[X86][AVX] Simplified diff between AVX1 and SSE42 fp stack folding tests. NFC.

Changed the AVX1 tests register spill tail call to return a xmm like the SSE42 version - makes doing diffs between them a lot easier without affecting the spills themselves.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@226623 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/test/CodeGen/X86/stack-folding-fp-avx1.ll b/test/CodeGen/X86/stack-folding-fp-avx1.ll
index 2115d93193944499a618dd57ead72bf686093003..2a026159f716fdb35b86e09b72c5f110d1f41909 100644 (file)
--- a/test/CodeGen/X86/stack-folding-fp-avx1.ll
+++ b/test/CodeGen/X86/stack-folding-fp-avx1.ll
@@ -11,7 +11,7 @@ target triple = "x86_64-unknown-unknown"
 define <2 x double> @stack_fold_addpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_addpd
   ;CHECK:       vaddpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd <2 x double> %a0, %a1
   ret <2 x double> %2
 }
@@ -19,7 +19,7 @@ define <2 x double> @stack_fold_addpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_addpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_addpd_ymm
   ;CHECK:       vaddpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd <4 x double> %a0, %a1
   ret <4 x double> %2
 }
@@ -27,7 +27,7 @@ define <4 x double> @stack_fold_addpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_addps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_addps
   ;CHECK:       vaddps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd <4 x float> %a0, %a1
   ret <4 x float> %2
 }
@@ -35,7 +35,7 @@ define <4 x float> @stack_fold_addps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_addps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_addps_ymm
   ;CHECK:       vaddps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd <8 x float> %a0, %a1
   ret <8 x float> %2
 }
@@ -43,7 +43,7 @@ define <8 x float> @stack_fold_addps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define double @stack_fold_addsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_addsd
   ;CHECK:       vaddsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd double %a0, %a1
   ret double %2
 }
@@ -51,7 +51,7 @@ define double @stack_fold_addsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_addsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_addsd_int
   ;CHECK:       vaddsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.add.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -60,7 +60,7 @@ declare <2 x double> @llvm.x86.sse2.add.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_addss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_addss
   ;CHECK:       vaddss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fadd float %a0, %a1
   ret float %2
 }
@@ -68,7 +68,7 @@ define float @stack_fold_addss(float %a0, float %a1) {
 define <4 x float> @stack_fold_addss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_addss_int
   ;CHECK:       vaddss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.add.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -77,7 +77,7 @@ declare <4 x float> @llvm.x86.sse.add.ss(<4 x float>, <4 x float>) nounwind read
 define <2 x double> @stack_fold_addsubpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_addsubpd
   ;CHECK:       vaddsubpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse3.addsub.pd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -86,7 +86,7 @@ declare <2 x double> @llvm.x86.sse3.addsub.pd(<2 x double>, <2 x double>) nounwi
 define <4 x double> @stack_fold_addsubpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_addsubpd_ymm
   ;CHECK:       vaddsubpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.addsub.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret <4 x double> %2
 }
@@ -95,7 +95,7 @@ declare <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double>, <4 x double>) nou
 define <4 x float> @stack_fold_addsubps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_addsubps
   ;CHECK:       vaddsubps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse3.addsub.ps(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -104,7 +104,7 @@ declare <4 x float> @llvm.x86.sse3.addsub.ps(<4 x float>, <4 x float>) nounwind
 define <8 x float> @stack_fold_addsubps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_addsubps_ymm
   ;CHECK:       vaddsubps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.addsub.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret <8 x float> %2
 }
@@ -113,7 +113,7 @@ declare <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float>, <8 x float>) nounwi
 define <2 x double> @stack_fold_andnpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_andnpd
   ;CHECK:       vandnpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <2 x double> %a0 to <2 x i64>
   %3 = bitcast <2 x double> %a1 to <2 x i64>
   %4 = xor <2 x i64> %2, <i64 -1, i64 -1>
@@ -127,7 +127,7 @@ define <2 x double> @stack_fold_andnpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_andnpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_andnpd_ymm
   ;CHECK:       vandnpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x double> %a0 to <4 x i64>
   %3 = bitcast <4 x double> %a1 to <4 x i64>
   %4 = xor <4 x i64> %2, <i64 -1, i64 -1, i64 -1, i64 -1>
@@ -141,7 +141,7 @@ define <4 x double> @stack_fold_andnpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_andnps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_andnps
   ;CHECK:       vandnps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x float> %a0 to <2 x i64>
   %3 = bitcast <4 x float> %a1 to <2 x i64>
   %4 = xor <2 x i64> %2, <i64 -1, i64 -1>
@@ -153,7 +153,7 @@ define <4 x float> @stack_fold_andnps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_andnps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_andnps_ymm
   ;CHECK:       vandnps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <8 x float> %a0 to <4 x i64>
   %3 = bitcast <8 x float> %a1 to <4 x i64>
   %4 = xor <4 x i64> %2, <i64 -1, i64 -1, i64 -1, i64 -1>
@@ -165,7 +165,7 @@ define <8 x float> @stack_fold_andnps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_andpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_andpd
   ;CHECK:       vandpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <2 x double> %a0 to <2 x i64>
   %3 = bitcast <2 x double> %a1 to <2 x i64>
   %4 = and <2 x i64> %2, %3
@@ -178,7 +178,7 @@ define <2 x double> @stack_fold_andpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_andpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_andpd_ymm
   ;CHECK:       vandpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x double> %a0 to <4 x i64>
   %3 = bitcast <4 x double> %a1 to <4 x i64>
   %4 = and <4 x i64> %2, %3
@@ -191,7 +191,7 @@ define <4 x double> @stack_fold_andpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_andps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_andps
   ;CHECK:       vandps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x float> %a0 to <2 x i64>
   %3 = bitcast <4 x float> %a1 to <2 x i64>
   %4 = and <2 x i64> %2, %3
@@ -202,7 +202,7 @@ define <4 x float> @stack_fold_andps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_andps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_andps_ymm
   ;CHECK:       vandps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <8 x float> %a0 to <4 x i64>
   %3 = bitcast <8 x float> %a1 to <4 x i64>
   %4 = and <4 x i64> %2, %3
@@ -213,7 +213,7 @@ define <8 x float> @stack_fold_andps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_blendpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_blendpd
   ;CHECK:       vblendpd $2, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = select <2 x i1> <i1 1, i1 0>, <2 x double> %a0, <2 x double> %a1
   ret <2 x double> %2
 }
@@ -221,7 +221,7 @@ define <2 x double> @stack_fold_blendpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_blendpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_blendpd_ymm
   ;CHECK:       vblendpd $6, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = select <4 x i1> <i1 1, i1 0, i1 0, i1 1>, <4 x double> %a0, <4 x double> %a1
   ret <4 x double> %2
 }
@@ -229,7 +229,7 @@ define <4 x double> @stack_fold_blendpd_ymm(<4 x double> %a0, <4 x double> %a1)
 define <4 x float> @stack_fold_blendps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_blendps
