OpenE2K
/
gcc
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

This patch adds new builtins to check for cpu type and features. 

2012-04-24  Sriraman Tallam  <tmsriram@google.com>

This patch adds new builtins to check for cpu type and features. 

* __builtin_cpu_is ("<CPUNAME>")
* __builtin_cpu_supports ("<FEATURE>")

apart from the cpu init builtin, __builtin_cpu_init.

List of CPU names :

* "amd"
* "intel"
* "atom"
* "core2"
* "corei7"
* "nehalem"
* "westmere"
* "sandybridge"
* "amdfam10h"
* "barcelona"
* "shanghai"
* "istanbul"
* "bdver1"
* "bdver2"

List of CPU features :

* "cmov"
* "mmx"
* "popcnt"
* "sse"
* "sse2"
* "sse3"
* "ssse3"
* "sse4.1"
* "sse4.2"
* "avx"

        * config/i386/i386.c (build_processor_model_struct): New function.
        (make_var_decl): New function.
        (fold_builtin_cpu): New function.
        (ix86_fold_builtin): New function.
        (make_cpu_type_builtin): New function.
        (ix86_init_platform_type_builtins): New function.
        (ix86_expand_builtin): Expand new builtins by folding them.
        (ix86_init_builtins): Make new builtins to detect CPU type.
        (TARGET_FOLD_BUILTIN): New macro.
        (IX86_BUILTIN_CPU_INIT): New enum value.
        (IX86_BUILTIN_CPU_IS): New enum value.
        (IX86_BUILTIN_CPU_SUPPORTS): New enum value.
        * config/i386/i386-builtin-types.def: New function type.
        * testsuite/gcc.target/builtin_target.c: New testcase.
        * doc/extend.texi: Document builtins.

        * libgcc/config/i386/i386-cpuinfo.c: New file.
        * libgcc/config/i386/t-cpuinfo: New file.
        * libgcc/config.host: Include t-cpuinfo.
        * libgcc/config/i386/libgcc-glibc.ver: Version symbol __cpu_model.

From-SVN: r186789
This commit is contained in:
Sriraman Tallam 2012-04-25 00:08:37 +00:00 committed by Sriraman Tallam
parent e1be98a467
commit 792317cc77
10 changed files with 921 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
gcc/ChangeLog
View File

@ -1,3 +1,21 @@
2012-04-24 Sriraman Tallam <tmsriram@google.com>
* config/i386/i386.c (build_processor_model_struct): New function.
(make_var_decl): New function.
(fold_builtin_cpu): New function.
(ix86_fold_builtin): New function.
(make_cpu_type_builtin): New function.
(ix86_init_platform_type_builtins): New function.
(ix86_expand_builtin): Expand new builtins by folding them.
(ix86_init_builtins): Make new builtins to detect CPU type.
(TARGET_FOLD_BUILTIN): New macro.
(IX86_BUILTIN_CPU_INIT): New enum value.
(IX86_BUILTIN_CPU_IS): New enum value.
(IX86_BUILTIN_CPU_SUPPORTS): New enum value.
* config/i386/i386-builtin-types.def: New function type.
* testsuite/gcc.target/builtin_target.c: New testcase.
* doc/extend.texi: Document builtins.
2012-04-24 Olivier Hainque <hainque@adacore.com>
* common.opt (gdwarf-): Initialize dwarf_version to -1 instead of 2.

1
gcc/config/i386/i386-builtin-types.def
View File

@ -155,6 +155,7 @@ DEF_FUNCTION_TYPE (INT, V4SF)
DEF_FUNCTION_TYPE (INT, V8QI)
DEF_FUNCTION_TYPE (INT, V8SF)
DEF_FUNCTION_TYPE (INT, V32QI)
DEF_FUNCTION_TYPE (INT, PCCHAR)
DEF_FUNCTION_TYPE (INT64, INT64)
DEF_FUNCTION_TYPE (INT64, V2DF)
DEF_FUNCTION_TYPE (INT64, V4SF)

