[all-commits] [llvm/llvm-project] 14858c: [mlir][Conversion/GPUCommon] Fix bug in conversion...

[Chris via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 14858cf05dc7cbc0f34629d693b0039c3d15c34f
      https://github.com/llvm/llvm-project/commit/14858cf05dc7cbc0f34629d693b0039c3d15c34f
  Author: Christopher Bate <cbate at nvidia.com>
  Date:   2023-07-03 (Mon, 03 Jul 2023)

  Changed paths:
    M mlir/lib/Conversion/GPUCommon/GPUOpsLowering.cpp
    M mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
    M mlir/test/Conversion/GPUToROCDL/gpu-to-rocdl.mlir

  Log Message:
  -----------
  [mlir][Conversion/GPUCommon] Fix bug in conversion of `math` ops

The common GPU operation transformation that lowers `math` operations
to function calls in the `gpu-to-nvvm` and `gpu-to-rocdl` passes handles
`vector` types by applying the function to each scalar and returning a
new vector. However, there was a typo that results in incorrectly
accumulating the result vector, and the rewrite returns an `llvm.mlir.undef`
result instead of the correct vector. A patch is added and tests are
strengthened.

Reviewed By: ThomasRaoux

Differential Revision: https://reviews.llvm.org/D154269