[all-commits] [llvm/llvm-project] eb5541: [mlir][Arith] Prevent IR modification for non-matc...

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: eb554128acef5f814d02d357f10e4666e32a699b
      https://github.com/llvm/llvm-project/commit/eb554128acef5f814d02d357f10e4666e32a699b
  Author: Krzysztof Drewniak <Krzysztof.Drewniak at amd.com>
  Date:   2025-07-22 (Tue, 22 Jul 2025)

  Changed paths:
    M mlir/lib/Dialect/Arith/Transforms/ExpandOps.cpp

  Log Message:
  -----------
  [mlir][Arith] Prevent IR modification for non-matching pattern (#150103)

The F4E2M1 truncation emulation was expanding or truncating operations
to F32 even when the pattern did not apply, causing non-convergent
rewrites when operating on doubles.

Also, fix a pair of whitespace issues that snuck in.



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