Show patches with: Submitter = liuhongt       |    Archived = No       |   632 patches
« 1 2 3 4 … 6 7 »
Patch Series A/F/R/T S/W/F Date Submitter Delegate State
Fix runtime error for nonlinear iv vectorization(step_mult). Fix runtime error for nonlinear iv vectorization(step_mult). - - - - --- 2024-03-21 liuhongt New
[V2] Document -fexcess-precision=16. [V2] Document -fexcess-precision=16. - - - - --- 2024-03-20 liuhongt New
Document -fexcess-precision=16. Document -fexcess-precision=16. - - - - --- 2024-03-18 liuhongt New
i386 [stv]: Handle REG_EH_REGION note [pr111822]. i386 [stv]: Handle REG_EH_REGION note [pr111822]. - - - - --- 2024-03-18 liuhongt New
Add missing <any_logic>hf/bf patterns. Add missing <any_logic>hf/bf patterns. - - - - --- 2024-03-18 liuhongt New
i386[stv]: Handle REG_EH_REGION note i386[stv]: Handle REG_EH_REGION note - - - - --- 2024-03-14 liuhongt New
sanitizer: [PR110027] Align asan_vec[0] to MAX (alignb, ASAN_RED_ZONE_SIZE) sanitizer: [PR110027] Align asan_vec[0] to MAX (alignb, ASAN_RED_ZONE_SIZE) - - - - --- 2024-03-12 liuhongt New
Fix testcase for platform without gnu/stubs-x32.h Fix testcase for platform without gnu/stubs-x32.h - - - - --- 2024-02-19 liuhongt New
[wwwdoc] Hardware-assisted AddressSanitizer now works for x86_64 with LAM_U57 [wwwdoc] Hardware-assisted AddressSanitizer now works for x86_64 with LAM_U57 - - - - --- 2024-02-08 liuhongt New
[2/2,x86] Enable -mlam=u57 by default when compiled with -fsanitize=hwaddress. [1/2] Adjust hwasan testcase for x86 target. - - - - --- 2024-01-23 liuhongt New
[1/2] Adjust hwasan testcase for x86 target. [1/2] Adjust hwasan testcase for x86 target. - - - - --- 2024-01-23 liuhongt New
Adjust testcase gcc.target/i386/part-vect-copysignhf.c. Adjust testcase gcc.target/i386/part-vect-copysignhf.c. - - - - --- 2024-01-19 liuhongt New
Fix testcase failure on many platforms which don't support vect_int_max. Fix testcase failure on many platforms which don't support vect_int_max. - - - - --- 2024-01-19 liuhongt New
Document refactoring of the option -fcf-protection=x. Document refactoring of the option -fcf-protection=x. - - - - --- 2024-01-10 liuhongt New
Update documents for fcf-protection= Update documents for fcf-protection= - - - - --- 2024-01-10 liuhongt New
Optimize A < B ? A : B to MIN_EXPR. Optimize A < B ? A : B to MIN_EXPR. - - - - --- 2024-01-09 liuhongt New
Optimize A < B ? A : B to MIN_EXPR. Optimize A < B ? A : B to MIN_EXPR. - - - - --- 2023-12-19 liuhongt New
Force broadcast constant to mem for vec_dup{v4di, v8si, v4df, v8df} when TARGET_AVX2 is not availab… Force broadcast constant to mem for vec_dup{v4di, v8si, v4df, v8df} when TARGET_AVX2 is not availab… - - - - --- 2023-12-13 liuhongt New
Adjust vectorized cost for reduction. Adjust vectorized cost for reduction. - - - - --- 2023-12-12 liuhongt New
[v3] Simplify vector ((VCE (a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE ((a cmp b) ? (VCE c) : (VC… [v3] Simplify vector ((VCE (a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE ((a cmp b) ? (VCE c) : (VC… - - - - --- 2023-12-11 liuhongt New
[ICE] Support vpcmov for V4HF/V4BF/V2HF/V2BF under TARGET_XOP. [ICE] Support vpcmov for V4HF/V4BF/V2HF/V2BF under TARGET_XOP. - - - - --- 2023-12-08 liuhongt New
