1; RUN: llc -O3 -mtriple=x86_64-unknown -mcpu=corei7 < %s | FileCheck %s
2
3define <8 x i16> @commute_fold_pblendw(<8 x i16> %a, <8 x i16>* %b) #0 {
4  %1 = load <8 x i16>* %b
5  %2 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %1, <8 x i16> %a, i8 17)
6  ret <8 x i16> %2
7
8  ;LABEL:      commute_fold_pblendw
9  ;CHECK:      pblendw {{.*#+}} xmm0 = xmm0[0],mem[1,2,3],xmm0[4],mem[5,6,7]
10  ;CHECK-NEXT: retq
11}
12declare <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16>, <8 x i16>, i8) nounwind readnone
13
14define <4 x float> @commute_fold_blendps(<4 x float> %a, <4 x float>* %b) #0 {
15  %1 = load <4 x float>* %b
16  %2 = call <4 x float> @llvm.x86.sse41.blendps(<4 x float> %1, <4 x float> %a, i8 3)
17  ret <4 x float> %2
18
19  ;LABEL:      commute_fold_blendps
20  ;CHECK:      blendps {{.*#+}} xmm0 = xmm0[0,1],mem[2,3]
21  ;CHECK-NEXT: retq
22}
23declare <4 x float> @llvm.x86.sse41.blendps(<4 x float>, <4 x float>, i8) nounwind readnone
24
25define <2 x double> @commute_fold_blendpd(<2 x double> %a, <2 x double>* %b) #0 {
26  %1 = load <2 x double>* %b
27  %2 = call <2 x double> @llvm.x86.sse41.blendpd(<2 x double> %1, <2 x double> %a, i8 1)
28  ret <2 x double> %2
29
30  ;LABEL:      commute_fold_vblendpd
31  ;CHECK:      blendpd {{.*#+}} xmm0 = xmm0[0],mem[1]
32  ;CHECK-NEXT: retq
33}
34declare <2 x double> @llvm.x86.sse41.blendpd(<2 x double>, <2 x double>, i8) nounwind readnone
35