361
gcc/config/i386/i386.c
View File

@ -25855,6 +25855,11 @@ enum ix86_builtins
/* CFString built-in for darwin */
IX86_BUILTIN_CFSTRING,
/* Builtins to get CPU type and supported features. */
IX86_BUILTIN_CPU_INIT,
IX86_BUILTIN_CPU_IS,
IX86_BUILTIN_CPU_SUPPORTS,
IX86_BUILTIN_MAX
};
@ -27673,6 +27678,334 @@ ix86_init_mmx_sse_builtins (void)
}
}
/* This builds the processor_model struct type defined in
libgcc/config/i386/i386-cpuinfo.c */
static tree
build_processor_model_struct (void)
{
const char *field_name[] = {"__cpu_vendor", "__cpu_type", "__cpu_subtype",
"__cpu_features"};
tree field = NULL_TREE, field_chain = NULL_TREE;
int i;
tree type = make_node (RECORD_TYPE);
/* The first 3 fields are unsigned int. */
for (i = 0; i < 3; ++i)
{
field = build_decl (UNKNOWN_LOCATION, FIELD_DECL,
get_identifier (field_name[i]), unsigned_type_node);
if (field_chain != NULL_TREE)
DECL_CHAIN (field) = field_chain;
field_chain = field;
}
/* The last field is an array of unsigned integers of size one. */
field = build_decl (UNKNOWN_LOCATION, FIELD_DECL,
get_identifier (field_name[3]),
build_array_type (unsigned_type_node,
build_index_type (size_one_node)));
if (field_chain != NULL_TREE)
DECL_CHAIN (field) = field_chain;
field_chain = field;
finish_builtin_struct (type, "__processor_model", field_chain, NULL_TREE);
return type;
}
/* Returns a extern, comdat VAR_DECL of type TYPE and name NAME. */
static tree
make_var_decl (tree type, const char *name)
{
tree new_decl;
new_decl = build_decl (UNKNOWN_LOCATION,
VAR_DECL,
get_identifier(name),
type);
DECL_EXTERNAL (new_decl) = 1;
TREE_STATIC (new_decl) = 1;
TREE_PUBLIC (new_decl) = 1;
DECL_INITIAL (new_decl) = 0;
DECL_ARTIFICIAL (new_decl) = 0;
DECL_PRESERVE_P (new_decl) = 1;
make_decl_one_only (new_decl, DECL_ASSEMBLER_NAME (new_decl));
assemble_variable (new_decl, 0, 0, 0);
return new_decl;
}
/* FNDECL is a __builtin_cpu_is or a __builtin_cpu_supports call that is folded
into an integer defined in libgcc/config/i386/i386-cpuinfo.c */
static tree
fold_builtin_cpu (tree fndecl, tree *args)
{
unsigned int i;
enum ix86_builtins fn_code = (enum ix86_builtins)
DECL_FUNCTION_CODE (fndecl);
tree param_string_cst = NULL;
/* This is the order of bit-fields in __processor_features in
i386-cpuinfo.c */
enum processor_features
{
F_CMOV = 0,
F_MMX,
F_POPCNT,
F_SSE,
F_SSE2,
F_SSE3,
F_SSSE3,
F_SSE4_1,
F_SSE4_2,
F_AVX,
F_MAX
};
/* These are the values for vendor types and cpu types and subtypes
in i386-cpuinfo.c. Cpu types and subtypes should be subtracted by
the corresponding start value. */
enum processor_model
{
M_INTEL = 1,
M_AMD,
M_CPU_TYPE_START,
M_INTEL_ATOM,
M_INTEL_CORE2,
M_INTEL_COREI7,
M_AMDFAM10H,
M_AMDFAM15H,
M_CPU_SUBTYPE_START,
M_INTEL_COREI7_NEHALEM,
M_INTEL_COREI7_WESTMERE,
M_INTEL_COREI7_SANDYBRIDGE,
M_AMDFAM10H_BARCELONA,
M_AMDFAM10H_SHANGHAI,
M_AMDFAM10H_ISTANBUL,
M_AMDFAM15H_BDVER1,
M_AMDFAM15H_BDVER2
};
static struct _arch_names_table
{
const char *const name;
const enum processor_model model;
}
const arch_names_table[] =
{
{"amd", M_AMD},
{"intel", M_INTEL},
{"atom", M_INTEL_ATOM},
{"core2", M_INTEL_CORE2},
{"corei7", M_INTEL_COREI7},
{"nehalem", M_INTEL_COREI7_NEHALEM},
{"westmere", M_INTEL_COREI7_WESTMERE},
{"sandybridge", M_INTEL_COREI7_SANDYBRIDGE},
{"amdfam10h", M_AMDFAM10H},
{"barcelona", M_AMDFAM10H_BARCELONA},
{"shanghai", M_AMDFAM10H_SHANGHAI},
{"istanbul", M_AMDFAM10H_ISTANBUL},
