[all-commits] [llvm/llvm-project] 25622a: [mlir][AMDGPU] Add gfx950 MFMAs to the amdgpu.mfma...

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 25622aa745671d8d1885250f559c6dac0a8c2aeb
      https://github.com/llvm/llvm-project/commit/25622aa745671d8d1885250f559c6dac0a8c2aeb
  Author: Krzysztof Drewniak <krzysdrewniak at gmail.com>
  Date:   2025-04-01 (Tue, 01 Apr 2025)

  Changed paths:
    M mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
    M mlir/lib/Conversion/AMDGPUToROCDL/AMDGPUToROCDL.cpp
    M mlir/lib/Dialect/AMDGPU/IR/AMDGPUDialect.cpp
    A mlir/test/Conversion/AMDGPUToROCDL/mfma-gfx950.mlir
    M mlir/test/Dialect/AMDGPU/invalid.mlir

  Log Message:
  -----------
  [mlir][AMDGPU] Add gfx950 MFMAs to the amdgpu.mfma op (#133553)

This commit extends the lowering of amdgpu.mfma to handle the new
double-rate MFMAs in gfx950 and adds tests for these operations.

It also adds support for MFMAs on small floats (f6 and f4), which are
implented using the "scaled" MFMA intrinsic with a scale value of 0 in
order to have an unscaled MFMA.

This commit does not add a `amdgpu.scaled_mfma` operation, as that is
future work.

---------

Co-authored-by: Jakub Kuderski <kubakuderski at gmail.com>



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