Show patches with: Submitter = Hongtao Liu       |    State = Action Required       |    Archived = No       |   534 patches
« 1 2 ... 3 4 5 6 »
Patch Series A/F/R/T S/W/F Date Submitter Delegate State
Adjust testcase for O2 vectorization enabling Adjust testcase for O2 vectorization enabling - - - - --- 2021-10-11 Liu, Hongtao New
Adjust more testcases for O2 vectorization enabling. Adjust more testcases for O2 vectorization enabling. - - - - --- 2021-10-09 Liu, Hongtao New
Refine movhfcc. Refine movhfcc. - - - - --- 2021-10-08 Liu, Hongtao New
[GCC-12] Mention O2 vectorization enabling. [GCC-12] Mention O2 vectorization enabling. - - - - --- 2021-10-08 Liu, Hongtao New
[i386] Support reduc_{plus,smax,smin,umax,min}_scal_v4hi. [i386] Support reduc_{plus,smax,smin,umax,min}_scal_v4hi. - - - - --- 2021-09-28 Liu, Hongtao New
Support 128/256/512-bit vector _Float16 plus/smin/smax reduce. Support 128/256/512-bit vector _Float16 plus/smin/smax reduce. - - - - --- 2021-09-27 Liu, Hongtao New
Revert "Optimize v4sf reduction.". Revert "Optimize v4sf reduction.". - - - - --- 2021-09-27 Liu, Hongtao New
Enable auto-vectorization at O2 with very-cheap cost model. Enable auto-vectorization at O2 with very-cheap cost model. - - - - --- 2021-09-26 Liu, Hongtao New
[i386] Remove storage only description for _Float16 w/o avx512fp16. [i386] Remove storage only description for _Float16 w/o avx512fp16. - - - - --- 2021-09-25 Liu, Hongtao New
[GIMPLE] Simplify (_Float16) ceil ((double) x) to .CEIL (x) when available. [GIMPLE] Simplify (_Float16) ceil ((double) x) to .CEIL (x) when available. - - - - --- 2021-09-24 Liu, Hongtao New
[GCC12] Mention Intel AVX512-FP16 and _Float16 support. [GCC12] Mention Intel AVX512-FP16 and _Float16 support. - - - - --- 2021-09-24 Liu, Hongtao New
[7/7] AVX512FP16: Enable vec_cmpmn/vcondmn expanders for HF modes. AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[6/7] AVX512FP16: add truncmn2/extendmn2 expanders AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[5/7] AVX512FP16: Add float(uns)?mn2 expander AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[4/7] AVX512FP16: Add fix(uns)?_truncmn2 for HF scalar and vector modes AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[3/7] AVX512FP16: Add expander for smin/maxhf3. AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[2/7] AVX512FP16: Add expander for fmahf4 AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
[1/7] AVX512FP16: Add expander for rint/nearbyinthf2. AVX512FP16: Support bunch of expanders for HFmode and vector HFmodes - - - - --- 2021-09-23 Liu, Hongtao New
wwwdocs: [GCC12] Mention Intel AVX512-FP16. wwwdocs: [GCC12] Mention Intel AVX512-FP16. - - - - --- 2021-09-23 Liu, Hongtao New
[i386] Adjust testcase. [i386] Adjust testcase. - - - - --- 2021-09-22 Liu, Hongtao New
Support 64bit fma/fms/fnma/fnms under avx512vl. Support 64bit fma/fms/fnma/fnms under avx512vl. - - - - --- 2021-09-22 Liu, Hongtao New
[i386] Fix ICE in pass_rpad. [i386] Fix ICE in pass_rpad. - - - - --- 2021-09-18 Liu, Hongtao New
[AVX512FP16] Support embedded broadcast for AVX512FP16 instructions. [AVX512FP16] Support embedded broadcast for AVX512FP16 instructions. - - - - --- 2021-09-16 Liu, Hongtao New
Check mask type when doing cond_op related gimple simplification. Check mask type when doing cond_op related gimple simplification. - - - - --- 2021-09-16 Liu, Hongtao New
Enable auto-vectorization at O2 with very-cheap cost model. Enable auto-vectorization at O2 with very-cheap cost model. - - - - --- 2021-09-16 Liu, Hongtao New
Optimize for V{8,16,32}HFmode vec_set/extract/init. Optimize for V{8,16,32}HFmode vec_set/extract/init. - - - - --- 2021-09-15 Liu, Hongtao New
Output vextract{i, f}{32x4, 64x2} for (vec_select:(reg:Vmode) idx) when byte_offset of idx % 16 == … Output vextract{i, f}{32x4, 64x2} for (vec_select:(reg:Vmode) idx) when byte_offset of idx % 16 == … - - - - --- 2021-09-15 Liu, Hongtao New
Remove UNSPEC_{COPYSIGN,XORSIGN}. Remove UNSPEC_{COPYSIGN,XORSIGN}. - - - - --- 2021-09-13 Liu, Hongtao New
[2/2] validate_subreg before call gen_lowpart to avoid ICE. Revert r12-3277 since it caused regressions on many other targets. - - - - --- 2021-09-10 Liu, Hongtao New
[1/2] Revert "Get rid of all float-int special cases in validate_subreg." Revert r12-3277 since it caused regressions on many other targets. - - - - --- 2021-09-10 Liu, Hongtao New
Disallow paradoxical subregs when outer mode is SCALAR_FLOAT_MODE_P. Disallow paradoxical subregs when outer mode is SCALAR_FLOAT_MODE_P. - - - - --- 2021-09-10 Liu, Hongtao New
Relax condition of (vec_concat:M(vec_select op0 idx0)(vec_select op0 idx1)) to allow different mode… Relax condition of (vec_concat:M(vec_select op0 idx0)(vec_select op0 idx1)) to allow different mode… - - - - --- 2021-09-10 Liu, Hongtao New
[i386] Remove copysign post_reload splitter for scalar modes. [i386] Remove copysign post_reload splitter for scalar modes. - - - - --- 2021-09-09 Liu, Hongtao New
Optimize vec_extract for 256/512-bit vector when index exceeds the lower 128 bits. Optimize vec_extract for 256/512-bit vector when index exceeds the lower 128 bits. - - - - --- 2021-09-08 Liu, Hongtao New
[i386] Optimize v4sf reduction. [i386] Optimize v4sf reduction. - - - - --- 2021-09-08 Liu, Hongtao New
Avoid FROM being overwritten in expand_fix. Avoid FROM being overwritten in expand_fix. - - - - --- 2021-09-06 Liu, Hongtao New
Adjust the wording for x86 _Float16 type. Adjust the wording for x86 _Float16 type. - - - - --- 2021-09-06 Liu, Hongtao New
Enable auto-vectorization at O2 with very-cheap cost model. Enable auto-vectorization at O2 with very-cheap cost model. - - - - --- 2021-09-06 Liu, Hongtao New
Explicitly add -msse2 to compile HF related libgcc source file. Explicitly add -msse2 to compile HF related libgcc source file. - - - - --- 2021-09-03 Liu, Hongtao New
Remove macro check for __AMX_BF16/INT8/TILE__ in header file. Remove macro check for __AMX_BF16/INT8/TILE__ in header file. - - - - --- 2021-09-02 Liu, Hongtao New
[2/2] Get rid of all float-int special cases in validate_subreg. Get rid of all float-int special cases in validate_subreg. - - - - --- 2021-08-31 Liu, Hongtao New
[1/2] Revert "Make sure we're playing with integral modes before call extract_integral_bit_field." Get rid of all float-int special cases in validate_subreg. - - - - --- 2021-08-31 Liu, Hongtao New
[i386] Unify UNSPEC_MASKED_EQ/GT to the form of UNSPEC_PCMP. [i386] Unify UNSPEC_MASKED_EQ/GT to the form of UNSPEC_PCMP. - - - - --- 2021-08-30 Liu, Hongtao New
Check the type of mask while generating cond_op in gimple simplication. Check the type of mask while generating cond_op in gimple simplication. - - - - --- 2021-08-27 Liu, Hongtao New
Fold more shuffle builtins to VEC_PERM_EXPR. Fold more shuffle builtins to VEC_PERM_EXPR. - - - - --- 2021-08-26 Liu, Hongtao New
Adjust testcases to avoid new failures brought by r12-3108 when compiled w -march=cascadelake. Adjust testcases to avoid new failures brought by r12-3108 when compiled w -march=cascadelake. - - - - --- 2021-08-25 Liu, Hongtao New
[i386] Enable avx512 embedde broadcast for vpternlog. [i386] Enable avx512 embedde broadcast for vpternlog. - - - - --- 2021-08-24 Liu, Hongtao New
Change illegitimate constant into memref of constant pool in change_zero_ext. Change illegitimate constant into memref of constant pool in change_zero_ext. - - - - --- 2021-08-24 Liu, Hongtao New
[i386] Optimize (a & b) | (c & ~b) to vpternlog instruction. [i386] Optimize (a & b) | (c & ~b) to vpternlog instruction. - - - - --- 2021-08-24 Liu, Hongtao New
