--- /dev/null
+; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx,+aes < %s | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-unknown"
+
+; Stack reload folding tests.
+;
+; By including a nop call with sideeffects we can force a partial register spill of the
+; relevant registers and check that the reload is correctly folded into the instruction.
+
+define <2 x i64> @stack_fold_aesdec(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesdec
+ ;CHECK: vaesdec {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesdec(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesdec(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesdeclast(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesdeclast
+ ;CHECK: vaesdeclast {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesdeclast(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesdeclast(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesenc(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesenc
+ ;CHECK: vaesenc {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesenc(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesenc(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesenclast(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesenclast
+ ;CHECK: vaesenclast {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesenclast(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesenclast(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesimc(<2 x i64> %a0) {
+ ;CHECK-LABEL: stack_fold_aesimc
+ ;CHECK: vaesimc {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesimc(<2 x i64> %a0)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesimc(<2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aeskeygenassist(<2 x i64> %a0) {
+ ;CHECK-LABEL: stack_fold_aeskeygenassist
+ ;CHECK: vaeskeygenassist $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aeskeygenassist(<2 x i64> %a0, i8 7)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aeskeygenassist(<2 x i64>, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pabsb(<16 x i8> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsb
+ ;CHECK: vpabsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.pabs.b.128(<16 x i8> %a0)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.pabs.b.128(<16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pabsd(<4 x i32> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsd
+ ;CHECK: vpabsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.pabs.d.128(<4 x i32> %a0)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.pabs.d.128(<4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pabsw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsw
+ ;CHECK: vpabsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pabs.w.128(<8 x i16> %a0)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pabs.w.128(<8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_packssdw(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_packssdw
+ ;CHECK: vpackssdw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_packsswb(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_packsswb
+ ;CHECK: vpacksswb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_packusdw(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_packusdw
+ ;CHECK: vpackusdw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.packusdw(<4 x i32> %a0, <4 x i32> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.packusdw(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_packuswb(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_packuswb
+ ;CHECK: vpackuswb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_paddb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddb
+ ;CHECK: vpaddb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <16 x i8> %a0, %a1
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_paddd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_paddd
+ ;CHECK: vpaddd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_paddq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_paddq
+ ;CHECK: vpaddq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <2 x i64> %a0, %a1
+ ret <2 x i64> %2
+}
+
+define <16 x i8> @stack_fold_paddsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddsb
+ ;CHECK: vpaddsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.padds.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.padds.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddsw
+ ;CHECK: vpaddsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.padds.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.padds.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_paddusb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddusb
+ ;CHECK: vpaddusb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.paddus.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.paddus.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddusw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddusw
+ ;CHECK: vpaddusw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.paddus.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.paddus.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddw
+ ;CHECK: vpaddw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_palignr(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_palignr
+ ;CHECK: vpalignr $1, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a1, <16 x i8> %a0, <16 x i32> <i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16>
+ ret <16 x i8> %2
+}
+
+define <16 x i8> @stack_fold_pand(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pand
+ ;CHECK: vpand {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = and <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
+
+define <16 x i8> @stack_fold_pandn(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pandn
+ ;CHECK: vpandn {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = xor <16 x i8> %a0, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+ %3 = and <16 x i8> %2, %a1
