuite/gas/i386/avx512f_gfni.s: Likewise. * testsuite/gas/i386/avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.d: Likewise. * testsuite/gas/i386/avx512vl_gfni.s: Likewise. * testsuite/gas/i386/gfni-intel.d: Likewise. * testsuite/gas/i386/gfni.d: Likewise. * testsuite/gas/i386/gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512f_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx512vl_gfni.s: Likewise. * testsuite/gas/i386/x86-64-avx_gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.d: Likewise. * testsuite/gas/i386/x86-64-avx_gfni.s: Likewise. * testsuite/gas/i386/x86-64-gfni-intel.d: Likewise. * testsuite/gas/i386/x86-64-gfni.d: Likewise. * testsuite/gas/i386/x86-64-gfni.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F38CF, PREFIX_0F3ACE, PREFIX_0F3ACF, PREFIX_VEX_0F38CF, PREFIX_VEX_0F3ACE, PREFIX_VEX_0F3ACF, PREFIX_EVEX_0F38CF, PREFIX_EVEX_0F3ACE, PREFIX_EVEX_0F3ACF. (enum): Add VEX_W_0F38CF_P_2, VEX_W_0F3ACE_P_2, VEX_W_0F3ACF_P_2, EVEX_W_0F3ACE_P_2, EVEX_W_0F3ACF_P_2. (prefix_table): Updated (see prefixes above). (three_byte_table): Likewise. (vex_w_table): Likewise. * i386-dis-evex.h: Likewise. * i386-gen.c (cpu_flag_init): Add CPU_GFNI_FLAGS, CpuGFNI. (cpu_flags): Add CpuGFNI. * i386-opc.h (enum): Add CpuGFNI. (i386_cpu_flags): Add cpugfni. * i386-opc.tbl: Add Intel GFNI instructions. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2017-10-20 22:26:11 +02:00
// AVX512 + GFNI instructions end
Enable Intel VAES instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VAES. * doc/c-i386.texi: Document VAES. * testsuite/gas/i386/i386.exp: Run VAES tests. * testsuite/gas/i386/avx512f_vaes-intel.d: New test. * testsuite/gas/i386/avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vaes.d: Ditto. * testsuite/gas/i386/avx512f_vaes.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.s: Ditto. * testsuite/gas/i386/vaes-intel.d: Ditto. * testsuite/gas/i386/vaes.d: Ditto. * testsuite/gas/i386/vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.s: Ditto. * testsuite/gas/i386/x86-64-vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-vaes.d: Ditto. * testsuite/gas/i386/x86-64-vaes.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F38DC, PREFIX_EVEX_0F38DD, PREFIX_EVEX_0F38DE, PREFIX_EVEX_0F38DF. (enum): Remove VEX_LEN_0F38DC_P_2, VEX_LEN_0F38DD_P_2, VEX_LEN_0F38DE_P_2, VEX_LEN_0F38DF_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F38DC_P_2, VEX_W_0F38DD_P_2, VEX_W_0F38DE_P_2, VEX_W_0F38DF_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VAES. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. * i386-opc.tbl (vaes{enc,dec}{last,}): New instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:35:45 +02:00
// AVX512 + VAES instructions
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vaesdec, 3, 0x66de, None, 1, CpuVAES|CpuAVX512F, Modrm|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vaesdeclast, 3, 0x66df, None, 1, CpuVAES|CpuAVX512F, Modrm|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vaesenc, 3, 0x66dc, None, 1, CpuVAES|CpuAVX512F, Modrm|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vaesenclast, 3, 0x66dd, None, 1, CpuVAES|CpuAVX512F, Modrm|VexOpcode=1|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel VAES instructions. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VAES. * doc/c-i386.texi: Document VAES. * testsuite/gas/i386/i386.exp: Run VAES tests. * testsuite/gas/i386/avx512f_vaes-intel.d: New test. * testsuite/gas/i386/avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vaes.d: Ditto. * testsuite/gas/i386/avx512f_vaes.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.d: Ditto. * testsuite/gas/i386/avx512vl_vaes.s: Ditto. * testsuite/gas/i386/vaes-intel.d: Ditto. * testsuite/gas/i386/vaes.d: Ditto. * testsuite/gas/i386/vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vaes.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vaes.s: Ditto. * testsuite/gas/i386/x86-64-vaes-intel.d: Ditto. * testsuite/gas/i386/x86-64-vaes.d: Ditto. * testsuite/gas/i386/x86-64-vaes.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F38DC, PREFIX_EVEX_0F38DD, PREFIX_EVEX_0F38DE, PREFIX_EVEX_0F38DF. (enum): Remove VEX_LEN_0F38DC_P_2, VEX_LEN_0F38DD_P_2, VEX_LEN_0F38DE_P_2, VEX_LEN_0F38DF_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F38DC_P_2, VEX_W_0F38DD_P_2, VEX_W_0F38DE_P_2, VEX_W_0F38DF_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VAES. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. * i386-opc.tbl (vaes{enc,dec}{last,}): New instructions. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:35:45 +02:00
// AVX512 + VAES instructions end
Enable Intel VPCLMULQDQ instruction. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VPCLMULQDQ. * doc/c-i386.texi: Document VPCLMULQDQ. * testsuite/gas/i386/i386.exp: Run VPCLMULQDQ tests. * testsuite/gas/i386/avx512f_vpclmulqdq-intel.d: New test. * testsuite/gas/i386/avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/vpclmulqdq.d: Ditto. * testsuite/gas/i386/vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3A44. (enum): Remove VEX_LEN_0F3A44_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F3A44_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VPCLMULQDQ. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. (CpuUnused): Comment out to avoid zero-width field problem. * i386-opc.tbl (vpclmulqdq): New instruction. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:42:40 +02:00
// AVX512 + VPCLMULQDQ instructions
x86: Replace VexW=3 with VexWIG * i386-opc.tbl (VexWIG): New. Replace VexW=3 with VexWIG.
2018-09-17 15:11:54 +02:00
vpclmulqdq, 4, 0x6644, None, 1, CpuVPCLMULQDQ|CpuAVX512F, Modrm|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Imm8, RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: add missing pseudo ops for VPCLMULQDQ ISA extension While the ISA extensions doc suggests them to be made available just like the SDM does for the PCLMULQDQ ISA extension, these weren't added when supposrt for the new extension was introduced. Also make sure the 64-bit non-AVX512 test actually tests VEX encodings, not EVEX ones.
2019-07-01 08:28:58 +02:00
vpclmullqlqdq, 3, 0x6644, 0x00, 1, CpuVPCLMULQDQ|CpuAVX512F, Modrm|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpclmulhqlqdq, 3, 0x6644, 0x01, 1, CpuVPCLMULQDQ|CpuAVX512F, Modrm|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpclmullqhqdq, 3, 0x6644, 0x10, 1, CpuVPCLMULQDQ|CpuAVX512F, Modrm|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
vpclmulhqhqdq, 3, 0x6644, 0x11, 1, CpuVPCLMULQDQ|CpuAVX512F, Modrm|VexOpcode=2|VexWIG|VexVVVV=1|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ImmExt, { RegXMM|RegYMM|RegZMM|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
Enable Intel VPCLMULQDQ instruction. Intel has disclosed a set of new instructions. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf gas/ * config/tc-i386.c (cpu_arch): Add VPCLMULQDQ. * doc/c-i386.texi: Document VPCLMULQDQ. * testsuite/gas/i386/i386.exp: Run VPCLMULQDQ tests. * testsuite/gas/i386/avx512f_vpclmulqdq-intel.d: New test. * testsuite/gas/i386/avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/vpclmulqdq.d: Ditto. * testsuite/gas/i386/vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512f_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig.s: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1-intel.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq-wig1.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-avx512vl_vpclmulqdq.s: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq-intel.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.d: Ditto. * testsuite/gas/i386/x86-64-vpclmulqdq.s: Ditto. opcodes/ * i386-dis.c (enum): Add PREFIX_EVEX_0F3A44. (enum): Remove VEX_LEN_0F3A44_P_2. (vex_len_table): Ditto. (enum): Remove VEX_W_0F3A44_P_2. (vew_w_table): Ditto. (prefix_table): Adjust instructions (see prefixes above). * i386-dis-evex.h (evex_table): Add new instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add VPCLMULQDQ. (bitfield_cpu_flags): Ditto. * i386-opc.h (enum): Ditto. (i386_cpu_flags): Ditto. (CpuUnused): Comment out to avoid zero-width field problem. * i386-opc.tbl (vpclmulqdq): New instruction. * i386-init.h: Regenerate. * i386-tbl.h: Ditto.
2017-10-20 22:42:40 +02:00
// AVX512 + VPCLMULQDQ instructions end
Add znver1 processor
2015-03-17 17:19:15 +01:00
// CLZERO instructions
clzero, 0, 0xf01fc, None, 3, CpuCLZERO, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
clzero, 1, 0xf01fc, None, 3, CpuCLZERO, AddrPrefixOpReg, { Acc|Word|Dword|Qword }
Add znver1 processor
2015-03-17 17:19:15 +01:00
// CLZERO instructions end
Add support for monitorx/mwaitx instructions gas/ * config/tc-i386.c (cpu_arch): Add .mwaitx. (process_immext): Check operands for monitorx/mwaitx instructions. * doc/c-i386.texi: Document mwaitx. gas/testsuite/ * gas/i386/i386.exp: Add new mwaitx test cases. * gas/i386/mwaitx.s: New. * gas/i386/mwaitx-bdver4.d: New. * gas/i386/x86-64-mwaitx.s: New. * gas/i386/x86-64-mwaitx-bdver4.d: New. * gas/i386/mwaitx-reg.s: New. * gas/i386/mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.s: New. * gas/i386/arch-13.s: Updated. * gas/i386/arch-13.d: Updated. * gas/i386/arch-13-znver1.d: Updated. * gas/i386/x86-64-arch-3.s: Updated. * gas/i386/x86-64-arch-3.d: Updated. * gas/i386/x86-64-arch-3-znver1.d: Updated. opcodes/ * i386-dis.c (OP_Mwaitx): New. (rm_table): Add monitorx/mwaitx. * i386-gen.c (cpu_flag_init): Add CpuMWAITX to CPU_BDVER4_FLAGS and CPU_ZNVER1_FLAGS. Add CPU_MWAITX_FLAGS. (operand_type_init): Add CpuMWAITX. * i386-opc.h (CpuMWAITX): New. (i386_cpu_flags): Add cpumwaitx. * i386-opc.tbl: Add monitorx and mwaitx. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2015-06-30 08:41:52 +02:00
// MONITORX/MWAITX instructions
x86: eliminate ImmExt abuse Drop the remaining instances left in place by commit c3949f432f ("x86: limit ImmExt abuse), now that we have a way to specify specific GPRs. Take the opportunity and also introduce proper 16-bit forms of applicable SVME insns as well as 1-operand forms of CLZERO.
