Lines Matching refs:xop
2 ; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx,+xop < %s | FileCheck …
22 %2 = call <2 x double> @llvm.x86.xop.vfrcz.pd(<2 x double> %a0)
25 declare <2 x double> @llvm.x86.xop.vfrcz.pd(<2 x double>) nounwind readnone
37 %2 = call <4 x double> @llvm.x86.xop.vfrcz.pd.256(<4 x double> %a0)
40 declare <4 x double> @llvm.x86.xop.vfrcz.pd.256(<4 x double>) nounwind readnone
52 %2 = call <4 x float> @llvm.x86.xop.vfrcz.ps(<4 x float> %a0)
55 declare <4 x float> @llvm.x86.xop.vfrcz.ps(<4 x float>) nounwind readnone
67 %2 = call <8 x float> @llvm.x86.xop.vfrcz.ps.256(<8 x float> %a0)
70 declare <8 x float> @llvm.x86.xop.vfrcz.ps.256(<8 x float>) nounwind readnone
82 %2 = call <2 x double> @llvm.x86.xop.vfrcz.sd(<2 x double> %a0)
85 declare <2 x double> @llvm.x86.xop.vfrcz.sd(<2 x double>) nounwind readnone
97 %2 = call <4 x float> @llvm.x86.xop.vfrcz.ss(<4 x float> %a0)
100 declare <4 x float> @llvm.x86.xop.vfrcz.ss(<4 x float>) nounwind readnone
112 %2 = call <2 x i64> @llvm.x86.xop.vpcmov(<2 x i64> %a0, <2 x i64> %a1, <2 x i64> %a2)
125 %2 = call <2 x i64> @llvm.x86.xop.vpcmov(<2 x i64> %a0, <2 x i64> %a2, <2 x i64> %a1)
128 declare <2 x i64> @llvm.x86.xop.vpcmov(<2 x i64>, <2 x i64>, <2 x i64>) nounwind readnone
140 %2 = call <4 x i64> @llvm.x86.xop.vpcmov.256(<4 x i64> %a0, <4 x i64> %a1, <4 x i64> %a2)
153 %2 = call <4 x i64> @llvm.x86.xop.vpcmov.256(<4 x i64> %a0, <4 x i64> %a2, <4 x i64> %a1)
156 declare <4 x i64> @llvm.x86.xop.vpcmov.256(<4 x i64>, <4 x i64>, <4 x i64>) nounwind readnone
168 %2 = call <16 x i8> @llvm.x86.xop.vpcomb(<16 x i8> %a0, <16 x i8> %a1, i8 0)
171 declare <16 x i8> @llvm.x86.xop.vpcomb(<16 x i8>, <16 x i8>, i8) nounwind readnone
183 %2 = call <4 x i32> @llvm.x86.xop.vpcomd(<4 x i32> %a0, <4 x i32> %a1, i8 0)
186 declare <4 x i32> @llvm.x86.xop.vpcomd(<4 x i32>, <4 x i32>, i8) nounwind readnone
198 %2 = call <2 x i64> @llvm.x86.xop.vpcomq(<2 x i64> %a0, <2 x i64> %a1, i8 0)
201 declare <2 x i64> @llvm.x86.xop.vpcomq(<2 x i64>, <2 x i64>, i8) nounwind readnone
213 %2 = call <16 x i8> @llvm.x86.xop.vpcomub(<16 x i8> %a0, <16 x i8> %a1, i8 0)
216 declare <16 x i8> @llvm.x86.xop.vpcomub(<16 x i8>, <16 x i8>, i8) nounwind readnone
228 %2 = call <4 x i32> @llvm.x86.xop.vpcomud(<4 x i32> %a0, <4 x i32> %a1, i8 0)
231 declare <4 x i32> @llvm.x86.xop.vpcomud(<4 x i32>, <4 x i32>, i8) nounwind readnone
243 %2 = call <2 x i64> @llvm.x86.xop.vpcomuq(<2 x i64> %a0, <2 x i64> %a1, i8 0)
