1; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
2; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse4.1 | FileCheck %s
3
4define <8 x i16> @commute_fold_pblendw(<8 x i16> %a, <8 x i16>* %b) #0 {
5; CHECK-LABEL: commute_fold_pblendw:
6; CHECK:       # BB#0:
7; CHECK-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],mem[1,2,3],xmm0[4],mem[5,6,7]
8; CHECK-NEXT:    retq
9  %1 = load <8 x i16>, <8 x i16>* %b
10  %2 = call <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16> %1, <8 x i16> %a, i8 17)
11  ret <8 x i16> %2
12}
13declare <8 x i16> @llvm.x86.sse41.pblendw(<8 x i16>, <8 x i16>, i8) nounwind readnone
14
15define <4 x float> @commute_fold_blendps(<4 x float> %a, <4 x float>* %b) #0 {
16; CHECK-LABEL: commute_fold_blendps:
17; CHECK:       # BB#0:
18; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],mem[1],xmm0[2],mem[3]
19; CHECK-NEXT:    retq
20  %1 = load <4 x float>, <4 x float>* %b
21  %2 = call <4 x float> @llvm.x86.sse41.blendps(<4 x float> %1, <4 x float> %a, i8 5)
22  ret <4 x float> %2
23}
24declare <4 x float> @llvm.x86.sse41.blendps(<4 x float>, <4 x float>, i8) nounwind readnone
25
26define <2 x double> @commute_fold_blendpd(<2 x double> %a, <2 x double>* %b) #0 {
27; CHECK-LABEL: commute_fold_blendpd:
28; CHECK:       # BB#0:
29; CHECK-NEXT:    blendpd {{.*#+}} xmm0 = xmm0[0],mem[1]
30; CHECK-NEXT:    retq
31  %1 = load <2 x double>, <2 x double>* %b
32  %2 = call <2 x double> @llvm.x86.sse41.blendpd(<2 x double> %1, <2 x double> %a, i8 1)
33  ret <2 x double> %2
34}
35declare <2 x double> @llvm.x86.sse41.blendpd(<2 x double>, <2 x double>, i8) nounwind readnone
36