2019-11-12 09:08:32 +01:00
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
monitorx, 0, 0xf01fa, None, 3, CpuMWAITX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
monitorx, 3, 0xf01fa, None, 3, CpuMWAITX, AddrPrefixOpReg, { Acc|Word|Dword|Qword, RegC|Dword, RegD|Dword }
x86: adjust monitor/mwait templates Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or 32-bit register outside of 64-bit mode, and a 64- or 32-bit register inside 64-bit mode. The other register operands, including all of them for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain the original 64-bit MONITOR{,X} templates for compatibility only, and fold the MWAIT{,X} ones.
2018-07-11 10:30:00 +02:00
// The 64-bit form exists only for compatibility with older gas.
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
monitorx, 3, 0xf01fa, None, 3, CpuMWAITX|Cpu64, AddrPrefixOpReg, { Acc|Dword|Qword, RegC|Qword, RegD|Qword }
Add support for monitorx/mwaitx instructions gas/ * config/tc-i386.c (cpu_arch): Add .mwaitx. (process_immext): Check operands for monitorx/mwaitx instructions. * doc/c-i386.texi: Document mwaitx. gas/testsuite/ * gas/i386/i386.exp: Add new mwaitx test cases. * gas/i386/mwaitx.s: New. * gas/i386/mwaitx-bdver4.d: New. * gas/i386/x86-64-mwaitx.s: New. * gas/i386/x86-64-mwaitx-bdver4.d: New. * gas/i386/mwaitx-reg.s: New. * gas/i386/mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.s: New. * gas/i386/arch-13.s: Updated. * gas/i386/arch-13.d: Updated. * gas/i386/arch-13-znver1.d: Updated. * gas/i386/x86-64-arch-3.s: Updated. * gas/i386/x86-64-arch-3.d: Updated. * gas/i386/x86-64-arch-3-znver1.d: Updated. opcodes/ * i386-dis.c (OP_Mwaitx): New. (rm_table): Add monitorx/mwaitx. * i386-gen.c (cpu_flag_init): Add CpuMWAITX to CPU_BDVER4_FLAGS and CPU_ZNVER1_FLAGS. Add CPU_MWAITX_FLAGS. (operand_type_init): Add CpuMWAITX. * i386-opc.h (CpuMWAITX): New. (i386_cpu_flags): Add cpumwaitx. * i386-opc.tbl: Add monitorx and mwaitx. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2015-06-30 08:41:52 +02:00
x86: limit ImmExt abuse In quite a few cases ImmExt gets used when there's not really any immediate, but rather a degenerate ModR/M byte. ENCL{S,U} show how this case is supposed to be dealt with. Eliminate most abuses, leaving in place (for now) only ones where process_immext() is involved.
2019-07-01 08:37:40 +02:00
mwaitx, 0, 0xf01fb, None, 3, CpuMWAITX, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
x86: adjust monitor/mwait templates Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or 32-bit register outside of 64-bit mode, and a 64- or 32-bit register inside 64-bit mode. The other register operands, including all of them for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain the original 64-bit MONITOR{,X} templates for compatibility only, and fold the MWAIT{,X} ones.
2018-07-11 10:30:00 +02:00
// The 64-bit form exists only for compatibility with older gas.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
mwaitx, 3, 0xf01fb, None, 3, CpuMWAITX, CheckRegSize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Acc|Dword|Qword, RegC|Dword|Qword, RegB|Dword|Qword }
x86: eliminate ImmExt abuse Drop the remaining instances left in place by commit c3949f432f ("x86: limit ImmExt abuse), now that we have a way to specify specific GPRs. Take the opportunity and also introduce proper 16-bit forms of applicable SVME insns as well as 1-operand forms of CLZERO.
2019-11-12 09:08:32 +01:00
Add support for monitorx/mwaitx instructions gas/ * config/tc-i386.c (cpu_arch): Add .mwaitx. (process_immext): Check operands for monitorx/mwaitx instructions. * doc/c-i386.texi: Document mwaitx. gas/testsuite/ * gas/i386/i386.exp: Add new mwaitx test cases. * gas/i386/mwaitx.s: New. * gas/i386/mwaitx-bdver4.d: New. * gas/i386/x86-64-mwaitx.s: New. * gas/i386/x86-64-mwaitx-bdver4.d: New. * gas/i386/mwaitx-reg.s: New. * gas/i386/mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.l: New. * gas/i386/x86-64-mwaitx-reg.s: New. * gas/i386/arch-13.s: Updated. * gas/i386/arch-13.d: Updated. * gas/i386/arch-13-znver1.d: Updated. * gas/i386/x86-64-arch-3.s: Updated. * gas/i386/x86-64-arch-3.d: Updated. * gas/i386/x86-64-arch-3-znver1.d: Updated. opcodes/ * i386-dis.c (OP_Mwaitx): New. (rm_table): Add monitorx/mwaitx. * i386-gen.c (cpu_flag_init): Add CpuMWAITX to CPU_BDVER4_FLAGS and CPU_ZNVER1_FLAGS. Add CPU_MWAITX_FLAGS. (operand_type_init): Add CpuMWAITX. * i386-opc.h (CpuMWAITX): New. (i386_cpu_flags): Add cpumwaitx. * i386-opc.tbl: Add monitorx and mwaitx. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2015-06-30 08:41:52 +02:00
// MONITORX/MWAITX instructions end
Implement Intel OSPKE instructions This patch implements Intel OSPKE instructions documented in Intel64 and IA-32 Architectures Software Developer’s Manual Volume 2, September 2015. gas/testsuite/ * gas/i386/i386.exp: Run ospke and x86-64-ospke. * gas/i386/ospke.d: New file. * gas/i386/ospke.s: Likewise. * gas/i386/x86-64-ospke.d: Likewise. opcodes/ * i386-dis.c (MOD_0F01_REG_5): New. (RM_0F01_REG_5): Likewise. (reg_table): Use MOD_0F01_REG_5. (mod_table): Add MOD_0F01_REG_5. (rm_table): Add RM_0F01_REG_5. * i386-gen.c (cpu_flag_init): Add CPU_OSPKE_FLAGS. (cpu_flags): Add CpuOSPKE. * i386-opc.h (CpuOSPKE): New. (i386_cpu_flags): Add cpuospke. * i386-opc.tbl: Add rdpkru and wrpkru instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2015-12-09 17:01:57 +01:00
// OSPKE instructions.
rdpkru, 0, 0xf01ee, None, 3, CpuOSPKE, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
wrpkru, 0, 0xf01ef, None, 3, CpuOSPKE, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// OSPKE instructions end.
Enable Intel RDPID instruction. This patch enables Intel RDPID instruction described in Intel64 and IA-32 Architectures Software Developer's Manual, April 2016. gas/ * config/tc-i386.c (cpu_arch): Add RDPID. * doc/c-i386.texi: Document RDPID. gas/testsuite/ * gas/i386/i386.exp: Run RDPID tests. * gas/i386/prefix.d: Adjust. * gas/i386/rdpid.s: New test. * gas/i386/rdpid.d: Ditto. * gas/i386/rdpid-intel.d: Ditto. * gas/i386/x86-64-rdpid.s: Ditto. * gas/i386/x86-64-rdpid.d: Ditto. * gas/i386/x86-64-rdpid-intel.d: Ditto. opcodes/ * i386-dis.c (prefix_table): Add RDPID instruction. * i386-gen.c (cpu_flag_init): Add RDPID flag. (cpu_flags): Add RDPID bitfield. * i386-opc.h (enum): Add RDPID element. (i386_cpu_flags): Add RDPID field. * i386-opc.tbl: Add RDPID instruction. * i386-init.h: Regenerate. * i386-tbl.h: Regenerate.
2016-05-10 20:35:52 +02:00
// RDPID instructions.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
rdpid, 1, 0xf30fc7, 0x7, 2, CpuRDPID|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32 }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
rdpid, 1, 0xf30fc7, 0x7, 2, CpuRDPID|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Reg64 }
Enable Intel RDPID instruction. This patch enables Intel RDPID instruction described in Intel64 and IA-32 Architectures Software Developer's Manual, April 2016. gas/ * config/tc-i386.c (cpu_arch): Add RDPID. * doc/c-i386.texi: Document RDPID. gas/testsuite/ * gas/i386/i386.exp: Run RDPID tests. * gas/i386/prefix.d: Adjust. * gas/i386/rdpid.s: New test. * gas/i386/rdpid.d: Ditto. * gas/i386/rdpid-intel.d: Ditto. * gas/i386/x86-64-rdpid.s: Ditto. * gas/i386/x86-64-rdpid.d: Ditto. * gas/i386/x86-64-rdpid-intel.d: Ditto. opcodes/ * i386-dis.c (prefix_table): Add RDPID instruction. * i386-gen.c (cpu_flag_init): Add RDPID flag. (cpu_flags): Add RDPID bitfield. * i386-opc.h (enum): Add RDPID element. (i386_cpu_flags): Add RDPID field. * i386-opc.tbl: Add RDPID instruction. * i386-init.h: Regenerate. * i386-tbl.h: Regenerate.
2016-05-10 20:35:52 +02:00
// RDPID instructions end.
