2022-01-03 |
Jakub Jelinek | Update copyright years. |
blob | commitdiff | raw |
2021-12-14 |
Uros Bizjak | i386: Implement VxHF vector set/insert/extract with... |
blob | commitdiff | raw | diff to current |
2021-12-03 |
H.J. Lu | x86: Add -mmove-max=bits and -mstore-max=bits |
blob | commitdiff | raw | diff to current |
2021-12-03 |
H.J. Lu | Add TARGET_IFUNC_REF_LOCAL_OK |
blob | commitdiff | raw | diff to current |
2021-12-01 |
Uros Bizjak | i386: Improve V8HI and V8HF inserts [PR102811] |
blob | commitdiff | raw | diff to current |
2021-11-29 |
Roger Sayle | x86_64: Improved V1TImode rotations by non-constant... |
blob | commitdiff | raw | diff to current |
2021-11-18 |
Hongyu Wang | i386: Fix wrong codegen for -mrelax-cmpxchg-loop |
blob | commitdiff | raw | diff to current |
2021-11-15 |
Hongyu Wang | PR target/103069: Relax cmpxchg loop for x86 target |
blob | commitdiff | raw | diff to current |
2021-11-02 |
Roger Sayle | x86_64: Expand ashrv1ti (and PR target/102986) |
blob | commitdiff | raw | diff to current |
2021-10-29 |
liuhongt | Enable vectorization for _Float16 floor/ceil/trunc... |
blob | commitdiff | raw | diff to current |
2021-10-26 |
Roger Sayle | x86_64: Implement V1TI mode shifts/rotates by a constant |
blob | commitdiff | raw | diff to current |
2021-10-15 |
Hongyu Wang | AVX512FP16: Enhance vector shuffle builtins |
blob | commitdiff | raw | diff to current |
2021-10-15 |
Hongyu Wang | AVX512FP16: Fix ICE for 2 v4hf vector concat |
blob | commitdiff | raw | diff to current |
2021-10-14 |
Hongyu Wang | AVX512FP16: Adjust builtin for mask complex fma |
blob | commitdiff | raw | diff to current |
2021-10-13 |
Roger Sayle | x86_64: Some SUBREG related optimization tweaks to... |
blob | commitdiff | raw | diff to current |
2021-10-12 |
liuhongt | Support reduc_{plus,smax,smin,umax,umin}_scal_v4qi. |
blob | commitdiff | raw | diff to current |
2021-10-09 |
liuhongt | Refine movhfcc. |
blob | commitdiff | raw | diff to current |
2021-10-08 |
liuhongt | Support reduc_{plus,smax,smin,umax,min}_scal_v4hi. |
blob | commitdiff | raw | diff to current |
2021-09-28 |
liuhongt | Support 128/256/512-bit vector plus/smin/smax reduction... |
blob | commitdiff | raw | diff to current |
2021-09-23 |
Hongyu Wang | AVX512FP16: Enable vec_cmpmn/vcondmn expanders for... |
blob | commitdiff | raw | diff to current |
2021-09-22 |
liuhongt | AVX512FP16: Add vfcmaddcph/vfmaddcph/vfcmulcph/vfmulcph |
blob | commitdiff | raw | diff to current |
2021-09-18 |
liuhongt | AVX512FP16: Add scalar fma instructions. |
blob | commitdiff | raw | diff to current |
2021-09-18 |
H.J. Lu | AVX512FP16: Add scalar/vector bitwise operations, including |
blob | commitdiff | raw | diff to current |
2021-09-17 |
liuhongt | AVX512FP16: Add vcvtsh2ss/vcvtsh2sd/vcvtss2sh/vcvtsd2sh. |
blob | commitdiff | raw | diff to current |
2021-09-17 |
liuhongt | AVX512FP16: Add vcvtph2pd/vcvtph2psx/vcvtpd2ph/vcvtps2phx. |
blob | commitdiff | raw | diff to current |
2021-09-17 |
liuhongt | AVX512FP16: Add vcvtsh2si/vcvtsh2usi/vcvtsi2sh/vcvtusi2sh. |