+ ; add forces execution domain
+ %4 = add <16 x i8> %3, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %4
+}
+
+define <16 x i8> @stack_fold_pavgb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pavgb
+ ;CHECK: vpavgb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_pavgw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pavgw
+ ;CHECK: vpavgw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pblendvb
+declare <16 x i8> @llvm.x86.sse41.pblendvb(<16 x i8>, <16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_pblendw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pblendw
+ ;CHECK: vpblendw $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %a0, <8 x i16> %a1, i8 7)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16>, <8 x i16>, i8) nounwind readnone
+
+; TODO stack_fold_pclmulqdq
+declare <2 x i64> @llvm.x86.pclmulqdq(<2 x i64>, <2 x i64>, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pcmpeqb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqb
+ ;CHECK: vpcmpeqb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <16 x i8> %a0, %a1
+ %3 = sext <16 x i1> %2 to <16 x i8>
+ ret <16 x i8> %3
+}
+
+define <4 x i32> @stack_fold_pcmpeqd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqd
+ ;CHECK: vpcmpeqd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <4 x i32> %a0, %a1
+ %3 = sext <4 x i1> %2 to <4 x i32>
+ ret <4 x i32> %3
+}
+
+define <2 x i64> @stack_fold_pcmpeqq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqq
+ ;CHECK: vpcmpeqq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <2 x i64> %a0, %a1
+ %3 = sext <2 x i1> %2 to <2 x i64>
+ ret <2 x i64> %3
+}
+
+define <8 x i16> @stack_fold_pcmpeqw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqw
+ ;CHECK: vpcmpeqw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <8 x i16> %a0, %a1
+ %3 = sext <8 x i1> %2 to <8 x i16>
+ ret <8 x i16> %3
+}
+
+; TODO stack_fold_pcmpestri
+declare i32 @llvm.x86.sse42.pcmpestri128(<16 x i8>, i32, <16 x i8>, i32, i8) nounwind readnone
+
+; TODO stack_fold_pcmpestrm
+declare <16 x i8> @llvm.x86.sse42.pcmpestrm128(<16 x i8>, i32, <16 x i8>, i32, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pcmpgtb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtb
+ ;CHECK: vpcmpgtb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <16 x i8> %a0, %a1
+ %3 = sext <16 x i1> %2 to <16 x i8>
+ ret <16 x i8> %3
+}
+
+define <4 x i32> @stack_fold_pcmpgtd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtd
+ ;CHECK: vpcmpgtd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <4 x i32> %a0, %a1
+ %3 = sext <4 x i1> %2 to <4 x i32>
+ ret <4 x i32> %3
+}
+
+define <2 x i64> @stack_fold_pcmpgtq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtq
+ ;CHECK: vpcmpgtq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <2 x i64> %a0, %a1
+ %3 = sext <2 x i1> %2 to <2 x i64>
+ ret <2 x i64> %3
+}
+
+define <8 x i16> @stack_fold_pcmpgtw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtw
+ ;CHECK: vpcmpgtw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <8 x i16> %a0, %a1
+ %3 = sext <8 x i1> %2 to <8 x i16>
+ ret <8 x i16> %3
+}
+
+; TODO stack_fold_pcmpistri
+declare i32 @llvm.x86.sse42.pcmpistri128(<16 x i8>, <16 x i8>, i8) nounwind readnone
+
+; TODO stack_fold_pcmpistrm
+declare <16 x i8> @llvm.x86.sse42.pcmpistrm128(<16 x i8>, <16 x i8>, i8) nounwind readnone
+
+; TODO stack_fold_pextrb
+; TODO stack_fold_pextrd
+; TODO stack_fold_pextrq
+; TODO stack_fold_pextrw
+
+define <4 x i32> @stack_fold_phaddd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddd
+ ;CHECK: vphaddd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.phadd.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.phadd.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_phaddsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddsw
+ ;CHECK: vphaddsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phadd.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phadd.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_phaddw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddw
+ ;CHECK: vphaddw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phadd.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phadd.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_phminposuw
+declare <8 x i16> @llvm.x86.sse41.phminposuw(<8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_phsubd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubd
+ ;CHECK: vphsubd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.phsub.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.phsub.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_phsubsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubsw
+ ;CHECK: vphsubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phsub.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phsub.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_phsubw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubw
+ ;CHECK: vphsubw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phsub.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phsub.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pinsrb
+; TODO stack_fold_pinsrd
+; TODO stack_fold_pinsrq
+; TODO stack_fold_pinsrw
+
+define <8 x i16> @stack_fold_pmaddubsw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaddubsw
+ ;CHECK: vpmaddubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pmadd.ub.sw.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pmadd.ub.sw.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaddwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaddwd
+ ;CHECK: vpmaddwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16> %a0, <8 x i16> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pmaxsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsb
+ ;CHECK: vpmaxsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse41.pmaxsb(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse41.pmaxsb(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaxsd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsd
