[Mlir-commits] [mlir] [mlir][scf] Fix `FoldTensorCastOfOutputIntoForallOp` for multi-result scf.forall (PR #173271)

[Oleksandr Alex Zinenko]

ftynse wrote:

Can we clarify and document semantics of these while we are here? Potentially encode it in the verifier, e.g., there should be exactly one user of a `shared_outs` argument in the `in_parallel` region. I see @joker-eph already commented in that direction on the issue

https://github.com/llvm/llvm-project/pull/173271