X86: Add ptwrite instruction Implement ptwrite instruction defined in Intel64 and IA-32 Architectures Software Developer’s Manual, June 2016. gas/ * config/tc-i386.c (cpu_arch): Add .ptwrite. * doc/c-i386.texi: Document ptwrite and .ptwrite. * testsuite/gas/i386/i386.exp: Run ptwrite, ptwrite-intel, x86-64-ptwrite and x86-64-ptwrite-intel. * testsuite/gas/i386/ptwrite-intel.d: New file. * testsuite/gas/i386/ptwrite.d: Likewise. * testsuite/gas/i386/ptwrite.s: Likewise. * testsuite/gas/i386/x86-64-ptwrite-intel.d: Likewise. * testsuite/gas/i386/x86-64-ptwrite.d: Likewise. * testsuite/gas/i386/x86-64-ptwrite.s: Likewise. opcodes/ * i386-dis.c (PREFIX_MOD_0_0FAE_REG_4): New. (PREFIX_MOD_3_0FAE_REG_4): Likewise. (prefix_table): Add PREFIX_MOD_0_0FAE_REG_4 and PREFIX_MOD_3_0FAE_REG_4. (mod_table): Use PREFIX_MOD_0_0FAE_REG_4 and PREFIX_MOD_3_0FAE_REG_4. * i386-gen.c (cpu_flag_init): Add CPU_PTWRITE_FLAGS. (cpu_flags): Add CpuPTWRITE. * i386-opc.h (CpuPTWRITE): New. (i386_cpu_flags): Add cpuptwrite. * i386-opc.tbl: Add ptwrite instruction. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2016-08-25 00:27:11 +02:00
// PTWRITE instructions.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
ptwrite, 1, 0xf30fae, 0x4, 2, CpuPTWRITE|CpuNo64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Unspecified|BaseIndex }
ptwrite, 1, 0xf30fae, 0x4, 2, CpuPTWRITE|Cpu64, Modrm|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64|Unspecified|BaseIndex }
X86: Add ptwrite instruction Implement ptwrite instruction defined in Intel64 and IA-32 Architectures Software Developer’s Manual, June 2016. gas/ * config/tc-i386.c (cpu_arch): Add .ptwrite. * doc/c-i386.texi: Document ptwrite and .ptwrite. * testsuite/gas/i386/i386.exp: Run ptwrite, ptwrite-intel, x86-64-ptwrite and x86-64-ptwrite-intel. * testsuite/gas/i386/ptwrite-intel.d: New file. * testsuite/gas/i386/ptwrite.d: Likewise. * testsuite/gas/i386/ptwrite.s: Likewise. * testsuite/gas/i386/x86-64-ptwrite-intel.d: Likewise. * testsuite/gas/i386/x86-64-ptwrite.d: Likewise. * testsuite/gas/i386/x86-64-ptwrite.s: Likewise. opcodes/ * i386-dis.c (PREFIX_MOD_0_0FAE_REG_4): New. (PREFIX_MOD_3_0FAE_REG_4): Likewise. (prefix_table): Add PREFIX_MOD_0_0FAE_REG_4 and PREFIX_MOD_3_0FAE_REG_4. (mod_table): Use PREFIX_MOD_0_0FAE_REG_4 and PREFIX_MOD_3_0FAE_REG_4. * i386-gen.c (cpu_flag_init): Add CPU_PTWRITE_FLAGS. (cpu_flags): Add CpuPTWRITE. * i386-opc.h (CpuPTWRITE): New. (i386_cpu_flags): Add cpuptwrite. * i386-opc.tbl: Add ptwrite instruction. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2016-08-25 00:27:11 +02:00
// PTWRITE instructions end.
Add support for Intel CET instructions Support Intel Control-flow Enforcement Technology (CET) instructions: https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf gas/ * config/tc-i386.c (cpu_arch): Add .cet. * doc/c-i386.texi: Document cet. * testsuite/gas/i386/cet-intel.d: New file. * testsuite/gas/i386/cet.d: Likewise. * testsuite/gas/i386/cet.s: Likewise. * testsuite/gas/i386/x86-64-cet-intel.d: Likewise. * testsuite/gas/i386/x86-64-cet.d: Likewise. * testsuite/gas/i386/x86-64-cet.s: Likewise. * testsuite/gas/i386/i386.exp: Run Intel CET tests. opcodes/ * i386-dis.c (REG_0F1E_MOD_3): New enum. (MOD_0F1E_PREFIX_1): Likewise. (MOD_0F38F5_PREFIX_2): Likewise. (MOD_0F38F6_PREFIX_0): Likewise. (RM_0F1E_MOD_3_REG_7): Likewise. (PREFIX_MOD_0_0F01_REG_5): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_1): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_2): Likewise. (PREFIX_0F1E): Likewise. (PREFIX_MOD_0_0FAE_REG_5): Likewise. (PREFIX_0F38F5): Likewise. (dis386_twobyte): Use PREFIX_0F1E. (reg_table): Add REG_0F1E_MOD_3. (prefix_table): Add PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_3_0F01_REG_5_RM_1, PREFIX_MOD_3_0F01_REG_5_RM_2, PREFIX_0F1E, PREFIX_MOD_0_0FAE_REG_5 and PREFIX_0F38F5. Update PREFIX_0FAE_REG_6 and PREFIX_0F38F6. (three_byte_table): Use PREFIX_0F38F5. (mod_table): Use PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_0_0FAE_REG_5. Add MOD_0F1E_PREFIX_1, MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0. (rm_table): Add MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0, RM_0F1E_MOD_3_REG_7. Use PREFIX_MOD_3_0F01_REG_5_RM_1 and PREFIX_MOD_3_0F01_REG_5_RM_2. * i386-gen.c (cpu_flag_init): Add CPU_CET_FLAGS. (cpu_flags): Add CpuCET. * i386-opc.h (CpuCET): New enum. (CpuUnused): Commented out. (i386_cpu_flags): Add cpucet. * i386-opc.tbl: Add Intel CET instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2017-03-07 00:26:37 +01:00
// CET instructions.
Replace CET bit with IBT and SHSTK bits. The latest specification for Intel CET technology defined two new bits instead of previously used CET bit. These are IBT and SHSTK bits. The patch replaces CET bit with IBT and SHSTK bits. gas/ * config/tc-i386.c (cpu_arch): Delete .cet. Add .ibt, .shstk. (cpu_noarch): Add noibt, noshstk. (parse_insn): Change cpucet to cpuibt. * doc/c-i386.texi: Delete .cet. Add .ibt, .shstk. * testsuite/gas/i386/cet-ibt-inval.l: New test. * testsuite/gas/i386/cet-ibt-inval.s: Likewise. * testsuite/gas/i386/cet-shstk-inval.l: Likewise. * testsuite/gas/i386/cet-shstk-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.s: Likewise. opcodes/ * i386-gen.c (cpu_flag_init): Delete CPU_CET_FLAGS, CpuCET. Add CPU_IBT_FLAGS, CPU_SHSTK_FLAGS, CPY_ANY_IBT_FLAGS, CPU_ANY_SHSTK_FLAGS, CpuIBT, CpuSHSTK. (cpu_flags): Add CpuIBT, CpuSHSTK. * i386-opc.h (enum): Add CpuIBT, CpuSHSTK. (i386_cpu_flags): Add cpuibt, cpushstk. * i386-opc.tbl: Change CpuCET to CpuSHSTK and CpuIBT. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-17 17:45:52 +01:00
incsspd, 1, 0xf30fae, 0x5, 2, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32 }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
incsspq, 1, 0xf30fae, 0x5, 2, CpuSHSTK|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg64 }
Replace CET bit with IBT and SHSTK bits. The latest specification for Intel CET technology defined two new bits instead of previously used CET bit. These are IBT and SHSTK bits. The patch replaces CET bit with IBT and SHSTK bits. gas/ * config/tc-i386.c (cpu_arch): Delete .cet. Add .ibt, .shstk. (cpu_noarch): Add noibt, noshstk. (parse_insn): Change cpucet to cpuibt. * doc/c-i386.texi: Delete .cet. Add .ibt, .shstk. * testsuite/gas/i386/cet-ibt-inval.l: New test. * testsuite/gas/i386/cet-ibt-inval.s: Likewise. * testsuite/gas/i386/cet-shstk-inval.l: Likewise. * testsuite/gas/i386/cet-shstk-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.s: Likewise. opcodes/ * i386-gen.c (cpu_flag_init): Delete CPU_CET_FLAGS, CpuCET. Add CPU_IBT_FLAGS, CPU_SHSTK_FLAGS, CPY_ANY_IBT_FLAGS, CPU_ANY_SHSTK_FLAGS, CpuIBT, CpuSHSTK. (cpu_flags): Add CpuIBT, CpuSHSTK. * i386-opc.h (enum): Add CpuIBT, CpuSHSTK. (i386_cpu_flags): Add cpuibt, cpushstk. * i386-opc.tbl: Change CpuCET to CpuSHSTK and CpuIBT. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-17 17:45:52 +01:00
rdsspd, 1, 0xf30f1e, 0x1, 2, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32 }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
rdsspq, 1, 0xf30f1e, 0x1, 2, CpuSHSTK|Cpu64, Modrm|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg64 }
saveprevssp, 0, 0xf30f01ea, None, 3, CpuSHSTK, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Replace CET bit with IBT and SHSTK bits. The latest specification for Intel CET technology defined two new bits instead of previously used CET bit. These are IBT and SHSTK bits. The patch replaces CET bit with IBT and SHSTK bits. gas/ * config/tc-i386.c (cpu_arch): Delete .cet. Add .ibt, .shstk. (cpu_noarch): Add noibt, noshstk. (parse_insn): Change cpucet to cpuibt. * doc/c-i386.texi: Delete .cet. Add .ibt, .shstk. * testsuite/gas/i386/cet-ibt-inval.l: New test. * testsuite/gas/i386/cet-ibt-inval.s: Likewise. * testsuite/gas/i386/cet-shstk-inval.l: Likewise. * testsuite/gas/i386/cet-shstk-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.s: Likewise. opcodes/ * i386-gen.c (cpu_flag_init): Delete CPU_CET_FLAGS, CpuCET. Add CPU_IBT_FLAGS, CPU_SHSTK_FLAGS, CPY_ANY_IBT_FLAGS, CPU_ANY_SHSTK_FLAGS, CpuIBT, CpuSHSTK. (cpu_flags): Add CpuIBT, CpuSHSTK. * i386-opc.h (enum): Add CpuIBT, CpuSHSTK. (i386_cpu_flags): Add cpuibt, cpushstk. * i386-opc.tbl: Change CpuCET to CpuSHSTK and CpuIBT. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-17 17:45:52 +01:00
rstorssp, 1, 0xf30f01, 0x5, 2, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86/Intel: accept memory operand size specifiers for CET insns
2018-07-11 10:25:40 +02:00
wrssd, 2, 0x0f38f6, None, 3, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, Dword|Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
wrssq, 2, 0x0f38f6, None, 3, CpuSHSTK|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64, Qword|Unspecified|BaseIndex }
x86/Intel: accept memory operand size specifiers for CET insns
2018-07-11 10:25:40 +02:00
wrussd, 2, 0x660f38f5, None, 3, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, Dword|Unspecified|BaseIndex }
x86: drop Rex64 attribute It is almost entirely redundant with Size64, and the sole case (CRC32) where direct replacement isn't possible can easily be taken care of in another way.