{"amdfam15h", M_AMDFAM15H},
{"bdver1", M_AMDFAM15H_BDVER1},
{"bdver2", M_AMDFAM15H_BDVER2},
};
static struct _isa_names_table
{
const char *const name;
const enum processor_features feature;
}
const isa_names_table[] =
{
{"cmov", F_CMOV},
{"mmx", F_MMX},
{"popcnt", F_POPCNT},
{"sse", F_SSE},
{"sse2", F_SSE2},
{"sse3", F_SSE3},
{"ssse3", F_SSSE3},
{"sse4.1", F_SSE4_1},
{"sse4.2", F_SSE4_2},
{"avx", F_AVX}
};
static tree __processor_model_type = NULL_TREE;
static tree __cpu_model_var = NULL_TREE;
if (__processor_model_type == NULL_TREE)
__processor_model_type = build_processor_model_struct ();
if (__cpu_model_var == NULL_TREE)
__cpu_model_var = make_var_decl (__processor_model_type,
"__cpu_model");
gcc_assert ((args != NULL) && (*args != NULL));
param_string_cst = *args;
while (param_string_cst
&& TREE_CODE (param_string_cst) != STRING_CST)
{
/* *args must be a expr that can contain other EXPRS leading to a
STRING_CST. */
if (!EXPR_P (param_string_cst))
{
error ("Parameter to builtin must be a string constant or literal");
return integer_zero_node;
}
param_string_cst = TREE_OPERAND (EXPR_CHECK (param_string_cst), 0);
}
gcc_assert (param_string_cst);
if (fn_code == IX86_BUILTIN_CPU_IS)
{
tree ref;
tree field;
unsigned int field_val = 0;
unsigned int NUM_ARCH_NAMES
= sizeof (arch_names_table) / sizeof (struct _arch_names_table);
for (i = 0; i < NUM_ARCH_NAMES; i++)
if (strcmp (arch_names_table[i].name,
TREE_STRING_POINTER (param_string_cst)) == 0)
break;
if (i == NUM_ARCH_NAMES)
{
error ("Parameter to builtin not valid: %s",
TREE_STRING_POINTER (param_string_cst));
return integer_zero_node;
}
field = TYPE_FIELDS (__processor_model_type);
field_val = arch_names_table[i].model;
/* CPU types are stored in the next field. */
if (field_val > M_CPU_TYPE_START
&& field_val < M_CPU_SUBTYPE_START)
{
field = DECL_CHAIN (field);
field_val -= M_CPU_TYPE_START;
}
/* CPU subtypes are stored in the next field. */
if (field_val > M_CPU_SUBTYPE_START)
{
field = DECL_CHAIN ( DECL_CHAIN (field));
field_val -= M_CPU_SUBTYPE_START;
}
/* Get the appropriate field in __cpu_model. */
ref = build3 (COMPONENT_REF, TREE_TYPE (field), __cpu_model_var,
field, NULL_TREE);
/* Check the value. */
return build2 (EQ_EXPR, unsigned_type_node, ref,
build_int_cstu (unsigned_type_node, field_val));
}
else if (fn_code == IX86_BUILTIN_CPU_SUPPORTS)
{
tree ref;
tree array_elt;
tree field;
unsigned int field_val = 0;
unsigned int NUM_ISA_NAMES
= sizeof (isa_names_table) / sizeof (struct _isa_names_table);
for (i = 0; i < NUM_ISA_NAMES; i++)
if (strcmp (isa_names_table[i].name,
TREE_STRING_POINTER (param_string_cst)) == 0)
break;
if (i == NUM_ISA_NAMES)
{
error ("Parameter to builtin not valid: %s",
TREE_STRING_POINTER (param_string_cst));
return integer_zero_node;
}
field = TYPE_FIELDS (__processor_model_type);
/* Get the last field, which is __cpu_features. */
while (DECL_CHAIN (field))
field = DECL_CHAIN (field);
/* Get the appropriate field: __cpu_model.__cpu_features */
ref = build3 (COMPONENT_REF, TREE_TYPE (field), __cpu_model_var,
field, NULL_TREE);
/* Access the 0th element of __cpu_features array. */
array_elt = build4 (ARRAY_REF, unsigned_type_node, ref,
integer_zero_node, NULL_TREE, NULL_TREE);
field_val = (1 << isa_names_table[i].feature);
/* Return __cpu_model.__cpu_features[0] & field_val */
return build2 (BIT_AND_EXPR, unsigned_type_node, array_elt,