   ;CHECK:       vblendps $6, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = select <4 x i1> <i1 1, i1 0, i1 0, i1 1>, <4 x float> %a0, <4 x float> %a1
   ret <4 x float> %2
 }
@@ -237,7 +237,7 @@ define <4 x float> @stack_fold_blendps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_blendps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_blendps_ymm
   ;CHECK:       vblendps $102, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = select <8 x i1> <i1 1, i1 0, i1 0, i1 1, i1 1, i1 0, i1 0, i1 1>, <8 x float> %a0, <8 x float> %a1
   ret <8 x float> %2
 }
@@ -245,7 +245,7 @@ define <8 x float> @stack_fold_blendps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_blendvpd(<2 x double> %a0, <2 x double> %a1, <2 x double> %c) {
   ;CHECK-LABEL: stack_fold_blendvpd
   ;CHECK:       vblendvpd {{%xmm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
   %2 = call <2 x double> @llvm.x86.sse41.blendvpd(<2 x double> %a1, <2 x double> %c, <2 x double> %a0)
   ret <2 x double> %2
 }
@@ -254,7 +254,7 @@ declare <2 x double> @llvm.x86.sse41.blendvpd(<2 x double>, <2 x double>, <2 x d
 define <4 x double> @stack_fold_blendvpd_ymm(<4 x double> %a0, <4 x double> %a1, <4 x double> %c) {
   ;CHECK-LABEL: stack_fold_blendvpd_ymm
   ;CHECK:       vblendvpd {{%ymm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
   %2 = call <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double> %a1, <4 x double> %c, <4 x double> %a0)
   ret <4 x double> %2
 }
@@ -263,7 +263,7 @@ declare <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double>, <4 x double>, <4
 define <4 x float> @stack_fold_blendvps(<4 x float> %a0, <4 x float> %a1, <4 x float> %c) {
   ;CHECK-LABEL: stack_fold_blendvps
   ;CHECK:       vblendvps {{%xmm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
   %2 = call <4 x float> @llvm.x86.sse41.blendvps(<4 x float> %a1, <4 x float> %c, <4 x float> %a0)
   ret <4 x float> %2
 }
@@ -272,7 +272,7 @@ declare <4 x float> @llvm.x86.sse41.blendvps(<4 x float>, <4 x float>, <4 x floa
 define <8 x float> @stack_fold_blendvps_ymm(<8 x float> %a0, <8 x float> %a1, <8 x float> %c) {
   ;CHECK-LABEL: stack_fold_blendvps_ymm
   ;CHECK:       vblendvps {{%ymm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
   %2 = call <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float> %a1, <8 x float> %c, <8 x float> %a0)
   ret <8 x float> %2
 }
@@ -281,7 +281,7 @@ declare <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float>, <8 x float>, <8 x f
 define <2 x double> @stack_fold_cmppd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_cmppd
   ;CHECK:       vcmpeqpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.cmp.pd(<2 x double> %a0, <2 x double> %a1, i8 0)
   ret <2 x double> %2
 }
@@ -290,7 +290,7 @@ declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8) nounw
 define <4 x double> @stack_fold_cmppd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_cmppd_ymm
   ;CHECK:       vcmpeqpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.cmp.pd.256(<4 x double> %a0, <4 x double> %a1, i8 0)
   ret <4 x double> %2
 }
@@ -299,7 +299,7 @@ declare <4 x double> @llvm.x86.avx.cmp.pd.256(<4 x double>, <4 x double>, i8) no
 define <4 x float> @stack_fold_cmpps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_cmpps
   ;CHECK:       vcmpeqps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.cmp.ps(<4 x float> %a0, <4 x float> %a1, i8 0)
   ret <4 x float> %2
 }
@@ -308,7 +308,7 @@ declare <4 x float> @llvm.x86.sse.cmp.ps(<4 x float>, <4 x float>, i8) nounwind
 define <8 x float> @stack_fold_cmpps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_cmpps_ymm
   ;CHECK:       vcmpeqps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 0)
   ret <8 x float> %2
 }
@@ -317,7 +317,7 @@ declare <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float>, <8 x float>, i8) nounw
 define i32 @stack_fold_cmpsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_cmpsd
   ;CHECK:       vcmpeqsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp oeq double %a0, %a1
   %3 = zext i1 %2 to i32
   ret i32 %3
@@ -326,7 +326,7 @@ define i32 @stack_fold_cmpsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_cmpsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_cmpsd_int
   ;CHECK:       vcmpeqsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.cmp.sd(<2 x double> %a0, <2 x double> %a1, i8 0)
   ret <2 x double> %2
 }
@@ -335,7 +335,7 @@ declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounw
 define i32 @stack_fold_cmpss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_cmpss
   ;CHECK:       vcmpeqss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp oeq float %a0, %a1
   %3 = zext i1 %2 to i32
   ret i32 %3
@@ -344,7 +344,7 @@ define i32 @stack_fold_cmpss(float %a0, float %a1) {
 define <4 x float> @stack_fold_cmpss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_cmpss_int
   ;CHECK:       vcmpeqss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.cmp.ss(<4 x float> %a0, <4 x float> %a1, i8 0)
   ret <4 x float> %2
 }
@@ -355,7 +355,7 @@ declare <4 x float> @llvm.x86.sse.cmp.ss(<4 x float>, <4 x float>, i8) nounwind
 define i32 @stack_fold_comisd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_comisd_int
   ;CHECK:       vcomisd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse2.comieq.sd(<2 x double> %a0, <2 x double> %a1)
   ret i32 %2
 }
@@ -366,7 +366,7 @@ declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readno
 define i32 @stack_fold_comiss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_comiss_int
   ;CHECK:       vcomiss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse.comieq.ss(<4 x float> %a0, <4 x float> %a1)
   ret i32 %2
 }
@@ -375,7 +375,7 @@ declare i32 @llvm.x86.sse.comieq.ss(<4 x float>, <4 x float>) nounwind readnone
 define <2 x double> @stack_fold_cvtdq2pd(<4 x i32> %a0) {
   ;CHECK-LABEL: stack_fold_cvtdq2pd
   ;CHECK:   vcvtdq2pd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.cvtdq2pd(<4 x i32> %a0)
   ret <2 x double> %2
 }
@@ -384,7 +384,7 @@ declare <2 x double> @llvm.x86.sse2.cvtdq2pd(<4 x i32>) nounwind readnone
 define <4 x double> @stack_fold_cvtdq2pd_ymm(<4 x i32> %a0) {
   ;CHECK-LABEL: stack_fold_cvtdq2pd_ymm
   ;CHECK:   vcvtdq2pd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.cvtdq2.pd.256(<4 x i32> %a0)
   ret <4 x double> %2
 }
@@ -393,7 +393,7 @@ declare <4 x double> @llvm.x86.avx.cvtdq2.pd.256(<4 x i32>) nounwind readnone
 define <4 x float> @stack_fold_cvtdq2ps(<4 x i32> %a0) {
   ;CHECK-LABEL: stack_fold_cvtdq2ps
   ;CHECK:   vcvtdq2ps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = sitofp <4 x i32> %a0 to <4 x float>
   ret <4 x float> %2
 }
@@ -401,7 +401,7 @@ define <4 x float> @stack_fold_cvtdq2ps(<4 x i32> %a0) {
 define <8 x float> @stack_fold_cvtdq2ps_ymm(<8 x i32> %a0) {
   ;CHECK-LABEL: stack_fold_cvtdq2ps_ymm
   ;CHECK:   vcvtdq2ps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = sitofp <8 x i32> %a0 to <8 x float>
   ret <8 x float> %2
 }
@@ -409,7 +409,7 @@ define <8 x float> @stack_fold_cvtdq2ps_ymm(<8 x i32> %a0) {
 define <4 x i32> @stack_fold_cvtpd2dq(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtpd2dq
   ;CHECK:   vcvtpd2dqx {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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.sse2.cvtpd2dq(<2 x double> %a0)
   ret <4 x i32> %2
 }
@@ -418,7 +418,7 @@ declare <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double>) nounwind readnone
 define <4 x i32> @stack_fold_cvtpd2dq_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtpd2dq_ymm
   ;CHECK:   vcvtpd2dqy {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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.avx.cvt.pd2dq.256(<4 x double> %a0)
   ret <4 x i32> %2
 }
@@ -427,7 +427,7 @@ declare <4 x i32> @llvm.x86.avx.cvt.pd2dq.256(<4 x double>) nounwind readnone
 define <2 x float> @stack_fold_cvtpd2ps(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtpd2ps
   ;CHECK:   vcvtpd2psx {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptrunc <2 x double> %a0 to <2 x float>
   ret <2 x float> %2
 }
@@ -435,7 +435,7 @@ define <2 x float> @stack_fold_cvtpd2ps(<2 x double> %a0) {
 define <4 x float> @stack_fold_cvtpd2ps_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtpd2ps_ymm
