[all-commits] [llvm/llvm-project] 326287: Add missing FillOp to winograd lowering (#108181)

[RoboTux via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 326287fd5b7b38987dbfbe80013225485d261790
      https://github.com/llvm/llvm-project/commit/326287fd5b7b38987dbfbe80013225485d261790
  Author: Thomas Preud'homme <thomas.preudhomme at arm.com>
  Date:   2024-09-13 (Fri, 13 Sep 2024)

  Changed paths:
    M mlir/lib/Dialect/Linalg/Transforms/WinogradConv2D.cpp
    M mlir/test/Dialect/Linalg/transform-tile-and-winograd-rewrite.mlir
    M mlir/test/Dialect/Linalg/winograd-conv2d-rewrite.mlir
    M mlir/test/Dialect/Linalg/winograd-conv2d.mlir

  Log Message:
  -----------
  Add missing FillOp to winograd lowering (#108181)

Winograd lowering involves a number of matmul and batch_matmul which
are currently passed tensor.empty result as out parameter, thereby
are undefined behaviour. This commit adds the necessary linalg.fill.

---------

Co-authored-by: Max191 <44243577+Max191 at users.noreply.github.com>



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