[PATCH] D54552: [InstSimplify] delete shift-of-zero guard ops around funnel shifts

[Fabian Giesen via Phabricator via llvm-commits]

fabiang accepted this revision.
fabiang added a comment.
This revision is now accepted and ready to land.

Looks correct to me.

I don't see any issues with doing this at the target-independent level. It could be a problem if fshl/fshr had tricky boundary cases, but their shift amount is taken to be modulo bitwidth and the ShAmt=0 case is explicitly legal and well-defined.


https://reviews.llvm.org/D54552