1 ; RUN: llc < %s -march=x86 -mattr=sse | grep movaps
2 ; Test that the load is NOT folded into the intrinsic, which would zero the top
3 ; elts of the loaded vector.
5 target datalayout = "e-p:32:32"
6 target triple = "i686-apple-darwin8.7.2"
8 define <4 x float> @test(<4 x float> %A, <4 x float>* %B) nounwind {
9 %BV = load <4 x float>* %B ; <<4 x float>> [#uses=1]
10 %tmp28 = tail call <4 x float> @llvm.x86.sse.sub.ss( <4 x float> %A, <4 x float> %BV ) ; <<4 x float>> [#uses=1]
11 ret <4 x float> %tmp28
14 declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>)