[all-commits] [llvm/llvm-project] ff8775: [mlir][GPU] Add op for unrolling contractions to a...

[Quinn Dawkins via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ff8775f3fffea787922eb238c0af3ed5715485d9
      https://github.com/llvm/llvm-project/commit/ff8775f3fffea787922eb238c0af3ed5715485d9
  Author: Quinn Dawkins <quinn at nod-labs.com>
  Date:   2023-07-25 (Tue, 25 Jul 2023)

  Changed paths:
    M mlir/include/mlir/Dialect/GPU/TransformOps/GPUTransformOps.td
    M mlir/lib/Dialect/GPU/TransformOps/GPUTransformOps.cpp
    A mlir/test/Dialect/GPU/subgroup-mma-vector-unroll.mlir

  Log Message:
  -----------
  [mlir][GPU] Add op for unrolling contractions to a native size

Adds `apply_patterns.gpu.unroll_vectors_subgroup_mma` which allows
specifying a native MMA shape of `m`, `n`, and `k` to unroll to,
greedily unrolling the inner most dimension of contractions and other
vector operations based on expected usage.

Differential Revision: https://reviews.llvm.org/D156079