[llvm-branch-commits] [mlir] [mlir][linalg] Enable scalable vectorization of linalg.unpack (PR #149293)

[Andrzej Warzyński via llvm-branch-commits]

banach-space wrote:

**UPDATE: 31/7/25**

* Implemented https://github.com/llvm/llvm-project/pull/151503 (effectively NFC) and rebased this PR on top of it to help keep the noise level contained.
* The latest [commit](https://github.com/llvm/llvm-project/pull/149293/commits/97fbccdc118f1f3da28b91be129ada6fe4f9b0ea) updates `vectorizeAsTensorUnpackOp` to only require "read-vector-sizes".

I’m quite happy with the current state - `vectorizeAsTensorUnpackOp` is much simpler now. Many thanks for the discussion, @hanhanW 🙏🏻

Also, kudos to @egebeysel - your [comment](https://github.com/llvm/llvm-project/pull/149293#discussion_r2237499014) prompted me to rethink the approach. That's the kind of review feedback every contributor hopes for!

https://github.com/llvm/llvm-project/pull/149293