[PATCH] D97818: [OpenMP] Replace GPU globalization calls with shared memory in the middle-end

Joseph Huber via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Jun 7 07:57:03 PDT 2021


jhuber6 updated this revision to Diff 350293.
jhuber6 edited the summary of this revision.
jhuber6 added a comment.
Herald added a subscriber: ormris.

Change the implementation to use the Attributor. This is because the attributor after registering HeapToStack in D102197 <https://reviews.llvm.org/D102197> would change the call sites. Putting this pass inside an attributor instance allows both to run and now globalization will only be replaced if HeapToStack doesn't remove it first.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D97818/new/

https://reviews.llvm.org/D97818

Files:
  llvm/lib/Transforms/IPO/OpenMPOpt.cpp
  llvm/test/Transforms/OpenMP/replace_globalization.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D97818.350293.patch
Type: text/x-patch
Size: 15069 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210607/c3cb7557/attachment.bin>


More information about the llvm-commits mailing list