2020-03-06 08:52:12 +01:00
wrussq, 2, 0x660f38f5, None, 3, CpuSHSTK|Cpu64, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|Size64, { Reg64, Qword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
setssbsy, 0, 0xf30f01e8, None, 3, CpuSHSTK, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Replace CET bit with IBT and SHSTK bits. The latest specification for Intel CET technology defined two new bits instead of previously used CET bit. These are IBT and SHSTK bits. The patch replaces CET bit with IBT and SHSTK bits. gas/ * config/tc-i386.c (cpu_arch): Delete .cet. Add .ibt, .shstk. (cpu_noarch): Add noibt, noshstk. (parse_insn): Change cpucet to cpuibt. * doc/c-i386.texi: Delete .cet. Add .ibt, .shstk. * testsuite/gas/i386/cet-ibt-inval.l: New test. * testsuite/gas/i386/cet-ibt-inval.s: Likewise. * testsuite/gas/i386/cet-shstk-inval.l: Likewise. * testsuite/gas/i386/cet-shstk-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.s: Likewise. opcodes/ * i386-gen.c (cpu_flag_init): Delete CPU_CET_FLAGS, CpuCET. Add CPU_IBT_FLAGS, CPU_SHSTK_FLAGS, CPY_ANY_IBT_FLAGS, CPU_ANY_SHSTK_FLAGS, CpuIBT, CpuSHSTK. (cpu_flags): Add CpuIBT, CpuSHSTK. * i386-opc.h (enum): Add CpuIBT, CpuSHSTK. (i386_cpu_flags): Add cpuibt, cpushstk. * i386-opc.tbl: Change CpuCET to CpuSHSTK and CpuIBT. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-17 17:45:52 +01:00
clrssbsy, 1, 0xf30fae, 0x6, 2, CpuSHSTK, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|NoRex64, { Qword|Unspecified|BaseIndex }
x86: drop bogus IgnoreSize from a few further insns
2018-09-13 11:26:06 +02:00
endbr64, 0, 0xf30f1efa, None, 3, CpuIBT, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
endbr32, 0, 0xf30f1efb, None, 3, CpuIBT, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
Add support for Intel CET instructions Support Intel Control-flow Enforcement Technology (CET) instructions: https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf gas/ * config/tc-i386.c (cpu_arch): Add .cet. * doc/c-i386.texi: Document cet. * testsuite/gas/i386/cet-intel.d: New file. * testsuite/gas/i386/cet.d: Likewise. * testsuite/gas/i386/cet.s: Likewise. * testsuite/gas/i386/x86-64-cet-intel.d: Likewise. * testsuite/gas/i386/x86-64-cet.d: Likewise. * testsuite/gas/i386/x86-64-cet.s: Likewise. * testsuite/gas/i386/i386.exp: Run Intel CET tests. opcodes/ * i386-dis.c (REG_0F1E_MOD_3): New enum. (MOD_0F1E_PREFIX_1): Likewise. (MOD_0F38F5_PREFIX_2): Likewise. (MOD_0F38F6_PREFIX_0): Likewise. (RM_0F1E_MOD_3_REG_7): Likewise. (PREFIX_MOD_0_0F01_REG_5): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_1): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_2): Likewise. (PREFIX_0F1E): Likewise. (PREFIX_MOD_0_0FAE_REG_5): Likewise. (PREFIX_0F38F5): Likewise. (dis386_twobyte): Use PREFIX_0F1E. (reg_table): Add REG_0F1E_MOD_3. (prefix_table): Add PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_3_0F01_REG_5_RM_1, PREFIX_MOD_3_0F01_REG_5_RM_2, PREFIX_0F1E, PREFIX_MOD_0_0FAE_REG_5 and PREFIX_0F38F5. Update PREFIX_0FAE_REG_6 and PREFIX_0F38F6. (three_byte_table): Use PREFIX_0F38F5. (mod_table): Use PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_0_0FAE_REG_5. Add MOD_0F1E_PREFIX_1, MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0. (rm_table): Add MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0, RM_0F1E_MOD_3_REG_7. Use PREFIX_MOD_3_0F01_REG_5_RM_1 and PREFIX_MOD_3_0F01_REG_5_RM_2. * i386-gen.c (cpu_flag_init): Add CPU_CET_FLAGS. (cpu_flags): Add CpuCET. * i386-opc.h (CpuCET): New enum. (CpuUnused): Commented out. (i386_cpu_flags): Add cpucet. * i386-opc.tbl: Add Intel CET instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2017-03-07 00:26:37 +01:00
x86: Add NOTRACK prefix support For register indirect branches, NOTRACK prefix (0x3e), which is also the DS segment register prefix, can be used to ignore the CET indirect branch track. gas/ * config/tc-i386.c (REX_PREFIX): Changed to 7. (NOTRACK_PREFIX): New. (MAX_PREFIXES): Changed to 8. (_i386_insn): Add notrack_prefix. (PREFIX_GROUP): Add PREFIX_DS. (add_prefix): Return PREFIX_DS for DS_PREFIX_OPCODE. (md_assemble): Check if NOTRACK prefix is supported. (parse_insn): Set notrack_prefix and issue an error for other prefixes after NOTRACK prefix. * testsuite/gas/i386/i386.exp: Run tests for NOTRACK prefix. * testsuite/gas/i386/notrack-intel.d: New file. * testsuite/gas/i386/notrack.d: Likewise. * testsuite/gas/i386/notrack.s: Likewise. * testsuite/gas/i386/notrackbad.l: Likewise. * testsuite/gas/i386/notrackbad.s: Likewise. * testsuite/gas/i386/x86-64-notrack-intel.d: Likewise. * testsuite/gas/i386/x86-64-notrack.d: Likewise. * testsuite/gas/i386/x86-64-notrack.s: Likewise. * testsuite/gas/i386/x86-64-notrackbad.l: Likewise. * testsuite/gas/i386/x86-64-notrackbad.s: Likewise. include/ * include/opcode/i386.h (NOTRACK_PREFIX_OPCODE): New. opcodes/ * i386-dis.c (NOTRACK_Fixup): New. (NOTRACK): Likewise. (NOTRACK_PREFIX): Likewise. (last_active_prefix): Likewise. (reg_table): Use NOTRACK on indirect call and jmp. (ckprefix): Set last_active_prefix. (prefix_name): Return "notrack" for NOTRACK_PREFIX. * i386-gen.c (opcode_modifiers): Add NoTrackPrefixOk. * i386-opc.h (NoTrackPrefixOk): New. (i386_opcode_modifier): Add notrackprefixok. * i386-opc.tbl: Add NoTrackPrefixOk to indirect call and jmp. Add notrack. * i386-tbl.h: Regenerated.
2017-05-22 20:02:46 +02:00
// notrack prefix
Replace CET bit with IBT and SHSTK bits. The latest specification for Intel CET technology defined two new bits instead of previously used CET bit. These are IBT and SHSTK bits. The patch replaces CET bit with IBT and SHSTK bits. gas/ * config/tc-i386.c (cpu_arch): Delete .cet. Add .ibt, .shstk. (cpu_noarch): Add noibt, noshstk. (parse_insn): Change cpucet to cpuibt. * doc/c-i386.texi: Delete .cet. Add .ibt, .shstk. * testsuite/gas/i386/cet-ibt-inval.l: New test. * testsuite/gas/i386/cet-ibt-inval.s: Likewise. * testsuite/gas/i386/cet-shstk-inval.l: Likewise. * testsuite/gas/i386/cet-shstk-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-ibt-inval.s: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.l: Likewise. * testsuite/gas/i386/x86-64-cet-shstk-inval.s: Likewise. opcodes/ * i386-gen.c (cpu_flag_init): Delete CPU_CET_FLAGS, CpuCET. Add CPU_IBT_FLAGS, CPU_SHSTK_FLAGS, CPY_ANY_IBT_FLAGS, CPU_ANY_SHSTK_FLAGS, CpuIBT, CpuSHSTK. (cpu_flags): Add CpuIBT, CpuSHSTK. * i386-opc.h (enum): Add CpuIBT, CpuSHSTK. (i386_cpu_flags): Add cpuibt, cpushstk. * i386-opc.tbl: Change CpuCET to CpuSHSTK and CpuIBT. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-17 17:45:52 +01:00
notrack, 0, 0x3e, None, 1, CpuIBT, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IsPrefix, { 0 }
x86: Add NOTRACK prefix support For register indirect branches, NOTRACK prefix (0x3e), which is also the DS segment register prefix, can be used to ignore the CET indirect branch track. gas/ * config/tc-i386.c (REX_PREFIX): Changed to 7. (NOTRACK_PREFIX): New. (MAX_PREFIXES): Changed to 8. (_i386_insn): Add notrack_prefix. (PREFIX_GROUP): Add PREFIX_DS. (add_prefix): Return PREFIX_DS for DS_PREFIX_OPCODE. (md_assemble): Check if NOTRACK prefix is supported. (parse_insn): Set notrack_prefix and issue an error for other prefixes after NOTRACK prefix. * testsuite/gas/i386/i386.exp: Run tests for NOTRACK prefix. * testsuite/gas/i386/notrack-intel.d: New file. * testsuite/gas/i386/notrack.d: Likewise. * testsuite/gas/i386/notrack.s: Likewise. * testsuite/gas/i386/notrackbad.l: Likewise. * testsuite/gas/i386/notrackbad.s: Likewise. * testsuite/gas/i386/x86-64-notrack-intel.d: Likewise. * testsuite/gas/i386/x86-64-notrack.d: Likewise. * testsuite/gas/i386/x86-64-notrack.s: Likewise. * testsuite/gas/i386/x86-64-notrackbad.l: Likewise. * testsuite/gas/i386/x86-64-notrackbad.s: Likewise. include/ * include/opcode/i386.h (NOTRACK_PREFIX_OPCODE): New. opcodes/ * i386-dis.c (NOTRACK_Fixup): New. (NOTRACK): Likewise. (NOTRACK_PREFIX): Likewise. (last_active_prefix): Likewise. (reg_table): Use NOTRACK on indirect call and jmp. (ckprefix): Set last_active_prefix. (prefix_name): Return "notrack" for NOTRACK_PREFIX. * i386-gen.c (opcode_modifiers): Add NoTrackPrefixOk. * i386-opc.h (NoTrackPrefixOk): New. (i386_opcode_modifier): Add notrackprefixok. * i386-opc.tbl: Add NoTrackPrefixOk to indirect call and jmp. Add notrack. * i386-tbl.h: Regenerated.
