[PATCH] D50636: [DAGCombiner] Add X / X -> 1 & X % X -> 0 folds.

[Jonas Paulsson via Phabricator via llvm-commits]

jonpa added a comment.

The origin for that test case I believe is that I had found a crash during testing and I reported this as PR32372.

The test changes here look OK to me, but when I reverted the fix for that original bug it seems the new test does no longer fail, while the unmodified one still does... :-/

Unfortunately, I don't seem to have the unreduced test failure available, so I can't make a new reduced version or even try if the original test case still fails...

Not sure if we should remove this test then, or keep it in case it may trigger somehow in the future...? Of course, the assert that the test used to trigger is still there, so removing it isn't all that unthinkable to me...


Repository:
  rL LLVM

https://reviews.llvm.org/D50636