+ ;CHECK: vpmaxsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pmaxsd(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pmaxsd(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmaxsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsw
+ ;CHECK: vpmaxsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pmaxub(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxub
+ ;CHECK: vpmaxub {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaxud(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxud
+ ;CHECK: vpmaxud {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pmaxud(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pmaxud(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmaxuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxuw
+ ;CHECK: vpmaxuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pmaxuw(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pmaxuw(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pminsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsb
+ ;CHECK: vpminsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse41.pminsb(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse41.pminsb(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pminsd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsd
+ ;CHECK: vpminsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pminsd(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pminsd(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pminsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsw
+ ;CHECK: vpminsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pminub(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pminub
+ ;CHECK: vpminub {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pminud(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pminud
+ ;CHECK: vpminud {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pminud(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pminud(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pminuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pminuw
+ ;CHECK: vpminuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pminuw(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pminuw(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovsxbd
+declare <4 x i32> @llvm.x86.sse41.pmovsxbd(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxbq
+declare <2 x i64> @llvm.x86.sse41.pmovsxbq(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxbw
+declare <8 x i16> @llvm.x86.sse41.pmovsxbw(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxdq
+declare <2 x i64> @llvm.x86.sse41.pmovsxdq(<4 x i32>) nounwind readnone
+
+; TODO stack_fold_pmovsxwd
+declare <4 x i32> @llvm.x86.sse41.pmovsxwd(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovsxwq
+declare <2 x i64> @llvm.x86.sse41.pmovsxwq(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovzxbd
+declare <4 x i32> @llvm.x86.sse41.pmovzxbd(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxbq
+declare <2 x i64> @llvm.x86.sse41.pmovzxbq(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxbw
+declare <8 x i16> @llvm.x86.sse41.pmovzxbw(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxdq
+declare <2 x i64> @llvm.x86.sse41.pmovzxdq(<4 x i32>) nounwind readnone
+
+; TODO stack_fold_pmovzxwd
+declare <4 x i32> @llvm.x86.sse41.pmovzxwd(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovzxwq
+declare <2 x i64> @llvm.x86.sse41.pmovzxwq(<8 x i16>) nounwind readnone
+
+define <2 x i64> @stack_fold_pmuldq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmuldq
+ ;CHECK: vpmuldq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse41.pmuldq(<4 x i32> %a0, <4 x i32> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse41.pmuldq(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhrsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhrsw
+ ;CHECK: vpmulhrsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pmul.hr.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pmul.hr.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhuw
+ ;CHECK: vpmulhuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhw
+ ;CHECK: vpmulhw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmulld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulld
+ ;CHECK: vpmulld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = mul <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+define <8 x i16> @stack_fold_pmullw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmullw
+ ;CHECK: vpmullw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = mul <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+define <2 x i64> @stack_fold_pmuludq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmuludq
+ ;CHECK: vpmuludq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.pmulu.dq(<4 x i32> %a0, <4 x i32> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.pmulu.dq(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_por(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_por
+ ;CHECK: vpor {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = or <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
+
+define <2 x i64> @stack_fold_psadbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psadbw
+ ;CHECK: vpsadbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8> %a0, <16 x i8> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <16 x i8> @stack_fold_pshufb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pshufb
+ ;CHECK: vpshufb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pshufd(<4 x i32> %a0) {
+ ;CHECK-LABEL: stack_fold_pshufd
+ ;CHECK: vpshufd $27, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
+ ret <4 x i32> %2
+}
+
+define <8 x i16> @stack_fold_pshufhw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pshufhw
+ ;CHECK: vpshufhw $11, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 7, i32 6, i32 4, i32 4>
