[all-commits] [llvm/llvm-project] 42bba9: [mlir] Extend CombineTransferReadOpTranspose patte...

[harsh-nod via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 42bba97fc24f045f593fc26f998bac9b08633255
      https://github.com/llvm/llvm-project/commit/42bba97fc24f045f593fc26f998bac9b08633255
  Author: harsh-nod <harsh at nod-labs.com>
  Date:   2023-12-07 (Thu, 07 Dec 2023)

  Changed paths:
    M mlir/lib/Conversion/VectorToGPU/VectorToGPU.cpp
    M mlir/test/Conversion/VectorToGPU/vector-to-mma-ops.mlir

  Log Message:
  -----------
  [mlir] Extend CombineTransferReadOpTranspose pattern to handle extf ops. (#74754)

This patch modifies the CombineTransferReadOpTranspose pattern to handle
extf ops. Also adds a test which shows the transpose getting folded into
the transfer_read.