[Openmp-commits] [PATCH] D56274: [OPENMP][NVPTX]Fix incompatibility of __syncthreads with LLVM, NFC.

[Johannes Doerfert via Phabricator via Openmp-commits]

jdoerfert added a comment.
Herald added a project: LLVM.

Do I understand correctly that the intrinsic is apparently "misoptimized" by LLVM?
If so, would that also be a problem for other CUDA code, for example, User code that uses this intrinsic?


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56274/new/

https://reviews.llvm.org/D56274