246 declare <2 x i64> @llvm.x86.xop.vpcomuq(<2 x i64>, <2 x i64>, i8) nounwind readnone
258 %2 = call <8 x i16> @llvm.x86.xop.vpcomuw(<8 x i16> %a0, <8 x i16> %a1, i8 0)
261 declare <8 x i16> @llvm.x86.xop.vpcomuw(<8 x i16>, <8 x i16>, i8) nounwind readnone
273 %2 = call <8 x i16> @llvm.x86.xop.vpcomw(<8 x i16> %a0, <8 x i16> %a1, i8 0)
276 declare <8 x i16> @llvm.x86.xop.vpcomw(<8 x i16>, <8 x i16>, i8) nounwind readnone
288 …%2 = call <2 x double> @llvm.x86.xop.vpermil2pd(<2 x double> %a0, <2 x double> %a1, <2 x i64> %a2,…
301 …%2 = call <2 x double> @llvm.x86.xop.vpermil2pd(<2 x double> %a0, <2 x double> %a2, <2 x i64> %a1,…
304 declare <2 x double> @llvm.x86.xop.vpermil2pd(<2 x double>, <2 x double>, <2 x i64>, i8) nounwind r…
316 …%2 = call <4 x double> @llvm.x86.xop.vpermil2pd.256(<4 x double> %a0, <4 x double> %a1, <4 x i64> …
329 …%2 = call <4 x double> @llvm.x86.xop.vpermil2pd.256(<4 x double> %a0, <4 x double> %a2, <4 x i64> …
332 declare <4 x double> @llvm.x86.xop.vpermil2pd.256(<4 x double>, <4 x double>, <4 x i64>, i8) nounwi…
344 …%2 = call <4 x float> @llvm.x86.xop.vpermil2ps(<4 x float> %a0, <4 x float> %a1, <4 x i32> %a2, i8…
357 …%2 = call <4 x float> @llvm.x86.xop.vpermil2ps(<4 x float> %a0, <4 x float> %a2, <4 x i32> %a1, i8…
360 declare <4 x float> @llvm.x86.xop.vpermil2ps(<4 x float>, <4 x float>, <4 x i32>, i8) nounwind read…
372 …%2 = call <8 x float> @llvm.x86.xop.vpermil2ps.256(<8 x float> %a0, <8 x float> %a1, <8 x i32> %a2…
385 …%2 = call <8 x float> @llvm.x86.xop.vpermil2ps.256(<8 x float> %a0, <8 x float> %a2, <8 x i32> %a1…
388 declare <8 x float> @llvm.x86.xop.vpermil2ps.256(<8 x float>, <8 x float>, <8 x i32>, i8) nounwind …
400 %2 = call <4 x i32> @llvm.x86.xop.vphaddbd(<16 x i8> %a0)
403 declare <4 x i32> @llvm.x86.xop.vphaddbd(<16 x i8>) nounwind readnone
415 %2 = call <2 x i64> @llvm.x86.xop.vphaddbq(<16 x i8> %a0)
418 declare <2 x i64> @llvm.x86.xop.vphaddbq(<16 x i8>) nounwind readnone
430 %2 = call <8 x i16> @llvm.x86.xop.vphaddbw(<16 x i8> %a0)
433 declare <8 x i16> @llvm.x86.xop.vphaddbw(<16 x i8>) nounwind readnone
445 %2 = call <2 x i64> @llvm.x86.xop.vphadddq(<4 x i32> %a0)
448 declare <2 x i64> @llvm.x86.xop.vphadddq(<4 x i32>) nounwind readnone
460 %2 = call <4 x i32> @llvm.x86.xop.vphaddubd(<16 x i8> %a0)
463 declare <4 x i32> @llvm.x86.xop.vphaddubd(<16 x i8>) nounwind readnone
475 %2 = call <2 x i64> @llvm.x86.xop.vphaddubq(<16 x i8> %a0)
478 declare <2 x i64> @llvm.x86.xop.vphaddubq(<16 x i8>) nounwind readnone
490 %2 = call <8 x i16> @llvm.x86.xop.vphaddubw(<16 x i8> %a0)
493 declare <8 x i16> @llvm.x86.xop.vphaddubw(<16 x i8>) nounwind readnone
505 %2 = call <2 x i64> @llvm.x86.xop.vphaddudq(<4 x i32> %a0)
508 declare <2 x i64> @llvm.x86.xop.vphaddudq(<4 x i32>) nounwind readnone
520 %2 = call <4 x i32> @llvm.x86.xop.vphadduwd(<8 x i16> %a0)
523 declare <4 x i32> @llvm.x86.xop.vphadduwd(<8 x i16>) nounwind readnone
535 %2 = call <2 x i64> @llvm.x86.xop.vphadduwq(<8 x i16> %a0)
538 declare <2 x i64> @llvm.x86.xop.vphadduwq(<8 x i16>) nounwind readnone
550 %2 = call <4 x i32> @llvm.x86.xop.vphaddwd(<8 x i16> %a0)
553 declare <4 x i32> @llvm.x86.xop.vphaddwd(<8 x i16>) nounwind readnone
565 %2 = call <2 x i64> @llvm.x86.xop.vphaddwq(<8 x i16> %a0)
568 declare <2 x i64> @llvm.x86.xop.vphaddwq(<8 x i16>) nounwind readnone
580 %2 = call <8 x i16> @llvm.x86.xop.vphsubbw(<16 x i8> %a0)
583 declare <8 x i16> @llvm.x86.xop.vphsubbw(<16 x i8>) nounwind readnone
595 %2 = call <2 x i64> @llvm.x86.xop.vphsubdq(<4 x i32> %a0)
598 declare <2 x i64> @llvm.x86.xop.vphsubdq(<4 x i32>) nounwind readnone
610 %2 = call <4 x i32> @llvm.x86.xop.vphsubwd(<8 x i16> %a0)
613 declare <4 x i32> @llvm.x86.xop.vphsubwd(<8 x i16>) nounwind readnone
627 %2 = call <4 x i32> @llvm.x86.xop.vpmacsdd(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
630 declare <4 x i32> @llvm.x86.xop.vpmacsdd(<4 x i32>, <4 x i32>, <4 x i32>) nounwind readnone
644 %2 = call <2 x i64> @llvm.x86.xop.vpmacsdqh(<4 x i32> %a0, <4 x i32> %a1, <2 x i64> %a2)
647 declare <2 x i64> @llvm.x86.xop.vpmacsdqh(<4 x i32>, <4 x i32>, <2 x i64>) nounwind readnone
661 %2 = call <2 x i64> @llvm.x86.xop.vpmacsdql(<4 x i32> %a0, <4 x i32> %a1, <2 x i64> %a2)
664 declare <2 x i64> @llvm.x86.xop.vpmacsdql(<4 x i32>, <4 x i32>, <2 x i64>) nounwind readnone
678 %2 = call <4 x i32> @llvm.x86.xop.vpmacssdd(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
681 declare <4 x i32> @llvm.x86.xop.vpmacssdd(<4 x i32>, <4 x i32>, <4 x i32>) nounwind readnone
695 %2 = call <2 x i64> @llvm.x86.xop.vpmacssdqh(<4 x i32> %a0, <4 x i32> %a1, <2 x i64> %a2)
698 declare <2 x i64> @llvm.x86.xop.vpmacssdqh(<4 x i32>, <4 x i32>, <2 x i64>) nounwind readnone
712 %2 = call <2 x i64> @llvm.x86.xop.vpmacssdql(<4 x i32> %a0, <4 x i32> %a1, <2 x i64> %a2)
