[all-commits] [llvm/llvm-project] ba7f3e: [mlir][Transform] Fix support for mapping to GPU w...

[Nicolas Vasilache via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ba7f3e1d1e50212bdc8cc438185519fd7257aa44
      https://github.com/llvm/llvm-project/commit/ba7f3e1d1e50212bdc8cc438185519fd7257aa44
  Author: Nicolas Vasilache <nicolas.vasilache at gmail.com>
  Date:   2023-03-20 (Mon, 20 Mar 2023)

  Changed paths:
    M mlir/lib/Dialect/GPU/TransformOps/GPUTransformOps.cpp
    M mlir/test/Dialect/GPU/transform-gpu.mlir

  Log Message:
  -----------
  [mlir][Transform] Fix support for mapping to GPU warps and to linear ids

c59465e1203dd78d06e15f7ddf62141807dbd5a7 introduced mapping to warps and
linear GPU ids.

In the implementation, the delinearization basis is reversed from [x, y, z]
to [z, y x] order to properly compute the strides and allow delinearization.

Prior to this commit, we forgot to reverse it back to [x, y, z] order
before materializing the indices.

Fix this oversight.