2017-05-22 20:02:46 +02:00
Add support for Intel CET instructions Support Intel Control-flow Enforcement Technology (CET) instructions: https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf gas/ * config/tc-i386.c (cpu_arch): Add .cet. * doc/c-i386.texi: Document cet. * testsuite/gas/i386/cet-intel.d: New file. * testsuite/gas/i386/cet.d: Likewise. * testsuite/gas/i386/cet.s: Likewise. * testsuite/gas/i386/x86-64-cet-intel.d: Likewise. * testsuite/gas/i386/x86-64-cet.d: Likewise. * testsuite/gas/i386/x86-64-cet.s: Likewise. * testsuite/gas/i386/i386.exp: Run Intel CET tests. opcodes/ * i386-dis.c (REG_0F1E_MOD_3): New enum. (MOD_0F1E_PREFIX_1): Likewise. (MOD_0F38F5_PREFIX_2): Likewise. (MOD_0F38F6_PREFIX_0): Likewise. (RM_0F1E_MOD_3_REG_7): Likewise. (PREFIX_MOD_0_0F01_REG_5): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_1): Likewise. (PREFIX_MOD_3_0F01_REG_5_RM_2): Likewise. (PREFIX_0F1E): Likewise. (PREFIX_MOD_0_0FAE_REG_5): Likewise. (PREFIX_0F38F5): Likewise. (dis386_twobyte): Use PREFIX_0F1E. (reg_table): Add REG_0F1E_MOD_3. (prefix_table): Add PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_3_0F01_REG_5_RM_1, PREFIX_MOD_3_0F01_REG_5_RM_2, PREFIX_0F1E, PREFIX_MOD_0_0FAE_REG_5 and PREFIX_0F38F5. Update PREFIX_0FAE_REG_6 and PREFIX_0F38F6. (three_byte_table): Use PREFIX_0F38F5. (mod_table): Use PREFIX_MOD_0_0F01_REG_5, PREFIX_MOD_0_0FAE_REG_5. Add MOD_0F1E_PREFIX_1, MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0. (rm_table): Add MOD_0F38F5_PREFIX_2, MOD_0F38F6_PREFIX_0, RM_0F1E_MOD_3_REG_7. Use PREFIX_MOD_3_0F01_REG_5_RM_1 and PREFIX_MOD_3_0F01_REG_5_RM_2. * i386-gen.c (cpu_flag_init): Add CPU_CET_FLAGS. (cpu_flags): Add CpuCET. * i386-opc.h (CpuCET): New enum. (CpuUnused): Commented out. (i386_cpu_flags): Add cpucet. * i386-opc.tbl: Add Intel CET instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2017-03-07 00:26:37 +01:00
// CET instructions end.
Enable Intel WBNOINVD instruction. Intel has disclosed a set of new instructions for Icelake processor. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf This patch enables Intel WBNOINVD instruction. gas/ * config/tc-i386.c (cpu_arch): Add .wbnoinvd. * doc/c-i386.texi: Document .wbnoinvd. * testsuite/gas/i386/i386.exp: Add WBNOINVD tests. * testsuite/gas/i386/wbnoinvd-intel.d: New test. * testsuite/gas/i386/wbnoinvd.d: Likewise. * testsuite/gas/i386/wbnoinvd.s: Likewise. * testsuite/gas/i386/x86-64-wbnoinvd-intel.d: Likewise. * testsuite/gas/i386/x86-64-wbnoinvd.d: Likewise. * testsuite/gas/i386/x86-64-wbnoinvd.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F09. * i386-gen.c (cpu_flag_init): Add CPU_WBNOINVD_FLAGS. (cpu_flags): Add CpuWBNOINVD. * i386-opc.h (enum): Add CpuWBNOINVD. (i386_cpu_flags): Add cpuwbnoinvd. * i386-opc.tbl: Add WBNOINVD instruction. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-23 17:39:05 +01:00
// WBNOINVD instruction.
wbnoinvd, 0, 0xf30f09, None, 2, CpuWBNOINVD, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// WBNOINVD instruction end.
Enable Intel PCONFIG instruction. Intel has disclosed a set of new instructions for Icelake processor. The spec is https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf This patch enables Intel PCONFIG instruction. gas/ * config/tc-i386.c (cpu_arch): Add .pconfig. * doc/c-i386.texi: Document .pconfig. * testsuite/gas/i386/i386.exp: Add PCONFIG tests. * testsuite/gas/i386/pconfig-intel.d: New test. * testsuite/gas/i386/pconfig.d: Likewise. * testsuite/gas/i386/pconfig.s: Likewise. * testsuite/gas/i386/x86-64-pconfig-intel.d: Likewise. * testsuite/gas/i386/x86-64-pconfig.d: Likewise. * testsuite/gas/i386/x86-64-pconfig.s: Likewise. opcodes/ * i386-dis.c (enum): Add pconfig. * i386-gen.c (cpu_flag_init): Add CPU_PCONFIG_FLAGS. (cpu_flags): Add CpuPCONFIG. * i386-opc.h (enum): Add CpuPCONFIG. (i386_cpu_flags): Add cpupconfig. * i386-opc.tbl: Add PCONFIG instruction. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-01-23 17:56:30 +01:00
// PCONFIG instruction.
pconfig, 0, 0x0f01c5, None, 3, CpuPCONFIG, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// PCONFIG instruction end.
Enable Intel WAITPKG instructions. Intel has disclosed a set of new instructions for Tremont processor. The spec is https://software.intel.com/en-us/intel-architecture-instruction-set-extensions-programming-reference This patch enables Intel WAITPKG instructions. gas/ * config/tc-i386.c (cpu_arch): Add WAITPKG. (cpu_noarch): Likewise. * doc/c-i386.texi: Document WAITPKG. * i386/i386.exp: Run WAITPKG tests. * testsuite/gas/i386/waitpkg-intel.d: New test. * testsuite/gas/i386/waitpkg.d: Likewise. * testsuite/gas/i386/waitpkg.s: Likewise. * testsuite/gas/i386/x86-64-waitpkg-intel.d: Likewise. * testsuite/gas/i386/x86-64-waitpkg.d: Likewise. * testsuite/gas/i386/x86-64-waitpkg.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_MOD_0_0FAE_REG_6, PREFIX_MOD_1_0FAE_REG_6. (va_mode): New. (OP_E_register): Use va_mode. * i386-dis-evex.h (prefix_table): New instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add WAITPKG. (cpu_flags): Likewise. * i386-opc.h (enum): Likewise. (i386_cpu_flags): Likewise. * i386-opc.tbl: Add umonitor, umwait, tpause. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-04-09 12:58:50 +02:00
// WAITPKG instructions.
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
umonitor, 1, 0xf30fae, 0x6, 2, CpuWAITPKG, Modrm|AddrPrefixOpReg, { Reg16|Reg32|Reg64 }
x86: refine TPAUSE and UMWAIT Allowing 64-bit registers is misleading here: Elsewhere these get allowed when there's no difference between either variant, because of 32-bit destination registers having their upper halves zeroed in 64-bit mode. Here, however, they're source registers, and hence specifying 64-bit registers would lead to the ambiguity of whether the upper 32 bits actually matter. Additionally, for proper code generation in 16-bit mode, IgnoreSize is needed on both. And finally, just like for e.g. MONITOR/MWAIT, add variants with all input registers explicitly specified.
2020-03-06 08:48:48 +01:00
tpause, 1, 0x660fae, 0x6, 2, CpuWAITPKG, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32 }
tpause, 3, 0x660fae, 0x6, 2, CpuWAITPKG, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegD|Dword, Acc|Dword }
umwait, 1, 0xf20fae, 0x6, 2, CpuWAITPKG, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32 }
umwait, 3, 0xf20fae, 0x6, 2, CpuWAITPKG, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { Reg32, RegD|Dword, Acc|Dword }
Enable Intel WAITPKG instructions. Intel has disclosed a set of new instructions for Tremont processor. The spec is https://software.intel.com/en-us/intel-architecture-instruction-set-extensions-programming-reference This patch enables Intel WAITPKG instructions. gas/ * config/tc-i386.c (cpu_arch): Add WAITPKG. (cpu_noarch): Likewise. * doc/c-i386.texi: Document WAITPKG. * i386/i386.exp: Run WAITPKG tests. * testsuite/gas/i386/waitpkg-intel.d: New test. * testsuite/gas/i386/waitpkg.d: Likewise. * testsuite/gas/i386/waitpkg.s: Likewise. * testsuite/gas/i386/x86-64-waitpkg-intel.d: Likewise. * testsuite/gas/i386/x86-64-waitpkg.d: Likewise. * testsuite/gas/i386/x86-64-waitpkg.s: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_MOD_0_0FAE_REG_6, PREFIX_MOD_1_0FAE_REG_6. (va_mode): New. (OP_E_register): Use va_mode. * i386-dis-evex.h (prefix_table): New instructions (see prefixes above). * i386-gen.c (cpu_flag_init): Add WAITPKG. (cpu_flags): Likewise. * i386-opc.h (enum): Likewise. (i386_cpu_flags): Likewise. * i386-opc.tbl: Add umonitor, umwait, tpause. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2018-04-09 12:58:50 +02:00
// WAITPKG instructions end.
