[all-commits] [llvm/llvm-project] 27ccf0: [MLIR] Provide bare pointer memref lowering option...

[Uday Bondhugula via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 27ccf0f40771047f5e6a20122cd53157df30f593
      https://github.com/llvm/llvm-project/commit/27ccf0f40771047f5e6a20122cd53157df30f593
  Author: Uday Bondhugula <uday at polymagelabs.com>
  Date:   2023-06-18 (Sun, 18 Jun 2023)

  Changed paths:
    M mlir/include/mlir/Conversion/Passes.td
    M mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
    A mlir/test/Conversion/GPUToNVVM/memref.mlir
    A mlir/test/Integration/GPU/CUDA/TensorCore/wmma-matmul-f32-bare-ptr.mlir
    M mlir/test/Integration/GPU/CUDA/all-reduce-and.mlir

  Log Message:
  -----------
  [MLIR] Provide bare pointer memref lowering option on gpu-to-nvvm pass

Provide the bare pointer memref lowering option on gpu-to-nvvm pass.
This is needed whenever we lower memrefs on the host function side and
the kernel calls on the host-side (gpu-to-llvm) with the bare ptr
convention. The GPU module side of the lowering should also "align" and
use the bare pointer convention.

Reviewed By: krzysz00

Differential Revision: https://reviews.llvm.org/D152480