diff options
Diffstat (limited to 'test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll')
-rw-r--r-- | test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll | 47 |
1 files changed, 42 insertions, 5 deletions
diff --git a/test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll b/test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll index ef616eb63801d..5c6663dbbdab9 100644 --- a/test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll +++ b/test/CodeGen/AMDGPU/si-annotate-cf-noloop.ll @@ -6,10 +6,10 @@ ; OPT-NOT: call i1 @llvm.amdgcn.loop ; GCN-LABEL: {{^}}annotate_unreachable_noloop: -; GCN: s_cbranch_vccnz +; GCN: s_cbranch_scc1 ; GCN-NOT: s_endpgm ; GCN: .Lfunc_end0 -define void @annotate_unreachable_noloop(<4 x float> addrspace(1)* noalias nocapture readonly %arg) #0 { +define amdgpu_kernel void @annotate_unreachable_noloop(<4 x float> addrspace(1)* noalias nocapture readonly %arg) #0 { bb: %tmp = tail call i32 @llvm.amdgcn.workitem.id.x() br label %bb1 @@ -37,12 +37,49 @@ bb5: ; preds = %bb3, %bb1 ; OPT-NOT: call i1 @llvm.amdgcn.loop ; GCN-LABEL: {{^}}annotate_ret_noloop: +; GCN: load_dwordx4 +; GCN: v_cmp_nlt_f32 +; GCN: s_and_saveexec_b64 +; GCN: ; mask branch [[UNIFIED_RET:BB[0-9]+_[0-9]+]] +; GCN-NEXT: [[UNIFIED_RET]]: +; GCN-NEXT: s_or_b64 exec, exec +; GCN-NEXT: s_endpgm +; GCN: .Lfunc_end +define amdgpu_kernel void @annotate_ret_noloop(<4 x float> addrspace(1)* noalias nocapture readonly %arg) #0 { +bb: + %tmp = tail call i32 @llvm.amdgcn.workitem.id.x() + br label %bb1 + +bb1: ; preds = %bb + %tmp2 = sext i32 %tmp to i64 + %tmp3 = getelementptr inbounds <4 x float>, <4 x float> addrspace(1)* %arg, i64 %tmp2 + %tmp4 = load <4 x float>, <4 x float> addrspace(1)* %tmp3, align 16 + %tmp5 = extractelement <4 x float> %tmp4, i32 1 + store volatile <4 x float> %tmp4, <4 x float> addrspace(1)* undef + %cmp = fcmp ogt float %tmp5, 1.0 + br i1 %cmp, label %bb5, label %bb3 + +bb3: ; preds = %bb1 + %tmp6 = extractelement <4 x float> %tmp4, i32 2 + %tmp7 = fcmp olt float %tmp6, 0.000000e+00 + br i1 %tmp7, label %bb4, label %bb5 ; crash goes away if these are swapped + +bb4: ; preds = %bb3 + ret void + +bb5: ; preds = %bb3, %bb1 + ret void +} + +; OPT-LABEL: @uniform_annotate_ret_noloop( +; OPT-NOT: call i1 @llvm.amdgcn.loop + +; GCN-LABEL: {{^}}uniform_annotate_ret_noloop: ; GCN: s_cbranch_scc1 ; GCN: s_endpgm -; GCN: .Lfunc_end1 -define void @annotate_ret_noloop(<4 x float> addrspace(1)* noalias nocapture readonly %arg) #0 { +; GCN: .Lfunc_end +define amdgpu_kernel void @uniform_annotate_ret_noloop(<4 x float> addrspace(1)* noalias nocapture readonly %arg, i32 %tmp) #0 { bb: - %tmp = tail call i32 @llvm.amdgcn.workitem.id.x() br label %bb1 bb1: ; preds = %bb |