[all-commits] [llvm/llvm-project] 181d96: [mlir][vectorize] Support affine.apply in SuperVec...

[Hsiangkai Wang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 181d9602f35c476201ecc72c26c00042ad949544
      https://github.com/llvm/llvm-project/commit/181d9602f35c476201ecc72c26c00042ad949544
  Author: Hsiangkai Wang <hsiangkai.wang at arm.com>
  Date:   2024-02-16 (Fri, 16 Feb 2024)

  Changed paths:
    M mlir/lib/Dialect/Affine/Transforms/SuperVectorize.cpp
    A mlir/test/Dialect/Affine/SuperVectorize/vectorize_affine_apply.mlir

  Log Message:
  -----------
  [mlir][vectorize] Support affine.apply in SuperVectorize (#77968)

We have no need to vectorize affine.apply inside the vectorizing loop.
However, we still need to generate it in the original scalar form. We
have to replace all its operands with the generated scalar operands in
the vectorizing loop, e.g., induction variables.