[all-commits] [llvm/llvm-project] fcffd0: [Matrix] Fold the transpose into the matmul operan...

[anemet via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fcffd087c6bc1364678c1549ffc7a26fd8bea930
      https://github.com/llvm/llvm-project/commit/fcffd087c6bc1364678c1549ffc7a26fd8bea930
  Author: Adam Nemet <anemet at apple.com>
  Date:   2021-05-17 (Mon, 17 May 2021)

  Changed paths:
    M llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
    M llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-row-major.ll
    A llvm/test/Transforms/LowerMatrixIntrinsics/multiply-left-transpose-row-major.ll
    A llvm/test/Transforms/LowerMatrixIntrinsics/multiply-right-transpose.ll

  Log Message:
  -----------
  [Matrix] Fold the transpose into the matmul operand used to fetch scalars

For column-major this is:
  A * B^t
whereas for row-major:
  A^t * B

Differential Revision: https://reviews.llvm.org/D101762