[all-commits] [llvm/llvm-project] 9630d7: [MLIR][XeGPU] add blocking support for reduce, bro...

[Jianhui Li via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9630d7cb92f1a95ed47e8e336c46a6e8bceb45dd
      https://github.com/llvm/llvm-project/commit/9630d7cb92f1a95ed47e8e336c46a6e8bceb45dd
  Author: Jianhui Li <jian.hui.li at intel.com>
  Date:   2025-06-10 (Tue, 10 Jun 2025)

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

  Log Message:
  -----------
  [MLIR][XeGPU] add blocking support for reduce, broadcast, and transpose (#143389)

This PR adds blocking support for vector dialect operations (`reduce`,
`broadcast`, and `transpose`) in the XeGPU based IR. It simply assigned
the shape specified by "inst_data" as its target shape of the unrolling
to implement the blocking. It is based on
https://github.com/llvm/llvm-project/pull/140163.



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