   ;CHECK:   vcvtpd2psy {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptrunc <4 x double> %a0 to <4 x float>
   ret <4 x float> %2
 }
@@ -443,7 +443,7 @@ define <4 x float> @stack_fold_cvtpd2ps_ymm(<4 x double> %a0) {
 define <4 x float> @stack_fold_cvtph2ps(<8 x i16> %a0) {
   ;CHECK-LABEL: stack_fold_cvtph2ps
   ;CHECK:   vcvtph2ps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.vcvtph2ps.128(<8 x i16> %a0)
   ret <4 x float> %2
 }
@@ -452,7 +452,7 @@ declare <4 x float> @llvm.x86.vcvtph2ps.128(<8 x i16>) nounwind readonly
 define <8 x float> @stack_fold_cvtph2ps_ymm(<8 x i16> %a0) {
   ;CHECK-LABEL: stack_fold_cvtph2ps_ymm
   ;CHECK:   vcvtph2ps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.vcvtph2ps.256(<8 x i16> %a0)
   ret <8 x float> %2
 }
@@ -461,7 +461,7 @@ declare <8 x float> @llvm.x86.vcvtph2ps.256(<8 x i16>) nounwind readonly
 define <4 x i32> @stack_fold_cvtps2dq(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2dq
   ;CHECK:  vcvtps2dq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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.sse2.cvtps2dq(<4 x float> %a0)
   ret <4 x i32> %2
 }
@@ -470,7 +470,7 @@ declare <4 x i32> @llvm.x86.sse2.cvtps2dq(<4 x float>) nounwind readnone
 define <8 x i32> @stack_fold_cvtps2dq_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2dq_ymm
   ;CHECK:  vcvtps2dq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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.avx.cvt.ps2dq.256(<8 x float> %a0)
   ret <8 x i32> %2
 }
@@ -479,7 +479,7 @@ declare <8 x i32> @llvm.x86.avx.cvt.ps2dq.256(<8 x float>) nounwind readnone
 define <2 x double> @stack_fold_cvtps2pd(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2pd
   ;CHECK:   vcvtps2pd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.cvtps2pd(<4 x float> %a0)
   ret <2 x double> %2
 }
@@ -488,7 +488,7 @@ declare <2 x double> @llvm.x86.sse2.cvtps2pd(<4 x float>) nounwind readnone
 define <4 x double> @stack_fold_cvtps2pd_ymm(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2pd_ymm
   ;CHECK:   vcvtps2pd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.cvt.ps2.pd.256(<4 x float> %a0)
   ret <4 x double> %2
 }
@@ -498,7 +498,7 @@ define <8 x i16> @stack_fold_cvtps2ph(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2ph
   ;CHECK:   vcvtps2ph $0, {{%xmm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp) {{.*#+}} 16-byte Folded Spill
   %1 = call <8 x i16> @llvm.x86.vcvtps2ph.128(<4 x float> %a0, i32 0)
-  %2 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %2 = 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}"()
   ret <8 x i16> %1
 }
 declare <8 x i16> @llvm.x86.vcvtps2ph.128(<4 x float>, i32) nounwind readonly
@@ -507,7 +507,7 @@ define <8 x i16> @stack_fold_cvtps2ph_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtps2ph_ymm
   ;CHECK:   vcvtps2ph $0, {{%ymm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp) {{.*#+}} 16-byte Folded Spill
   %1 = call <8 x i16> @llvm.x86.vcvtps2ph.256(<8 x float> %a0, i32 0)
-  %2 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %2 = 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}"()
   ret <8 x i16> %1
 }
 declare <8 x i16> @llvm.x86.vcvtps2ph.256(<8 x float>, i32) nounwind readonly
@@ -517,7 +517,7 @@ declare <8 x i16> @llvm.x86.vcvtps2ph.256(<8 x float>, i32) nounwind readonly
 define i32 @stack_fold_cvtsd2si_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtsd2si_int
   ;CHECK:  cvtsd2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %a0)
   ret i32 %2
 }
@@ -528,7 +528,7 @@ declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>) nounwind readnone
 define i64 @stack_fold_cvtsd2si64_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtsd2si64_int
   ;CHECK:  cvtsd2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i64 @llvm.x86.sse2.cvtsd2si64(<2 x double> %a0)
   ret i64 %2
 }
@@ -539,7 +539,7 @@ declare i64 @llvm.x86.sse2.cvtsd2si64(<2 x double>) nounwind readnone
 define <4 x float> @stack_fold_cvtsd2ss_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvtsd2ss_int
   ;CHECK:  cvtsd2ss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse2.cvtsd2ss(<4 x float> <float 0x0, float 0x0, float 0x0, float 0x0>, <2 x double> %a0)
   ret <4 x float> %2
 }
@@ -548,7 +548,7 @@ declare <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float>, <2 x double>) nounwind
 define double @stack_fold_cvtsi2sd(i32 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi2sd
   ;CHECK:  cvtsi2sdl {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = sitofp i32 %a0 to double
   ret double %2
 }
@@ -556,7 +556,7 @@ define double @stack_fold_cvtsi2sd(i32 %a0) {
 define <2 x double> @stack_fold_cvtsi2sd_int(i32 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi2sd_int
   ;CHECK:  cvtsi2sdl {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = call <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double> <double 0x0, double 0x0>, i32 %a0)
   ret <2 x double> %2
 }
@@ -565,7 +565,7 @@ declare <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double>, i32) nounwind readnon
 define double @stack_fold_cvtsi642sd(i64 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi642sd
   ;CHECK:  cvtsi2sdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = sitofp i64 %a0 to double
   ret double %2
 }
@@ -573,7 +573,7 @@ define double @stack_fold_cvtsi642sd(i64 %a0) {
 define <2 x double> @stack_fold_cvtsi642sd_int(i64 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi642sd_int
   ;CHECK:  cvtsi2sdq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = call <2 x double> @llvm.x86.sse2.cvtsi642sd(<2 x double> <double 0x0, double 0x0>, i64 %a0)
   ret <2 x double> %2
 }
@@ -582,7 +582,7 @@ declare <2 x double> @llvm.x86.sse2.cvtsi642sd(<2 x double>, i64) nounwind readn
 define float @stack_fold_cvtsi2ss(i32 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi2ss
   ;CHECK:  cvtsi2ssl {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = sitofp i32 %a0 to float
   ret float %2
 }
@@ -590,7 +590,7 @@ define float @stack_fold_cvtsi2ss(i32 %a0) {
 define <4 x float> @stack_fold_cvtsi2ss_int(i32 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi2ss_int
   ;CHECK:  cvtsi2ssl {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = call <4 x float> @llvm.x86.sse.cvtsi2ss(<4 x float> <float 0x0, float 0x0, float 0x0, float 0x0>, i32 %a0)
   ret <4 x float> %2
 }
@@ -599,7 +599,7 @@ declare <4 x float> @llvm.x86.sse.cvtsi2ss(<4 x float>, i32) nounwind readnone
 define float @stack_fold_cvtsi642ss(i64 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi642ss
   ;CHECK:  cvtsi2ssq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = sitofp i64 %a0 to float
   ret float %2
 }
@@ -607,7 +607,7 @@ define float @stack_fold_cvtsi642ss(i64 %a0) {
 define <4 x float> @stack_fold_cvtsi642ss_int(i64 %a0) {
   ;CHECK-LABEL: stack_fold_cvtsi642ss_int
   ;CHECK:  cvtsi2ssq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   %2 = call <4 x float> @llvm.x86.sse.cvtsi642ss(<4 x float> <float 0x0, float 0x0, float 0x0, float 0x0>, i64 %a0)
   ret <4 x float> %2
 }
@@ -618,7 +618,7 @@ declare <4 x float> @llvm.x86.sse.cvtsi642ss(<4 x float>, i64) nounwind readnone
 define <2 x double> @stack_fold_cvtss2sd_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtss2sd_int
   ;CHECK:  cvtss2sd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.cvtss2sd(<2 x double> <double 0x0, double 0x0>, <4 x float> %a0)
   ret <2 x double> %2
 }
@@ -629,7 +629,7 @@ declare <2 x double> @llvm.x86.sse2.cvtss2sd(<2 x double>, <4 x float>) nounwind
 define i32 @stack_fold_cvtss2si_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtss2si_int
   ;CHECK:  vcvtss2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse.cvtss2si(<4 x float> %a0)
   ret i32 %2
 }
@@ -640,7 +640,7 @@ declare i32 @llvm.x86.sse.cvtss2si(<4 x float>) nounwind readnone
 define i64 @stack_fold_cvtss2si64_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvtss2si64_int
   ;CHECK:  vcvtss2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i64 @llvm.x86.sse.cvtss2si64(<4 x float> %a0)
   ret i64 %2
 }
@@ -649,7 +649,7 @@ declare i64 @llvm.x86.sse.cvtss2si64(<4 x float>) nounwind readnone
 define <4 x i32> @stack_fold_cvttpd2dq(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvttpd2dq
