X-Git-Url: http://plrg.eecs.uci.edu/git/?a=blobdiff_plain;f=test%2FCodeGen%2FX86%2Fvec_extract-sse4.ll;h=9f4210f7847ed7f88aac09ad0a17536074eee42e;hb=39cf5554292b7a82552b704c3905ab9e7557a9db;hp=f4876543d329a083db3933dd944a573c291e3dbb;hpb=42febc6e9963f82d5c56c3c7e6afe5e00769af41;p=oota-llvm.git diff --git a/test/CodeGen/X86/vec_extract-sse4.ll b/test/CodeGen/X86/vec_extract-sse4.ll index f4876543d32..9f4210f7847 100644 --- a/test/CodeGen/X86/vec_extract-sse4.ll +++ b/test/CodeGen/X86/vec_extract-sse4.ll @@ -1,31 +1,60 @@ -; RUN: llc < %s -mcpu=corei7 -march=x86 -mattr=+sse41 -o %t -; RUN: not grep extractps %t -; RUN: not grep pextrd %t -; RUN: not grep pshufd %t -; RUN: grep movss %t | count 2 +; RUN: llc < %s -mcpu=corei7 -march=x86 -mattr=+sse4.1 | FileCheck %s define void @t1(float* %R, <4 x float>* %P1) nounwind { - %X = load <4 x float>* %P1 +; CHECK-LABEL: t1: +; CHECK: # BB#0: +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %ecx +; CHECK-NEXT: movss 12(%ecx), %xmm0 +; CHECK-NEXT: movss %xmm0, (%eax) +; CHECK-NEXT: retl + + %X = load <4 x float>, <4 x float>* %P1 %tmp = extractelement <4 x float> %X, i32 3 store float %tmp, float* %R ret void } define float @t2(<4 x float>* %P1) nounwind { - %X = load <4 x float>* %P1 +; CHECK-LABEL: t2: +; CHECK: # BB#0: +; CHECK-NEXT: pushl %eax +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax +; CHECK-NEXT: movapd (%eax), %xmm0 +; CHECK-NEXT: shufpd {{.*#+}} xmm0 = xmm0[1,0] +; CHECK-NEXT: movss %xmm0, (%esp) +; CHECK-NEXT: flds (%esp) +; CHECK-NEXT: popl %eax +; CHECK-NEXT: retl + + %X = load <4 x float>, <4 x float>* %P1 %tmp = extractelement <4 x float> %X, i32 2 ret float %tmp } define void @t3(i32* %R, <4 x i32>* %P1) nounwind { - %X = load <4 x i32>* %P1 +; CHECK-LABEL: t3: +; CHECK: # BB#0: +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %ecx +; CHECK-NEXT: movl 12(%ecx), %ecx +; CHECK-NEXT: movl %ecx, (%eax) +; CHECK-NEXT: retl + + %X = load <4 x i32>, <4 x i32>* %P1 %tmp = extractelement <4 x i32> %X, i32 3 store i32 %tmp, i32* %R ret void } define i32 @t4(<4 x i32>* %P1) nounwind { - %X = load <4 x i32>* %P1 +; CHECK-LABEL: t4: +; CHECK: # BB#0: +; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax +; CHECK-NEXT: movl 12(%eax), %eax +; CHECK-NEXT: retl + + %X = load <4 x i32>, <4 x i32>* %P1 %tmp = extractelement <4 x i32> %X, i32 3 ret i32 %tmp }