diff --git a/test/CodeGen/X86/stack-folding-int-avx2.ll b/test/CodeGen/X86/stack-folding-int-avx2.ll
new file mode 100644
index 00000000000..8a93a14f3ac
--- /dev/null
+++ b/test/CodeGen/X86/stack-folding-int-avx2.ll
@@ -0,0 +1,295 @@
+; RUN: llc -O3 -disable-peephole -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx2 < %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 <8 x i32> @stack_fold_inserti128(<4 x i32> %a0, <4 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_inserti128
+  ;CHECK:       vinserti128 $1, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[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, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ; add forces execution domain
+  %3 = add <8 x i32> %2, <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+  ret <8 x i32> %3
+}
+
+define <32 x i8> @stack_fold_pabsb(<32 x i8> %a0) {
+  ;CHECK-LABEL: stack_fold_pabsb
+  ;CHECK:       vpabsb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8> %a0)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8>) nounwind readnone
+
+define <8 x i32> @stack_fold_pabsd(<8 x i32> %a0) {
+  ;CHECK-LABEL: stack_fold_pabsd
+  ;CHECK:       vpabsd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i32> @llvm.x86.avx2.pabs.d(<8 x i32> %a0)
+  ret <8 x i32> %2
+}
+declare <8 x i32> @llvm.x86.avx2.pabs.d(<8 x i32>) nounwind readnone
+
+define <16 x i16> @stack_fold_pabsw(<16 x i16> %a0) {
+  ;CHECK-LABEL: stack_fold_pabsw
+  ;CHECK:       vpabsw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.pabs.w(<16 x i16> %a0)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.pabs.w(<16 x i16>) nounwind readnone
+
+define <16 x i16> @stack_fold_packssdw(<8 x i32> %a0, <8 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_packssdw
+  ;CHECK:       vpackssdw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.packssdw(<8 x i32> %a0, <8 x i32> %a1)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.packssdw(<8 x i32>, <8 x i32>) nounwind readnone
+
+define <32 x i8> @stack_fold_packsswb(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_packsswb
+  ;CHECK:       vpacksswb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16> %a0, <16 x i16> %a1)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16>, <16 x i16>) nounwind readnone
+
+define <16 x i16> @stack_fold_packusdw(<8 x i32> %a0, <8 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_packusdw
+  ;CHECK:       vpackusdw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.packusdw(<8 x i32> %a0, <8 x i32> %a1)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.packusdw(<8 x i32>, <8 x i32>) nounwind readnone
+
+define <32 x i8> @stack_fold_packuswb(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_packuswb
+  ;CHECK:       vpackuswb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16> %a0, <16 x i16> %a1)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16>, <16 x i16>) nounwind readnone
+
+define <32 x i8> @stack_fold_paddb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_paddb
+  ;CHECK:       vpaddb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> %a0, %a1
+  ret <32 x i8> %2
+}
+
+define <8 x i32> @stack_fold_paddd(<8 x i32> %a0, <8 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_paddd
+  ;CHECK:       vpaddd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i32> %a0, %a1
+  ret <8 x i32> %2
+}
+
+define <4 x i64> @stack_fold_paddq(<4 x i64> %a0, <4 x i64> %a1) {
+  ;CHECK-LABEL: stack_fold_paddq
+  ;CHECK:       vpaddq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i64> %a0, %a1
+  ret <4 x i64> %2
+}
+
+define <32 x i8> @stack_fold_paddsb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_paddsb
+  ;CHECK:       vpaddsb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8> %a0, <32 x i8> %a1)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8>, <32 x i8>) nounwind readnone
+
+define <16 x i16> @stack_fold_paddsw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_paddsw
+  ;CHECK:       vpaddsw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.padds.w(<16 x i16> %a0, <16 x i16> %a1)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.padds.w(<16 x i16>, <16 x i16>) nounwind readnone
+
+define <32 x i8> @stack_fold_paddusb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_paddusb
+  ;CHECK:       vpaddusb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.paddus.b(<32 x i8> %a0, <32 x i8> %a1)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.paddus.b(<32 x i8>, <32 x i8>) nounwind readnone
+
+define <16 x i16> @stack_fold_paddusw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_paddusw
+  ;CHECK:       vpaddusw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.paddus.w(<16 x i16> %a0, <16 x i16> %a1)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.paddus.w(<16 x i16>, <16 x i16>) nounwind readnone
+