   ;CHECK:  vcvttpd2dqx {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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.sse2.cvttpd2dq(<2 x double> %a0)
   ret <4 x i32> %2
 }
@@ -658,7 +658,7 @@ declare <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double>) nounwind readnone
 define <4 x i32> @stack_fold_cvttpd2dq_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvttpd2dq_ymm
   ;CHECK:  vcvttpd2dqy {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi <4 x double> %a0 to <4 x i32>
   ret <4 x i32> %2
 }
@@ -666,7 +666,7 @@ define <4 x i32> @stack_fold_cvttpd2dq_ymm(<4 x double> %a0) {
 define <4 x i32> @stack_fold_cvttps2dq(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvttps2dq
   ;CHECK:  vcvttps2dq {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi <4 x float> %a0 to <4 x i32>
   ret <4 x i32> %2
 }
@@ -674,7 +674,7 @@ define <4 x i32> @stack_fold_cvttps2dq(<4 x float> %a0) {
 define <8 x i32> @stack_fold_cvttps2dq_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvttps2dq_ymm
   ;CHECK:  vcvttps2dq {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi <8 x float> %a0 to <8 x i32>
   ret <8 x i32> %2
 }
@@ -682,7 +682,7 @@ define <8 x i32> @stack_fold_cvttps2dq_ymm(<8 x float> %a0) {
 define i32 @stack_fold_cvttsd2si(double %a0) {
   ;CHECK-LABEL: stack_fold_cvttsd2si
   ;CHECK:  vcvttsd2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi double %a0 to i32
   ret i32 %2
 }
@@ -690,7 +690,7 @@ define i32 @stack_fold_cvttsd2si(double %a0) {
 define i32 @stack_fold_cvttsd2si_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvttsd2si_int
   ;CHECK:  vcvttsd2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse2.cvttsd2si(<2 x double> %a0)
   ret i32 %2
 }
@@ -699,7 +699,7 @@ declare i32 @llvm.x86.sse2.cvttsd2si(<2 x double>) nounwind readnone
 define i64 @stack_fold_cvttsd2si64(double %a0) {
   ;CHECK-LABEL: stack_fold_cvttsd2si64
   ;CHECK:  vcvttsd2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi double %a0 to i64
   ret i64 %2
 }
@@ -707,7 +707,7 @@ define i64 @stack_fold_cvttsd2si64(double %a0) {
 define i64 @stack_fold_cvttsd2si64_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_cvttsd2si64_int
   ;CHECK:  vcvttsd2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i64 @llvm.x86.sse2.cvttsd2si64(<2 x double> %a0)
   ret i64 %2
 }
@@ -716,7 +716,7 @@ declare i64 @llvm.x86.sse2.cvttsd2si64(<2 x double>) nounwind readnone
 define i32 @stack_fold_cvttss2si(float %a0) {
   ;CHECK-LABEL: stack_fold_cvttss2si
   ;CHECK:  vcvttss2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi float %a0 to i32
   ret i32 %2
 }
@@ -724,7 +724,7 @@ define i32 @stack_fold_cvttss2si(float %a0) {
 define i32 @stack_fold_cvttss2si_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvttss2si_int
   ;CHECK:  vcvttss2si {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse.cvttss2si(<4 x float> %a0)
   ret i32 %2
 }
@@ -733,7 +733,7 @@ declare i32 @llvm.x86.sse.cvttss2si(<4 x float>) nounwind readnone
 define i64 @stack_fold_cvttss2si64(float %a0) {
   ;CHECK-LABEL: stack_fold_cvttss2si64
   ;CHECK:  vcvttss2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fptosi float %a0 to i64
   ret i64 %2
 }
@@ -741,7 +741,7 @@ define i64 @stack_fold_cvttss2si64(float %a0) {
 define i64 @stack_fold_cvttss2si64_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_cvttss2si64_int
   ;CHECK:  cvttss2si {{-?[0-9]*}}(%rsp), %rax {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i64 @llvm.x86.sse.cvttss2si64(<4 x float> %a0)
   ret i64 %2
 }
@@ -750,7 +750,7 @@ declare i64 @llvm.x86.sse.cvttss2si64(<4 x float>) nounwind readnone
 define <2 x double> @stack_fold_divpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_divpd
   ;CHECK:       vdivpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv <2 x double> %a0, %a1
   ret <2 x double> %2
 }
@@ -758,7 +758,7 @@ define <2 x double> @stack_fold_divpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_divpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_divpd_ymm
   ;CHECK:       vdivpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv <4 x double> %a0, %a1
   ret <4 x double> %2
 }
@@ -766,7 +766,7 @@ define <4 x double> @stack_fold_divpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_divps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_divps
   ;CHECK:       vdivps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv <4 x float> %a0, %a1
   ret <4 x float> %2
 }
@@ -774,7 +774,7 @@ define <4 x float> @stack_fold_divps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_divps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_divps_ymm
   ;CHECK:       vdivps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv <8 x float> %a0, %a1
   ret <8 x float> %2
 }
@@ -782,7 +782,7 @@ define <8 x float> @stack_fold_divps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define double @stack_fold_divsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_divsd
   ;CHECK:       vdivsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv double %a0, %a1
   ret double %2
 }
@@ -790,7 +790,7 @@ define double @stack_fold_divsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_divsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_divsd_int
   ;CHECK:       vdivsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.div.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -799,7 +799,7 @@ declare <2 x double> @llvm.x86.sse2.div.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_divss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_divss
   ;CHECK:       vdivss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fdiv float %a0, %a1
   ret float %2
 }
@@ -807,7 +807,7 @@ define float @stack_fold_divss(float %a0, float %a1) {
 define <4 x float> @stack_fold_divss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_divss_int
   ;CHECK:       vdivss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.div.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -816,7 +816,7 @@ declare <4 x float> @llvm.x86.sse.div.ss(<4 x float>, <4 x float>) nounwind read
 define <2 x double> @stack_fold_dppd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_dppd
   ;CHECK:       vdppd $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse41.dppd(<2 x double> %a0, <2 x double> %a1, i8 7)
   ret <2 x double> %2
 }
@@ -825,7 +825,7 @@ declare <2 x double> @llvm.x86.sse41.dppd(<2 x double>, <2 x double>, i8) nounwi
 define <4 x float> @stack_fold_dpps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_dpps
   ;CHECK:       vdpps $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse41.dpps(<4 x float> %a0, <4 x float> %a1, i8 7)
   ret <4 x float> %2
 }
@@ -834,7 +834,7 @@ declare <4 x float> @llvm.x86.sse41.dpps(<4 x float>, <4 x float>, i8) nounwind
 define <8 x float> @stack_fold_dpps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_dpps_ymm
   ;CHECK:       vdpps $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.dp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 7)
   ret <8 x float> %2
 }
@@ -845,7 +845,7 @@ define <4 x float> @stack_fold_extractf128(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK:       vextractf128 $1, {{%ymm[0-9][0-9]*}}, {{-?[0-9]*}}(%rsp) {{.*#+}} 16-byte Folded Spill
   ;CHECK:       vmovaps {{-?[0-9]*}}(%rsp), %xmm0 {{.*#+}} 16-byte Reload
   %1 = shufflevector <8 x float> %a0, <8 x float> %a1, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-  %2 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %2 = 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}"()
   ret <4 x float> %1
 }
 
@@ -855,14 +855,14 @@ define i32 @stack_fold_extractps(<4 x float> %a0) {
   ;CHECK:       movl    {{-?[0-9]*}}(%rsp), %eax {{.*#+}} 4-byte Reload
   %1 = extractelement <4 x float> %a0, i32 1
   %2 = bitcast float %1 to i32
-  %3 = tail call <4 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
+  %3 = tail call <2 x i64> asm sideeffect "nop", "=x,~{rax},~{rbx},~{rcx},~{rdx},~{rsi},~{rdi},~{rbp},~{r8},~{r9},~{r10},~{r11},~{r12},~{r13},~{r14},~{r15}"()
   ret i32 %2
 }
 
 define <2 x double> @stack_fold_haddpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_haddpd
   ;CHECK:       vhaddpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse3.hadd.pd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -871,7 +871,7 @@ declare <2 x double> @llvm.x86.sse3.hadd.pd(<2 x double>, <2 x double>) nounwind