build_int_cstu (unsigned_type_node, field_val));
}
gcc_unreachable ();
}
static tree
ix86_fold_builtin (tree fndecl, int n_args,
tree *args, bool ignore ATTRIBUTE_UNUSED)
{
if (DECL_BUILT_IN_CLASS (fndecl) == BUILT_IN_MD)
{
enum ix86_builtins fn_code = (enum ix86_builtins)
DECL_FUNCTION_CODE (fndecl);
if (fn_code == IX86_BUILTIN_CPU_IS
|| fn_code == IX86_BUILTIN_CPU_SUPPORTS)
{
gcc_assert (n_args == 1);
return fold_builtin_cpu (fndecl, args);
}
}
return NULL_TREE;
}
/* Make builtins to detect cpu type and features supported. NAME is
the builtin name, CODE is the builtin code, and FTYPE is the function
type of the builtin. */
static void
make_cpu_type_builtin (const char* name, int code,
enum ix86_builtin_func_type ftype, bool is_const)
{
tree decl;
tree type;
type = ix86_get_builtin_func_type (ftype);
decl = add_builtin_function (name, type, code, BUILT_IN_MD,
NULL, NULL_TREE);
gcc_assert (decl != NULL_TREE);
ix86_builtins[(int) code] = decl;
TREE_READONLY (decl) = is_const;
}
/* Make builtins to get CPU type and features supported. The created
builtins are :
__builtin_cpu_init (), to detect cpu type and features,
__builtin_cpu_is ("<CPUNAME>"), to check if cpu is of type <CPUNAME>,
__builtin_cpu_supports ("<FEATURE>"), to check if cpu supports <FEATURE>
*/
static void
ix86_init_platform_type_builtins (void)
{
make_cpu_type_builtin ("__builtin_cpu_init", IX86_BUILTIN_CPU_INIT,
INT_FTYPE_VOID, false);
make_cpu_type_builtin ("__builtin_cpu_is", IX86_BUILTIN_CPU_IS,
INT_FTYPE_PCCHAR, true);
make_cpu_type_builtin ("__builtin_cpu_supports", IX86_BUILTIN_CPU_SUPPORTS,
INT_FTYPE_PCCHAR, true);
}
/* Internal method for ix86_init_builtins. */
static void
@ -27756,6 +28089,9 @@ ix86_init_builtins (void)
ix86_init_builtin_types ();
/* Builtins to get CPU type and features. */
ix86_init_platform_type_builtins ();
/* TFmode support builtins. */
def_builtin_const (0, "__builtin_infq",
FLOAT128_FTYPE_VOID, IX86_BUILTIN_INFQ);
@ -29374,6 +29710,28 @@ ix86_expand_builtin (tree exp, rtx target, rtx subtarget ATTRIBUTE_UNUSED,
enum machine_mode mode0, mode1, mode2, mode3, mode4;
unsigned int fcode = DECL_FUNCTION_CODE (fndecl);
/* For CPU builtins that can be folded, fold first and expand the fold. */
switch (fcode)
{
case IX86_BUILTIN_CPU_INIT:
{
/* Make it call __cpu_indicator_init in libgcc. */
tree call_expr, fndecl, type;
type = build_function_type_list (integer_type_node, NULL_TREE);
fndecl = build_fn_decl ("__cpu_indicator_init", type);
call_expr = build_call_expr (fndecl, 0);
return expand_expr (call_expr, target, mode, EXPAND_NORMAL);
}
case IX86_BUILTIN_CPU_IS:
case IX86_BUILTIN_CPU_SUPPORTS:
{
tree arg0 = CALL_EXPR_ARG (exp, 0);
tree fold_expr = fold_builtin_cpu (fndecl, &arg0);
gcc_assert (fold_expr != NULL_TREE);
return expand_expr (fold_expr, target, mode, EXPAND_NORMAL);
}
}
/* Determine whether the builtin function is available under the current ISA.
Originally the builtin was not created if it wasn't applicable to the
current ISA based on the command line switches. With function specific
@ -39100,6 +39458,9 @@ ix86_autovectorize_vector_sizes (void)
#undef TARGET_BUILD_BUILTIN_VA_LIST
#define TARGET_BUILD_BUILTIN_VA_LIST ix86_build_builtin_va_list
#undef TARGET_FOLD_BUILTIN
#define TARGET_FOLD_BUILTIN ix86_fold_builtin
#undef TARGET_ENUM_VA_LIST_P
#define TARGET_ENUM_VA_LIST_P ix86_enum_va_list

