[all-commits] [llvm/llvm-project] a3cd2e: [mlir][nvgpu] Add a nvgpu.rewrite_copy_as_tma tran...

[Nicolas Vasilache via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a3cd2eeb2d2a69b2c4d1fe9b634123b9f9943d0b
      https://github.com/llvm/llvm-project/commit/a3cd2eeb2d2a69b2c4d1fe9b634123b9f9943d0b
  Author: Nicolas Vasilache <nicolas.vasilache at gmail.com>
  Date:   2023-08-08 (Tue, 08 Aug 2023)

  Changed paths:
    M mlir/include/mlir/Dialect/NVGPU/TransformOps/NVGPUTransformOps.td
    M mlir/lib/Dialect/NVGPU/TransformOps/NVGPUTransformOps.cpp
    M mlir/lib/Dialect/Utils/IndexingUtils.cpp
    A mlir/test/Dialect/NVGPU/tmaload-transform.mlir
    A mlir/test/Integration/GPU/CUDA/sm90/tmaload-transform.mlir

  Log Message:
  -----------
  [mlir][nvgpu] Add a nvgpu.rewrite_copy_as_tma transform operation.

This revision adds support for direct lowering of a linalg.copy on buffers between global and shared memory to a tma async load + synchronization operations.
This uses the recently introduced Hopper NVVM and NVGPU abstraction to connect things end to end.

Differential Revision: https://reviews.llvm.org/D157087