 define <4 x double> @stack_fold_haddpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_haddpd_ymm
   ;CHECK:       vhaddpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.hadd.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret <4 x double> %2
 }
@@ -880,7 +880,7 @@ declare <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double>, <4 x double>) nounw
 define <4 x float> @stack_fold_haddps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_haddps
   ;CHECK:       vhaddps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse3.hadd.ps(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -889,7 +889,7 @@ declare <4 x float> @llvm.x86.sse3.hadd.ps(<4 x float>, <4 x float>) nounwind re
 define <8 x float> @stack_fold_haddps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_haddps_ymm
   ;CHECK:       vhaddps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.hadd.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret <8 x float> %2
 }
@@ -898,7 +898,7 @@ declare <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float>, <8 x float>) nounwind
 define <2 x double> @stack_fold_hsubpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_hsubpd
   ;CHECK:       vhsubpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse3.hsub.pd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -907,7 +907,7 @@ declare <2 x double> @llvm.x86.sse3.hsub.pd(<2 x double>, <2 x double>) nounwind
 define <4 x double> @stack_fold_hsubpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_hsubpd_ymm
   ;CHECK:       vhsubpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.hsub.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret <4 x double> %2
 }
@@ -916,7 +916,7 @@ declare <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double>, <4 x double>) nounw
 define <4 x float> @stack_fold_hsubps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_hsubps
   ;CHECK:       vhsubps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse3.hsub.ps(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -925,7 +925,7 @@ declare <4 x float> @llvm.x86.sse3.hsub.ps(<4 x float>, <4 x float>) nounwind re
 define <8 x float> @stack_fold_hsubps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_hsubps_ymm
   ;CHECK:       vhsubps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.hsub.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret <8 x float> %2
 }
@@ -934,7 +934,7 @@ declare <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float>, <8 x float>) nounwind
 define <8 x float> @stack_fold_insertf128(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_insertf128
   ;CHECK:       vinsertf128 $1, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> %a1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   ret <8 x float> %2
 }
@@ -944,7 +944,7 @@ define <8 x float> @stack_fold_insertf128(<4 x float> %a0, <4 x float> %a1) {
 define <2 x double> @stack_fold_maxpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_maxpd
   ;CHECK:       vmaxpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.max.pd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -953,7 +953,7 @@ declare <2 x double> @llvm.x86.sse2.max.pd(<2 x double>, <2 x double>) nounwind
 define <4 x double> @stack_fold_maxpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_maxpd_ymm
   ;CHECK:       vmaxpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.max.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret <4 x double> %2
 }
@@ -962,7 +962,7 @@ declare <4 x double> @llvm.x86.avx.max.pd.256(<4 x double>, <4 x double>) nounwi
 define <4 x float> @stack_fold_maxps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_maxps
   ;CHECK:       vmaxps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.max.ps(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -971,7 +971,7 @@ declare <4 x float> @llvm.x86.sse.max.ps(<4 x float>, <4 x float>) nounwind read
 define <8 x float> @stack_fold_maxps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_maxps_ymm
   ;CHECK:       vmaxps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.max.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret <8 x float> %2
 }
@@ -980,7 +980,7 @@ declare <8 x float> @llvm.x86.avx.max.ps.256(<8 x float>, <8 x float>) nounwind
 define double @stack_fold_maxsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_maxsd
   ;CHECK:       vmaxsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp ogt double %a0, %a1
   %3 = select i1 %2, double %a0, double %a1
   ret double %3
@@ -989,7 +989,7 @@ define double @stack_fold_maxsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_maxsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_maxsd_int
   ;CHECK:       vmaxsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.max.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -998,7 +998,7 @@ declare <2 x double> @llvm.x86.sse2.max.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_maxss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_maxss
   ;CHECK:       vmaxss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp ogt float %a0, %a1
   %3 = select i1 %2, float %a0, float %a1
   ret float %3
@@ -1007,7 +1007,7 @@ define float @stack_fold_maxss(float %a0, float %a1) {
 define <4 x float> @stack_fold_maxss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_maxss_int
   ;CHECK:       vmaxss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.max.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -1016,7 +1016,7 @@ declare <4 x float> @llvm.x86.sse.max.ss(<4 x float>, <4 x float>) nounwind read
 define <2 x double> @stack_fold_minpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_minpd
   ;CHECK:       vminpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.min.pd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -1025,7 +1025,7 @@ declare <2 x double> @llvm.x86.sse2.min.pd(<2 x double>, <2 x double>) nounwind
 define <4 x double> @stack_fold_minpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_minpd_ymm
   ;CHECK:       vminpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.min.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret <4 x double> %2
 }
@@ -1034,7 +1034,7 @@ declare <4 x double> @llvm.x86.avx.min.pd.256(<4 x double>, <4 x double>) nounwi
 define <4 x float> @stack_fold_minps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_minps
   ;CHECK:       vminps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.min.ps(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -1043,7 +1043,7 @@ declare <4 x float> @llvm.x86.sse.min.ps(<4 x float>, <4 x float>) nounwind read
 define <8 x float> @stack_fold_minps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_minps_ymm
   ;CHECK:       vminps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.min.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret <8 x float> %2
 }
@@ -1052,7 +1052,7 @@ declare <8 x float> @llvm.x86.avx.min.ps.256(<8 x float>, <8 x float>) nounwind
 define double @stack_fold_minsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_minsd
   ;CHECK:       vminsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp olt double %a0, %a1
   %3 = select i1 %2, double %a0, double %a1
   ret double %3
@@ -1061,7 +1061,7 @@ define double @stack_fold_minsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_minsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_minsd_int
   ;CHECK:       vminsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.min.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -1070,7 +1070,7 @@ declare <2 x double> @llvm.x86.sse2.min.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_minss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_minss
   ;CHECK:       vminss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp olt float %a0, %a1
   %3 = select i1 %2, float %a0, float %a1
   ret float %3
@@ -1079,7 +1079,7 @@ define float @stack_fold_minss(float %a0, float %a1) {
 define <4 x float> @stack_fold_minss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_minss_int
   ;CHECK:       vminss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.min.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -1103,7 +1103,7 @@ declare <4 x float> @llvm.x86.sse.min.ss(<4 x float>, <4 x float>) nounwind read
 define <4 x float> @stack_fold_movshdup(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_movshdup