Enable Intel CLDEMOTE instruction. gas/ * config/tc-i386.c (cpu_arch): Add .cldemote. * doc/c-i386.texi: Document cldemote/.cldemote. * testsuite/gas/i386/cldemote-intel.d: New. * testsuite/gas/i386/cldemote.d: Likewise. * testsuite/gas/i386/cldemote.s: Likewise. * testsuite/gas/i386/i386.exp: Run new tests. * testsuite/gas/i386/x86-64-cldemote-intel.d: New. * testsuite/gas/i386/x86-64-cldemote.d: Likewise. * testsuite/gas/i386/x86-64-cldemote.s: Likewise. * testsuite/gas/i386/ilp32/x86-64-nops.d: Remove 0x0f1c NOP encoding that maps to cldemote. * testsuite/gas/i386/nops.d: Likewise. * testsuite/gas/i386/nops.s: Likewise. * testsuite/gas/i386/x86-64-nops.d: Likewise. * testsuite/gas/i386/x86-64-nops.s: Likewise. opcode/ * i386-dis.c: Add REG_0F1C_MOD_0, MOD_0F1C_PREFIX_0, PREFIX_0F1C. * i386-gen.c (cpu_flag_init): Add CPU_CLDEMOTE_FLAGS, (cpu_flags): Add CpuCLDEMOTE. * i386-init.h: Regenerate. * i386-opc.h (enum): Add CpuCLDEMOTE, (i386_cpu_flags): Add cpucldemote. * i386-opc.tbl: Add cldemote. * i386-tbl.h: Regenerate.
2018-04-16 23:09:01 +02:00
// CLDEMOTE instructions.
x86: make AnySize an insn attribute ... instead of an operand one. Which operand it applies to can be determined from other operand properties, but as it turns out the only place it is actually used at doesn't even need further qualification.
2019-11-14 08:46:19 +01:00
cldemote, 1, 0x0f1c, 0x0, 2, CpuCLDEMOTE, Modrm|Anysize|IgnoreSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { BaseIndex }
Enable Intel CLDEMOTE instruction. gas/ * config/tc-i386.c (cpu_arch): Add .cldemote. * doc/c-i386.texi: Document cldemote/.cldemote. * testsuite/gas/i386/cldemote-intel.d: New. * testsuite/gas/i386/cldemote.d: Likewise. * testsuite/gas/i386/cldemote.s: Likewise. * testsuite/gas/i386/i386.exp: Run new tests. * testsuite/gas/i386/x86-64-cldemote-intel.d: New. * testsuite/gas/i386/x86-64-cldemote.d: Likewise. * testsuite/gas/i386/x86-64-cldemote.s: Likewise. * testsuite/gas/i386/ilp32/x86-64-nops.d: Remove 0x0f1c NOP encoding that maps to cldemote. * testsuite/gas/i386/nops.d: Likewise. * testsuite/gas/i386/nops.s: Likewise. * testsuite/gas/i386/x86-64-nops.d: Likewise. * testsuite/gas/i386/x86-64-nops.s: Likewise. opcode/ * i386-dis.c: Add REG_0F1C_MOD_0, MOD_0F1C_PREFIX_0, PREFIX_0F1C. * i386-gen.c (cpu_flag_init): Add CPU_CLDEMOTE_FLAGS, (cpu_flags): Add CpuCLDEMOTE. * i386-init.h: Regenerate. * i386-opc.h (enum): Add CpuCLDEMOTE, (i386_cpu_flags): Add cpucldemote. * i386-opc.tbl: Add cldemote. * i386-tbl.h: Regenerate.
2018-04-16 23:09:01 +02:00
// CLDEMOTE instructions end.
Enable Intel MOVDIRI, MOVDIR64B instructions gas/ * config/tc-i386.c (cpu_arch): Add .movdir, .movdir64b. (cpu_noarch): Likewise. (process_suffix): Add check for register size. * doc/c-i386.texi: Document movdiri, movdir64b. * testsuite/gas/i386/i386.exp: Run MOVDIR{I,64B} tests. * testsuite/gas/i386/movdir-intel.d: New file. * testsuite/gas/i386/movdir.d: Likewise. * testsuite/gas/i386/movdir.s: Likewise. * testsuite/gas/i386/movdir64b-reg.s: Likewise. * testsuite/gas/i386/movdir64b-reg.l: Likewise. * testsuite/gas/i386/x86-64-movdir-intel.d: Likewise. * testsuite/gas/i386/x86-64-movdir.d: Likewise. * testsuite/gas/i386/x86-64-movdir.s: Likewise. * testsuite/gas/i386/x86-64-movdir64b-reg.s: Likewise. * testsuite/gas/i386/x86-64-movdir64b-reg.l: Likewise. opcodes/ * i386-dis.c (Gva): New. (enum): Add PREFIX_0F38F8, PREFIX_0F38F9, MOD_0F38F8_PREFIX_2, MOD_0F38F9_PREFIX_0. (prefix_table): New instructions (see prefix above). (mod_table): New instructions (see prefix above). (OP_G): Handle va_mode. * i386-gen.c (cpu_flag_init): Add CPU_MOVDIRI_FLAGS, CPU_MOVDIR64B_FLAGS. (cpu_flags): Add CpuMOVDIRI and CpuMOVDIR64B. * i386-opc.h (enum): Add CpuMOVDIRI, CpuMOVDIR64B. (i386_cpu_flags): Add cpumovdiri and cpumovdir64b. * i386-opc.tbl: Add movidir{i,64b}. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2018-05-07 18:30:02 +02:00
// MOVDIR[I,64B] instructions.
x86: add missing IgnoreSize For proper code generation in 16-bit mode (or to avoid the "same type of prefix used twice" diagnostic there), IgnoreSize is needed on certain templates allowing for just 32-(and maybe 64-)bit operands. Beyond adding tests for the previously broken cases, also add ones for the previously working cases where IgnoreSize is needed for the same reason (leaving out MPX for now, as that'll require an assembler change first). Some minor adjustments to tests get done such that re-use of the same code for 16-bit code generation testing becomes easier.
2020-03-06 08:49:45 +01:00
movdiri, 2, 0xf38f9, None, 3, CpuMOVDIRI, Modrm|IgnoreSize|No_bSuf|No_wSuf|No_sSuf|No_ldSuf, { Reg32|Reg64, Dword|Qword|Unspecified|BaseIndex }
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
movdir64b, 2, 0x660f38f8, None, 3, CpuMOVDIR64B, Modrm|AddrPrefixOpReg, { Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
Enable Intel MOVDIRI, MOVDIR64B instructions gas/ * config/tc-i386.c (cpu_arch): Add .movdir, .movdir64b. (cpu_noarch): Likewise. (process_suffix): Add check for register size. * doc/c-i386.texi: Document movdiri, movdir64b. * testsuite/gas/i386/i386.exp: Run MOVDIR{I,64B} tests. * testsuite/gas/i386/movdir-intel.d: New file. * testsuite/gas/i386/movdir.d: Likewise. * testsuite/gas/i386/movdir.s: Likewise. * testsuite/gas/i386/movdir64b-reg.s: Likewise. * testsuite/gas/i386/movdir64b-reg.l: Likewise. * testsuite/gas/i386/x86-64-movdir-intel.d: Likewise. * testsuite/gas/i386/x86-64-movdir.d: Likewise. * testsuite/gas/i386/x86-64-movdir.s: Likewise. * testsuite/gas/i386/x86-64-movdir64b-reg.s: Likewise. * testsuite/gas/i386/x86-64-movdir64b-reg.l: Likewise. opcodes/ * i386-dis.c (Gva): New. (enum): Add PREFIX_0F38F8, PREFIX_0F38F9, MOD_0F38F8_PREFIX_2, MOD_0F38F9_PREFIX_0. (prefix_table): New instructions (see prefix above). (mod_table): New instructions (see prefix above). (OP_G): Handle va_mode. * i386-gen.c (cpu_flag_init): Add CPU_MOVDIRI_FLAGS, CPU_MOVDIR64B_FLAGS. (cpu_flags): Add CpuMOVDIRI and CpuMOVDIR64B. * i386-opc.h (enum): Add CpuMOVDIRI, CpuMOVDIR64B. (i386_cpu_flags): Add cpumovdiri and cpumovdir64b. * i386-opc.tbl: Add movidir{i,64b}. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2018-05-07 18:30:02 +02:00
// MOVEDIR instructions end.
x86: Support Intel AVX512 BF16 Add assembler and disassembler support Intel AVX512 BF16: https://software.intel.com/en-us/download/intel-architecture-instruction-set-extensions-programming-reference gas/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_bf16. (cpu_noarch): Add noavx512_bf16. * doc/c-i386.texi: Document avx512_bf16. * testsuite/gas/i386/avx512_bf16.d: New file. * testsuite/gas/i386/avx512_bf16.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.l: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.l: Likesie. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/i386.exp: Add BF16 related tests. opcodes/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * i386-dis-evex.h (evex_table): Updated to support BF16 instructions. * i386-dis.c (enum): Add EVEX_W_0F3852_P_1, EVEX_W_0F3872_P_1 and EVEX_W_0F3872_P_3. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BF16_FLAGS. (cpu_flags): Add bitfield for CpuAVX512_BF16. * i386-opc.h (enum): Add CpuAVX512_BF16. (i386_cpu_flags): Add bitfield for cpuavx512_bf16. * i386-opc.tbl: Add AVX512 BF16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2019-04-05 20:03:01 +02:00
// AVX512_BF16 instructions.
x86: Add CheckRegSize to AVX512_BF16 instructions with Disp8ShiftVL For AVX512 instructions with Disp8ShiftVL and Broadcast, we may need to add CheckRegSize to check if broadcast matches the destination register size. gas/ PR gas/24625 * testsuite/gas/i386/inval-avx512f.s: Add tests for AVX512_BF16 instructions with invalid broadcast. * testsuite/gas/i386/x86-64-inval-avx512f.s: Likewise. * testsuite/gas/i386/inval-avx512f.l: Updated. * testsuite/gas/i386/x86-64-inval-avx512f.l: Likewise. opcodes/ PR gas/24625 * i386-opc.tbl: Add CheckRegSize to AVX512_BF16 instructions with Disp8ShiftVL. * i386-tbl.h: Regenerated.
