[all-commits] [llvm/llvm-project] a3388f: [mlir] Introduce a pattern to lower `gpu.subgroup_...

[Guray Ozen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a3388f3e2a30eaca4123f14a6444fd5aa03b24c5
      https://github.com/llvm/llvm-project/commit/a3388f3e2a30eaca4123f14a6444fd5aa03b24c5
  Author: Guray Ozen <guray.ozen at gmail.com>
  Date:   2023-01-20 (Fri, 20 Jan 2023)

  Changed paths:
    M mlir/include/mlir/Conversion/GPUToNVVM/GPUToNVVMPass.h
    M mlir/include/mlir/Conversion/Passes.td
    M mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
    M mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir

  Log Message:
  -----------
  [mlir] Introduce a pattern to lower `gpu.subgroup_reduce` to `nvvm.redux_op`

This revision introduces a pattern to lower `gpu.subgroup_reduce` op into to the `nvvm.redux_sync` op. The op must be run by the entire subgroup, otherwise it is undefined behaviour.

It also adds a flag and populate function, because the op is not avaiable for every gpu (sm80+), so it can be used when it is desired.

Depends on D142088

Reviewed By: nicolasvasilache

Differential Revision: https://reviews.llvm.org/D142103