[Mlir-commits] [mlir] [MLIR][SCF] Sink scf.if from scf.while before region into after region. (PR #165216)

[Mehdi Amini]

joker-eph wrote:

> I’m not opposed to moving it out of canonicalization, but I believe the reason for temporarily moving it out of canonicalization is that this transformation is complex and expensive, which introduces instability and uncertainty., rather than it being unsuitable for canonicalization. Once this optimization becomes sufficiently stable, could we move it back into the canonicalization stage?

If there is "instability and uncertainty", that's a problem to add this in the first place. 
Otherwise I'd like to understand the path to get this into canonicalization? What does it mean for this to "become sufficiently stable"?

https://github.com/llvm/llvm-project/pull/165216