[PATCH] D16986: [LICM] Don't assert on volatile accesses

[Philip Reames via llvm-commits]

reames added a comment.

The changes to LICM in this patch are almost certainly not the right approach.  The removed assertions give a good hint on where to look; this is likely an AliasSetTracker bug.

p.s. Finding another test case which doesn't involve UB - even just using a non-zero address space - would be good.


http://reviews.llvm.org/D16986