1; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s
2
3declare float @llvm.AMDGPU.dp4(<4 x float>, <4 x float>) nounwind readnone
4
5define void @test_dp4(float addrspace(1)* %out, <4 x float> addrspace(1)* %a, <4 x float> addrspace(1)* %b) nounwind {
6  %src0 = load <4 x float> addrspace(1)* %a, align 16
7  %src1 = load <4 x float> addrspace(1)* %b, align 16
8  %dp4 = call float @llvm.AMDGPU.dp4(<4 x float> %src0, <4 x float> %src1) nounwind readnone
9  store float %dp4, float addrspace(1)* %out, align 4
10  ret void
11}
12