[all-commits] [llvm/llvm-project] 510380: [mlir][xegpu] Support for nD memrefs in `vector.tr...

[Charitha Saumya via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 510380a43a3d61f13d50c4118036d1f51f22ccd2
      https://github.com/llvm/llvm-project/commit/510380a43a3d61f13d50c4118036d1f51f22ccd2
  Author: Charitha Saumya <136391709+charithaintc at users.noreply.github.com>
  Date:   2026-05-01 (Fri, 01 May 2026)

  Changed paths:
    M mlir/lib/Conversion/VectorToXeGPU/VectorToXeGPU.cpp
    M mlir/test/Conversion/VectorToXeGPU/transfer-read-to-xegpu.mlir

  Log Message:
  -----------
  [mlir][xegpu] Support for nD memrefs in `vector.transfer_read` with transposed permutation maps.  (#195197)

Current implementation fails when loading from a 3d memref with perm map
(d0, d1, d2) -> (d2, d1).



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