blob | commitdiff | raw | diff to current |
2021-09-16 |
liuhongt | AVX512FP16: Add vcvtuw2ph/vcvtw2ph/vcvtdq2ph/vcvtudq2ph... |
blob | commitdiff | raw | diff to current |
2021-09-16 |
liuhongt | AVX512FP16: Add vcvtph2dq/vcvtph2qq/vcvtph2w/vcvtph2uw... |
blob | commitdiff | raw | diff to current |
2021-09-16 |
liuhongt | AVX512FP16: Add vmovw/vmovsh. |
blob | commitdiff | raw | diff to current |
2021-09-15 |
liuhongt | Optimize for V{8,16,32}HFmode vec_set/extract/init. |
blob | commitdiff | raw | diff to current |
2021-09-14 |
liuhongt | AVX512FP16: Add fpclass/getexp/getmant instructions. |
blob | commitdiff | raw | diff to current |
2021-09-14 |
liuhongt | AVX512FP16: Add vreduceph/vreducesh/vrndscaleph/vrndsca... |
blob | commitdiff | raw | diff to current |
2021-09-14 |
liuhongt | AVX512FP16: Add vsqrtph/vrsqrtph/vsqrtsh/vrsqrtsh. |
blob | commitdiff | raw | diff to current |
2021-09-10 |
liuhongt | AVX512FP16: Add vcmpph/vcmpsh/vcomish/vucomish. |
blob | commitdiff | raw | diff to current |
2021-09-10 |
liuhongt | AVX512FP16: Add vmaxph/vminph/vmaxsh/vminsh. |
blob | commitdiff | raw | diff to current |
2021-09-10 |
Liu, Hongtao | AVX512FP16: Add vaddsh/vsubsh/vmulsh/vdivsh. |
blob | commitdiff | raw | diff to current |
2021-09-10 |
H.J. Lu | AVX512FP16: Enable _Float16 autovectorization |
blob | commitdiff | raw | diff to current |
2021-09-10 |
liuhongt | Remove copysign post_reload splitter for scalar modes. |
blob | commitdiff | raw | diff to current |
2021-09-09 |
liuhongt | AVX512FP16: Add vaddph/vsubph/vdivph/vmulph. |
blob | commitdiff | raw | diff to current |
2021-09-08 |
Jakub Jelinek | i386: Fix up xorsign for AVX [PR89984] |
blob | commitdiff | raw | diff to current |
2021-09-08 |
Jakub Jelinek | i386: Fix up @xorsign<mode>3_1 [PR102224] |
blob | commitdiff | raw | diff to current |
2021-09-08 |
liuhongt | AVX512FP16: Support vector init/broadcast/set/extract... |
blob | commitdiff | raw | diff to current |
2021-09-08 |
Guo, Xuepeng | AVX512FP16: Initial support for AVX512FP16 feature... |
blob | commitdiff | raw | diff to current |
2021-09-07 |
H.J. Lu | x86: Enable FMA in unsigned SI to SF expanders |
blob | commitdiff | raw | diff to current |
2021-09-06 |
H.J. Lu | x86: Add non-destructive source to @xorsign<mode>3_1 |
blob | commitdiff | raw | diff to current |
2021-08-24 |
H.J. Lu | x86: Broadcast from integer to a pseudo vector register |
blob | commitdiff | raw | diff to current |
2021-08-16 |
liuhongt | Optimize __builtin_shuffle_vector. |
blob | commitdiff | raw | diff to current |
2021-08-14 |
Jakub Jelinek | i386: Fix ICE with V64QImode broadcast permutation... |
blob | commitdiff | raw | diff to current |
2021-08-12 |
Jakub Jelinek | i386: Fix up V32HImode permutations with -mno-avx512bw... |
blob | commitdiff | raw | diff to current |
2021-08-10 |
Jakub Jelinek | i386: Allow some V32HImode and V64QImode permutations... |
blob | commitdiff | raw | diff to current |
2021-08-04 |
H.J. Lu | x86: Avoid stack realignment when copying data with... |
blob | commitdiff | raw | diff to current |
