[PATCH] D153992: AMDGPU: Avoid computing dominator tree for printf lowering

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: AMDGPU, vikramRH, sameerds, rampitec, Pierre-vh, cdevadas.
Herald added subscribers: foad, kerbowa, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl.
Herald added a project: All.
arsenm requested review of this revision.
Herald added a subscriber: wdng.
Herald added a project: LLVM.

This was only used to assist simplifyInstruction. I think it's
unlikely to help anything introduced here. We should probably run this
earlier anyway for other optimizations to clean up. I think we
actually run this twice when invoked from clang, once in
EarlySimplificationEPCallback and then again in addIRPasses. Clang
should probably be adding this in the first place.


https://reviews.llvm.org/D153992

Files:
  llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp
  llvm/test/CodeGen/AMDGPU/llc-pipeline.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D153992.535468.patch
Type: text/x-patch
Size: 7398 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230628/544822f3/attachment.bin>