[all-commits] [llvm/llvm-project] 15bcc3: [mlir][gpu] Move async copy ops to NVGPU and add c...

[Thomas via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 15bcc36eede13b24470e554dfa932f1fc40dd4ba
      https://github.com/llvm/llvm-project/commit/15bcc36eede13b24470e554dfa932f1fc40dd4ba
  Author: Thomas Raoux <thomasraoux at google.com>
  Date:   2022-05-10 (Tue, 10 May 2022)

  Changed paths:
    M mlir/include/mlir/Dialect/GPU/GPUBase.td
    M mlir/include/mlir/Dialect/GPU/GPUDialect.h
    M mlir/include/mlir/Dialect/GPU/GPUOps.td
    M mlir/include/mlir/Dialect/LLVMIR/NVVMDialect.h
    M mlir/include/mlir/Dialect/NVGPU/NVGPU.td
    M mlir/include/mlir/Dialect/NVGPU/NVGPUDialect.h
    M mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
    M mlir/lib/Conversion/NVGPUToNVVM/CMakeLists.txt
    M mlir/lib/Conversion/NVGPUToNVVM/NVGPUToNVVM.cpp
    M mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
    M mlir/lib/Dialect/NVGPU/IR/CMakeLists.txt
    M mlir/lib/Dialect/NVGPU/IR/NVGPUDialect.cpp
    M mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
    R mlir/test/Conversion/NVGPUToNVVM/mma-sync-to-nvvm.mlir
    A mlir/test/Conversion/NVGPUToNVVM/nvgpu-to-nvvm.mlir
    M mlir/test/Dialect/GPU/invalid.mlir
    M mlir/test/Dialect/GPU/ops.mlir
    A mlir/test/Dialect/NVGPU/invalid.mlir
    M mlir/test/Dialect/NVGPU/roundtrip.mlir
    M utils/bazel/llvm-project-overlay/mlir/BUILD.bazel

  Log Message:
  -----------
  [mlir][gpu] Move async copy ops to NVGPU and add caching hints

Move async copy operations to NVGPU as they only exist on NV target and are
designed to match ptx semantic. This allows us to also add more fine grain
caching hint attribute to the op.
Add hint to bypass L1 and hook it up to NVVM op.

Differential Revision: https://reviews.llvm.org/D125244