[all-commits] [llvm/llvm-project] 2af186: [mlir][gpu] Add patterns to break down subgroup re...

[Jakub Kuderski via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2af186f9bd13e2c3772e6fbc55eacb5bfdd4b5a9
      https://github.com/llvm/llvm-project/commit/2af186f9bd13e2c3772e6fbc55eacb5bfdd4b5a9
  Author: Jakub Kuderski <jakub at nod-labs.com>
  Date:   2023-12-28 (Thu, 28 Dec 2023)

  Changed paths:
    M mlir/include/mlir/Dialect/GPU/Transforms/Passes.h
    M mlir/lib/Dialect/GPU/CMakeLists.txt
    A mlir/lib/Dialect/GPU/Transforms/SubgroupReduceLowering.cpp
    A mlir/test/Dialect/GPU/subgroup-redule-lowering.mlir
    M mlir/test/lib/Dialect/GPU/CMakeLists.txt
    M mlir/test/lib/Dialect/GPU/TestGpuRewrite.cpp
    M mlir/tools/mlir-opt/mlir-opt.cpp

  Log Message:
  -----------
  [mlir][gpu] Add patterns to break down subgroup reduce (#76271)

The new patterns break down subgroup reduce ops with vector values into
a sequence of subgroup reductions that fit the native shuffle size. The
maximum/native shuffle size is parametrized.

The overall goal is to be able to perform multi-element reductions with
a sequence of `gpu.shuffle` ops.