138
gcc/doc/extend.texi
View File

@ -9419,6 +9419,144 @@ Similar to @code{__builtin_huge_val}, except the return type is @code{__float128
@findex __builtin_huge_valq
@end table
The following built-in functions are always available and can be used to
check the target platform type.
@deftypefn {Built-in Function} void __builtin_cpu_init (void)
This function runs the CPU detection code to check the type of CPU and the
features supported. This builtin needs to be invoked along with the builtins
to check CPU type and features, @code{__builtin_cpu_is} and
@code{__builtin_cpu_supports}, only when used in a function that will be
executed before any constructors are called. The CPU detection code is
automatically executed in a very high priority constructor.
For example, this function has to be used in @code{ifunc} resolvers which
check for CPU type using the builtins, @code{__builtin_cpu_is}
and @code{__builtin_cpu_supports}.
@smallexample
static void (*resolve_memcpy (void)) (void)
@{
// ifunc resolvers fire before constructors, explicitly call the init
// function.
__builtin_cpu_init ();
if (__builtin_cpu_supports ("ssse3"))
return ssse3_memcpy; // super fast memcpy with ssse3 instructions.
else
return default_memcpy;
@}
void *memcpy (void *, const void *, size_t)
__attribute__ ((ifunc ("resolve_memcpy")));
@end smallexample
@end deftypefn
@deftypefn {Built-in Function} int __builtin_cpu_is (const char *@var{cpuname})
This function returns a positive integer if the runtime cpu is of type @var{cpuname}
and returns @code{0} otherwise. The following cpu names can be detected:
@table @samp
@item intel
Intel CPU.
@item atom
Intel ATOM CPU.
@item core2
Intel Core2 CPU.
@item corei7
Intel Corei7 CPU.
@item nehalem
Intel Corei7 Nehalem CPU.
@item westmere
Intel Corei7 Westmere CPU.
@item sandybridge
Intel Corei7 Sandybridge CPU.
@item amd
AMD CPU.
@item amdfam10h
AMD family 10h CPU.
@item barcelona
AMD family 10h Barcelona CPU.
@item shanghai
AMD family 10h Shanghai CPU.
@item istanbul
AMD family 10h Istanbul CPU.
@item amdfam15h
AMD family 15h CPU.
@item bdver1
AMD family 15h Bulldozer version 1.
@item bdver2
AMD family 15h Bulldozer version 2.
@end table
Here is an example:
@smallexample
if (__builtin_cpu_is ("corei7"))
@{
do_corei7 (); //Corei7 specific implementation.
@}
else
@{
do_generic (); //Generic implementation.
@}
@end smallexample
@end deftypefn
@deftypefn {Built-in Function} int __builtin_cpu_supports (const char *@var{feature})
This function returns a postive integer if the runtime cpu supports @var{feature}
and returns @code{0} otherwise. The following features can be detected:
@table @samp
@item cmov
CMOV instruction.
@item mmx
MMX instructions.
@item popcnt
POPCNT instruction.
@item sse
SSE instructions.
@item sse2
SSE2 instructions.
@item sse3
SSE3 instructions.
@item ssse3
SSSE3 instructions.
@item sse4.1
SSE4.1 instructions.
@item sse4.2
SSE4.2 instructions.
@item avx
AVX instructions.
@end table
Here is an example:
@smallexample
if (__builtin_cpu_supports ("popcnt"))
@{
asm("popcnt %1,%0" : "=r"(count) : "rm"(n) : "cc");
@}
else
@{
count = generic_countbits (n); //generic implementation.
@}
@end smallexample
@end deftypefn
The following built-in functions are made available by @option{-mmmx}.
All of them generate the machine instruction that is part of the name.

