[Mlir-commits] [mlir] [mlir][linalg] Add support for scalable vectorization of `linalg.batch_mmt4d` (PR #152984)

[Renato Golin]

rengolin wrote:

That's what I meant. I've misused the term _specialization_. It's still not clear to me how we do _aliasing_ (wrt pattern matching, class structure, rewrites). 

https://github.com/llvm/llvm-project/pull/152984