diff --git a/llvm/test/CodeGen/X86/avx2-nontemporal.ll b/llvm/test/CodeGen/X86/avx2-nontemporal.ll
index 4d28a979712a..544c096c52df 100644
--- a/llvm/test/CodeGen/X86/avx2-nontemporal.ll
+++ b/llvm/test/CodeGen/X86/avx2-nontemporal.ll
@@ -1,21 +1,18 @@
 ; RUN: llc < %s -march=x86 -mattr=+avx2 | FileCheck %s
 
-define void @f(<8 x float> %A, i8* %B, <4 x double> %C, i32 %D, <4 x i64> %E) {
-; CHECK: vmovntps
+define void @f(<8 x float> %A, i8* %B, <4 x double> %C, <4 x i64> %E) {
+; CHECK: vmovntps %y
   %cast = bitcast i8* %B to <8 x float>*
   %A2 = fadd <8 x float> %A, <float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x0, float 0x4200000000000000>
   store <8 x float> %A2, <8 x float>* %cast, align 16, !nontemporal !0
-; CHECK: vmovntdq
+; CHECK: vmovntdq %y
   %cast1 = bitcast i8* %B to <4 x i64>*
   %E2 = add <4 x i64> %E, <i64 1, i64 2, i64 3, i64 4>
   store <4 x i64> %E2, <4 x i64>* %cast1, align 16, !nontemporal !0
-; CHECK: vmovntpd
+; CHECK: vmovntpd %y
   %cast2 = bitcast i8* %B to <4 x double>*
   %C2 = fadd <4 x double> %C, <double 0x0, double 0x0, double 0x0, double 0x4200000000000000>
   store <4 x double> %C2, <4 x double>* %cast2, align 16, !nontemporal !0
-; CHECK: movnti
-  %cast3 = bitcast i8* %B to i32*
-  store i32 %D, i32* %cast3, align 16, !nontemporal !0
   ret void
 }
 
diff --git a/llvm/test/CodeGen/X86/nontemporal.ll b/llvm/test/CodeGen/X86/nontemporal.ll
index f9385df36421..9a2f23596f79 100644
--- a/llvm/test/CodeGen/X86/nontemporal.ll
+++ b/llvm/test/CodeGen/X86/nontemporal.ll
@@ -1,6 +1,6 @@
-; RUN: llc < %s -march=x86 -mattr=+sse2 | FileCheck %s
+; RUN: llc < %s -mtriple x86_64-unknown-unknown | FileCheck %s
 
-define void @f(<4 x float> %A, i8* %B, <2 x double> %C, i32 %D, <2 x i64> %E) {
+define void @f(<4 x float> %A, i8* %B, <2 x double> %C, i32 %D, <2 x i64> %E, i64 %F) {
 ; CHECK: movntps
   %cast = bitcast i8* %B to <4 x float>*
   %A2 = fadd <4 x float> %A, <float 0x0, float 0x0, float 0x0, float 0x4200000000000000>
@@ -13,9 +13,12 @@ define void @f(<4 x float> %A, i8* %B, <2 x double> %C, i32 %D, <2 x i64> %E) {
   %cast2 = bitcast i8* %B to <2 x double>*
   %C2 = fadd <2 x double> %C, <double 0x0, double 0x4200000000000000>
   store <2 x double> %C2, <2 x double>* %cast2, align 16, !nontemporal !0
-; CHECK: movnti
+; CHECK: movntil
   %cast3 = bitcast i8* %B to i32*
-  store i32 %D, i32* %cast3, align 16, !nontemporal !0
+  store i32 %D, i32* %cast3, align 1, !nontemporal !0
+; CHECK: movntiq
+  %cast4 = bitcast i8* %B to i64*
+  store i64 %F, i64* %cast4, align 1, !nontemporal !0
   ret void
 }