[PATCH] D90217: [AMDGPU][GlobalISel] Fold a chain of two shift instructions with constant operands
Mirko Brkusanin via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Nov 6 08:50:12 PST 2020
mbrkusanin updated this revision to Diff 303457.
mbrkusanin added a comment.
- Avoid combine for G_USHLSAT when sum exceeds scalar size.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D90217/new/
https://reviews.llvm.org/D90217
Files:
llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h
llvm/include/llvm/Target/GlobalISel/Combine.td
llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
llvm/test/CodeGen/AMDGPU/GlobalISel/ashr.ll
llvm/test/CodeGen/AMDGPU/GlobalISel/combine-shift-imm-chain-illegal-types.mir
llvm/test/CodeGen/AMDGPU/GlobalISel/combine-shift-imm-chain-shlsat.mir
llvm/test/CodeGen/AMDGPU/GlobalISel/combine-shift-imm-chain.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D90217.303457.patch
Type: text/x-patch
Size: 33773 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201106/30eb4e9f/attachment.bin>
More information about the llvm-commits
mailing list