70
gcc/testsuite/gcc.target/i386/builtin_target.c Normal file
View File

@ -0,0 +1,70 @@
/* This test checks if the __builtin_cpu_is and __builtin_cpu_supports calls
are recognized. */
/* { dg-do run } */
#include <assert.h>
int
fn1 ()
{
/* Check CPU Features. */
assert (__builtin_cpu_supports ("cmov") >= 0);
assert (__builtin_cpu_supports ("mmx") >= 0);
assert (__builtin_cpu_supports ("popcnt") >= 0);
assert (__builtin_cpu_supports ("sse") >= 0);
assert (__builtin_cpu_supports ("sse2") >= 0);
assert (__builtin_cpu_supports ("sse3") >= 0);
assert (__builtin_cpu_supports ("ssse3") >= 0);
assert (__builtin_cpu_supports ("sse4.1") >= 0);
assert (__builtin_cpu_supports ("sse4.2") >= 0);
assert (__builtin_cpu_supports ("avx") >= 0);
/* Check CPU type. */
assert (__builtin_cpu_is ("amd") >= 0);
assert (__builtin_cpu_is ("intel") >= 0);
assert (__builtin_cpu_is ("atom") >= 0);
assert (__builtin_cpu_is ("core2") >= 0);
assert (__builtin_cpu_is ("corei7") >= 0);
assert (__builtin_cpu_is ("nehalem") >= 0);
assert (__builtin_cpu_is ("westmere") >= 0);
assert (__builtin_cpu_is ("sandybridge") >= 0);
assert (__builtin_cpu_is ("amdfam10h") >= 0);
assert (__builtin_cpu_is ("barcelona") >= 0);
assert (__builtin_cpu_is ("shanghai") >= 0);
assert (__builtin_cpu_is ("istanbul") >= 0);
assert (__builtin_cpu_is ("amdfam15h") >= 0);
assert (__builtin_cpu_is ("bdver1") >= 0);
assert (__builtin_cpu_is ("bdver2") >= 0);
return 0;
}
int main ()
{
__builtin_cpu_init ();
return fn1 ();
}

7
libgcc/ChangeLog
View File

@ -1,3 +1,10 @@
2012-04-24 Sriraman Tallam <tmsriram@google.com>
* libgcc/config/i386/i386-cpuinfo.c: New file.
* libgcc/config/i386/t-cpuinfo: New file.
* libgcc/config.host: Include t-cpuinfo.
* libgcc/config/i386/libgcc-glibc.ver: Version symbol __cpu_model.
2012-04-24 Chao-ying Fu <fu@mips.com>
* unwind-dw2-fde-dip.c: Define USE_PT_GNU_EH_FRAME for BIONIC.

2
libgcc/config.host
View File

@ -1130,7 +1130,7 @@ i[34567]86-*-linux* | x86_64-*-linux* | \
i[34567]86-*-kfreebsd*-gnu | x86_64-*-kfreebsd*-gnu | \
i[34567]86-*-knetbsd*-gnu | \
i[34567]86-*-gnu*)
tmake_file="${tmake_file} t-tls i386/t-linux"
tmake_file="${tmake_file} t-tls i386/t-linux i386/t-cpuinfo"
if test "$libgcc_cv_cfi" = "yes"; then
tmake_file="${tmake_file} t-stack i386/t-stack-i386"
fi