   ;CHECK:   vmovshdup {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> undef, <4 x i32> <i32 1, i32 1, i32 3, i32 3>
   ret <4 x float> %2
 }
@@ -1113,7 +1113,7 @@ define <4 x float> @stack_fold_movshdup(<4 x float> %a0) {
 define <4 x float> @stack_fold_movsldup(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_movsldup
   ;CHECK:   vmovsldup {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
   ret <4 x float> %2
 }
@@ -1123,7 +1123,7 @@ define <4 x float> @stack_fold_movsldup(<4 x float> %a0) {
 define <2 x double> @stack_fold_mulpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_mulpd
   ;CHECK:       vmulpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul <2 x double> %a0, %a1
   ret <2 x double> %2
 }
@@ -1131,7 +1131,7 @@ define <2 x double> @stack_fold_mulpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_mulpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_mulpd_ymm
   ;CHECK:       vmulpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul <4 x double> %a0, %a1
   ret <4 x double> %2
 }
@@ -1139,7 +1139,7 @@ define <4 x double> @stack_fold_mulpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_mulps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_mulps
   ;CHECK:       vmulps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul <4 x float> %a0, %a1
   ret <4 x float> %2
 }
@@ -1147,7 +1147,7 @@ define <4 x float> @stack_fold_mulps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_mulps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_mulps_ymm
   ;CHECK:       vmulps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul <8 x float> %a0, %a1
   ret <8 x float> %2
 }
@@ -1155,7 +1155,7 @@ define <8 x float> @stack_fold_mulps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define double @stack_fold_mulsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_mulsd
   ;CHECK:       vmulsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul double %a0, %a1
   ret double %2
 }
@@ -1163,7 +1163,7 @@ define double @stack_fold_mulsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_mulsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_mulsd_int
   ;CHECK:       vmulsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.mul.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -1172,7 +1172,7 @@ declare <2 x double> @llvm.x86.sse2.mul.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_mulss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_mulss
   ;CHECK:       vmulss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fmul float %a0, %a1
   ret float %2
 }
@@ -1180,7 +1180,7 @@ define float @stack_fold_mulss(float %a0, float %a1) {
 define <4 x float> @stack_fold_mulss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_mulss_int
   ;CHECK:       vmulss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.mul.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -1189,7 +1189,7 @@ declare <4 x float> @llvm.x86.sse.mul.ss(<4 x float>, <4 x float>) nounwind read
 define <2 x double> @stack_fold_orpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_orpd
   ;CHECK:       vorpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <2 x double> %a0 to <2 x i64>
   %3 = bitcast <2 x double> %a1 to <2 x i64>
   %4 = or <2 x i64> %2, %3
@@ -1202,7 +1202,7 @@ define <2 x double> @stack_fold_orpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_orpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_orpd_ymm
   ;CHECK:       vorpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x double> %a0 to <4 x i64>
   %3 = bitcast <4 x double> %a1 to <4 x i64>
   %4 = or <4 x i64> %2, %3
@@ -1215,7 +1215,7 @@ define <4 x double> @stack_fold_orpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_orps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_orps
   ;CHECK:       vorps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x float> %a0 to <2 x i64>
   %3 = bitcast <4 x float> %a1 to <2 x i64>
   %4 = or <2 x i64> %2, %3
@@ -1226,7 +1226,7 @@ define <4 x float> @stack_fold_orps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_orps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_orps_ymm
   ;CHECK:       vorps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <8 x float> %a0 to <4 x i64>
   %3 = bitcast <8 x float> %a1 to <4 x i64>
   %4 = or <4 x i64> %2, %3
@@ -1237,7 +1237,7 @@ define <8 x float> @stack_fold_orps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <8 x float> @stack_fold_perm2f128(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_perm2f128
   ;CHECK:   vperm2f128 $33, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <8 x float> %a1, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11>
   ret <8 x float> %2
 }
@@ -1245,7 +1245,7 @@ define <8 x float> @stack_fold_perm2f128(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_permilpd(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_permilpd
   ;CHECK:   vpermilpd $1, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 <2 x double> %a0, <2 x double> undef, <2 x i32> <i32 1, i32 0>
   ret <2 x double> %2
 }
@@ -1253,7 +1253,7 @@ define <2 x double> @stack_fold_permilpd(<2 x double> %a0) {
 define <4 x double> @stack_fold_permilpd_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_permilpd_ymm
   ;CHECK:   vpermilpd $5, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <4 x double> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
   ret <4 x double> %2
 }
@@ -1261,7 +1261,7 @@ define <4 x double> @stack_fold_permilpd_ymm(<4 x double> %a0) {
 define <2 x double> @stack_fold_permilpdvar(<2 x double> %a0, <2 x i64> %a1) {
   ;CHECK-LABEL: stack_fold_permilpdvar
   ;CHECK:       vpermilpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %a0, <2 x i64> %a1)
   ret <2 x double> %2
 }
@@ -1270,7 +1270,7 @@ declare <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double>, <2 x i64>) nounwi
 define <4 x double> @stack_fold_permilpdvar_ymm(<4 x double> %a0, <4 x i64> %a1) {
   ;CHECK-LABEL: stack_fold_permilpdvar_ymm
   ;CHECK:       vpermilpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %a0, <4 x i64> %a1)
   ret <4 x double> %2
 }
@@ -1279,7 +1279,7 @@ declare <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double>, <4 x i64>) no
 define <4 x float> @stack_fold_permilps(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_permilps
   ;CHECK:   vpermilps $27, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
   ret <4 x float> %2
 }
@@ -1287,7 +1287,7 @@ define <4 x float> @stack_fold_permilps(<4 x float> %a0) {
 define <8 x float> @stack_fold_permilps_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_permilps_ymm
   ;CHECK:   vpermilps $27, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <8 x float> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
   ret <8 x float> %2
 }
@@ -1295,7 +1295,7 @@ define <8 x float> @stack_fold_permilps_ymm(<8 x float> %a0) {
 define <4 x float> @stack_fold_permilpsvar(<4 x float> %a0, <4 x i32> %a1) {
   ;CHECK-LABEL: stack_fold_permilpsvar
   ;CHECK:       vpermilps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.vpermilvar.ps(<4 x float> %a0, <4 x i32> %a1)
   ret <4 x float> %2
 }
@@ -1304,7 +1304,7 @@ declare <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float>, <4 x i32>) nounwind
 define <8 x float> @stack_fold_permilpsvar_ymm(<8 x float> %a0, <8 x i32> %a1) {
   ;CHECK-LABEL: stack_fold_permilpsvar_ymm
   ;CHECK:       vpermilps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float> %a0, <8 x i32> %a1)
   ret <8 x float> %2
 }
@@ -1315,7 +1315,7 @@ declare <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float>, <8 x i32>) noun
 define <4 x float> @stack_fold_rcpps_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rcpps_int
   ;CHECK:       vrcpps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.rcp.ps(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1326,7 +1326,7 @@ declare <4 x float> @llvm.x86.sse.rcp.ps(<4 x float>) nounwind readnone
 define <8 x float> @stack_fold_rcpps_ymm_int(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rcpps_ymm_int
   ;CHECK:       vrcpps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.rcp.ps.256(<8 x float> %a0)
   ret <8 x float> %2
 }
@@ -1337,7 +1337,7 @@ declare <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float>) nounwind readnone