+ ret <8 x i16> %2
+}
+
+define <8 x i16> @stack_fold_pshuflw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pshuflw
+ ;CHECK: vpshuflw $27, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_psignb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psignb
+ ;CHECK: vpsignb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.psign.b.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.psign.b.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_psignd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psignd
+ ;CHECK: vpsignd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.psign.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.psign.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_psignw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psignw
+ ;CHECK: vpsignw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.psign.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.psign.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_pslld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pslld
+ ;CHECK: vpslld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psllq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psllq
+ ;CHECK: vpsllq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <8 x i16> @stack_fold_psllw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psllw
+ ;CHECK: vpsllw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_psrad(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrad
+ ;CHECK: vpsrad {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_psraw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psraw
+ ;CHECK: vpsraw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_psrld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrld
+ ;CHECK: vpsrld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psrlq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlq
+ ;CHECK: vpsrlq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <8 x i16> @stack_fold_psrlw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlw
+ ;CHECK: vpsrlw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_psubb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psubb
+ ;CHECK: vpsubb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <16 x i8> %a0, %a1
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_psubd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psubd
+ ;CHECK: vpsubd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+; TODO stack_fold_psubq
+
+define <16 x i8> @stack_fold_psubsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psubsb
+ ;CHECK: vpsubsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.psubs.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.psubs.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_psubsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psubsw
+ ;CHECK: vpsubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psubs.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psubs.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_psubusb
+declare <16 x i8> @llvm.x86.sse2.psubus.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+; TODO stack_fold_psubusw
+declare <8 x i16> @llvm.x86.sse2.psubus.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_psubw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psubw
+ ;CHECK: vpsubw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+; TODO stack_fold_ptest
+declare i32 @llvm.x86.sse41.ptestc(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <16 x i8> @stack_fold_punpckhbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhbw
+ ;CHECK: vpunpckhbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 8, i32 24, i32 9, i32 25, i32 10, i32 26, i32 11, i32 27, i32 12, i32 28, i32 13, i32 29, i32 14, i32 30, i32 15, i32 31>
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_punpckhdq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhdq
+ ;CHECK: vpunpckhdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> %a1, <4 x i32> <i32 2, i32 6, i32 3, i32 7>
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_punpckhqdq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhqdq
+ ;CHECK: vpunpckhqdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <2 x i64> %a0, <2 x i64> %a1, <2 x i32> <i32 1, i32 3>
+ ret <2 x i64> %2
+}
+
+define <8 x i16> @stack_fold_punpckhwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhwd
+ ;CHECK: vpunpckhwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> %a1, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_punpcklbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklbw
+ ;CHECK: vpunpcklbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23>
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_punpckldq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckldq
+ ;CHECK: vpunpckldq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> %a1, <4 x i32> <i32 0, i32 4, i32 1, i32 5>
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_punpcklqdq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklqdq
+ ;CHECK: vpunpcklqdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <2 x i64> %a0, <2 x i64> %a1, <2 x i32> <i32 0, i32 2>
+ ret <2 x i64> %2
+}
+
+define <8 x i16> @stack_fold_punpcklwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklwd
+ ;CHECK: vpunpcklwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> %a1, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_pxor(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pxor
+ ;CHECK: vpxor {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = xor <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
--- /dev/null
+; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+sse4.2,+aes < %s | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-unknown"
+
+; Stack reload folding tests.
+;
+; By including a nop call with sideeffects we can force a partial register spill of the
+; relevant registers and check that the reload is correctly folded into the instruction.