316
libgcc/config/i386/i386-cpuinfo.c Normal file
View File

@ -0,0 +1,316 @@
/* Get CPU type and Features for x86 processors.
Copyright (C) 2012 Free Software Foundation, Inc.
Contributed by Sriraman Tallam (tmsriram@google.com)
This file is part of GCC.
GCC 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.
GCC 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 GCC; see the file COPYING3. If not see
<http://www.gnu.org/licenses/>. */
#include "cpuid.h"
#include "tsystem.h"
int __cpu_indicator_init (void) __attribute__ ((constructor (101)));
enum vendor_signatures
{
SIG_INTEL = 0x756e6547 /* Genu */,
SIG_AMD = 0x68747541 /* Auth */
};
/* Processor Vendor and Models. */
enum processor_vendor
{
VENDOR_INTEL = 1,
VENDOR_AMD,
VENDOR_OTHER,
VENDOR_MAX
};
enum processor_types
{
INTEL_ATOM = 1,
INTEL_CORE2,
INTEL_COREI7,
AMDFAM10H,
AMDFAM15H,
CPU_TYPE_MAX
};
enum processor_subtypes
{
INTEL_COREI7_NEHALEM = 1,
INTEL_COREI7_WESTMERE,
INTEL_COREI7_SANDYBRIDGE,
AMDFAM10H_BARCELONA,
AMDFAM10H_SHANGHAI,
AMDFAM10H_ISTANBUL,
AMDFAM15H_BDVER1,
AMDFAM15H_BDVER2,
CPU_SUBTYPE_MAX
};
/* ISA Features supported. */
enum processor_features
{
FEATURE_CMOV = 0,
FEATURE_MMX,
FEATURE_POPCNT,
FEATURE_SSE,
FEATURE_SSE2,
FEATURE_SSE3,
FEATURE_SSSE3,
FEATURE_SSE4_1,
FEATURE_SSE4_2,
FEATURE_AVX
};
struct __processor_model
{
unsigned int __cpu_vendor;
unsigned int __cpu_type;
unsigned int __cpu_subtype;
unsigned int __cpu_features[1];
} __cpu_model;
/* Get the specific type of AMD CPU. */
static void
get_amd_cpu (unsigned int family, unsigned int model)
{
switch (family)
{
/* AMD Family 10h. */
case 0x10:
switch (model)
{
case 0x2:
/* Barcelona. */
__cpu_model.__cpu_type = AMDFAM10H;
__cpu_model.__cpu_subtype = AMDFAM10H_BARCELONA;
break;
case 0x4:
/* Shanghai. */
__cpu_model.__cpu_type = AMDFAM10H;
__cpu_model.__cpu_subtype = AMDFAM10H_SHANGHAI;
break;
case 0x8:
/* Istanbul. */
__cpu_model.__cpu_type = AMDFAM10H;
__cpu_model.__cpu_subtype = AMDFAM10H_ISTANBUL;
break;
default:
break;
}
break;
/* AMD Family 15h. */
case 0x15:
__cpu_model.__cpu_type = AMDFAM15H;
/* Bulldozer version 1. */
if ( model <= 0xf)
__cpu_model.__cpu_subtype = AMDFAM15H_BDVER1;
/* Bulldozer version 2. */
if (model >= 0x10 && model <= 0x1f)
__cpu_model.__cpu_subtype = AMDFAM15H_BDVER2;
break;
default:
break;
}
}
/* Get the specific type of Intel CPU. */
static void
get_intel_cpu (unsigned int family, unsigned int model, unsigned int brand_id)
{
/* Parse family and model only if brand ID is 0. */
if (brand_id == 0)
{
switch (family)
{
case 0x5:
/* Pentium. */
break;
case 0x6:
switch (model)
{
case 0x1c:
case 0x26:
/* Atom. */
__cpu_model.__cpu_type = INTEL_ATOM;
break;
case 0x1a:
case 0x1e:
case 0x1f:
case 0x2e:
/* Nehalem. */
__cpu_model.__cpu_type = INTEL_COREI7;
__cpu_model.__cpu_subtype = INTEL_COREI7_NEHALEM;
break;
case 0x25:
case 0x2c:
case 0x2f:
/* Westmere. */
__cpu_model.__cpu_type = INTEL_COREI7;
__cpu_model.__cpu_subtype = INTEL_COREI7_WESTMERE;
break;
case 0x2a:
/* Sandy Bridge. */
__cpu_model.__cpu_type = INTEL_COREI7;
__cpu_model.__cpu_subtype = INTEL_COREI7_SANDYBRIDGE;
break;
case 0x17:
case 0x1d:
/* Penryn. */
case 0x0f:
/* Merom. */
__cpu_model.__cpu_type = INTEL_CORE2;
break;
default:
break;
}
break;
default:
/* We have no idea. */
break;
}
}
}
static void
get_available_features (unsigned int ecx, unsigned int edx)
{
unsigned int features = 0;