 define <4 x float> @stack_fold_rcpss_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rcpss_int
   ;CHECK:       vrcpss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.rcp.ss(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1346,7 +1346,7 @@ declare <4 x float> @llvm.x86.sse.rcp.ss(<4 x float>) nounwind readnone
 define <2 x double> @stack_fold_roundpd(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_roundpd
   ;CHECK:  vroundpd $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse41.round.pd(<2 x double> %a0, i32 7)
   ret <2 x double> %2
 }
@@ -1355,7 +1355,7 @@ declare <2 x double> @llvm.x86.sse41.round.pd(<2 x double>, i32) nounwind readno
 define <4 x double> @stack_fold_roundpd_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_roundpd_ymm
   ;CHECK:  vroundpd $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.round.pd.256(<4 x double> %a0, i32 7)
   ret <4 x double> %2
 }
@@ -1364,7 +1364,7 @@ declare <4 x double> @llvm.x86.avx.round.pd.256(<4 x double>, i32) nounwind read
 define <4 x float> @stack_fold_roundps(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_roundps
   ;CHECK:  vroundps $7, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse41.round.ps(<4 x float> %a0, i32 7)
   ret <4 x float> %2
 }
@@ -1373,7 +1373,7 @@ declare <4 x float> @llvm.x86.sse41.round.ps(<4 x float>, i32) nounwind readnone
 define <8 x float> @stack_fold_roundps_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_roundps_ymm
   ;CHECK:  vroundps $7, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.round.ps.256(<8 x float> %a0, i32 7)
   ret <8 x float> %2
 }
@@ -1394,7 +1394,7 @@ declare <4 x float> @llvm.x86.sse41.round.ss(<4 x float>, <4 x float>, i32) noun
 define <4 x float> @stack_fold_rsqrtps_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rsqrtps_int
   ;CHECK:       vrsqrtps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.rsqrt.ps(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1405,7 +1405,7 @@ declare <4 x float> @llvm.x86.sse.rsqrt.ps(<4 x float>) nounwind readnone
 define <8 x float> @stack_fold_rsqrtps_ymm_int(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rsqrtps_ymm_int
   ;CHECK:       vrsqrtps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.rsqrt.ps.256(<8 x float> %a0)
   ret <8 x float> %2
 }
@@ -1416,7 +1416,7 @@ declare <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float>) nounwind readnone
 define <4 x float> @stack_fold_rsqrtss_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_rsqrtss_int
   ;CHECK:       vrsqrtss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.rsqrt.ss(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1425,7 +1425,7 @@ declare <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float>) nounwind readnone
 define <2 x double> @stack_fold_shufpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_shufpd
   ;CHECK:       vshufpd $1, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <2 x double> %a1, <2 x i32> <i32 1, i32 2>
   ret <2 x double> %2
 }
@@ -1433,7 +1433,7 @@ define <2 x double> @stack_fold_shufpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_shufpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_shufpd_ymm
   ;CHECK:       vshufpd $5, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <4 x double> %a1, <4 x i32> <i32 1, i32 4, i32 3, i32 6>
   ret <4 x double> %2
 }
@@ -1441,7 +1441,7 @@ define <4 x double> @stack_fold_shufpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_shufps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_shufps
   ;CHECK:       vshufps $-56, {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> %a1, <4 x i32> <i32 0, i32 2, i32 4, i32 7>
   ret <4 x float> %2
 }
@@ -1449,7 +1449,7 @@ define <4 x float> @stack_fold_shufps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_shufps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_shufps_ymm
   ;CHECK:       vshufps $-108, {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <8 x float> %a1, <8 x i32> <i32 0, i32 1, i32 9, i32 10, i32 undef, i32 undef, i32 undef, i32 undef>
   ret <8 x float> %2
 }
@@ -1457,7 +1457,7 @@ define <8 x float> @stack_fold_shufps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_sqrtpd(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtpd
   ;CHECK:       vsqrtpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.sqrt.pd(<2 x double> %a0)
   ret <2 x double> %2
 }
@@ -1466,7 +1466,7 @@ declare <2 x double> @llvm.x86.sse2.sqrt.pd(<2 x double>) nounwind readnone
 define <4 x double> @stack_fold_sqrtpd_ymm(<4 x double> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtpd_ymm
   ;CHECK:       vsqrtpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.avx.sqrt.pd.256(<4 x double> %a0)
   ret <4 x double> %2
 }
@@ -1475,7 +1475,7 @@ declare <4 x double> @llvm.x86.avx.sqrt.pd.256(<4 x double>) nounwind readnone
 define <4 x float> @stack_fold_sqrtps(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtps
   ;CHECK:       vsqrtps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.sqrt.ps(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1484,7 +1484,7 @@ declare <4 x float> @llvm.x86.sse.sqrt.ps(<4 x float>) nounwind readnone
 define <8 x float> @stack_fold_sqrtps_ymm(<8 x float> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtps_ymm
   ;CHECK:       vsqrtps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.avx.sqrt.ps.256(<8 x float> %a0)
   ret <8 x float> %2
 }
@@ -1493,7 +1493,7 @@ declare <8 x float> @llvm.x86.avx.sqrt.ps.256(<8 x float>) nounwind readnone
 define double @stack_fold_sqrtsd(double %a0) {
   ;CHECK-LABEL: stack_fold_sqrtsd
   ;CHECK:       vsqrtsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double @llvm.sqrt.f64(double %a0)
   ret double %2
 }
@@ -1502,7 +1502,7 @@ declare double @llvm.sqrt.f64(double) nounwind readnone
 define <2 x double> @stack_fold_sqrtsd_int(<2 x double> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtsd_int
   ;CHECK:       vsqrtsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.sqrt.sd(<2 x double> %a0)
   ret <2 x double> %2
 }
@@ -1511,7 +1511,7 @@ declare <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double>) nounwind readnone
 define float @stack_fold_sqrtss(float %a0) {
   ;CHECK-LABEL: stack_fold_sqrtss
   ;CHECK:       vsqrtss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float @llvm.sqrt.f32(float %a0)
   ret float %2
 }
@@ -1520,7 +1520,7 @@ declare float @llvm.sqrt.f32(float) nounwind readnone
 define <4 x float> @stack_fold_sqrtss_int(<4 x float> %a0) {
   ;CHECK-LABEL: stack_fold_sqrtss_int
   ;CHECK:       vsqrtss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.sqrt.ss(<4 x float> %a0)
   ret <4 x float> %2
 }
@@ -1529,7 +1529,7 @@ declare <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float>) nounwind readnone
 define <2 x double> @stack_fold_subpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_subpd
   ;CHECK:       vsubpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub <2 x double> %a0, %a1
   ret <2 x double> %2
 }
@@ -1537,7 +1537,7 @@ define <2 x double> @stack_fold_subpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_subpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_subpd_ymm
   ;CHECK:       vsubpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub <4 x double> %a0, %a1
   ret <4 x double> %2
 }
@@ -1545,7 +1545,7 @@ define <4 x double> @stack_fold_subpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_subps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_subps
   ;CHECK:       vsubps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub <4 x float> %a0, %a1
   ret <4 x float> %2
 }
@@ -1553,7 +1553,7 @@ define <4 x float> @stack_fold_subps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_subps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_subps_ymm
   ;CHECK:       vsubps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub <8 x float> %a0, %a1
   ret <8 x float> %2
 }
@@ -1561,7 +1561,7 @@ define <8 x float> @stack_fold_subps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define double @stack_fold_subsd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_subsd
   ;CHECK:       vsubsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub double %a0, %a1
   ret double %2
 }
@@ -1569,7 +1569,7 @@ define double @stack_fold_subsd(double %a0, double %a1) {
