[PATCH] D73794: [MLIR] Remove all-reduce lowering from GPU to NVVM. Use in-dialect lowering instead.

[Christian Sigg via Phabricator via llvm-commits]

csigg updated this revision to Diff 249084.
csigg added a comment.
Herald added a reviewer: mravishankar.

Avoid DCE in tests by returning values from functions.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D73794/new/

https://reviews.llvm.org/D73794

Files:
  mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
  mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D73794.249084.patch
Type: text/x-patch
Size: 29743 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200309/3058209d/attachment.bin>