if (edx & bit_CMOV)
features |= (1 << FEATURE_CMOV);
if (edx & bit_MMX)
features |= (1 << FEATURE_MMX);
if (edx & bit_SSE)
features |= (1 << FEATURE_SSE);
if (edx & bit_SSE2)
features |= (1 << FEATURE_SSE2);
if (ecx & bit_POPCNT)
features |= (1 << FEATURE_POPCNT);
if (ecx & bit_SSE3)
features |= (1 << FEATURE_SSE3);
if (ecx & bit_SSSE3)
features |= (1 << FEATURE_SSSE3);
if (ecx & bit_SSE4_1)
features |= (1 << FEATURE_SSE4_1);
if (ecx & bit_SSE4_2)
features |= (1 << FEATURE_SSE4_2);
if (ecx & bit_AVX)
features |= (1 << FEATURE_AVX);
__cpu_model.__cpu_features[0] = features;
}
/* A noinline function calling __get_cpuid. Having many calls to
cpuid in one function in 32-bit mode causes GCC to complain:
"can't find a register in class CLOBBERED_REGS". This is
related to PR rtl-optimization 44174. */
static int __attribute__ ((noinline))
__get_cpuid_output (unsigned int __level,
unsigned int *__eax, unsigned int *__ebx,
unsigned int *__ecx, unsigned int *__edx)
{
return __get_cpuid (__level, __eax, __ebx, __ecx, __edx);
}
/* A constructor function that is sets __cpu_model and __cpu_features with
the right values. This needs to run only once. This constructor is
given the highest priority and it should run before constructors without
the priority set. However, it still runs after ifunc initializers and
needs to be called explicitly there. */
int __attribute__ ((constructor (101)))
__cpu_indicator_init (void)
{
unsigned int eax, ebx, ecx, edx;
int max_level = 5;
unsigned int vendor;
unsigned int model, family, brand_id;
unsigned int extended_model, extended_family;
/* This function needs to run just once. */
if (__cpu_model.__cpu_vendor)
return 0;
/* Assume cpuid insn present. Run in level 0 to get vendor id. */
if (!__get_cpuid_output (0, &eax, &ebx, &ecx, &edx))
return -1;
vendor = ebx;
max_level = eax;
if (max_level < 1)
return -1;
if (!__get_cpuid_output (1, &eax, &ebx, &ecx, &edx))
return -1;
model = (eax >> 4) & 0x0f;
family = (eax >> 8) & 0x0f;
brand_id = ebx & 0xff;
extended_model = (eax >> 12) & 0xf0;
extended_family = (eax >> 20) & 0xff;
if (vendor == SIG_INTEL)
{
/* Adjust model and family for Intel CPUS. */
if (family == 0x0f)
{
family += extended_family;
model += extended_model;
}
else if (family == 0x06)
model += extended_model;
/* Get CPU type. */
get_intel_cpu (family, model, brand_id);
/* Find available features. */
get_available_features (ecx, edx);
__cpu_model.__cpu_vendor = VENDOR_INTEL;
}
else if (vendor == SIG_AMD)
{
/* Adjust model and family for AMD CPUS. */
if (family == 0x0f)
{
family += extended_family;
model += (extended_model << 4);
}
/* Get CPU type. */
get_amd_cpu (family, model);
/* Find available features. */
get_available_features (ecx, edx);
__cpu_model.__cpu_vendor = VENDOR_AMD;
}
else
__cpu_model.__cpu_vendor = VENDOR_OTHER;
gcc_assert (__cpu_model.__cpu_vendor < VENDOR_MAX);
gcc_assert (__cpu_model.__cpu_type < CPU_TYPE_MAX);
gcc_assert (__cpu_model.__cpu_subtype < CPU_SUBTYPE_MAX);
return 0;
}

8
libgcc/config/i386/libgcc-glibc.ver
View File

@ -147,6 +147,10 @@ GCC_4.3.0 {
__trunctfxf2
__unordtf2
}
GCC_4.8.0 {
__cpu_model
}
%else
GCC_4.4.0 {
__addtf3
@ -183,4 +187,8 @@ GCC_4.4.0 {
GCC_4.5.0 {
__extendxftf2
}
GCC_4.8.0 {
__cpu_model
}
%endif

1
libgcc/config/i386/t-cpuinfo Normal file
View File

@ -0,0 +1 @@
LIB2ADD += $(srcdir)/config/i386/i386-cpuinfo.c