[all-commits] [llvm/llvm-project] 4b59b7: [mlir][Linalg] Fix fusing of indexed linalg consum...

[Simone Pellegrini via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 4b59b7b94608ddbd21d14bec68400f2eb21f510d
      https://github.com/llvm/llvm-project/commit/4b59b7b94608ddbd21d14bec68400f2eb21f510d
  Author: Simone Pellegrini <simone.pellegrini at arm.com>
  Date:   2025-06-13 (Fri, 13 Jun 2025)

  Changed paths:
    M mlir/lib/Dialect/Linalg/Transforms/ElementwiseOpFusion.cpp
    M mlir/test/Dialect/Linalg/fusion-elementwise-ops.mlir

  Log Message:
  -----------
  [mlir][Linalg] Fix fusing of indexed linalg consumer with different axes (#140892)

When fusing two `linalg.genericOp`, where the producer has index
semantics, invalid `affine.apply` ops can be generated where the number
of indices do not match the number of loops in the fused genericOp.

This patch fixes the issue by directly using the number of loops from
the generated fused op.



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