[PATCH] D51995: AMDGPU: Generate VALU ThreeOp Integer instructions

Nicolai Hähnle via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Sep 13 02:02:06 PDT 2018


nhaehnle updated this revision to Diff 165218.
nhaehnle marked 2 inline comments as done.
nhaehnle added a comment.

- Cleaned up the pattern condition to take inline constants into account
- Add tests for inline constants and uniform values ending up in VGPRs
- update_llc-ify the tests


Repository:
  rL LLVM

https://reviews.llvm.org/D51995

Files:
  lib/Target/AMDGPU/VOP3Instructions.td
  test/CodeGen/AMDGPU/add3.ll
  test/CodeGen/AMDGPU/add_shl.ll
  test/CodeGen/AMDGPU/and_or.ll
  test/CodeGen/AMDGPU/or3.ll
  test/CodeGen/AMDGPU/shl_add.ll
  test/CodeGen/AMDGPU/shl_or.ll
  test/CodeGen/AMDGPU/sibling-call.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D51995.165218.patch
Type: text/x-patch
Size: 28416 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180913/ad5a3963/attachment.bin>


More information about the llvm-commits mailing list