Searched refs:avx2 (Results 1 – 25 of 156) sorted by relevance
1234567
1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -march=x86 -mattr=avx2 | FileCheck %s5 …%res = call <16 x i16> @llvm.x86.avx2.packssdw(<8 x i32> %a0, <8 x i32> %a1) ; <<16 x i16>> [#uses…8 declare <16 x i16> @llvm.x86.avx2.packssdw(<8 x i32>, <8 x i32>) nounwind readnone13 …%res = call <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16> %a0, <16 x i16> %a1) ; <<32 x i8>> [#uses…16 declare <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16>, <16 x i16>) nounwind readnone21 …%res = call <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16> %a0, <16 x i16> %a1) ; <<32 x i8>> [#uses…24 declare <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16>, <16 x i16>) nounwind readnone29 %res = call <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]32 declare <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8>, <32 x i8>) nounwind readnone37 …%res = call <16 x i16> @llvm.x86.avx2.padds.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#use…[all …]
1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -march=x86 -mattr=avx2 | FileCheck %s5 …%res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i32 7) ; <<16 x i16>…8 declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i32) nounwind readnone13 …%res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i32 7) ; <<4 x i32>…16 declare <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32>, <4 x i32>, i32) nounwind readnone21 …%res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i32 7) ; <<8 x i32>…24 declare <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32>, <8 x i32>, i32) nounwind readnone29 …%res = call <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8> %a0, <32 x i8> %a1, i32 7) ; <<16 x i16>> …32 declare <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8>, <32 x i8>, i32) nounwind readnone37 %res = call <4 x i64> @llvm.x86.avx2.psll.dq.bs(<4 x i64> %a0, i32 7) ; <<4 x i64>> [#uses=1][all …]
1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=x86-64 -mattr=+avx2 | FileCheck %s7 %2 = call <16 x i16> @llvm.x86.avx2.pmovsxbw(<16 x i8> %1)15 %2 = call <8 x i32> @llvm.x86.avx2.pmovsxbd(<16 x i8> %1)23 %2 = call <4 x i64> @llvm.x86.avx2.pmovsxbq(<16 x i8> %1)31 %2 = call <8 x i32> @llvm.x86.avx2.pmovsxwd(<8 x i16> %1)39 %2 = call <4 x i64> @llvm.x86.avx2.pmovsxwq(<8 x i16> %1)47 %2 = call <4 x i64> @llvm.x86.avx2.pmovsxdq(<4 x i32> %1)55 %2 = call <16 x i16> @llvm.x86.avx2.pmovzxbw(<16 x i8> %1)63 %2 = call <8 x i32> @llvm.x86.avx2.pmovzxbd(<16 x i8> %1)71 %2 = call <4 x i64> @llvm.x86.avx2.pmovzxbq(<16 x i8> %1)[all …]
1 ; RUN: llc < %s -march=x86-64 -mcpu=core-avx2 | FileCheck %s7 %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a0, i32 7)16 %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a0, i32 7)25 %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a0, i32 7)34 %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i32 0)43 %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i32 0)52 %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i32 0)61 %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i32 -1)70 %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i32 -1)79 %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i32 -1)[all …]
2 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core-avx2 | FileCheck %s4 declare <4 x float> @llvm.x86.avx2.gather.d.ps(<4 x float>, i8*,9 %res = call <4 x float> @llvm.x86.avx2.gather.d.ps(<4 x float> undef,21 declare <2 x double> @llvm.x86.avx2.gather.d.pd(<2 x double>, i8*,26 %res = call <2 x double> @llvm.x86.avx2.gather.d.pd(<2 x double> undef,36 declare <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float>, i8*,41 %res = call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> undef,49 declare <4 x double> @llvm.x86.avx2.gather.d.pd.256(<4 x double>, i8*,54 %res = call <4 x double> @llvm.x86.avx2.gather.d.pd.256(<4 x double> undef,