2021-08-02 |
H.J. Lu | x86: Avoid stack realignment when copying data |
blob | commitdiff | raw | diff to current |
2021-07-22 |
liuhongt | Remove pass_cpb which is related to enable avx512 embed... |
blob | commitdiff | raw | diff to current |
2021-07-09 |
Iain Sandoe | Darwin, X86: Adjust call clobbers to allow for lazy... |
blob | commitdiff | raw | diff to current |
2021-07-08 |
Uros Bizjak | i386: Add pack/unpack patterns for 32bit vectors [PR100637] |
blob | commitdiff | raw | diff to current |
2021-07-05 |
Uros Bizjak | i386: Implement 4-byte vector (V4QI/V2HI) constant... |
blob | commitdiff | raw | diff to current |
2021-07-02 |
Jakub Jelinek | i386: Punt on broadcasts from TImode integers [PR101286] |
blob | commitdiff | raw | diff to current |
2021-07-02 |
Hongyu Wang | Clear odata for aes(enc|dec)(wide)?kl intrinsics when... |
blob | commitdiff | raw | diff to current |
2021-07-01 |
H.J. Lu | x86: Add vec_duplicate<mode> expander |
blob | commitdiff | raw | diff to current |
2021-07-01 |
H.J. Lu | x86: Convert CONST_WIDE_INT/CONST_VECTOR to broadcast |
blob | commitdiff | raw | diff to current |
2021-07-01 |
Uros Bizjak | Change the type of predicates to bool. |
blob | commitdiff | raw | diff to current |
2021-06-24 |
Uros Bizjak | i386: Add pack/unpack patterns for 64bit vectors [PR89021] |
blob | commitdiff | raw | diff to current |
2021-06-24 |
liuhongt | i386: Add vashlm3/vashrm3/vlshrm3 to enable vectorizati... |
blob | commitdiff | raw | diff to current |
2021-06-23 |
Uros Bizjak | i386: Add PPERM two-operand 64bit vector permutation... |
blob | commitdiff | raw | diff to current |
2021-06-22 |
Jakub Jelinek | i386: Use xor to write zero to memory with -Os even... |
blob | commitdiff | raw | diff to current |
2021-06-22 |
liuhongt | Fix ICE for vpexpand*. |
blob | commitdiff | raw | diff to current |
2021-06-17 |
Uros Bizjak | i386: Add variable vec_set for 64bit vectors [PR97194] |
blob | commitdiff | raw | diff to current |
2021-06-16 |
Uros Bizjak | ii386: Add missing two element 64bit vector permutation... |
blob | commitdiff | raw | diff to current |
2021-06-14 |
Uros Bizjak | i386: Split V2HImode *punpckwd to SSE instruction ... |
blob | commitdiff | raw | diff to current |
2021-06-13 |
Uros Bizjak | i386: Improve variable permutation insn avoidance ... |
blob | commitdiff | raw | diff to current |
2021-06-11 |
Uros Bizjak | i386: Try to avoid variable permutation instruction... |
blob | commitdiff | raw | diff to current |
2021-06-10 |
Peter Bergner | i386: Add V8QI and other 64bit vector permutations... |
blob | commitdiff | raw | diff to current |
2021-06-07 |
Uros Bizjak | i386: Add init pattern for V4QI vectors [PR100637] |
blob | commitdiff | raw | diff to current |
2021-06-07 |
liuhongt | Fix _mm256_zeroupper by representing the instructions... |
blob | commitdiff | raw | diff to current |
2021-06-04 |
Uros Bizjak | i386: Add init pattern for V2HI vectors [PR100637] |
blob | commitdiff | raw | diff to current |
2021-06-04 |
Jakub Jelinek | x86: Fix ix86_expand_vector_init for V*TImode [PR100887] |
