1;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
2
3;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
4;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
5;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
6;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
7;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:UUNN
8;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYZZ}} RID:0 SID:0 CT:NNNN
9;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYZZ}} RID:0 SID:0 CT:NNNN
10;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYZZ}} RID:0 SID:0 CT:UUNN
11;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYYW}} RID:0 SID:0 CT:NNUN
12;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNUN
13;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYYZ}} RID:0 SID:0 CT:NNUN
14;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNUN
15;CHECK: TEX_SAMPLE_C T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
16;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
17;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNNN
18;CHECK: TEX_SAMPLE T{{[0-9]+\.XYZW, T[0-9]+\.XYZW}} RID:0 SID:0 CT:NNUN
19
20define amdgpu_kernel void @test(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)* %in) {
21bb:
22  %addr = load <4 x float>, <4 x float> addrspace(1)* %in
23  %tmp = shufflevector <4 x float> %addr, <4 x float> %addr, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
24  %tmp1 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
25  %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> %tmp1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
26  %tmp3 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp2, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
27  %tmp4 = shufflevector <4 x float> %tmp3, <4 x float> %tmp3, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
28  %tmp5 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp4, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
29  %tmp6 = shufflevector <4 x float> %tmp5, <4 x float> %tmp5, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
30  %tmp7 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp6, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
31  %tmp8 = shufflevector <4 x float> %tmp7, <4 x float> %tmp7, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
32  %tmp9 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp8, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1)
33  %tmp10 = shufflevector <4 x float> %tmp9, <4 x float> %tmp9, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
34  %tmp11 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp10, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
35  %tmp12 = shufflevector <4 x float> %tmp11, <4 x float> %tmp11, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
36  %tmp13 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp12, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
37  %tmp14 = shufflevector <4 x float> %tmp13, <4 x float> %tmp13, <4 x i32> <i32 0, i32 1, i32 2, i32 2>
38  %tmp15 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp14, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1)
39  %tmp16 = shufflevector <4 x float> %tmp15, <4 x float> %tmp15, <4 x i32> <i32 0, i32 1, i32 1, i32 3>
40  %tmp17 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp16, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 0, i32 1)
41  %tmp18 = shufflevector <4 x float> %tmp17, <4 x float> %tmp17, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
42  %tmp19 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp18, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 0, i32 1)
43  %tmp20 = shufflevector <4 x float> %tmp19, <4 x float> %tmp19, <4 x i32> <i32 0, i32 1, i32 1, i32 2>
44  %tmp21 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp20, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 0, i32 1)
45  %tmp22 = shufflevector <4 x float> %tmp21, <4 x float> %tmp21, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
46  %tmp23 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp22, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 0, i32 1)
47  %tmp24 = shufflevector <4 x float> %tmp23, <4 x float> %tmp23, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
48  %tmp25 = call <4 x float> @llvm.r600.texc(<4 x float> %tmp24, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
49  %tmp26 = shufflevector <4 x float> %tmp25, <4 x float> %tmp25, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
50  %tmp27 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp26, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
51  %tmp28 = shufflevector <4 x float> %tmp27, <4 x float> %tmp27, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
52  %tmp29 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp28, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 1, i32 1)
53  %tmp30 = shufflevector <4 x float> %tmp29, <4 x float> %tmp29, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
54  %tmp31 = call <4 x float> @llvm.r600.tex(<4 x float> %tmp30, i32 0, i32 0, i32 0, i32 0, i32 0, i32 1, i32 1, i32 0, i32 1)
55  store <4 x float> %tmp31, <4 x float> addrspace(1)* %out
56  ret void
57}
58
59; Function Attrs: readnone
60declare <4 x float> @llvm.r600.tex(<4 x float>, i32, i32, i32, i32, i32, i32, i32, i32, i32) #0
61
62; Function Attrs: readnone
63declare <4 x float> @llvm.r600.texc(<4 x float>, i32, i32, i32, i32, i32, i32, i32, i32, i32) #0
64
65attributes #0 = { nounwind readnone }
66