715 declare <2 x i64> @llvm.x86.xop.vpmacssdql(<4 x i32>, <4 x i32>, <2 x i64>) nounwind readnone
729 %2 = call <4 x i32> @llvm.x86.xop.vpmacsswd(<8 x i16> %a0, <8 x i16> %a1, <4 x i32> %a2)
732 declare <4 x i32> @llvm.x86.xop.vpmacsswd(<8 x i16>, <8 x i16>, <4 x i32>) nounwind readnone
746 %2 = call <8 x i16> @llvm.x86.xop.vpmacssww(<8 x i16> %a0, <8 x i16> %a1, <8 x i16> %a2)
749 declare <8 x i16> @llvm.x86.xop.vpmacssww(<8 x i16>, <8 x i16>, <8 x i16>) nounwind readnone
763 %2 = call <4 x i32> @llvm.x86.xop.vpmacswd(<8 x i16> %a0, <8 x i16> %a1, <4 x i32> %a2)
766 declare <4 x i32> @llvm.x86.xop.vpmacswd(<8 x i16>, <8 x i16>, <4 x i32>) nounwind readnone
780 %2 = call <8 x i16> @llvm.x86.xop.vpmacsww(<8 x i16> %a0, <8 x i16> %a1, <8 x i16> %a2)
783 declare <8 x i16> @llvm.x86.xop.vpmacsww(<8 x i16>, <8 x i16>, <8 x i16>) nounwind readnone
797 %2 = call <4 x i32> @llvm.x86.xop.vpmadcsswd(<8 x i16> %a0, <8 x i16> %a1, <4 x i32> %a2)
800 declare <4 x i32> @llvm.x86.xop.vpmadcsswd(<8 x i16>, <8 x i16>, <4 x i32>) nounwind readnone
814 %2 = call <4 x i32> @llvm.x86.xop.vpmadcswd(<8 x i16> %a0, <8 x i16> %a1, <4 x i32> %a2)
817 declare <4 x i32> @llvm.x86.xop.vpmadcswd(<8 x i16>, <8 x i16>, <4 x i32>) nounwind readnone
829 %2 = call <16 x i8> @llvm.x86.xop.vpperm(<16 x i8> %a0, <16 x i8> %a1, <16 x i8> %a2)
842 %2 = call <16 x i8> @llvm.x86.xop.vpperm(<16 x i8> %a0, <16 x i8> %a2, <16 x i8> %a1)
845 declare <16 x i8> @llvm.x86.xop.vpperm(<16 x i8>, <16 x i8>, <16 x i8>) nounwind readnone
857 %2 = call <16 x i8> @llvm.x86.xop.vprotbi(<16 x i8> %a0, i8 7)
860 declare <16 x i8> @llvm.x86.xop.vprotbi(<16 x i8>, i8) nounwind readnone
872 %2 = call <16 x i8> @llvm.x86.xop.vprotb(<16 x i8> %a0, <16 x i8> %a1)
885 %2 = call <16 x i8> @llvm.x86.xop.vprotb(<16 x i8> %a1, <16 x i8> %a0)
888 declare <16 x i8> @llvm.x86.xop.vprotb(<16 x i8>, <16 x i8>) nounwind readnone
900 %2 = call <4 x i32> @llvm.x86.xop.vprotdi(<4 x i32> %a0, i8 7)
903 declare <4 x i32> @llvm.x86.xop.vprotdi(<4 x i32>, i8) nounwind readnone
915 %2 = call <4 x i32> @llvm.x86.xop.vprotd(<4 x i32> %a0, <4 x i32> %a1)
928 %2 = call <4 x i32> @llvm.x86.xop.vprotd(<4 x i32> %a1, <4 x i32> %a0)
931 declare <4 x i32> @llvm.x86.xop.vprotd(<4 x i32>, <4 x i32>) nounwind readnone
943 %2 = call <2 x i64> @llvm.x86.xop.vprotqi(<2 x i64> %a0, i8 7)
946 declare <2 x i64> @llvm.x86.xop.vprotqi(<2 x i64>, i8) nounwind readnone
958 %2 = call <2 x i64> @llvm.x86.xop.vprotq(<2 x i64> %a0, <2 x i64> %a1)
