i386.md (unnamed peephole2): Do not force memory operands of arith or logical instructions into registers...

* config/i386/i386.md (unnamed peephole2): Do not force memory
	operands of arith or logical instructions into registers for
	non-TARGET_READ_MODIFY targets.

	(ffs_cmove): Change operand 0 predicate to register_operand.
	Change operand 1 predicate to nonimmediate_operand.
	(ffsdi2): Ditto.
	(*ffs_no_cmove): Change operand 0 predicate to register_operand.

From-SVN: r141213
This commit is contained in:
Uros Bizjak 2008-10-18 20:15:14 +02:00
parent 73a5f1995f
commit 3cdf0c6237
4 changed files with 86 additions and 75 deletions

View File

@ -220,8 +220,8 @@ auto-vectorizer Dorit Nuzman dorit@il.ibm.com
loop infrastructure Zdenek Dvorak ook@ucw.cz
OpenMP Jakub Jelinek jakub@redhat.com
Note that individuals who maintain parts of the compiler need approval to check
in changes outside of the parts of the compiler they maintain.
Note that individuals who maintain parts of the compiler need approval to
check in changes outside of the parts of the compiler they maintain.
Non-Algorithmic Maintainers

View File

@ -1,3 +1,14 @@
2008-10-18 Uros Bizjak <ubizjak@gmail.com>
* config/i386/i386.md (unnamed peephole2): Do not force memory
operands of arith or logical instructions into registers for
non-TARGET_READ_MODIFY targets.
(ffs_cmove): Change operand 0 predicate to register_operand.
Change operand 1 predicate to nonimmediate_operand.
(ffsdi2): Ditto.
(*ffs_no_cmove): Change operand 0 predicate to register_operand.
2008-10-18 Richard Guenther <rguenther@suse.de>
* tree-ssa-forwprop.c (forward_propagate_addr_expr_1): Guard
@ -69,8 +80,7 @@
mark_hard_reg_early_clobbers): New functions.
(process_bb_node_lives): Call
make_early_clobber_and_input_conflicts and
mark_hard_reg_early_clobbers. Make hard register inputs live
again.
mark_hard_reg_early_clobbers. Make hard register inputs live again.
* doc/rtl.texi (clobber): Change descriotion of RA behaviour for
early clobbers of pseudo-registers.
@ -107,7 +117,7 @@
* config/mn10300/constraints.md (S): Allow UNSPEC_GOTSYM_OFF.
2008-10-15 Jan Sjodin <jan.sjodin@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
PR tree-optimization/37485
* graphite.c (gmp_cst_to_tree): Moved.
@ -143,7 +153,7 @@
(limit_scops): Same.
2008-10-15 Sebastian Pop <sebastian.pop@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
PR tree-optimization/37828
* graphite.c (graphite_trans_loop_block): Do not loop block
@ -215,12 +225,12 @@
2008-10-14 Douglas Gregor <doug.gregor@gmail.com>
PR c++/37553
* tree.c (build_type_attribute_qual_variant): Hash on the
unqualified type, and don't overwrite an existing
(type_hash_eq): Make the TYPE_NAME of the types significant, to
allow distinguishing between wchar_t and its underlying type. This
also means that we'll retain a little more typedef information.
PR c++/37553
* tree.c (build_type_attribute_qual_variant): Hash on the
unqualified type, and don't overwrite an existing
(type_hash_eq): Make the TYPE_NAME of the types significant, to
allow distinguishing between wchar_t and its underlying type. This
also means that we'll retain a little more typedef information.
2008-10-14 Andrey Belevantsev <abel@ispras.ru>
Dmitry Melnik <dm@ispras.ru>
@ -238,67 +248,67 @@
(ia64_gen_spec_insn): Removed.
(get_spec_check_gen_function, insn_can_be_in_speculative_p,
ia64_gen_spec_check): New static functions.
(ia64_alloc_sched_context, ia64_init_sched_context,
ia64_set_sched_context, ia64_clear_sched_context,
ia64_free_sched_context, ia64_get_insn_spec_ds,
(ia64_alloc_sched_context, ia64_init_sched_context,
ia64_set_sched_context, ia64_clear_sched_context,
ia64_free_sched_context, ia64_get_insn_spec_ds,
ia64_get_insn_checked_ds, ia64_skip_rtx_p): Declare functions.
(ia64_needs_block_p): Change prototype.
(ia64_gen_check): Rename to ia64_gen_spec_check.
(ia64_adjust_cost): Rename to ia64_adjust_cost_2. Add new parameter
(ia64_adjust_cost): Rename to ia64_adjust_cost_2. Add new parameter
into declaration, add special memory dependencies handling.
(TARGET_SCHED_ALLOC_SCHED_CONTEXT, TARGET_SCHED_INIT_SCHED_CONTEXT,
TARGET_SCHED_SET_SCHED_CONTEXT, TARGET_SCHED_CLEAR_SCHED_CONTEXT,
TARGET_SCHED_FREE_SCHED_CONTEXT, TARGET_SCHED_GET_INSN_SPEC_DS,
TARGET_SCHED_GET_INSN_CHECKED_DS, TARGET_SCHED_SKIP_RTX_P):
TARGET_SCHED_GET_INSN_CHECKED_DS, TARGET_SCHED_SKIP_RTX_P):
Define new target hooks.
(TARGET_SCHED_GEN_CHECK): Rename to TARGET_SCHED_GEN_SPEC_CHECK.
