[all-commits] [llvm/llvm-project] 48c9a8: [MLIR][XeGPU] Enable blocking for scatter ops with...

[Nishant Patel via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 48c9a8a9c859701d3bd401dc1e8c5002c9bd9c86
      https://github.com/llvm/llvm-project/commit/48c9a8a9c859701d3bd401dc1e8c5002c9bd9c86
  Author: Nishant Patel <nishant.b.patel at intel.com>
  Date:   2025-10-13 (Mon, 13 Oct 2025)

  Changed paths:
    M mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp
    M mlir/test/Dialect/XeGPU/xegpu-blocking.mlir

  Log Message:
  -----------
  [MLIR][XeGPU] Enable blocking for scatter ops with offsets (#162896)

The unroll patterns for these ops were added in the previous PR but the
getTileShape method was not changed to handle these ops and hence
blocking pass was not kicking in.



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