+define <16 x i16> @stack_fold_paddw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_paddw
+  ;CHECK:       vpaddw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> %a0, %a1
+  ret <16 x i16> %2
+}
+
+; TODO
+; define <32 x i8> @stack_fold_palignr(<32 x i8> %a0, <32 x i8> %a1)
+
+define <32 x i8> @stack_fold_pand(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pand
+  ;CHECK:       vpand {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> %a0, %a1
+  ; add forces execution domain
+  %3 = add <32 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, 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 <32 x i8> %3
+}
+
+define <32 x i8> @stack_fold_pandn(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pandn
+  ;CHECK:       vpandn {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 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, 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 <32 x i8> %2, %a1
+  ; add forces execution domain
+  %4 = add <32 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, 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 <32 x i8> %4
+}
+
+define <32 x i8> @stack_fold_pavgb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pavgb
+  ;CHECK:       vpavgb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> @llvm.x86.avx2.pavg.b(<32 x i8> %a0, <32 x i8> %a1)
+  ret <32 x i8> %2
+}
+declare <32 x i8> @llvm.x86.avx2.pavg.b(<32 x i8>, <32 x i8>) nounwind readnone
+
+define <16 x i16> @stack_fold_pavgw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_pavgw
+  ;CHECK:       vpavgw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.pavg.w(<16 x i16> %a0, <16 x i16> %a1)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.pavg.w(<16 x i16>, <16 x i16>) nounwind readnone
+
+define <16 x i16> @stack_fold_pblendw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_pblendw
+  ;CHECK:       vpblendw $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i8 7)
+  ret <16 x i16> %2
+}
+declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i8) nounwind readnone
+
+define <32 x i8> @stack_fold_pcmpeqb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpeqb
+  ;CHECK:       vpcmpeqb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> %a0, %a1
+  %3 = sext <32 x i1> %2 to <32 x i8>
+  ret <32 x i8> %3
+}
+
+define <8 x i32> @stack_fold_pcmpeqd(<8 x i32> %a0, <8 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpeqd
+  ;CHECK:       vpcmpeqd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i32> %a0, %a1
+  %3 = sext <8 x i1> %2 to <8 x i32>
+  ret <8 x i32> %3
+}
+
+define <4 x i64> @stack_fold_pcmpeqq(<4 x i64> %a0, <4 x i64> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpeqq
+  ;CHECK:       vpcmpeqq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i64> %a0, %a1
+  %3 = sext <4 x i1> %2 to <4 x i64>
+  ret <4 x i64> %3
+}
+
+define <16 x i16> @stack_fold_pcmpeqw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpeqw
+  ;CHECK:       vpcmpeqw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> %a0, %a1
+  %3 = sext <16 x i1> %2 to <16 x i16>
+  ret <16 x i16> %3
+}
+
+define <32 x i8> @stack_fold_pcmpgtb(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpgtb
+  ;CHECK:       vpcmpgtb {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> %a0, %a1
+  %3 = sext <32 x i1> %2 to <32 x i8>
+  ret <32 x i8> %3
+}
+
+define <8 x i32> @stack_fold_pcmpgtd(<8 x i32> %a0, <8 x i32> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpgtd
+  ;CHECK:       vpcmpgtd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i32> %a0, %a1
+  %3 = sext <8 x i1> %2 to <8 x i32>
+  ret <8 x i32> %3
+}
+
+define <4 x i64> @stack_fold_pcmpgtq(<4 x i64> %a0, <4 x i64> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpgtq
+  ;CHECK:       vpcmpgtq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i64> %a0, %a1
+  %3 = sext <4 x i1> %2 to <4 x i64>
+  ret <4 x i64> %3
+}
+
+define <16 x i16> @stack_fold_pcmpgtw(<16 x i16> %a0, <16 x i16> %a1) {
+  ;CHECK-LABEL: stack_fold_pcmpgtw
+  ;CHECK:       vpcmpgtw {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i16> %a0, %a1
+  %3 = sext <16 x i1> %2 to <16 x i16>
+  ret <16 x i16> %3
+}
+
+define <8 x i32> @stack_fold_pshufd(<8 x i32> %a0) {
+  ;CHECK-LABEL: stack_fold_pshufd
+  ;CHECK:       vpshufd $27, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 i32> %a0, <8 x i32> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
+  ret <8 x i32> %2
+}
+
+;TODO stack_fold_pshufhw
+
+;TODO stack_fold_pshuflw
+
+define <32 x i8> @stack_fold_pxor(<32 x i8> %a0, <32 x i8> %a1) {
+  ;CHECK-LABEL: stack_fold_pxor
+  ;CHECK:       vpxor {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-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 <32 x i8> %a0, %a1
+  ; add forces execution domain
+  %3 = add <32 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, 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 <32 x i8> %3
+}