blob | commitdiff | raw | diff to current |
2021-06-03 |
Uros Bizjak | i386: Add insert and extract patterns for 4-byte vector... |
blob | commitdiff | raw | diff to current |
2021-05-27 |
Uros Bizjak | i386: Add XOP comparisons for 4- and 8-byte vectors... |
blob | commitdiff | raw | diff to current |
2021-05-21 |
Uros Bizjak | i386: Add comparisons for 4-byte vectors [PR100637] |
blob | commitdiff | raw | diff to current |
2021-05-18 |
Uros Bizjak | i386: Fix split_double_mode with paradoxical subreg... |
blob | commitdiff | raw | diff to current |
2021-05-13 |
Jakub Jelinek | ix86: Support V{2, 4}DImode arithmetic right shifts... |
blob | commitdiff | raw | diff to current |
2021-05-13 |
Uros Bizjak | i386: Fix up V2SFmode vcond* with -mxop [PR100581] |
blob | commitdiff | raw | diff to current |
2021-05-12 |
Uros Bizjak | i386: Implement FP vector compares for V2SFmode [PR98218] |
blob | commitdiff | raw | diff to current |
2021-05-10 |
Uros Bizjak | i386: Force V2SI mode operands to registers in expand_s... |
blob | commitdiff | raw | diff to current |
2021-05-07 |
Uros Bizjak | i386: Implement mmx_pblendv to optimize SSE conditional... |
blob | commitdiff | raw | diff to current |
2021-05-07 |
Uros Bizjak | i386: Do not emit mask compares for mode sizes < 16... |
blob | commitdiff | raw | diff to current |
2021-05-05 |
Uros Bizjak | i386: Implement integer vector compares for 64bit vecto... |
blob | commitdiff | raw | diff to current |
2021-04-29 |
Uros Bizjak | i386: Optimize carry flag comparisons a bit |
blob | commitdiff | raw | diff to current |
2021-04-29 |
Uros Bizjak | i386: Mark x86 masked load builtins pure [PR100312] |
blob | commitdiff | raw | diff to current |
2021-04-22 |
Uros Bizjak | i386: Fix unsigned int -> double conversion on i386... |
blob | commitdiff | raw | diff to current |
2021-04-21 |
Martin Liska | Remove TARGET_foo (ix86_tune == PROCESSOR_foo) macros. |
blob | commitdiff | raw | diff to current |
2021-03-31 |
H.J. Lu | x86: Update memcpy/memset inline strategies for Ice... |
blob | commitdiff | raw | diff to current |
2021-03-16 |
Jakub Jelinek | i386: Avoid mutual recursion between two peephole2s... |
blob | commitdiff | raw | diff to current |
2021-03-16 |
Jakub Jelinek | i386: Fix up _mm256_vzeroupper() handling [PR99563] |
blob | commitdiff | raw | diff to current |
2021-02-05 |
liuhongt | Fix ICE: Don't generate integer mask comparision for... |
blob | commitdiff | raw | diff to current |
2021-01-22 |
liuhongt | Lower AVX512 vector comparison to AVX version when... |
blob | commitdiff | raw | diff to current |
2021-01-13 |
Jakub Jelinek | i386, expand: Optimize also 256-bit and 512-bit permuta... |
blob | commitdiff | raw | diff to current |
2021-01-12 |
liuhongt | Delete dead code in ix86_expand_sse_comi. |
blob | commitdiff | raw | diff to current |
2021-01-04 |
Jakub Jelinek | Update copyright years. |
blob | commitdiff | raw | diff to current |
2020-12-29 |
Uros Bizjak | i386: Rounding functions TLC |
blob | commitdiff | raw | diff to current |
next |