[Mlir-commits] [mlir] [mlir][vector] Canonicalize gathers/scatters with trivial offsets (PR #117939)

Diego Caballero llvmlistbot at llvm.org
Wed Dec 11 15:18:34 PST 2024


dcaballe wrote:

> Maybe, instead of adding special case for vector.step just fold all non-scalable vector.steps into arith.constants? It will probably help other canonicalizations as well.

I would say the other way around. We may want `vector.step` to be the canonical form for vector constants that fall into that category as `vector.step` holds valuable information that has to be inferred by inspecting the values of a constant op.

https://github.com/llvm/llvm-project/pull/117939


More information about the Mlir-commits mailing list