2019-05-28 19:05:28 +02:00
vcvtne2ps2bf16, 3, 0xf272, None, 1, CpuAVX512_BF16, Modrm|VexOpcode|VexVVVV|Masking=3|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: Support Intel AVX512 BF16 Add assembler and disassembler support Intel AVX512 BF16: https://software.intel.com/en-us/download/intel-architecture-instruction-set-extensions-programming-reference gas/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_bf16. (cpu_noarch): Add noavx512_bf16. * doc/c-i386.texi: Document avx512_bf16. * testsuite/gas/i386/avx512_bf16.d: New file. * testsuite/gas/i386/avx512_bf16.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.l: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.l: Likesie. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/i386.exp: Add BF16 related tests. opcodes/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * i386-dis-evex.h (evex_table): Updated to support BF16 instructions. * i386-dis.c (enum): Add EVEX_W_0F3852_P_1, EVEX_W_0F3872_P_1 and EVEX_W_0F3872_P_3. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BF16_FLAGS. (cpu_flags): Add bitfield for CpuAVX512_BF16. * i386-opc.h (enum): Add CpuAVX512_BF16. (i386_cpu_flags): Add bitfield for cpuavx512_bf16. * i386-opc.tbl: Add AVX512 BF16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2019-04-05 20:03:01 +02:00
x86: Consolidate AVX512 BF16 entries in i386-opc.tbl 1. Use single entry for vcvtne2ps2bf16 and vdpbf16ps with Disp8ShiftVL. 2. Use 5 entries, instead of 8, for vcvtneps2bf16. * i386-opc.tbl: Consolidate AVX512 BF16 entries. * i386-init.h: Regenerated.
2019-04-08 20:06:04 +02:00
vcvtneps2bf16, 2, 0xf372, None, 1, CpuAVX512_BF16, Modrm|VexOpcode|EVex512|Masking=3|VexW0|Broadcast|Disp8MemShift=6|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegZMM|Dword|Unspecified|BaseIndex, RegYMM }
x86/Intel: improve diagnostics for ambiguous VCVT* operands Conversions which shrink element size and which have a memory source can't be disambiguated between their 128- and 256-bit variants by looking at the register operand. "operand size mismatch", however, is a pretty misleading diagnostic. Generalize the logic introduced for VFPCLASSP{S,D} such that, with suitable similar adjustments to the respective templates, it'll cover these cases too. For VCVTNEPS2BF16 also fold the two previously separate AVX512VL templates to achieve the intended effect. This is then also accompanied by a respective addition to the inval-avx512f testcase.
2020-02-17 08:56:18 +01:00
vcvtneps2bf16, 2, 0xf372, None, 1, CpuAVX512_BF16|CpuAVX512VL, Modrm|VexOpcode|Masking=3|VexW0|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|IntelSyntax, { RegXMM|RegYMM|Dword|Unspecified|BaseIndex, RegXMM }
vcvtneps2bf16, 2, 0xf372, None, 1, CpuAVX512_BF16|CpuAVX512VL, Modrm|VexOpcode|Masking=3|VexW0|Broadcast|Disp8ShiftVL|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|RegYMM|Dword|BaseIndex, RegXMM }
x86: VCVTNEPS2BF16{X,Y} should permit broadcasting Just like other VCVT*{X,Y} templates do, and to allow the programmer flexibility (might be relevant in particular when heavily macro-izing code), the two templates should also have Broadcast set, just like their X/Y-suffix-less counterparts. This in turn requires them to also have * Dword set on their memory operands, to cover the logic added to i386gen by 4a1b91eabbe7 ("x86: Expand Broadcast to 3 bits"), * RegXMM/RegYMM set on their source operands, to satisfy broadcast sizing logic in gas itself. Otherwise ATTSyntax templates wouldn't need such operand size attributes. While extending the test cases, also add Intel syntax broadcast forms without explicit size specifiers.
2020-01-21 08:25:31 +01:00
vcvtneps2bf16x, 2, 0xf372, None, 1, CpuAVX512_BF16|CpuAVX512VL, Modrm|VexOpcode|EVex128|Masking=3|VexW0|Broadcast|Disp8MemShift=4|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegXMM|Dword|Unspecified|BaseIndex, RegXMM }
vcvtneps2bf16y, 2, 0xf372, None, 1, CpuAVX512_BF16|CpuAVX512VL, Modrm|VexOpcode|EVex256|Masking=3|VexW0|Broadcast|Disp8MemShift=5|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf|ATTSyntax, { RegYMM|Dword|Unspecified|BaseIndex, RegXMM }
x86: Support Intel AVX512 BF16 Add assembler and disassembler support Intel AVX512 BF16: https://software.intel.com/en-us/download/intel-architecture-instruction-set-extensions-programming-reference gas/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_bf16. (cpu_noarch): Add noavx512_bf16. * doc/c-i386.texi: Document avx512_bf16. * testsuite/gas/i386/avx512_bf16.d: New file. * testsuite/gas/i386/avx512_bf16.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.l: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.l: Likesie. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/i386.exp: Add BF16 related tests. opcodes/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * i386-dis-evex.h (evex_table): Updated to support BF16 instructions. * i386-dis.c (enum): Add EVEX_W_0F3852_P_1, EVEX_W_0F3872_P_1 and EVEX_W_0F3872_P_3. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BF16_FLAGS. (cpu_flags): Add bitfield for CpuAVX512_BF16. * i386-opc.h (enum): Add CpuAVX512_BF16. (i386_cpu_flags): Add bitfield for cpuavx512_bf16. * i386-opc.tbl: Add AVX512 BF16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2019-04-05 20:03:01 +02:00
x86: Add CheckRegSize to AVX512_BF16 instructions with Disp8ShiftVL For AVX512 instructions with Disp8ShiftVL and Broadcast, we may need to add CheckRegSize to check if broadcast matches the destination register size. gas/ PR gas/24625 * testsuite/gas/i386/inval-avx512f.s: Add tests for AVX512_BF16 instructions with invalid broadcast. * testsuite/gas/i386/x86-64-inval-avx512f.s: Likewise. * testsuite/gas/i386/inval-avx512f.l: Updated. * testsuite/gas/i386/x86-64-inval-avx512f.l: Likewise. opcodes/ PR gas/24625 * i386-opc.tbl: Add CheckRegSize to AVX512_BF16 instructions with Disp8ShiftVL. * i386-tbl.h: Regenerated.
2019-05-28 19:05:28 +02:00
vdpbf16ps, 3, 0xf352, None, 1, CpuAVX512_BF16, Modrm|VexOpcode|VexVVVV|Masking=3|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegXMM|RegYMM|RegZMM }
x86: Support Intel AVX512 BF16 Add assembler and disassembler support Intel AVX512 BF16: https://software.intel.com/en-us/download/intel-architecture-instruction-set-extensions-programming-reference gas/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_bf16. (cpu_noarch): Add noavx512_bf16. * doc/c-i386.texi: Document avx512_bf16. * testsuite/gas/i386/avx512_bf16.d: New file. * testsuite/gas/i386/avx512_bf16.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.l: Likewise. * testsuite/gas/i386/avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.l: Likesie. * testsuite/gas/i386/x86-64-avx512_bf16_vl-inval.s: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.d: Likewise. * testsuite/gas/i386/x86-64-avx512_bf16_vl.s: Likewise. * testsuite/gas/i386/i386.exp: Add BF16 related tests. opcodes/ 2019-04-05 Xuepeng Guo <xuepeng.guo@intel.com> * i386-dis-evex.h (evex_table): Updated to support BF16 instructions. * i386-dis.c (enum): Add EVEX_W_0F3852_P_1, EVEX_W_0F3872_P_1 and EVEX_W_0F3872_P_3. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_BF16_FLAGS. (cpu_flags): Add bitfield for CpuAVX512_BF16. * i386-opc.h (enum): Add CpuAVX512_BF16. (i386_cpu_flags): Add bitfield for cpuavx512_bf16. * i386-opc.tbl: Add AVX512 BF16 instructions. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2019-04-05 20:03:01 +02:00
// AVX512_BF16 instructions end.
Add support for Intel ENQCMD[S] instructions This patch enables support for ENQCMD[S] in binutils. Please refer to https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf for ENQCMD[S] details. Make check-gas is ok. gas/ChangeLog: 2019-06-04 Xuepeng Guo <xuepeng.guo@intel.com> Lili Cui <lili.cui@intel.com> * doc/c-i386.texi: Document enqcmd. * testsuite/gas/i386/enqcmd-intel.d: New file. * testsuite/gas/i386/enqcmd-inval.l: Likewise. * testsuite/gas/i386/enqcmd-inval.s: Likewise. * testsuite/gas/i386/enqcmd.d: Likewise. * testsuite/gas/i386/enqcmd.s: Likewise. * testsuite/gas/i386/x86-64-enqcmd-intel.d: Likewise. * testsuite/gas/i386/x86-64-enqcmd-inval.l: Likewise. * testsuite/gas/i386/x86-64-enqcmd-inval.s: Likewise. * testsuite/gas/i386/x86-64-enqcmd.d: Likewise. * testsuite/gas/i386/x86-64-enqcmd.s: Likewise. * testsuite/gas/i386/i386.exp: Run enqcmd-intel, enqcmd-inval, enqcmd, x86-64-enqcmd-intel, x86-64-enqcmd-inval, and x86-64-enqcmd. opcodes/ChangeLog: 2019-06-04 Xuepeng Guo <xuepeng.guo@intel.com> Lili Cui <lili.cui@intel.com> * i386-dis.c (enum): Add MOD_0F38F8_PREFIX_1 and MOD_0F38F8_PREFIX_3. (prefix_table): New instructions (see prefix above). (mod_table): New instructions (see prefix above). * i386-gen.c (cpu_flag_init): Add entries for enqcmd. (cpu_flags): Add a bitfield for enqmcd. * i386-init.h: Regenerated. * i386-opc.h (enum): Add CpuENQCMD. (i386_cpu_flags): Add a bitfield for cpuenqcmd. * i386-opc.tbl: Add enqcmd and enqcmds instructions. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated.