+
+define <2 x i64> @stack_fold_aesdec(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesdec
+ ;CHECK: aesdec {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesdec(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesdec(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesdeclast(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesdeclast
+ ;CHECK: aesdeclast {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesdeclast(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesdeclast(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesenc(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesenc
+ ;CHECK: aesenc {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesenc(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesenc(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesenclast(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_aesenclast
+ ;CHECK: aesenclast {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesenclast(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesenclast(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aesimc(<2 x i64> %a0) {
+ ;CHECK-LABEL: stack_fold_aesimc
+ ;CHECK: aesimc {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aesimc(<2 x i64> %a0)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aesimc(<2 x i64>) nounwind readnone
+
+define <2 x i64> @stack_fold_aeskeygenassist(<2 x i64> %a0) {
+ ;CHECK-LABEL: stack_fold_aeskeygenassist
+ ;CHECK: aeskeygenassist $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.aesni.aeskeygenassist(<2 x i64> %a0, i8 7)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.aesni.aeskeygenassist(<2 x i64>, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pabsb(<16 x i8> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsb
+ ;CHECK: pabsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.pabs.b.128(<16 x i8> %a0)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.pabs.b.128(<16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pabsd(<4 x i32> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsd
+ ;CHECK: pabsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.pabs.d.128(<4 x i32> %a0)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.pabs.d.128(<4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pabsw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pabsw
+ ;CHECK: pabsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pabs.w.128(<8 x i16> %a0)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pabs.w.128(<8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_packssdw(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_packssdw
+ ;CHECK: packssdw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_packsswb(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_packsswb
+ ;CHECK: packsswb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_packusdw(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_packusdw
+ ;CHECK: packusdw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.packusdw(<4 x i32> %a0, <4 x i32> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.packusdw(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_packuswb(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_packuswb
+ ;CHECK: packuswb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_paddb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddb
+ ;CHECK: paddb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <16 x i8> %a0, %a1
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_paddd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_paddd
+ ;CHECK: paddd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_paddq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_paddq
+ ;CHECK: paddq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <2 x i64> %a0, %a1
+ ret <2 x i64> %2
+}
+
+define <16 x i8> @stack_fold_paddsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddsb
+ ;CHECK: paddsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.padds.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.padds.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddsw
+ ;CHECK: paddsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.padds.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.padds.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_paddusb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_paddusb
+ ;CHECK: paddusb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.paddus.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.paddus.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddusw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddusw
+ ;CHECK: paddusw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.paddus.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.paddus.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_paddw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_paddw
+ ;CHECK: paddw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = add <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_palignr(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_palignr
+ ;CHECK: palignr $1, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a1, <16 x i8> %a0, <16 x i32> <i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16>
+ ret <16 x i8> %2
+}
+
+define <16 x i8> @stack_fold_pand(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pand
+ ;CHECK: pand {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = and <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
+
+define <16 x i8> @stack_fold_pandn(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pandn
+ ;CHECK: pandn {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = xor <16 x i8> %a0, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
+ %3 = and <16 x i8> %2, %a1
+ ; add forces execution domain
+ %4 = add <16 x i8> %3, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %4
+}
+
+define <16 x i8> @stack_fold_pavgb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pavgb
+ ;CHECK: pavgb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_pavgw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pavgw
+ ;CHECK: pavgw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pblendvb
