diff options
| author | Nicolai Haehnle <nhaehnle@gmail.com> | 2018-10-17 15:37:41 +0000 |
|---|---|---|
| committer | Nicolai Haehnle <nhaehnle@gmail.com> | 2018-10-17 15:37:41 +0000 |
| commit | 0823050b9f4ca2dd0aaf27ee34cb45bde5bf05ea (patch) | |
| tree | 85a15b1292b69c96e98dc91fcfc8f607f32bd8ee /llvm/test/CodeGen/AMDGPU/smrd.ll | |
| parent | c4a2ff095078ae2390ce216bd5b3522921b7f722 (diff) | |
| download | bcm5719-llvm-0823050b9f4ca2dd0aaf27ee34cb45bde5bf05ea.tar.gz bcm5719-llvm-0823050b9f4ca2dd0aaf27ee34cb45bde5bf05ea.zip | |
StructurizeCFG: Simplify inserted PHI nodes
Summary:
This improves subsequent divergence analysis in some cases.
Change-Id: I5e95e7ec7fd3fa80d414d1a53a02fea23e3d67d3
Reviewers: arsenm, rampitec
Subscribers: jvesely, wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D53316
llvm-svn: 344697
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/smrd.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/smrd.ll | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/smrd.ll b/llvm/test/CodeGen/AMDGPU/smrd.ll index 0e89f8516bf..f453cfdbd1f 100644 --- a/llvm/test/CodeGen/AMDGPU/smrd.ll +++ b/llvm/test/CodeGen/AMDGPU/smrd.ll @@ -592,11 +592,12 @@ exit: ; GCN-LABEL: {{^}}smrd_uniform_loop2: ; (this test differs from smrd_uniform_loop by the more complex structure of phis, -; which currently confuses the DivergenceAnalysis after structurization) +; which used to confuse the DivergenceAnalysis after structurization) ; -; TODO: this should use an s_buffer_load +; TODO: we should keep the loop counter in an SGPR ; -; GCN: buffer_load_dword +; GCN: v_readfirstlane_b32 +; GCN: s_buffer_load_dword define amdgpu_ps float @smrd_uniform_loop2(<4 x i32> inreg %desc, i32 %bound, i32 %bound.a) #0 { main_body: br label %loop |