Don't assume it's AVX_U128_CLEAN after call_insn whose abi.mode_clobber(V4DImode) deosn't contains … Don't assume it's AVX_U128_CLEAN after call_insn whose abi.mode_clobber(V4DImode) deosn't contains … - - - - --- 2023-12-08 liuhongt New
Support udot_prodv*qi with emulation sdot_prodv*hi Support udot_prodv*qi with emulation sdot_prodv*hi - - - - --- 2023-12-04 liuhongt New
Don't vectorize when vector stmts are only vec_contruct and stores Don't vectorize when vector stmts are only vec_contruct and stores - - - - --- 2023-12-04 liuhongt New
Take register pressure into account for vec_construct/scalar_to_vec when the components are not loa… Take register pressure into account for vec_construct/scalar_to_vec when the components are not loa… - - - - --- 2023-12-01 liuhongt New
Use vec_extact_lo instead of subreg in reduc_<code>_scal_m. Use vec_extact_lo instead of subreg in reduc_<code>_scal_m. - - - - --- 2023-11-29 liuhongt New
[x86] Support sdot_prodv*qi with emulation of sdot_prodv*hi. [x86] Support sdot_prodv*qi with emulation of sdot_prodv*hi. - - - - --- 2023-11-29 liuhongt New
Take register pressure into account for vec_construct when the components are not loaded from memor… Take register pressure into account for vec_construct when the components are not loaded from memor… - - - - --- 2023-11-28 liuhongt New
Set AVOID_256FMA_CHAINS TO m_GENERIC as it's generally good to new platforms Set AVOID_256FMA_CHAINS TO m_GENERIC as it's generally good to new platforms - - - - --- 2023-11-22 liuhongt New
[x86] Support reduc_{and, ior, xor}_scal_m for V4HI/V8QI/V4QImode [x86] Support reduc_{and, ior, xor}_scal_m for V4HI/V8QI/V4QImode - - - - --- 2023-11-20 liuhongt New
Support cbranchm for Vector HI/QImode. Support cbranchm for Vector HI/QImode. - - - - --- 2023-11-17 liuhongt New
[2/2] Add i?86-*-* and x86_64-*-* to vect_logical_reduc [1/2] Support reduc_{plus, xor, and, ior}_scal_m for vector integer mode. - - - - --- 2023-11-17 liuhongt New
[1/2] Support reduc_{plus, xor, and, ior}_scal_m for vector integer mode. [1/2] Support reduc_{plus, xor, and, ior}_scal_m for vector integer mode. - - - - --- 2023-11-17 liuhongt New
[V2] Simplify vector ((VCE (a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE ((a cmp b) ? (VCE c) : (VC… [V2] Simplify vector ((VCE (a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE ((a cmp b) ? (VCE c) : (VC… - - - - --- 2023-11-16 liuhongt New
Fix ICE of unrecognizable insn. Fix ICE of unrecognizable insn. - - - - --- 2023-11-16 liuhongt New
Fix ICE in vectorizable_nonlinear_induction with bitfield. Fix ICE in vectorizable_nonlinear_induction with bitfield. - - - - --- 2023-11-14 liuhongt New
Simplify vector ((VCE?(a cmp b ? -1 : 0)) < 0) ? c : d to just VCE:((a cmp b) ? (VCE c) : (VCE d)). Simplify vector ((VCE?(a cmp b ? -1 : 0)) < 0) ? c : d to just VCE:((a cmp b) ? (VCE c) : (VCE d)). - - - - --- 2023-11-10 liuhongt New
Support vec_set/vec_extract/vec_init for V4HF/V2HF. Support vec_set/vec_extract/vec_init for V4HF/V2HF. - - - - --- 2023-11-10 liuhongt New
Simplify vector ((VCE?(a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE:a cmp VCE:b) ? c : d. Simplify vector ((VCE?(a cmp b ? -1 : 0)) < 0) ? c : d to just (VCE:a cmp VCE:b) ? c : d. - - - - --- 2023-11-10 liuhongt New
