[all-commits] [llvm/llvm-project] ca1bad: [MLIR][GPU] Properly model step in parallel loop t...

[Stephan Herhut via All-commits]

  Branch: refs/heads/MASTER
  Home:   https://github.com/llvm/llvm-project
  Commit: ca1bad5253a18ea6ab6573abdc26b740ff4593c4
      https://github.com/llvm/llvm-project/commit/ca1bad5253a18ea6ab6573abdc26b740ff4593c4
  Author: Stephan Herhut <herhut at google.com>
  Date:   2020-02-25 (Tue, 25 Feb 2020)

  Changed paths:
    M mlir/include/mlir/Conversion/LoopsToGPU/LoopsToGPUPass.h
    M mlir/lib/Conversion/LoopsToGPU/LoopsToGPU.cpp
    M mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp
    M mlir/test/Conversion/LoopsToGPU/parallel_loop.mlir

  Log Message:
  -----------
  [MLIR][GPU] Properly model step in parallel loop to gpu conversion.

Summary:
The original patch had TODOs to add support for step computations,
which this commit addresses. The computations are expressed using
affine expressions so that the affine canonicalizers can simplify
the full bound and index computations.

Also cleans up the code a little and exposes the pass in the
header file.

Differential Revision: https://reviews.llvm.org/D75052