[all-commits] [llvm/llvm-project] 330a7e: [mlir][Vector] Make elementwise-on-broadcast sinki...

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 330a7e1136f536bf7cd642e460734d0bd6e0d0bb
      https://github.com/llvm/llvm-project/commit/330a7e1136f536bf7cd642e460734d0bd6e0d0bb
  Author: Krzysztof Drewniak <Krzysztof.Drewniak at amd.com>
  Date:   2025-07-29 (Tue, 29 Jul 2025)

  Changed paths:
    M mlir/lib/Dialect/Vector/Transforms/VectorTransforms.cpp
    M mlir/test/Dialect/Linalg/vectorization/extract-with-patterns.mlir
    M mlir/test/Dialect/Vector/vector-sink.mlir

  Log Message:
  -----------
  [mlir][Vector] Make elementwise-on-broadcast sinking handle splat consts (#150867)

There is a pattern that rewrites
elementwise_op(broadcast(x1 : T to U), broadcast(x2 : T to U), ...) to
broadcast(elementwise_op(x1, x2, ...) : T to U).

This pattern did not, however, account for the case where a broadcast
constant is represented as a SplatElementsAttr, which can safely be
reshaped or scalarized but is not a `vector.broadcast` or `vector.splat`
operation.

This patch fixes this oversight, prenting premature broadcasting.

This did result in the need to update some linalg dialect tests, which
now feature a less-broadcast computation and/or more constant folding.



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