+declare <16 x i8> @llvm.x86.sse41.pblendvb(<16 x i8>, <16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_pblendw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pblendw
+ ;CHECK: pblendw $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %a0, <8 x i16> %a1, i8 7)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16>, <8 x i16>, i8) nounwind readnone
+
+; TODO stack_fold_pclmulqdq
+declare <2 x i64> @llvm.x86.pclmulqdq(<2 x i64>, <2 x i64>, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pcmpeqb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqb
+ ;CHECK: pcmpeqb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <16 x i8> %a0, %a1
+ %3 = sext <16 x i1> %2 to <16 x i8>
+ ret <16 x i8> %3
+}
+
+define <4 x i32> @stack_fold_pcmpeqd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqd
+ ;CHECK: pcmpeqd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <4 x i32> %a0, %a1
+ %3 = sext <4 x i1> %2 to <4 x i32>
+ ret <4 x i32> %3
+}
+
+define <2 x i64> @stack_fold_pcmpeqq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqq
+ ;CHECK: pcmpeqq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <2 x i64> %a0, %a1
+ %3 = sext <2 x i1> %2 to <2 x i64>
+ ret <2 x i64> %3
+}
+
+define <8 x i16> @stack_fold_pcmpeqw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpeqw
+ ;CHECK: pcmpeqw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp eq <8 x i16> %a0, %a1
+ %3 = sext <8 x i1> %2 to <8 x i16>
+ ret <8 x i16> %3
+}
+
+; TODO stack_fold_pcmpestri
+declare i32 @llvm.x86.sse42.pcmpestri128(<16 x i8>, i32, <16 x i8>, i32, i8) nounwind readnone
+
+; TODO stack_fold_pcmpestrm
+declare <16 x i8> @llvm.x86.sse42.pcmpestrm128(<16 x i8>, i32, <16 x i8>, i32, i8) nounwind readnone
+
+define <16 x i8> @stack_fold_pcmpgtb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtb
+ ;CHECK: pcmpgtb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <16 x i8> %a0, %a1
+ %3 = sext <16 x i1> %2 to <16 x i8>
+ ret <16 x i8> %3
+}
+
+define <4 x i32> @stack_fold_pcmpgtd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtd
+ ;CHECK: pcmpgtd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <4 x i32> %a0, %a1
+ %3 = sext <4 x i1> %2 to <4 x i32>
+ ret <4 x i32> %3
+}
+
+define <2 x i64> @stack_fold_pcmpgtq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtq
+ ;CHECK: pcmpgtq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <2 x i64> %a0, %a1
+ %3 = sext <2 x i1> %2 to <2 x i64>
+ ret <2 x i64> %3
+}
+
+define <8 x i16> @stack_fold_pcmpgtw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pcmpgtw
+ ;CHECK: pcmpgtw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = icmp sgt <8 x i16> %a0, %a1
+ %3 = sext <8 x i1> %2 to <8 x i16>
+ ret <8 x i16> %3
+}
+
+; TODO stack_fold_pcmpistri
+declare i32 @llvm.x86.sse42.pcmpistri128(<16 x i8>, <16 x i8>, i8) nounwind readnone
+
+; TODO stack_fold_pcmpistrm
+declare <16 x i8> @llvm.x86.sse42.pcmpistrm128(<16 x i8>, <16 x i8>, i8) nounwind readnone
+
+; TODO stack_fold_pextrb
+; TODO stack_fold_pextrd
+; TODO stack_fold_pextrq
+; TODO stack_fold_pextrw
+
+define <4 x i32> @stack_fold_phaddd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddd
+ ;CHECK: phaddd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.phadd.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.phadd.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_phaddsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddsw
+ ;CHECK: phaddsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phadd.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phadd.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_phaddw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phaddw
+ ;CHECK: phaddw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phadd.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phadd.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_phminposuw
+declare <8 x i16> @llvm.x86.sse41.phminposuw(<8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_phsubd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubd
+ ;CHECK: phsubd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.phsub.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.phsub.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_phsubsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubsw
+ ;CHECK: phsubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phsub.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phsub.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_phsubw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_phsubw
+ ;CHECK: phsubw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.phsub.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.phsub.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pinsrb
+; TODO stack_fold_pinsrd
+; TODO stack_fold_pinsrq
+; TODO stack_fold_pinsrw
+
+define <8 x i16> @stack_fold_pmaddubsw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaddubsw
+ ;CHECK: pmaddubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pmadd.ub.sw.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pmadd.ub.sw.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaddwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaddwd
+ ;CHECK: pmaddwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16> %a0, <8 x i16> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pmaxsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsb
+ ;CHECK: pmaxsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse41.pmaxsb(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse41.pmaxsb(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaxsd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsd
+ ;CHECK: pmaxsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pmaxsd(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pmaxsd(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmaxsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxsw
+ ;CHECK: pmaxsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pmaxub(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxub
+ ;CHECK: pmaxub {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmaxud(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxud
+ ;CHECK: pmaxud {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pmaxud(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pmaxud(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmaxuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmaxuw
+ ;CHECK: pmaxuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pmaxuw(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pmaxuw(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pminsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsb
+ ;CHECK: pminsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse41.pminsb(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse41.pminsb(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pminsd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsd
+ ;CHECK: pminsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pminsd(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pminsd(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pminsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pminsw
+ ;CHECK: pminsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_pminub(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pminub
+ ;CHECK: pminub {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pminud(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pminud
+ ;CHECK: pminud {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse41.pminud(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse41.pminud(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pminuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pminuw
+ ;CHECK: pminuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse41.pminuw(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse41.pminuw(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovsxbd
+declare <4 x i32> @llvm.x86.sse41.pmovsxbd(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxbq
+declare <2 x i64> @llvm.x86.sse41.pmovsxbq(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxbw
+declare <8 x i16> @llvm.x86.sse41.pmovsxbw(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovsxdq
+declare <2 x i64> @llvm.x86.sse41.pmovsxdq(<4 x i32>) nounwind readnone
+
+; TODO stack_fold_pmovsxwd
+declare <4 x i32> @llvm.x86.sse41.pmovsxwd(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovsxwq
+declare <2 x i64> @llvm.x86.sse41.pmovsxwq(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovzxbd
+declare <4 x i32> @llvm.x86.sse41.pmovzxbd(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxbq
+declare <2 x i64> @llvm.x86.sse41.pmovzxbq(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxbw
+declare <8 x i16> @llvm.x86.sse41.pmovzxbw(<16 x i8>) nounwind readnone
+
+; TODO stack_fold_pmovzxdq
+declare <2 x i64> @llvm.x86.sse41.pmovzxdq(<4 x i32>) nounwind readnone
+
+; TODO stack_fold_pmovzxwd
+declare <4 x i32> @llvm.x86.sse41.pmovzxwd(<8 x i16>) nounwind readnone
+
+; TODO stack_fold_pmovzxwq
+declare <2 x i64> @llvm.x86.sse41.pmovzxwq(<8 x i16>) nounwind readnone
+
+define <2 x i64> @stack_fold_pmuldq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmuldq
+ ;CHECK: pmuldq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse41.pmuldq(<4 x i32> %a0, <4 x i32> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse41.pmuldq(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhrsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhrsw
+ ;CHECK: pmulhrsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.pmul.hr.sw.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.pmul.hr.sw.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhuw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhuw
+ ;CHECK: pmulhuw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_pmulhw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulhw
+ ;CHECK: pmulhw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_pmulld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmulld
+ ;CHECK: pmulld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = mul <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+define <8 x i16> @stack_fold_pmullw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_pmullw
+ ;CHECK: pmullw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = mul <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+define <2 x i64> @stack_fold_pmuludq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pmuludq
+ ;CHECK: pmuludq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.pmulu.dq(<4 x i32> %a0, <4 x i32> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.pmulu.dq(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <16 x i8> @stack_fold_por(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_por
+ ;CHECK: por {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = or <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
+
+define <2 x i64> @stack_fold_psadbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psadbw
+ ;CHECK: psadbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8> %a0, <16 x i8> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <16 x i8> @stack_fold_pshufb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pshufb
+ ;CHECK: pshufb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_pshufd(<4 x i32> %a0) {
+ ;CHECK-LABEL: stack_fold_pshufd
+ ;CHECK: pshufd $27, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
+ ret <4 x i32> %2
+}
+
+define <8 x i16> @stack_fold_pshufhw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pshufhw
+ ;CHECK: pshufhw $11, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 7, i32 6, i32 4, i32 4>
+ ret <8 x i16> %2
+}
+
+define <8 x i16> @stack_fold_pshuflw(<8 x i16> %a0) {
+ ;CHECK-LABEL: stack_fold_pshuflw
+ ;CHECK: pshuflw $27, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 4, i32 5, i32 6, i32 7>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_psignb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psignb
+ ;CHECK: psignb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.ssse3.psign.b.128(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.ssse3.psign.b.128(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <4 x i32> @stack_fold_psignd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psignd
