[Mlir-commits] [mlir] [mlir][ArmSME] Remove `ConvertIllegalShapeCastOpsToTransposes` (PR #139706)

Benjamin Maxwell llvmlistbot at llvm.org
Wed Jun 18 04:55:25 PDT 2025


MacDue wrote:

I think you could just update the `LiftIllegalVectorTransposeToMemory` rewrite to match `vector.shape_cast` for this case (rather than `vector.transpose`). That should lead to less duplication as the existing lowerings for `memref.transpose` and the vector transfer operations would introduce the necessary loops.

https://github.com/llvm/llvm-project/pull/139706


More information about the Mlir-commits mailing list