(ia64_optimization_options): Turn on selective scheduling with -O3,
(ia64_optimization_options): Turn on selective scheduling with -O3,
disable -fauto-inc-dec. Set mflag_sched_control_spec to true by default
with selective scheduling.
(ia64_override_options): Initialize align_loops and align_functions
to 32 and 64, respectively. Set global selective scheduling flags
(ia64_override_options): Initialize align_loops and align_functions
to 32 and 64, respectively. Set global selective scheduling flags
according to target-dependent flags.
(rtx_needs_barrier): Support UNSPEC_LDS_A.
(group_barrier_needed): Use new mstop-bit-before-check flag.
(group_barrier_needed): Use new mstop-bit-before-check flag.
Add heuristic.
(dfa_state_size): Make global.
(spec_check_no, max_uid): Remove.
(mem_ops_in_group, current_cycle): New variables.
(mem_ops_in_group, current_cycle): New variables.
(ia64_sched_init): Disable checks for !SCHED_GROUP_P after reload.
Initialize new variables.
(is_load_p, record_memory_reference): New functions.
(ia64_dfa_sched_reorder): Lower priority of loads when limit is
reached.
(ia64_variable_issue): Change use of current_sched_info to
Initialize new variables.
(is_load_p, record_memory_reference): New functions.
(ia64_dfa_sched_reorder): Lower priority of loads when limit is
reached.
(ia64_variable_issue): Change use of current_sched_info to
sched_deps_info. Update comment. Note if a load or a store is issued.
(ia64_first_cycle_multipass_dfa_lookahead_guard_spec): Require a cycle
advance if maximal number of loads or stores was issued on current
cycle.
(ia64_first_cycle_multipass_dfa_lookahead_guard_spec): Require a cycle
advance if maximal number of loads or stores was issued on current
cycle.
(scheduled_good_insn): New static helper function.
(ia64_dfa_new_cycle): Assert that last_scheduled_insn is set when
a group barrier is needed. Fix vertical spacing. Guard the code
doing state transition with last_scheduled_insn check.
Mark that a stop bit should be before current insn if there was a
cycle advance. Update current_cycle and mem_ops_in_group.
(ia64_dfa_new_cycle): Assert that last_scheduled_insn is set when
a group barrier is needed. Fix vertical spacing. Guard the code
doing state transition with last_scheduled_insn check.
Mark that a stop bit should be before current insn if there was a
cycle advance. Update current_cycle and mem_ops_in_group.
(ia64_h_i_d_extended): Change use of current_sched_info to
sched_deps_info. Reallocate stops_p by larger chunks.
sched_deps_info. Reallocate stops_p by larger chunks.
(struct _ia64_sched_context): New structure.
(ia64_sched_context_t): New typedef.
(ia64_alloc_sched_context, ia64_init_sched_context,
(ia64_alloc_sched_context, ia64_init_sched_context,
ia64_set_sched_context, ia64_clear_sched_context,
ia64_free_sched_context): New static functions.
(gen_func_t): New typedef.
(get_spec_load_gen_function): New function.
(SPEC_GEN_EXTEND_OFFSET): Declare.
(ia64_set_sched_flags): Check common_sched_info instead of *flags.
(get_mode_no_for_insn): Change the condition that prevents use of
(get_mode_no_for_insn): Change the condition that prevents use of
special hardware registers so it can now handle pseudos.
(get_spec_unspec_code): New function.
(ia64_skip_rtx_p, get_insn_spec_code, ia64_get_insn_spec_ds,
ia64_get_insn_checked_ds, ia64_gen_spec_load): New static functions.
(ia64_speculate_insn, ia64_needs_block_p): Support branchy checks
during selective scheduling.
(ia64_speculate_insn): Use ds_get_speculation_types when
(ia64_speculate_insn): Use ds_get_speculation_types when
determining whether we need to change the pattern.
(SPEC_GEN_LD_MAP, SPEC_GEN_CHECK_OFFSET): Declare.
(ia64_spec_check_src_p): Support new speculation/check codes.
@ -307,38 +317,39 @@
(insert_bundle_state): Minimize mid-bundle stop bits.
(important_for_bundling_p): New function.
(get_next_important_insn): Use important_for_bundling_p.
(bundling): When shifting TImode from unimportant insns, ignore
also group barriers. Assert that best state is found before
the backward bundling pass. Print number of mid-bundle stop bits.
Minimize mid-bundle stop bits. Check correct calculation of
(bundling): When shifting TImode from unimportant insns, ignore
also group barriers. Assert that best state is found before
the backward bundling pass. Print number of mid-bundle stop bits.
Minimize mid-bundle stop bits. Check correct calculation of