971 %2 = call <2 x i64> @llvm.x86.xop.vprotq(<2 x i64> %a1, <2 x i64> %a0)
974 declare <2 x i64> @llvm.x86.xop.vprotq(<2 x i64>, <2 x i64>) nounwind readnone
986 %2 = call <8 x i16> @llvm.x86.xop.vprotwi(<8 x i16> %a0, i8 7)
989 declare <8 x i16> @llvm.x86.xop.vprotwi(<8 x i16>, i8) nounwind readnone
1001 %2 = call <8 x i16> @llvm.x86.xop.vprotw(<8 x i16> %a0, <8 x i16> %a1)
1014 %2 = call <8 x i16> @llvm.x86.xop.vprotw(<8 x i16> %a1, <8 x i16> %a0)
1017 declare <8 x i16> @llvm.x86.xop.vprotw(<8 x i16>, <8 x i16>) nounwind readnone
1029 %2 = call <16 x i8> @llvm.x86.xop.vpshab(<16 x i8> %a0, <16 x i8> %a1)
1042 %2 = call <16 x i8> @llvm.x86.xop.vpshab(<16 x i8> %a1, <16 x i8> %a0)
1045 declare <16 x i8> @llvm.x86.xop.vpshab(<16 x i8>, <16 x i8>) nounwind readnone
1057 %2 = call <4 x i32> @llvm.x86.xop.vpshad(<4 x i32> %a0, <4 x i32> %a1)
1070 %2 = call <4 x i32> @llvm.x86.xop.vpshad(<4 x i32> %a1, <4 x i32> %a0)
1073 declare <4 x i32> @llvm.x86.xop.vpshad(<4 x i32>, <4 x i32>) nounwind readnone
1085 %2 = call <2 x i64> @llvm.x86.xop.vpshaq(<2 x i64> %a0, <2 x i64> %a1)
1098 %2 = call <2 x i64> @llvm.x86.xop.vpshaq(<2 x i64> %a1, <2 x i64> %a0)
1101 declare <2 x i64> @llvm.x86.xop.vpshaq(<2 x i64>, <2 x i64>) nounwind readnone
1113 %2 = call <8 x i16> @llvm.x86.xop.vpshaw(<8 x i16> %a0, <8 x i16> %a1)
1126 %2 = call <8 x i16> @llvm.x86.xop.vpshaw(<8 x i16> %a1, <8 x i16> %a0)
1129 declare <8 x i16> @llvm.x86.xop.vpshaw(<8 x i16>, <8 x i16>) nounwind readnone
1141 %2 = call <16 x i8> @llvm.x86.xop.vpshlb(<16 x i8> %a0, <16 x i8> %a1)
1154 %2 = call <16 x i8> @llvm.x86.xop.vpshlb(<16 x i8> %a1, <16 x i8> %a0)
1157 declare <16 x i8> @llvm.x86.xop.vpshlb(<16 x i8>, <16 x i8>) nounwind readnone
1169 %2 = call <4 x i32> @llvm.x86.xop.vpshld(<4 x i32> %a0, <4 x i32> %a1)
1182 %2 = call <4 x i32> @llvm.x86.xop.vpshld(<4 x i32> %a1, <4 x i32> %a0)
1185 declare <4 x i32> @llvm.x86.xop.vpshld(<4 x i32>, <4 x i32>) nounwind readnone
1197 %2 = call <2 x i64> @llvm.x86.xop.vpshlq(<2 x i64> %a0, <2 x i64> %a1)
1210 %2 = call <2 x i64> @llvm.x86.xop.vpshlq(<2 x i64> %a1, <2 x i64> %a0)
1213 declare <2 x i64> @llvm.x86.xop.vpshlq(<2 x i64>, <2 x i64>) nounwind readnone
1225 %2 = call <8 x i16> @llvm.x86.xop.vpshlw(<8 x i16> %a0, <8 x i16> %a1)
1238 %2 = call <8 x i16> @llvm.x86.xop.vpshlw(<8 x i16> %a1, <8 x i16> %a0)
1241 declare <8 x i16> @llvm.x86.xop.vpshlw(<8 x i16>, <8 x i16>) nounwind readnone