Fix wrong code due to vec_merge + pcmp to blendvb splitter. Fix wrong code due to vec_merge + pcmp to blendvb splitter. - - - - --- 2023-11-09 liuhongt New
[V2] Handle bitop with INTEGER_CST in analyze_and_compute_bitop_with_inv_effect. [V2] Handle bitop with INTEGER_CST in analyze_and_compute_bitop_with_inv_effect. - - - - --- 2023-11-07 liuhongt New
Avoid generating RTL code when d->testing_p. Avoid generating RTL code when d->testing_p. - - - - --- 2023-11-06 liuhongt New
Support cmul{_conj}v4hf3/cmla{_conj}v4hf4 with AVX512FP16 instruction. Support cmul{_conj}v4hf3/cmla{_conj}v4hf4 with AVX512FP16 instruction. - - - - --- 2023-11-01 liuhongt New
Handle bitop with INTEGER_CST in analyze_and_compute_bitop_with_inv_effect. Handle bitop with INTEGER_CST in analyze_and_compute_bitop_with_inv_effect. - - - - --- 2023-10-30 liuhongt New
Fix wrong code due to incorrest define_split Fix wrong code due to incorrest define_split - - - - --- 2023-10-30 liuhongt New
Improve memcmpeq for 512-bit vector with vpcmpeq + kortest. Improve memcmpeq for 512-bit vector with vpcmpeq + kortest. - - - - --- 2023-10-27 liuhongt New
[V2,2/2] Support vec_cmpmn/vcondmn for v2hf/v4hf. [V2,1/2] Pass type of comparison operands instead of comparison result to truth_type_for in build_v… - - - - --- 2023-10-26 liuhongt New
[V2,1/2] Pass type of comparison operands instead of comparison result to truth_type_for in build_v… [V2,1/2] Pass type of comparison operands instead of comparison result to truth_type_for in build_v… - - - - --- 2023-10-26 liuhongt New
[GCC13,backport] Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_… [GCC13,backport] Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_… - - - - --- 2023-10-24 liuhongt New
Support vec_cmpmn/vcondmn for v2hf/v4hf. Support vec_cmpmn/vcondmn for v2hf/v4hf. - - - - --- 2023-10-23 liuhongt New
[x86] Remove unused mmx_pinsrw. [x86] Remove unused mmx_pinsrw. - - - - --- 2023-10-20 liuhongt New
Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … - - - - --- 2023-10-20 liuhongt New
Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … - - - - --- 2023-10-19 liuhongt New
Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … Avoid compile time hog on vect_peel_nonlinear_iv_init for nonlinear induction vec_step_op_mul when … - - - - --- 2023-10-18 liuhongt New
Support 32/64-bit vectorization for _Float16 fma related operations. Support 32/64-bit vectorization for _Float16 fma related operations. - - - - --- 2023-10-17 liuhongt New
[2/2] Support 32/64-bit vectorization for conversion between _Float16 and integer/float. [1/2] Enable vectorization for V2HF/V4HF rounding operations and sqrt. - - - - --- 2023-10-12 liuhongt New
[1/2] Enable vectorization for V2HF/V4HF rounding operations and sqrt. [1/2] Enable vectorization for V2HF/V4HF rounding operations and sqrt. - - - - --- 2023-10-12 liuhongt New
[x86] Refine predicate of operands[2] in divv4hf3 with register_operand. [x86] Refine predicate of operands[2] in divv4hf3 with register_operand. - - - - --- 2023-10-10 liuhongt New
[2/2] Support signbit/xorsign/copysign/abs/neg/and/xor/ior/andn for V2HF/V4HF. [1/2,x86] Support smin/smax for V2HF/V4HF - - - - --- 2023-10-08 liuhongt New
[1/2,x86] Support smin/smax for V2HF/V4HF [1/2,x86] Support smin/smax for V2HF/V4HF - - - - --- 2023-10-08 liuhongt New
