[PATCH] D25200: [InstCombine] New opportunities for FoldAndOfICmp and FoldXorOfICmp

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

Sorry for not noticing this earlier, but InstCombine always canonicalizes the 'icmp ugt X, 0' cases to 'icmp ne X, 0' (see foldICmpUsingKnownBits()).

Can you just assert that the predicate is not 'ugt' in isAnyBitSet() and simplify that a bit more? Ie, we don't need a switch if we know that eq/ne are the only predicate possibilities for this transform.


https://reviews.llvm.org/D25200