2019-06-04 17:50:10 +02:00
// ENQCMD instructions.
x86: fold AddrPrefixOpReg templates There's no need to have separate Cpu64 and CpuNo64 templates: There already is special logic handling the attribute, and all that's needed is rejecting 16-bit address registers in 64-bit mode. Suppress suffix guessing and group all involved logic together, outside of suffix processing (arguably it doesn't even belong in process_suffix()). Also, since no AddrPrefixOpReg template permits any suffixes, move the No_*Suf specifiers for them to a central place. Along with this drop the no longer relevant NoRex64 from there.
2020-02-17 08:59:07 +01:00
enqcmd, 2, 0xf20f38f8, None, 3, CpuENQCMD, Modrm|AddrPrefixOpReg, { Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
enqcmds, 2, 0xf30f38f8, None, 3, CpuENQCMD, Modrm|AddrPrefixOpReg, { Unspecified|BaseIndex, Reg16|Reg32|Reg64 }
Add support for Intel ENQCMD[S] instructions This patch enables support for ENQCMD[S] in binutils. Please refer to https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf for ENQCMD[S] details. Make check-gas is ok. gas/ChangeLog: 2019-06-04 Xuepeng Guo <xuepeng.guo@intel.com> Lili Cui <lili.cui@intel.com> * doc/c-i386.texi: Document enqcmd. * testsuite/gas/i386/enqcmd-intel.d: New file. * testsuite/gas/i386/enqcmd-inval.l: Likewise. * testsuite/gas/i386/enqcmd-inval.s: Likewise. * testsuite/gas/i386/enqcmd.d: Likewise. * testsuite/gas/i386/enqcmd.s: Likewise. * testsuite/gas/i386/x86-64-enqcmd-intel.d: Likewise. * testsuite/gas/i386/x86-64-enqcmd-inval.l: Likewise. * testsuite/gas/i386/x86-64-enqcmd-inval.s: Likewise. * testsuite/gas/i386/x86-64-enqcmd.d: Likewise. * testsuite/gas/i386/x86-64-enqcmd.s: Likewise. * testsuite/gas/i386/i386.exp: Run enqcmd-intel, enqcmd-inval, enqcmd, x86-64-enqcmd-intel, x86-64-enqcmd-inval, and x86-64-enqcmd. opcodes/ChangeLog: 2019-06-04 Xuepeng Guo <xuepeng.guo@intel.com> Lili Cui <lili.cui@intel.com> * i386-dis.c (enum): Add MOD_0F38F8_PREFIX_1 and MOD_0F38F8_PREFIX_3. (prefix_table): New instructions (see prefix above). (mod_table): New instructions (see prefix above). * i386-gen.c (cpu_flag_init): Add entries for enqcmd. (cpu_flags): Add a bitfield for enqmcd. * i386-init.h: Regenerated. * i386-opc.h (enum): Add CpuENQCMD. (i386_cpu_flags): Add a bitfield for cpuenqcmd. * i386-opc.tbl: Add enqcmd and enqcmds instructions. * i386-init.h: Regenerated. * i386-tbl.h: Regenerated.
2019-06-04 17:50:10 +02:00
// ENQCMD instructions end.
Enable Intel AVX512_VP2INTERSECT insn This patch enables support for VP2INTERSECT in binutils. Please refer to https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf for VP2INTERSECT details. Make check-gas is ok. gas/ 2019-06-04 Igor Tsimbalist <igor.v.tsimbalist@intel.com> Lili Cui <lili.cui@intel.com> * config/tc-i386.c (cpu_arch): Add .avx512_vp2intersect. (cpu_noarch): Likewise. * doc/c-i386.texi: Document avx512_vp2intersect. * testsuite/gas/i386/i386.exp: Run vp2intersect tests. * testsuite/gas/i386/vp2intersect-intel.d: New test. * testsuite/gas/i386/vp2intersect.d: Likewise. * testsuite/gas/i386/vp2intersect.s: Likewise. * testsuite/gas/i386/vp2intersect-inval-bcast.l: Likewise. * testsuite/gas/i386/vp2intersect-inval-bcast.s: Likewise. * testsuite/gas/i386/x86-64-vp2intersect-intel.d: Likewise. * testsuite/gas/i386/x86-64-vp2intersect.d: Likewise. * testsuite/gas/i386/x86-64-vp2intersect.s: Likewise. * testsuite/gas/i386/x86-64-vp2intersect-inval-bcast.l: Likewise. * testsuite/gas/i386/x86-64-vp2intersect-inval-bcast.s: Likewise. opcodes/ 2019-06-04 Igor Tsimbalist <igor.v.tsimbalist@intel.com> Lili Cui <lili.cui@intel.com> * i386-dis.c (enum): Add PREFIX_EVEX_0F3868, EVEX_W_0F3868_P_3. * i386-dis-evex.h (evex_table): Add AVX512_VP2INTERSECT instructions. * i386-gen.c (cpu_flag_init): Add CPU_AVX512_VP2INTERSECT_FLAGS, CPU_ANY_AVX512_VP2INTERSECT_FLAGS. (cpu_flags): Add CpuAVX512_VP2INTERSECT. * i386-opc.h (enum): Add CpuAVX512_VP2INTERSECT. (i386_cpu_flags): Add cpuavx512_vp2intersect. * i386-opc.tbl: Add AVX512_VP2INTERSECT insns. * i386-init.h: Regenerated. * i386-tbl.h: Likewise.
2019-06-04 17:58:21 +02:00
// VP2INTERSECT instructions.
vp2intersectd, 3, 0xf268, None, 1, CpuAVX512_VP2INTERSECT, Modrm|VexOpcode|VexVVVV|VexW0|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Dword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
vp2intersectq, 3, 0xf268, None, 1, CpuAVX512_VP2INTERSECT, Modrm|VexOpcode|VexVVVV|VexW1|Broadcast|Disp8ShiftVL|CheckRegSize|No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { RegXMM|RegYMM|RegZMM|Qword|Unspecified|BaseIndex, RegXMM|RegYMM|RegZMM, RegMask }
// VP2INTERSECT instructions end.
x86: support further AMD Zen2 instructions Both RDPRU and MCOMMIT have been publicly documented meanwhile: https://www.amd.com/system/files/TechDocs/24594.pdf.
2019-11-07 09:29:14 +01:00
// MCOMMIT instruction
mcommit, 0, 0xf30f01fa, None, 3, CpuMCOMMIT, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// MCOMMIT instruction end
// RDPRU instruction
rdpru, 0, 0x0f01fd, None, 3, CpuRDPRU, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// RDPRU instruction end
Add support for intel SERIALIZE instruction gas/ * config/tc-i386.c (cpu_arch): Add .serialize. (cpu_noarch): Likewise. * doc/c-i386.texi: Document serialize. * testsuite/gas/i386/i386.exp: Run serialize tests * testsuite/gas/i386/serialize.d: Likewise. * testsuite/gas/i386/x86-64-serialize.d: Likewise. * testsuite/gas/i386/serialize.s: Likewise. opcodes/ * i386-dis.c (prefix_table): New instructions serialize. * i386-gen.c (cpu_flag_init): Add CPU_SERIALIZE_FLAGS, CPU_ANY_SERIALIZE_FLAGS. (cpu_flags): Add CpuSERIALIZE. * i386-opc.h (enum): Add CpuSERIALIZE. (i386_cpu_flags): Add cpuserialize. * i386-opc.tbl: Add SERIALIZE insns. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2020-03-06 07:38:15 +01:00
// SERIALIZE instruction.
serialize, 0, 0x0f01e8, None, 3, CpuSERIALIZE, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// SERIALIZE instruction end.
Add support for intel TSXLDTRK instructions$ gas/ * config/tc-i386.c (cpu_arch): Add .TSXLDTRK. (cpu_noarch): Likewise. * doc/c-i386.texi: Document TSXLDTRK. * testsuite/gas/i386/i386.exp: Run TSXLDTRK tests. * testsuite/gas/i386/tsxldtrk.d: Likewise. * testsuite/gas/i386/tsxldtrk.s: Likewise. * testsuite/gas/i386/x86-64-tsxldtrk.d: Likewise. opcodes/ * i386-dis.c (enum): Add PREFIX_0F01_REG_5_MOD_3_RM_1, (prefix_table): New instructions (see prefixes above). (rm_table): Likewise. * i386-gen.c (cpu_flag_init): Add CPU_TSXLDTRK_FLAGS, CPU_ANY_TSXLDTRK_FLAGS. (cpu_flags): Add CpuTSXLDTRK. * i386-opc.h (enum): Add CpuTSXLDTRK. (i386_cpu_flags): Add cputsxldtrk. * i386-opc.tbl: Add XSUSPLDTRK insns. * i386-init.h: Regenerate. * i386-tbl.h: Likewise.
2020-04-07 07:59:50 +02:00
// TSXLDTRK instructions.
xsuspldtrk, 0, 0xf20f01e8, None, 3, CpuTSXLDTRK, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
xresldtrk, 0, 0xf20f01e9, None, 3, CpuTSXLDTRK, No_bSuf|No_wSuf|No_lSuf|No_sSuf|No_qSuf|No_ldSuf, { 0 }
// TSXLDTRK instructions end.