[all-commits] [llvm/llvm-project] 459fd3: [MLIR][GPU] Detect bounds with `arith.minsi ` in l...

[Christian Sigg via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 459fd3fb342d565bbaff48673838c5ea138128f8
      https://github.com/llvm/llvm-project/commit/459fd3fb342d565bbaff48673838c5ea138128f8
  Author: Christian Sigg <csigg at google.com>
  Date:   2022-08-22 (Mon, 22 Aug 2022)

  Changed paths:
    M mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp
    M mlir/test/Conversion/SCFToGPU/parallel_loop.mlir

  Log Message:
  -----------
  [MLIR][GPU] Detect bounds with `arith.minsi ` in loops-to-gpu

Previously, `arith.constant`, `arith.muli` and `affine.min` were supported when deriving upper loop bounds when converting parallel loops to GPU.

Reviewed By: akuegel

Differential Revision: https://reviews.llvm.org/D132354