1 ; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mattr=+avx2 < %s | FileCheck %s20 declare <4 x double> @llvm.x86.avx2.vbroadcast.sd.pd.256(<2 x double>) nounwind readonly31 declare <4 x float> @llvm.x86.avx2.vbroadcast.ss.ps(<4 x float>) nounwind readonly42 declare <8 x float> @llvm.x86.avx2.vbroadcast.ss.ps.256(<4 x float>) nounwind readonly68 %2 = call <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8> %a0, <32 x i8> %a1, i8 7)71 declare <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8>, <32 x i8>, i8) nounwind readnone77 %2 = call <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8> %a0)80 declare <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8>) nounwind readnone86 %2 = call <8 x i32> @llvm.x86.avx2.pabs.d(<8 x i32> %a0)89 declare <8 x i32> @llvm.x86.avx2.pabs.d(<8 x i32>) nounwind readnone[all …]
1 ; RUN: llc -O3 -mtriple=x86_64-unknown -mcpu=core-avx2 -mattr=avx2 < %s | FileCheck %s16 %2 = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %1, <16 x i16> %a, i8 17)23 declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i8) nounwind readnone27 %2 = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %1, <4 x i32> %a, i8 1)34 declare <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32>, <4 x i32>, i8) nounwind readnone38 %2 = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %1, <8 x i32> %a, i8 129)45 declare <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32>, <8 x i32>, i8) nounwind readnone
1 ; RUN: llc -mattr=+avx2 < %s | FileCheck %s14 %3 = call <4 x i64> @llvm.x86.avx2.vbroadcasti128(i8* %2)18 declare <4 x i64> @llvm.x86.avx2.vbroadcasti128(i8*) #1
2 ; RUN: llc < %s -march x86-64 -mattr=+avx2 | FileCheck %s15 %2 = tail call <4 x i64> @llvm.x86.avx2.psad.bw(<32 x i8> %0, <32 x i8> %1)19 declare <4 x i64> @llvm.x86.avx2.psad.bw(<32 x i8>, <32 x i8>)
3 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=-avx2,+avx | FileCh…4 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mattr=-avx2,+avx -fast-isel -fast-isel-abort=1 | File…
1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core-avx2 -mattr=+avx2 | FileCheck %s
517 TARGET_BUILTIN(__builtin_ia32_mpsadbw256, "V32cV32cV32cIc", "", "avx2")518 TARGET_BUILTIN(__builtin_ia32_pabsb256, "V32cV32c", "", "avx2")519 TARGET_BUILTIN(__builtin_ia32_pabsw256, "V16sV16s", "", "avx2")520 TARGET_BUILTIN(__builtin_ia32_pabsd256, "V8iV8i", "", "avx2")521 TARGET_BUILTIN(__builtin_ia32_packsswb256, "V32cV16sV16s", "", "avx2")522 TARGET_BUILTIN(__builtin_ia32_packssdw256, "V16sV8iV8i", "", "avx2")523 TARGET_BUILTIN(__builtin_ia32_packuswb256, "V32cV16sV16s", "", "avx2")524 TARGET_BUILTIN(__builtin_ia32_packusdw256, "V16sV8iV8i", "", "avx2")525 TARGET_BUILTIN(__builtin_ia32_paddsb256, "V32cV32cV32c", "", "avx2")526 TARGET_BUILTIN(__builtin_ia32_paddsw256, "V16sV16sV16s", "", "avx2")[all …]
3 ; Check that llvm.x86.avx2.gather.d.ps.256 intrinsic is not eliminated as gather and store memory a…8 declare <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float>, i8*, <8 x i32>, <8 x float>, i8) #016 …%v1 = tail call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> undef, i8* %arr.ptr, <8 x i…19 …%v2 = tail call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> undef, i8* %arr.ptr, <8 x i…25 ; CHECK: llvm.x86.avx2.gather.d.ps.25627 ; CHECK: llvm.x86.avx2.gather.d.ps.25637 …%v1 = tail call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> undef, i8* %arr.ptr, <8 x i…40 …%v2 = tail call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> undef, i8* %arr.ptr, <8 x i…46 ; CHECK: llvm.x86.avx2.gather.d.ps.25648 ; CHECK-NOT: llvm.x86.avx2.gather.d.ps.256[all …]
57 %1 = tail call <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16> %v, i32 0)65 %1 = tail call <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16> %v, i32 15)73 %1 = tail call <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16> %v, i32 64)80 %1 = tail call <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32> %v, i32 0)88 %1 = tail call <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32> %v, i32 15)96 %1 = tail call <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32> %v, i32 64)173 %1 = tail call <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16> %v, i32 0)181 %1 = tail call <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16> %v, i32 15)188 %1 = tail call <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16> %v, i32 64)195 %1 = tail call <8 x i32> @llvm.x86.avx2.psrli.d(<8 x i32> %v, i32 0)[all …]
