[all-commits] [llvm/llvm-project] f1032f: [MLIR][NVVM][NVGPU] Combine prefetch and prefetch....

[Srinivasa Ravi via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: f1032f06e8bd6571a929e45ca8d0afc9f17957cc
      https://github.com/llvm/llvm-project/commit/f1032f06e8bd6571a929e45ca8d0afc9f17957cc
  Author: Srinivasa Ravi <srinivasar at nvidia.com>
  Date:   2025-09-01 (Mon, 01 Sep 2025)

  Changed paths:
    M mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
    M mlir/lib/Conversion/NVGPUToNVVM/NVGPUToNVVM.cpp
    M mlir/lib/Dialect/LLVMIR/IR/NVVMDialect.cpp
    M mlir/test/Conversion/NVGPUToNVVM/nvgpu-to-nvvm.mlir
    M mlir/test/Conversion/NVVMToLLVM/nvvm-to-llvm.mlir
    M mlir/test/Dialect/LLVMIR/nvvm.mlir
    M mlir/test/Target/LLVMIR/nvvm/prefetch.mlir
    M mlir/test/Target/LLVMIR/nvvmir-invalid.mlir

  Log Message:
  -----------
  [MLIR][NVVM][NVGPU] Combine prefetch and prefetch.tensormap (#153134)

This PR combines the `prefetch` and `prefetch.tensormap` NVVM Ops
to one `prefetch` Op. The `tensormap` variant is lowered through the
newly added intrinsics.

The lowering of the NVGPU `tma.prefetch.descriptor` Op is changed
from lowering to the `prefetch.tensormap` Op to `prefetch`.

PTX Spec Reference:
https://docs.nvidia.com/cuda/parallel-thread-execution/#data-movement-and-conversion-instructions-prefetch-prefetchu



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications