[Mlir-commits] [mlir] [mlir][tensor] Loosen restrictions on folding dynamic reshapes (PR #137963)

[Artem Gindinson]

AGindinson wrote:

@IanWood1 If this is a dynamic size-specific issue, I'd say it's caused by my change either way. That said, I wouldn't be surprised if the same happened for `2x8x128x2 -> 16x256 -> 16x128x2` with the original approach.

Not sure if I understand correctly, what exactly goes against the contiguous rule in this example?

https://github.com/llvm/llvm-project/pull/137963