aboutsummaryrefslogtreecommitdiff
path: root/gcc/config/i386/mmx.md
AgeCommit message (Expand)AuthorFilesLines
2022-01-17Change references of .c files to .cc filesMartin Liska1-4/+4
2022-01-13i386: Introduce V2QImode vectorized shifts [PR103861]Uros Bizjak1-4/+40
2022-01-13i386: Cleanup V2QI arithmetic instructionsUros Bizjak1-28/+28
2022-01-13ii386: Add 16-bit vector modes to xop_pcmov [PR104003]Uros Bizjak1-5/+5
2022-01-12i386: Add CC clobber and splits for 32-bit vector mode logic insns [PR100673,...Uros Bizjak1-106/+69
2022-01-11i386: Introduce V2QImode vector cmove for -msse4.1 [PR103861]Uros Bizjak1-27/+27
2022-01-10i386: Introduce V2QImode vector compares [PR103861]Uros Bizjak1-28/+28
2022-01-07i386: Robustify V2QI and V4QI move patternsUros Bizjak1-4/+12
2022-01-06i386: Improve HImode interunit movesUros Bizjak1-9/+9
2022-01-05i386: Introduce V2QImode minmax, abs and uavgv2hi3_ceil [PR103861]Uros Bizjak1-14/+41
2022-01-05i386: Fix type of one_cmplv2qi2 alternatives 1,2 [PR103915]Uros Bizjak1-1/+1
2022-01-04i386: Introduce V2QImode vectorized logic [PR103861]Uros Bizjak1-0/+146
2022-01-03i386: Always enable mov<V_32:mode> patterns [PR103894]Uros Bizjak1-17/+6
2022-01-03Update copyright years.Jakub Jelinek1-1/+1
2022-01-02i386: Introduce V2QImode vectorized arithmetic [PR103861]Uros Bizjak1-2/+282
2021-12-28i386: Fix handling of SUBREGs in divv2sf3 [PR103842]Jakub Jelinek1-4/+6
2021-12-24i386: Add V2SFmode DIV insn pattern [PR95046, PR103797]Uros Bizjak1-0/+20
2021-10-12Support reduc_{plus,smax,smin,umax,umin}_scal_v4qi.liuhongt1-0/+45
2021-10-08Support reduc_{plus,smax,smin,umax,min}_scal_v4hi.liuhongt1-0/+36
2021-09-28AVX512FP16: Support basic 64/32bit vector type and operation.Hongyu Wang1-9/+43
2021-09-22Support 64bit fma/fms/fnma/fnms under avx512vl.liuhongt1-8/+12
2021-08-09i386: Name V2SF logic insns [PR101812]Uros Bizjak1-1/+1
2021-07-12i386: Fix vec_set<mode> expanders [PR101424]Uros Bizjak1-1/+1
2021-07-08i386: Add pack/unpack patterns for 32bit vectors [PR100637]Uros Bizjak1-15/+85
2021-07-06i386: Add variable vec_set for 32bit vectors [PR97194]Uros Bizjak1-6/+12
2021-07-05i386: Implement 4-byte vector (V4QI/V2HI) constant permutationsUros Bizjak1-2/+84
2021-06-29i386: Add V2SFmode vec_addsub pattern [PR95046]Uros Bizjak1-0/+18
2021-06-24i386: Add pack/unpack patterns for 64bit vectors [PR89021]Uros Bizjak1-0/+72
2021-06-23i386: Add PPERM two-operand 64bit vector permutation [PR89021]Uros Bizjak1-0/+13
2021-06-17i386: Add variable vec_set for 64bit vectors [PR97194]Uros Bizjak1-12/+24
2021-06-16ii386: Add missing two element 64bit vector permutations [PR89021]Uros Bizjak1-0/+16
2021-06-14i386: Split V2HImode *punpckwd to SSE instruction [PR101058]Uros Bizjak1-6/+7
2021-06-10i386: Add V8QI and other 64bit vector permutations [PR89021]Peter Bergner1-8/+78
2021-06-07i386: Add init pattern for V4QI vectors [PR100637]Uros Bizjak1-1/+11
2021-06-04i386: Add init pattern for V2HI vectors [PR100637]Uros Bizjak1-0/+82
2021-06-03i386: Add insert and extract patterns for 4-byte vectors [PR100637]Uros Bizjak1-2/+174
2021-05-27i386: Remove unneeded binary operand fixup from expanders.Uros Bizjak1-147/+52
2021-05-27i386: Add XOP comparisons for 4- and 8-byte vectors [PR100637]Uros Bizjak1-0/+56
2021-05-27i386: Add uavg_ceil patterns for 4-byte vectors [PR100637]Uros Bizjak1-0/+41
2021-05-23i386: Add push insns for 4-byte vectors [PR100722]Uros Bizjak1-0/+33
2021-05-21i386: Add comparisons for 4-byte vectors [PR100637]Uros Bizjak1-1/+139
2021-05-21i386: Add minmax and abs patterns for 4-byte vectors [PR100637]Uros Bizjak1-28/+145
2021-05-20i386: Avoid integer logic insns for 32bit and 64bit vector modes [PR100701]Uros Bizjak1-34/+26
2021-05-20i386: Add mult-high and shift patterns for 4-byte vectors [PR100637]Uros Bizjak1-5/+73
2021-05-19i386: Allow 64bit vector modes in general registersUros Bizjak1-14/+19
2021-05-18i386: Implement 4-byte vector support [PR100637]Uros Bizjak1-4/+191
2021-05-13i386: Fix up V2SFmode vcond* with -mxop [PR100581]Uros Bizjak1-5/+5
2021-05-12i386: Implement FP vector compares for V2SFmode [PR98218]Uros Bizjak1-18/+112
2021-05-07i386: Implement mmx_pblendv to optimize SSE conditional moves [PR98218]Uros Bizjak1-0/+20
2021-05-07i386: Fix up 8-byte vcond* with -mxop [PR100445]Jakub Jelinek1-0/+11