[i386] Fix ICE. [i386] Fix ICE. - - - - --- 2021-08-23 Liu, Hongtao New
Disable slp in loop vectorizer when cost model is very-cheap. Disable slp in loop vectorizer when cost model is very-cheap. - - - - --- 2021-08-23 Liu, Hongtao New
Revert "Add the member integer_to_sse to processor_cost as a cost simulation for movd/pinsrd. It wi… Revert "Add the member integer_to_sse to processor_cost as a cost simulation for movd/pinsrd. It wi… - - - - --- 2021-08-17 Liu, Hongtao New
[i386] Add x86 tune to enable v2df vector reduction by paddpd. [i386] Add x86 tune to enable v2df vector reduction by paddpd. - - - - --- 2021-08-17 Liu, Hongtao New
[i386] Fix ICE. [i386] Fix ICE. - - - - --- 2021-08-16 Liu, Hongtao New
[i386] Optimize __builtin_shuffle_vector. [i386] Optimize __builtin_shuffle_vector. - - - - --- 2021-08-16 Liu, Hongtao New
[i386] Optimize vec_perm_expr to match vpmov{dw,qd,wb}. [i386] Optimize vec_perm_expr to match vpmov{dw,qd,wb}. - - - - --- 2021-08-12 Liu, Hongtao New
[i386] Introduce a scalar version of avx512f_vmscalef and adjust ldexp<mode>3 for it. [i386] Introduce a scalar version of avx512f_vmscalef and adjust ldexp<mode>3 for it. - - - - --- 2021-08-12 Liu, Hongtao New
[i386] Combine avx_vec_concatv16si and avx512f_zero_extendv16hiv16si2_1 to avx512f_zero_extendv16hi… [i386] Combine avx_vec_concatv16si and avx512f_zero_extendv16hiv16si2_1 to avx512f_zero_extendv16hi… - - - - --- 2021-08-11 Liu, Hongtao New
Extend ldexp{s, d}f3 to vscalefs{s, d} when TARGET_AVX512F and TARGET_SSE_MATH. Extend ldexp{s, d}f3 to vscalefs{s, d} when TARGET_AVX512F and TARGET_SSE_MATH. - - - - --- 2021-08-10 Liu, Hongtao New
[i386] Support cond_ashr/lshr/ashl for vector integer modes under AVX512. [i386] Support cond_ashr/lshr/ashl for vector integer modes under AVX512. - - - - --- 2021-08-09 Liu, Hongtao New
[rtl-optimization] Simplify vector shift/rotate with const_vec_duplicate to vector shift/rotate wit… [rtl-optimization] Simplify vector shift/rotate with const_vec_duplicate to vector shift/rotate wit… - - - - --- 2021-08-06 Liu, Hongtao New
Make sure we're playing with integral modes before call extract_integral_bit_field. Make sure we're playing with integral modes before call extract_integral_bit_field. - - - - --- 2021-08-06 Liu, Hongtao New
[3/3,i386] Support cond_{xor, ior, and} for vector integer mode under AVX512. Support cond_{smax, smin, umax, umin, xor, ior, and} for vector modes under AVX512 - - - - --- 2021-08-04 Liu, Hongtao New
[2/3,i386] Support cond_{smax, smin} for vector float/double modes under AVX512. Support cond_{smax, smin, umax, umin, xor, ior, and} for vector modes under AVX512 - - - - --- 2021-08-04 Liu, Hongtao New
[1/3,i386] Support cond_{smax, smin, umax, umin} for vector integer modes under AVX512. Support cond_{smax, smin, umax, umin, xor, ior, and} for vector modes under AVX512 - - - - --- 2021-08-04 Liu, Hongtao New
Add dg-require-effective-target for testcases. Add dg-require-effective-target for testcases. - - - - --- 2021-08-04 Liu, Hongtao New
[i386] Support cond_{fma, fms, fnma, fnms} for vector float/double under AVX512. [i386] Support cond_{fma, fms, fnma, fnms} for vector float/double under AVX512. - - - - --- 2021-08-04 Liu, Hongtao New
[i386] Refine predicate of peephole2 to general_reg_operand. [PR target/101743] [i386] Refine predicate of peephole2 to general_reg_operand. [PR target/101743] - - - - --- 2021-08-04 Liu, Hongtao New
Add cond_add/sub/mul for vector integer modes. Add cond_add/sub/mul for vector integer modes. - - - - --- 2021-08-03 Liu, Hongtao New
[5/6] AVX512FP16: Initial support for AVX512FP16 feature and scalar _Float16 instructions. Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
