1 ; RUN: llc < %s -march=x86-64 -mtriple=x86_64-apple-darwin -mcpu=knl | FileCheck %s
3 define <16 x i32> @test1(i32* %x) {
5 ; CHECK: vmovd (%rdi), %xmm
8 %y = load i32, i32* %x, align 4
9 %res = insertelement <16 x i32>zeroinitializer, i32 %y, i32 4
13 define <16 x i32> @test2(<16 x i32> %x) {
16 ; CHECK-NEXT: vpaddd {{.*}}(%rip){1to16}, %zmm0, %zmm0
18 %res = add <16 x i32><i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1>, %x
22 define <16 x float> @test3(<4 x float> %a) {
25 ; CHECK-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
26 ; CHECK-NEXT: vxorps %xmm2, %xmm2, %xmm2
27 ; CHECK-NEXT: vmovss %xmm0, %xmm2, %xmm0
28 ; CHECK-NEXT: vmovss %xmm1, %xmm2, %xmm1
29 ; CHECK-NEXT: vshufps {{.*#+}} xmm0 = xmm1[1,0],xmm0[0,1]
30 ; CHECK-NEXT: vinsertf128 $1, %xmm0, %ymm2, %ymm0
31 ; CHECK-NEXT: vxorps %ymm1, %ymm1, %ymm1
32 ; CHECK-NEXT: vinsertf64x4 $1, %ymm1, %zmm0, %zmm0
34 %b = extractelement <4 x float> %a, i32 2
35 %c = insertelement <16 x float> <float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float undef, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00>, float %b, i32 5
36 %b1 = extractelement <4 x float> %a, i32 0
37 %c1 = insertelement <16 x float> %c, float %b1, i32 6