[PATCH] D101187: [MachineCSE] Prevent CSE of non-local convergent instrs

[Michael Kitzan via Phabricator via llvm-commits]

mkitzan updated this revision to Diff 340176.
mkitzan added a comment.
Herald added subscribers: kerbowa, nhaehnle, jvesely.

Update: added handwritten MIR unit test for the MachineCSE change using AMDGPU's `DS_SWIZZLE_B32` instr (which is marked `isConvergent` in `llvm/lib/Target/AMDGPU/DSInstructions.td`)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101187/new/

https://reviews.llvm.org/D101187

Files:
  llvm/lib/CodeGen/MachineCSE.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/no-cse-nonlocal-convergent-instrs.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D101187.340176.patch
Type: text/x-patch
Size: 3709 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210423/7de5eb9d/attachment.bin>