 define <2 x double> @stack_fold_subsd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_subsd_int
   ;CHECK:       vsubsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> @llvm.x86.sse2.sub.sd(<2 x double> %a0, <2 x double> %a1)
   ret <2 x double> %2
 }
@@ -1578,7 +1578,7 @@ declare <2 x double> @llvm.x86.sse2.sub.sd(<2 x double>, <2 x double>) nounwind
 define float @stack_fold_subss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_subss
   ;CHECK:       vsubss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fsub float %a0, %a1
   ret float %2
 }
@@ -1586,7 +1586,7 @@ define float @stack_fold_subss(float %a0, float %a1) {
 define <4 x float> @stack_fold_subss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_subss_int
   ;CHECK:       vsubss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> @llvm.x86.sse.sub.ss(<4 x float> %a0, <4 x float> %a1)
   ret <4 x float> %2
 }
@@ -1595,7 +1595,7 @@ declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>) nounwind read
 define i32 @stack_fold_testpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_testpd
   ;CHECK:       vtestpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.avx.vtestc.pd(<2 x double> %a0, <2 x double> %a1)
   ret i32 %2
 }
@@ -1604,7 +1604,7 @@ declare i32 @llvm.x86.avx.vtestc.pd(<2 x double>, <2 x double>) nounwind readnon
 define i32 @stack_fold_testpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_testpd_ymm
   ;CHECK:       vtestpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.avx.vtestc.pd.256(<4 x double> %a0, <4 x double> %a1)
   ret i32 %2
 }
@@ -1613,7 +1613,7 @@ declare i32 @llvm.x86.avx.vtestc.pd.256(<4 x double>, <4 x double>) nounwind rea
 define i32 @stack_fold_testps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_testps
   ;CHECK:       vtestps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.avx.vtestc.ps(<4 x float> %a0, <4 x float> %a1)
   ret i32 %2
 }
@@ -1622,7 +1622,7 @@ declare i32 @llvm.x86.avx.vtestc.ps(<4 x float>, <4 x float>) nounwind readnone
 define i32 @stack_fold_testps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_testps_ymm
   ;CHECK:       vtestps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.avx.vtestc.ps.256(<8 x float> %a0, <8 x float> %a1)
   ret i32 %2
 }
@@ -1631,7 +1631,7 @@ declare i32 @llvm.x86.avx.vtestc.ps.256(<8 x float>, <8 x float>) nounwind readn
 define i32 @stack_fold_ucomisd(double %a0, double %a1) {
   ;CHECK-LABEL: stack_fold_ucomisd
   ;CHECK:       vucomisd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp ueq double %a0, %a1
   %3 = select i1 %2, i32 1, i32 -1
   ret i32 %3
@@ -1640,7 +1640,7 @@ define i32 @stack_fold_ucomisd(double %a0, double %a1) {
 define i32 @stack_fold_ucomisd_int(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_ucomisd_int
   ;CHECK:       vucomisd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse2.ucomieq.sd(<2 x double> %a0, <2 x double> %a1)
   ret i32 %2
 }
@@ -1649,7 +1649,7 @@ declare i32 @llvm.x86.sse2.ucomieq.sd(<2 x double>, <2 x double>) nounwind readn
 define i32 @stack_fold_ucomiss(float %a0, float %a1) {
   ;CHECK-LABEL: stack_fold_ucomiss
   ;CHECK:       vucomiss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = fcmp ueq float %a0, %a1
   %3 = select i1 %2, i32 1, i32 -1
   ret i32 %3
@@ -1658,7 +1658,7 @@ define i32 @stack_fold_ucomiss(float %a0, float %a1) {
 define i32 @stack_fold_ucomiss_int(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_ucomiss_int
   ;CHECK:       vucomiss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 i32 @llvm.x86.sse.ucomieq.ss(<4 x float> %a0, <4 x float> %a1)
   ret i32 %2
 }
@@ -1667,7 +1667,7 @@ declare i32 @llvm.x86.sse.ucomieq.ss(<4 x float>, <4 x float>) nounwind readnone
 define <2 x double> @stack_fold_unpckhpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_unpckhpd
   ;CHECK:       vunpckhpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <2 x double> %a1, <2 x i32> <i32 1, i32 3>
   ret <2 x double> %2
 }
@@ -1675,7 +1675,7 @@ define <2 x double> @stack_fold_unpckhpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_unpckhpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_unpckhpd_ymm
   ;CHECK:       vunpckhpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <4 x double> %a1, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
   ret <4 x double> %2
 }
@@ -1683,7 +1683,7 @@ define <4 x double> @stack_fold_unpckhpd_ymm(<4 x double> %a0, <4 x double> %a1)
 define <4 x float> @stack_fold_unpckhps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_unpckhps
   ;CHECK:       vunpckhps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> %a1, <4 x i32> <i32 2, i32 6, i32 3, i32 7>
   ret <4 x float> %2
 }
@@ -1691,7 +1691,7 @@ define <4 x float> @stack_fold_unpckhps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_unpckhps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_unpckhps_ymm
   ;CHECK:       vunpckhps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <8 x float> %a1, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
   ret <8 x float> %2
 }
@@ -1699,7 +1699,7 @@ define <8 x float> @stack_fold_unpckhps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_unpcklpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_unpcklpd
   ;CHECK:       vunpcklpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <2 x double> %a1, <2 x i32> <i32 0, i32 2>
   ret <2 x double> %2
 }
@@ -1707,7 +1707,7 @@ define <2 x double> @stack_fold_unpcklpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_unpcklpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_unpcklpd_ymm
   ;CHECK:       vunpcklpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 double> %a0, <4 x double> %a1, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
   ret <4 x double> %2
 }
@@ -1715,7 +1715,7 @@ define <4 x double> @stack_fold_unpcklpd_ymm(<4 x double> %a0, <4 x double> %a1)
 define <4 x float> @stack_fold_unpcklps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_unpcklps
   ;CHECK:       vunpcklps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <4 x float> %a1, <4 x i32> <i32 0, i32 4, i32 1, i32 5>
   ret <4 x float> %2
 }
@@ -1723,7 +1723,7 @@ define <4 x float> @stack_fold_unpcklps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_unpcklps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_unpcklps_ymm
   ;CHECK:       vunpcklps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 float> %a0, <8 x float> %a1, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
   ret <8 x float> %2
 }
@@ -1731,7 +1731,7 @@ define <8 x float> @stack_fold_unpcklps_ymm(<8 x float> %a0, <8 x float> %a1) {
 define <2 x double> @stack_fold_xorpd(<2 x double> %a0, <2 x double> %a1) {
   ;CHECK-LABEL: stack_fold_xorpd
   ;CHECK:       vxorpd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <2 x double> %a0 to <2 x i64>
   %3 = bitcast <2 x double> %a1 to <2 x i64>
   %4 = xor <2 x i64> %2, %3
@@ -1744,7 +1744,7 @@ define <2 x double> @stack_fold_xorpd(<2 x double> %a0, <2 x double> %a1) {
 define <4 x double> @stack_fold_xorpd_ymm(<4 x double> %a0, <4 x double> %a1) {
   ;CHECK-LABEL: stack_fold_xorpd_ymm
   ;CHECK:       vxorpd {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x double> %a0 to <4 x i64>
   %3 = bitcast <4 x double> %a1 to <4 x i64>
   %4 = xor <4 x i64> %2, %3
@@ -1757,7 +1757,7 @@ define <4 x double> @stack_fold_xorpd_ymm(<4 x double> %a0, <4 x double> %a1) {
 define <4 x float> @stack_fold_xorps(<4 x float> %a0, <4 x float> %a1) {
   ;CHECK-LABEL: stack_fold_xorps
   ;CHECK:       vxorps {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}}, {{%xmm[0-9][0-9]*}} {{.*#+}} 16-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <4 x float> %a0 to <2 x i64>
   %3 = bitcast <4 x float> %a1 to <2 x i64>
   %4 = xor <2 x i64> %2, %3
@@ -1768,7 +1768,7 @@ define <4 x float> @stack_fold_xorps(<4 x float> %a0, <4 x float> %a1) {
 define <8 x float> @stack_fold_xorps_ymm(<8 x float> %a0, <8 x float> %a1) {
   ;CHECK-LABEL: stack_fold_xorps_ymm
   ;CHECK:       vxorps {{-?[0-9]*}}(%rsp), {{%ymm[0-9][0-9]*}}, {{%ymm[0-9][0-9]*}} {{.*#+}} 32-byte Folded Reload
-  %1 = tail call <4 x i64> asm sideeffect "nop", "=x,~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+  %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 = bitcast <8 x float> %a0 to <4 x i64>
   %3 = bitcast <8 x float> %a1 to <4 x i64>
   %4 = xor <4 x i64> %2, %3