Remove constraint modifier % for fcmaddcph/fmaddcph/fcmulcph since there're not commutative. Remove constraint modifier % for fcmaddcph/fmaddcph/fcmulcph since there're not commutative. - - - - --- 2023-09-11 liuhongt New
Remove constraint modifier % for fcmaddcph/fcmulcph since there're not commutative. Remove constraint modifier % for fcmaddcph/fcmulcph since there're not commutative. - - - - --- 2023-09-08 liuhongt New
Support vpermw/vpermi2w/vpermt2w instructions for vector HF/BFmodes. Support vpermw/vpermi2w/vpermt2w instructions for vector HF/BFmodes. - - - - --- 2023-09-07 liuhongt New
Generate vmovsh instead of vpblendw for specific vec_merge. Generate vmovsh instead of vpblendw for specific vec_merge. - - - - --- 2023-09-04 liuhongt New
Adjust costing of emulated vectorized gather/scatter Adjust costing of emulated vectorized gather/scatter - - - - --- 2023-08-30 liuhongt New
Refactor vector HF/BF mode iterators and patterns. Refactor vector HF/BF mode iterators and patterns. - - - - --- 2023-08-30 liuhongt New
Use vmaskmov{ps, pd} for VI48_128_256 when TARGET_AVX2 is not available. Use vmaskmov{ps, pd} for VI48_128_256 when TARGET_AVX2 is not available. - - - - --- 2023-08-25 liuhongt New
[x86] Refactor mode iterator V_128 and V_128H, V_256 and V_256H [x86] Refactor mode iterator V_128 and V_128H, V_256 and V_256H - - - - --- 2023-08-24 liuhongt New
Fix target_clone ("arch=graniterapids-d") and target_clone ("arch=arrowlake-s") Fix target_clone ("arch=graniterapids-d") and target_clone ("arch=arrowlake-s") - - - - --- 2023-08-23 liuhongt New
[x86] Testcase fix. [x86] Testcase fix. - - - - --- 2023-08-22 liuhongt New
Adjust testcase for Intel GDS. Adjust testcase for Intel GDS. - - - - --- 2023-08-21 liuhongt New
Mention Intel -march=gracemont for Alderlake-N. Mention Intel -march=gracemont for Alderlake-N. - - - - --- 2023-08-21 liuhongt New
Support -march=gracemont Support -march=gracemont - - - - --- 2023-08-18 liuhongt New
Generate vmovapd instead of vmovsd for moving DFmode between SSE_REGS. Generate vmovapd instead of vmovsd for moving DFmode between SSE_REGS. - - - - --- 2023-08-14 liuhongt New
[V2] Support -m[no-]gather -m[no-]scatter to enable/disable vectorization for all gather/scatter in… [V2] Support -m[no-]gather -m[no-]scatter to enable/disable vectorization for all gather/scatter in… - - - - --- 2023-08-11 liuhongt New
Software mitigation: Disable gather generation in vectorization for GDS affected Intel Processors. Software mitigation: Disable gather generation in vectorization for GDS affected Intel Processors. - - - - --- 2023-08-11 liuhongt New
Support -m[no-]gather -m[no-]scatter to enable/disable vectorization for all gather/scatter instruc… Support -m[no-]gather -m[no-]scatter to enable/disable vectorization for all gather/scatter instruc… - - - - --- 2023-08-10 liuhongt New
i386: Do not sanitize upper part of V2HFmode and V4HFmode reg with -fno-trapping-math [PR110832] i386: Do not sanitize upper part of V2HFmode and V4HFmode reg with -fno-trapping-math [PR110832] - - - - --- 2023-08-10 liuhongt New
Rename local variable subleaf_level to max_subleaf_level. Rename local variable subleaf_level to max_subleaf_level. - - - - --- 2023-08-09 liuhongt New