mid-bundle stop bits.
(ia64_sched_finish, final_emit_insn_group_barriers): Fix formatting.
(final_emit_insn_group_barriers): Emit stop bits before insns starting
a new cycle.
(sel2_run): New variable.
(ia64_reorg): When flag_selective_scheduling2 is set, run the selective
scheduling pass instead of schedule_ebbs.
(ia64_reorg): When flag_selective_scheduling2 is set, run the selective
scheduling pass instead of schedule_ebbs.
* config/ia64/ia64.md (speculable1, speculable2): New attributes.
(UNSPEC_LDS_A): New UNSPEC.
(movqi_internal, movhi_internal, movsi_internal, movdi_internal,
movti_internal, movsf_internal, movdf_internal,
(movqi_internal, movhi_internal, movsi_internal, movdi_internal,
movti_internal, movsf_internal, movdf_internal,
movxf_internal): Make visible. Add speculable* attributes.
(output_c_nc): New mode attribute.
(mov<mode>_speculative_a, zero_extend<mode>di2_speculative_a,
mov<mode>_nc, zero_extend<mode>di2_nc,
mov<mode>_nc, zero_extend<mode>di2_nc,
advanced_load_check_nc_<mode>): New insns.
(zero_extend*): Add speculable* attributes.
* config/ia64/ia64.opt (msched_fp_mem_deps_zero_cost): New option.
(msched-stop-bits-after-every-cycle): Likewise.
(msched-max-memory-insns,
msched-max-memory-insns-hard-limit): Likewise.
(msched-max-memory-insns, msched-max-memory-insns-hard-limit):
Likewise.
(msched-spec-verbose): Remove.
(msched-prefer-non-data-spec-insns,
msched-prefer-non-control-spec-insns, msched-count-spec-in-critical-path,
msel-sched-dont-check-control-spec): Use Target
Report Var instead of Common Report Var.
msched-prefer-non-control-spec-insns,
msched-count-spec-in-critical-path,
msel-sched-dont-check-control-spec): Use Target Report Var
instead of Common Report Var.
* config/ia64/itanium2.md: Remove incorrect bypass.
@ -365,15 +376,16 @@
for hppa64-linux-gnu targets.
2008-10-13 Andrew Pinski <andrew_pinski@playstation.sony.com>
Kaushal Kantawala <Kaushal_Kantawala@playstation.sony.com>
Trevor Smigiel <Trevor_Smigiel@playstation.sony.com>
Grace Cao <grace_cao@playstation.sony.com>
Kaushal Kantawala <Kaushal_Kantawala@playstation.sony.com>
Trevor Smigiel <Trevor_Smigiel@playstation.sony.com>
Grace Cao <grace_cao@playstation.sony.com>
* doc/invoke.texi (-mgen-cell-microcode): Document.
(-mwarn-cell-microcode): Document.
* cfglayout.c (locator_location): Export.
* rtl.h (locator_location): Define prototype.
* config/rs6000/predicates.md (cc_reg_not_micro_cr0_operand): New predicate.
* config/rs6000/predicates.md (cc_reg_not_micro_cr0_operand):
New predicate.
* rs6000/rs6000-protos.h (rs6000_final_prescan_insn): Define prototype.
* config/rs6000/rs6000.opt (mgen-cell-microcode): New option.
(mwarn-cell-microcode): New option.
@ -386,9 +398,10 @@
* config/rs6000/rs6000.h (FINAL_PRESCAN_INSN): Define.
* config/rs6000/rs6000.md
Replace cc_reg_not_cr0_operand with cc_reg_not_micro_cr0_operand if
the instruction would have been microcoded on the Cell.
Set cell_micro to always on unnamed patterns for the string instructions.
(cell_micro): Update definition, remove load/store conditional microcoded.
the instruction would have been microcoded on the Cell. Set cell_micro
to always on unnamed patterns for the string instructions.
(cell_micro): Update definition, remove load/store conditional
microcoded.
(sign_extend:DI): Define new pattern for non microcoded version.
(sign_extend:SI): Likewise.
(compare (div:P)): Set cell_micro to not.
@ -522,9 +535,8 @@
DF_REF_CLASS, DF_REF_TYPE, DF_REF_CHAIN, DF_REF_ID, DF_REF_FLAGS,
DF_REF_ORDER, DF_REF_IS_ARTIFICIAL, DF_REF_NEXT_REG,
DF_REF_PREV_REG, DF_REF_EXTRACT_WIDTH, DF_REF_EXTRACT_OFFSET,
DF_REF_EXTRACT_MODE): Replaced definition to access union
df_ref_d.
(DF_MWS_REG_DEF_P, DF_MWS_REG_USE_P, DF_MWS_TYPE): New macros.
DF_REF_EXTRACT_MODE): Replaced definition to access union df_ref_d.
(DF_MWS_REG_DEF_P, DF_MWS_REG_USE_P, DF_MWS_TYPE): New macros.
(df_scan_bb_info, df_bb_regno_first_def_find,
df_bb_regno_last_def_find, df_find_def, df_find_use,
df_refs_chain_dump, df_regs_chain_dump, df_ref_debug,
@ -548,8 +560,7 @@
df_ref * with df_ref.
(df_ref_record, df_uses_record, df_ref_create_structure): Added
df_ref_class parameter.
(df_scan_problem_data): Added new pools for different types of
refs.
(df_scan_problem_data): Added new pools for different types of refs.
(df_scan_free_internal, df_scan_alloc, df_free_ref,
df_ref_create_structure): Processed new ref pools.
(df_scan_start_dump): Added counts of refs and insns.