+ ;CHECK: psignd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.ssse3.psign.d.128(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.ssse3.psign.d.128(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_psignw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psignw
+ ;CHECK: psignw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.ssse3.psign.w.128(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.ssse3.psign.w.128(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_pslld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_pslld
+ ;CHECK: pslld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psllq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psllq
+ ;CHECK: psllq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <8 x i16> @stack_fold_psllw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psllw
+ ;CHECK: psllw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_psrad(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrad
+ ;CHECK: psrad {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <8 x i16> @stack_fold_psraw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psraw
+ ;CHECK: psraw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <4 x i32> @stack_fold_psrld(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psrld
+ ;CHECK: psrld {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32> %a0, <4 x i32> %a1)
+ ret <4 x i32> %2
+}
+declare <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32>, <4 x i32>) nounwind readnone
+
+define <2 x i64> @stack_fold_psrlq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlq
+ ;CHECK: psrlq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64> %a0, <2 x i64> %a1)
+ ret <2 x i64> %2
+}
+declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <8 x i16> @stack_fold_psrlw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psrlw
+ ;CHECK: psrlw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <16 x i8> @stack_fold_psubb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psubb
+ ;CHECK: psubb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <16 x i8> %a0, %a1
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_psubd(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_psubd
+ ;CHECK: psubd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <4 x i32> %a0, %a1
+ ret <4 x i32> %2
+}
+
+; TODO stack_fold_psubq
+
+define <16 x i8> @stack_fold_psubsb(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_psubsb
+ ;CHECK: psubsb {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <16 x i8> @llvm.x86.sse2.psubs.b(<16 x i8> %a0, <16 x i8> %a1)
+ ret <16 x i8> %2
+}
+declare <16 x i8> @llvm.x86.sse2.psubs.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+define <8 x i16> @stack_fold_psubsw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psubsw
+ ;CHECK: psubsw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call <8 x i16> @llvm.x86.sse2.psubs.w(<8 x i16> %a0, <8 x i16> %a1)
+ ret <8 x i16> %2
+}
+declare <8 x i16> @llvm.x86.sse2.psubs.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+; TODO stack_fold_psubusb
+declare <16 x i8> @llvm.x86.sse2.psubus.b(<16 x i8>, <16 x i8>) nounwind readnone
+
+; TODO stack_fold_psubusw
+declare <8 x i16> @llvm.x86.sse2.psubus.w(<8 x i16>, <8 x i16>) nounwind readnone
+
+define <8 x i16> @stack_fold_psubw(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_psubw
+ ;CHECK: psubw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = sub <8 x i16> %a0, %a1
+ ret <8 x i16> %2
+}
+
+; TODO stack_fold_ptest
+declare i32 @llvm.x86.sse41.ptestc(<2 x i64>, <2 x i64>) nounwind readnone
+
+define <16 x i8> @stack_fold_punpckhbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhbw
+ ;CHECK: punpckhbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 8, i32 24, i32 9, i32 25, i32 10, i32 26, i32 11, i32 27, i32 12, i32 28, i32 13, i32 29, i32 14, i32 30, i32 15, i32 31>
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_punpckhdq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhdq
+ ;CHECK: punpckhdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> %a1, <4 x i32> <i32 2, i32 6, i32 3, i32 7>
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_punpckhqdq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhqdq
+ ;CHECK: punpckhqdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <2 x i64> %a0, <2 x i64> %a1, <2 x i32> <i32 1, i32 3>
+ ret <2 x i64> %2
+}
+
+define <8 x i16> @stack_fold_punpckhwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckhwd
+ ;CHECK: punpckhwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> %a1, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_punpcklbw(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklbw
+ ;CHECK: punpcklbw {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <16 x i8> %a0, <16 x i8> %a1, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23>
+ ret <16 x i8> %2
+}
+
+define <4 x i32> @stack_fold_punpckldq(<4 x i32> %a0, <4 x i32> %a1) {
+ ;CHECK-LABEL: stack_fold_punpckldq
+ ;CHECK: punpckldq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <4 x i32> %a0, <4 x i32> %a1, <4 x i32> <i32 0, i32 4, i32 1, i32 5>
+ ret <4 x i32> %2
+}
+
+define <2 x i64> @stack_fold_punpcklqdq(<2 x i64> %a0, <2 x i64> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklqdq
+ ;CHECK: punpcklqdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <2 x i64> %a0, <2 x i64> %a1, <2 x i32> <i32 0, i32 2>
+ ret <2 x i64> %2
+}
+
+define <8 x i16> @stack_fold_punpcklwd(<8 x i16> %a0, <8 x i16> %a1) {
+ ;CHECK-LABEL: stack_fold_punpcklwd
+ ;CHECK: punpcklwd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = shufflevector <8 x i16> %a0, <8 x i16> %a1, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
+ ret <8 x i16> %2
+}
+
+define <16 x i8> @stack_fold_pxor(<16 x i8> %a0, <16 x i8> %a1) {
+ ;CHECK-LABEL: stack_fold_pxor
+ ;CHECK: pxor {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = xor <16 x i8> %a0, %a1
+ ; add forces execution domain
+ %3 = add <16 x i8> %2, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
+ ret <16 x i8> %3
+}
projects / oota-llvm.git / commitdiff