[V2,X86] Workaround possible CPUID bug in Sandy Bridge. [V2,X86] Workaround possible CPUID bug in Sandy Bridge. - - - - --- 2023-08-09 liuhongt New
[X86] Workaround possible CPUID bug in Sandy Bridge. [X86] Workaround possible CPUID bug in Sandy Bridge. - - - - --- 2023-08-08 liuhongt New
i386: Clear upper bits of XMM register for V4HFmode/V2HFmode operations [PR110762] i386: Clear upper bits of XMM register for V4HFmode/V2HFmode operations [PR110762] - - - - --- 2023-08-07 liuhongt New
Fix ICE in rtl check when bootstrap. Fix ICE in rtl check when bootstrap. - - - - --- 2023-08-07 liuhongt New
Optimize vlddqu + inserti128 to vbroadcasti128 Optimize vlddqu + inserti128 to vbroadcasti128 - - - - --- 2023-08-02 liuhongt New
Support vec_fmaddsub/vec_fmsubadd for vector HFmode. Support vec_fmaddsub/vec_fmsubadd for vector HFmode. - - - - --- 2023-08-02 liuhongt New
Adjust testcase for more optimal codegen. Adjust testcase for more optimal codegen. - - - - --- 2023-08-01 liuhongt New
[x86] Add UNSPEC_MASKOP to vpbroadcastm pattern. [x86] Add UNSPEC_MASKOP to vpbroadcastm pattern. - - - - --- 2023-07-27 liuhongt New
Optimize vlddqu to vmovdqu for TARGET_AVX Optimize vlddqu to vmovdqu for TARGET_AVX - - - - --- 2023-07-20 liuhongt New
Fix fp16 related testcase failure for i686. Fix fp16 related testcase failure for i686. - - - - --- 2023-07-20 liuhongt New
Remove # from <mask_codefor>one_cmpl<mode>2<mask_name> assemble output. Remove # from <mask_codefor>one_cmpl<mode>2<mask_name> assemble output. - - - - --- 2023-07-17 liuhongt New
Fix typo in the testcase. Fix typo in the testcase. - - - - --- 2023-07-11 liuhongt New
Add peephole to eliminate redundant comparison after cmpccxadd. Add peephole to eliminate redundant comparison after cmpccxadd. - - - - --- 2023-07-11 liuhongt New
[v2] Break false dependence for vpternlog by inserting vpxor or setting constraint of input operand… [v2] Break false dependence for vpternlog by inserting vpxor or setting constraint of input operand… - - - - --- 2023-07-11 liuhongt New
Add peephole to eliminate redundant comparison after cmpccxadd. Add peephole to eliminate redundant comparison after cmpccxadd. - - - - --- 2023-07-11 liuhongt New
Break false dependence for vpternlog by inserting vpxor or setting constraint of input operand to '… Break false dependence for vpternlog by inserting vpxor or setting constraint of input operand to '… - - - - --- 2023-07-10 liuhongt New
[V2,x86] Add pre_reload splitter to detect fp min/max pattern. [V2,x86] Add pre_reload splitter to detect fp min/max pattern. - - - - --- 2023-07-07 liuhongt New
[2/2] Adjust rtx_cost for DF/SFmode AND/IOR/XOR/ANDN operations. [1/2,x86] Add pre_reload splitter to detect fp min/max pattern. - - - - --- 2023-07-06 liuhongt New
[1/2,x86] Add pre_reload splitter to detect fp min/max pattern. [1/2,x86] Add pre_reload splitter to detect fp min/max pattern. - - - - --- 2023-07-06 liuhongt New
Disparage slightly for the alternative which move DFmode between SSE_REGS and GENERAL_REGS. Disparage slightly for the alternative which move DFmode between SSE_REGS and GENERAL_REGS. - - - - --- 2023-07-06 liuhongt New
Break false dependence for vpternlog by inserting vpxor. Break false dependence for vpternlog by inserting vpxor. - - - - --- 2023-07-04 liuhongt New
« 1 2 3 4 … 6 7 »