[all-commits] [llvm/llvm-project] 3f2e3e: [mlir][AMDGPU][NFC] Fix overlapping masked load re...

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 3f2e3e67c11d3a86123aeb9ef5adfd9c9eb6f3ba
      https://github.com/llvm/llvm-project/commit/3f2e3e67c11d3a86123aeb9ef5adfd9c9eb6f3ba
  Author: Krzysztof Drewniak <Krzysztof.Drewniak at amd.com>
  Date:   2025-12-02 (Tue, 02 Dec 2025)

  Changed paths:
    M mlir/lib/Dialect/AMDGPU/Transforms/MaskedloadToLoad.cpp

  Log Message:
  -----------
  [mlir][AMDGPU][NFC] Fix overlapping masked load refinements (#159805)

The two paterns for handlig vector.maskedload on AMD GPUs had an overlap
- both the "scalar mask becomes an if statement" pattern and the "masked
loads become a normal load + a select on buffers" patterns could handle
a load with a broadcast mask on a fat buffer resource.

This commet add checks to resolve the overlap.



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