View File

@ -15366,9 +15366,9 @@
(define_expand "ffs_cmove"
[(set (match_dup 2) (const_int -1))
(parallel [(set (reg:CCZ FLAGS_REG)
(compare:CCZ (match_operand:SI 1 "register_operand" "")
(compare:CCZ (match_operand:SI 1 "nonimmediate_operand" "")
(const_int 0)))
(set (match_operand:SI 0 "nonimmediate_operand" "")
(set (match_operand:SI 0 "register_operand" "")
(ctz:SI (match_dup 1)))])
(set (match_dup 0) (if_then_else:SI
(eq (reg:CCZ FLAGS_REG) (const_int 0))
@ -15380,7 +15380,7 @@
"operands[2] = gen_reg_rtx (SImode);")
(define_insn_and_split "*ffs_no_cmove"
[(set (match_operand:SI 0 "nonimmediate_operand" "=r")
[(set (match_operand:SI 0 "register_operand" "=r")
(ffs:SI (match_operand:SI 1 "nonimmediate_operand" "rm")))
(clobber (match_scratch:SI 2 "=&q"))
(clobber (reg:CC FLAGS_REG))]
@ -15416,9 +15416,9 @@
(define_expand "ffsdi2"
[(set (match_dup 2) (const_int -1))
(parallel [(set (reg:CCZ FLAGS_REG)
(compare:CCZ (match_operand:DI 1 "register_operand" "")
(compare:CCZ (match_operand:DI 1 "nonimmediate_operand" "")
(const_int 0)))
(set (match_operand:DI 0 "nonimmediate_operand" "")
(set (match_operand:DI 0 "register_operand" "")
(ctz:DI (match_dup 1)))])
(set (match_dup 0) (if_then_else:DI
(eq (reg:CCZ FLAGS_REG) (const_int 0))
@ -20684,7 +20684,7 @@
[(match_dup 0)
(match_operand:SI 1 "memory_operand" "")]))
(clobber (reg:CC FLAGS_REG))])]
"optimize_insn_for_speed_p ()"
"optimize_insn_for_speed_p () && ! TARGET_READ_MODIFY"
[(set (match_dup 2) (match_dup 1))
(parallel [(set (match_dup 0)
(match_op_dup 3 [(match_dup 0) (match_dup 2)]))

View File

@ -42,13 +42,13 @@
* gfortran.dg/reshape_order_4.f90: Likewise.
2008-10-15 Jan Sjodin <jan.sjodin@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
PR tree-optimization/37485
* gcc.dg/graphite/pr37485.c: New.
2008-10-15 Sebastian Pop <sebastian.pop@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
Harsha Jagasia <harsha.jagasia@amd.com>
PR tree-optimization/37828
* testsuite/gcc.dg/graphite/pr37828.c: New.
@ -75,8 +75,8 @@
2008-10-14 Douglas Gregor <doug.gregor@gmail.com>
PR c++/37553
* g++.dg/ext/alias-canon2.C: New.
PR c++/37553
* g++.dg/ext/alias-canon2.C: New.
2008-10-13 Jerry DeLisle <jvdelisle@gcc.gnu.org