There's a lot of test case churn but the overall effect is to increase
the number of back-to-back v_sub,v_subbrev pairs, which can execute with
no delay even on gfx10.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Paths
| Differential D75999
[AMDGPU] Extend macro fusion for ADDC and SUBB to SUBBREV ClosedPublic Authored by foad on Mar 11 2020, 8:45 AM.
Details Summary There's a lot of test case churn but the overall effect is to increase
Diff Detail
Event TimelineThis revision is now accepted and ready to land.Mar 11 2020, 10:12 AM Closed by commit rGa46dba24fa35: [AMDGPU] Extend macro fusion for ADDC and SUBB to SUBBREV (authored by foad). · Explain WhyMar 11 2020, 11:20 AM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 249697 llvm/lib/Target/AMDGPU/AMDGPUMacroFusion.cpp
llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-idiv.ll
llvm/test/CodeGen/AMDGPU/bypass-div.ll
llvm/test/CodeGen/AMDGPU/macro-fusion-cluster-vcc-uses.mir
llvm/test/CodeGen/AMDGPU/srem64.ll
llvm/test/CodeGen/AMDGPU/sub-zext-cc-zext-cc.ll
llvm/test/CodeGen/AMDGPU/udiv64.ll
llvm/test/CodeGen/AMDGPU/urem64.ll
llvm/test/CodeGen/AMDGPU/wave32.ll
|