[6/6] AVX512FP16: Support vector init/broadcast/set/extract for FP16. Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
[4/6] Support -fexcess-precision=16 which will enable FLT_EVAL_METHOD_PROMOTE_TO_FLOAT16 when backe… Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
[3/6,i386] libgcc: Enable hfmode soft-sf/df/xf/tf extensions and truncations. Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
[2/6,i386] Enable _Float16 type for TARGET_SSE2 and above. Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
[1/6] Update hf soft-fp from glibc. Initial support for AVX512FP16 - - - - --- 2021-08-02 Liu, Hongtao New
Support cond_add/sub/mul/div for vector float/double. Support cond_add/sub/mul/div for vector float/double. - - - - --- 2021-08-02 Liu, Hongtao New
Adjust/Refine testcases. Adjust/Refine testcases. - - - - --- 2021-07-29 Liu, Hongtao New
[i386] Add a separate function to calculate cost for WIDEN_MULT_EXPR. [i386] Add a separate function to calculate cost for WIDEN_MULT_EXPR. - - - - --- 2021-07-28 Liu, Hongtao New
[10/10] AVX512FP16: Add abi test for zmm Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[09/10] AVX512FP16: Add ABI test for ymm. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[08/10] AVX512FP16: Add ABI tests for xmm. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[07/10] AVX512FP16: Add tests for vector passing in variable arguments. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[06/10] AVX512FP16: Add testcase for vector init and broadcast intrinsics. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[05/10] AVX512FP16: Support vector init/broadcast/set/extract for FP16. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[04/10] AVX512FP16: Initial support for AVX512FP16 feature and scalar _Float16 instructions. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[03/10,i386] libgcc: Enable hfmode soft-sf/df/xf/tf extensions and truncations. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[02/10,i386] Enable _Float16 type for TARGET_SSE2 and above. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
[01/10] Update hf soft-fp from glibc. Initial support for AVX512FP16 - - - - --- 2021-07-21 Liu, Hongtao New
Support logic shift left/right for avx512 mask type. Support logic shift left/right for avx512 mask type. - - - - --- 2021-07-20 Liu, Hongtao New
[i386] Remove pass_cpb which is related to enable avx512 embedded broadcast from constant pool. [i386] Remove pass_cpb which is related to enable avx512 embedded broadcast from constant pool. - - - - --- 2021-07-14 Liu, Hongtao New
Fix typo in standard pattern name of trunc<mode><pmov_dst_4>2. Fix typo in standard pattern name of trunc<mode><pmov_dst_4>2. - - - - --- 2021-07-01 Liu, Hongtao New
[62/62] AVX512FP16: Add permutation and mask blend intrinsics. Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[61/62] AVX512FP16: Add complex conjugation intrinsic instructions. Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[60/62] AVX512FP16: Add reduce operators(add/mul/min/max). Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[59/62] AVX512FP16: Support load/store/abs intrinsics. Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[58/62] AVX512FP16: Optimize for code like (_Float16) __builtin_ceif ((float) f16). Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[57/62] AVX512FP16: Add expander for fmahf4 Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[56/62] AVX512FP16: Optimize (_Float16) sqrtf ((float) f16) to sqrtf16 (f16). Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[55/62] AVX512FP16: Add expander for cstorehf4. Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
[54/62] AVX512FP16: Add expander for ceil/floor/trunc/roundeven. Support all AVX512FP16 intrinsics. - - - - --- 2021-07-01 Liu, Hongtao New
« 1 2 ... 3 4 5 6 »