10 declare <8 x i32> @llvm.x86.avx2.pmovsxbd(<16 x i8>) nounwind readnone11 declare <4 x i64> @llvm.x86.avx2.pmovsxbq(<16 x i8>) nounwind readnone12 declare <16 x i16> @llvm.x86.avx2.pmovsxbw(<16 x i8>) nounwind readnone13 declare <4 x i64> @llvm.x86.avx2.pmovsxdq(<4 x i32>) nounwind readnone14 declare <8 x i32> @llvm.x86.avx2.pmovsxwd(<8 x i16>) nounwind readnone15 declare <4 x i64> @llvm.x86.avx2.pmovsxwq(<8 x i16>) nounwind readnone87 %res = call <8 x i32> @llvm.x86.avx2.pmovsxbd(<16 x i8> %v)97 %res = call <4 x i64> @llvm.x86.avx2.pmovsxbq(<16 x i8> %v)106 %res = call <16 x i16> @llvm.x86.avx2.pmovsxbw(<16 x i8> %v)115 %res = call <4 x i64> @llvm.x86.avx2.pmovsxdq(<4 x i32> %v)[all …]
10 declare <8 x i32> @llvm.x86.avx2.pmovzxbd(<16 x i8>) nounwind readnone11 declare <4 x i64> @llvm.x86.avx2.pmovzxbq(<16 x i8>) nounwind readnone12 declare <16 x i16> @llvm.x86.avx2.pmovzxbw(<16 x i8>) nounwind readnone13 declare <4 x i64> @llvm.x86.avx2.pmovzxdq(<4 x i32>) nounwind readnone14 declare <8 x i32> @llvm.x86.avx2.pmovzxwd(<8 x i16>) nounwind readnone15 declare <4 x i64> @llvm.x86.avx2.pmovzxwq(<8 x i16>) nounwind readnone87 %res = call <8 x i32> @llvm.x86.avx2.pmovzxbd(<16 x i8> %v)97 %res = call <4 x i64> @llvm.x86.avx2.pmovzxbq(<16 x i8> %v)106 %res = call <16 x i16> @llvm.x86.avx2.pmovzxbw(<16 x i8> %v)115 %res = call <4 x i64> @llvm.x86.avx2.pmovzxdq(<4 x i32> %v)[all …]
17 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 0, i8 1, i8 2, i8 …36 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 -128, i8 -128, i8 …60 %1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> zeroinitializer)119 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 -128, i8 1, i8 -12…127 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 -128, i8 -128, i8 …135 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 -128, i8 -128, i8 …143 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 -128, i8 -128, i8 …151 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 0, i8 1, i8 2, i8 …159 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 0, i8 1, i8 -128, …176 …%1 = tail call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %InVec, <32 x i8> <i8 0, i8 1, i8 2, i8 …[all …]
9 declare <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32>, <8 x i32>)10 declare <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32>, <4 x i32>)76 %0 = tail call <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32> %x, <4 x i32> %y)83 ; CHECK: = call <4 x i32> @llvm.x86.avx2.psllv.d(85 ; CHECK: = tail call <4 x i32> @llvm.x86.avx2.psllv.d(90 %0 = tail call <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32> %x, <8 x i32> %y)97 ; CHECK: = call <8 x i32> @llvm.x86.avx2.psllv.d.256(99 ; CHECK: = tail call <8 x i32> @llvm.x86.avx2.psllv.d.256(
7 declare <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16> %a, <16 x i16> %b) nounwind readnone28 %c = tail call <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16> %a, <16 x i16> %b) nounwind37 ; CHECK-DAG: call <32 x i8> @llvm.x86.avx2.packsswb(38 ; CHECK-DAG: call <32 x i8> @llvm.x86.avx2.packuswb(
1 prog: avx2-12 prereq: test -x avx2-1 && ../../../tests/x86_amd64_features amd64-avx
27 avx2-1.vgtest avx2-1.stdout.exp avx2-1.stderr.exp \115 check_PROGRAMS += avx2-1
10 rsaz-avx2.${ASM_EXT}72 perlasm(rsaz-avx2.${ASM_EXT} asm/rsaz-avx2.pl)
1 ; RUN: opt < %s -mtriple=x86_64-apple-darwin -mcpu=broadwell -mattr=+avx2 -S | FileCheck %s6 …ounwind readnone ssp uwtable "target-cpu"="broadwell" "target-features"="+avx2" "use-soft-float"="…
60 info->ia32.avx2 = false; in disable_avx()69 info->x86.avx2 = false; in disable_avx()188 info->ia32.avx2 = (flags_ebx & FLAC__CPUINFO_IA32_CPUID_AVX2 )? true : false; in FLAC__cpu_info()205 fprintf(stderr, " AVX2 ....... %c\n", info->ia32.avx2 ? 'Y' : 'n'); in FLAC__cpu_info()366 info->x86.avx2 = (flags_ebx & FLAC__CPUINFO_IA32_CPUID_AVX2 )? true : false; in FLAC__cpu_info()378 fprintf(stderr, " AVX2 ....... %c\n", info->x86.avx2 ? 'Y' : 'n'); in FLAC__cpu_info()
131 FLAC__bool avx2; member143 FLAC__bool avx2; member