[Mlir-commits] [mlir] [mlir] Canonicalize tensor.extract_slice (linalg.fill) (PR #112619)

[llvmlistbot at llvm.org]

MaheshRavishankar wrote:

> Can you elaborate why this isn’t a good canonicalization? Is this losing semantics or anything?

Like Andrej said, it depends on whether for a compiler stack a linalg.fill -> extract_slice is better or a two linalg.fills are better. I think it is an opinionated decision. So I can see reasons not to do it as a canonicalization. Fill is a bit of a corner case though. IMO it is always better to do this, but that is also based on my use case.  I am not necessarily pushing back against this as a canonicalization, but if someone has an issue with it being one, then id be fine moving it out of canonicalization.

https://github.com/llvm/llvm-project/pull/112619