1 ; RUN: llc < %s -march=x86 -mattr=+sse2,-avx | FileCheck %s -check-prefix=X86-32
2 ; RUN: llc < %s -march=x86-64 -mattr=+sse2,-avx | FileCheck %s -check-prefix=X86-64
4 define <4 x i32> @t1(i32 %a) nounwind {
6 %tmp = insertelement <4 x i32> undef, i32 %a, i32 0
7 %tmp6 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %tmp, <4 x i32> < i32 4, i32 1, i32 2, i32 3 > ; <<4 x i32>> [#uses=1]
11 ; X86-32: movd 4(%esp), %xmm0
14 ; X86-64: movd %e{{..}}, %xmm0
17 define <2 x i64> @t2(i64 %a) nounwind {
19 %tmp = insertelement <2 x i64> undef, i64 %a, i32 0
20 %tmp6 = shufflevector <2 x i64> zeroinitializer, <2 x i64> %tmp, <2 x i32> < i32 2, i32 1 > ; <<4 x i32>> [#uses=1]
24 ; X86-32: movq 4(%esp), %xmm0
27 ; X86-64: movd %r{{..}}, %xmm0
30 define <2 x i64> @t3(<2 x i64>* %a) nounwind {
32 %tmp4 = load <2 x i64>* %a, align 16 ; <<2 x i64>> [#uses=1]
33 %tmp6 = bitcast <2 x i64> %tmp4 to <4 x i32> ; <<4 x i32>> [#uses=1]
34 %tmp7 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %tmp6, <4 x i32> < i32 4, i32 5, i32 2, i32 3 > ; <<4 x i32>> [#uses=1]
35 %tmp8 = bitcast <4 x i32> %tmp7 to <2 x i64> ; <<2 x i64>> [#uses=1]
39 ; X86-32: movl 4(%esp)
43 ; X86-64: movq ({{.*}}), %xmm0
46 define <2 x i64> @t4(<2 x i64> %a) nounwind {
48 %tmp5 = bitcast <2 x i64> %a to <4 x i32> ; <<4 x i32>> [#uses=1]
49 %tmp6 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %tmp5, <4 x i32> < i32 4, i32 5, i32 2, i32 3 > ; <<4 x i32>> [#uses=1]
50 %tmp7 = bitcast <4 x i32> %tmp6 to <2 x i64> ; <<2 x i64>> [#uses=1]
54 ; X86-32: movq %xmm0, %xmm0
57 ; X86-64: movq {{.*}}, %xmm0
60 define <2 x i64> @t5(<2 x i64> %a) nounwind {
62 %tmp6 = shufflevector <2 x i64> zeroinitializer, <2 x i64> %a, <2 x i32> < i32 2, i32 1 > ; <<4 x i32>> [#uses=1]
66 ; X86-32: movq %xmm0, %xmm0
69 ; X86-64: movq {{.*}}, %xmm0