[InstCombine] Fix infinite loop due to bitcast <-> phi transforms

Fix for https://bugs.llvm.org/show_bug.cgi?id=44245.

The optimizeBitCastFromPhi() and FoldPHIArgOpIntoPHI() end up
fighting against each other, because optimizeBitCastFromPhi()
assumes that bitcasts of loads will get folded. This doesn't happen
here, because a dangling phi node prevents the one-use fold in
https://github.com/llvm/llvm-project/blob/master/llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp#L620-L628 from triggering.

This patch fixes the issue by adding manually removing the old phis.

Differential Revision: https://reviews.llvm.org/D71164

1 files changed, 6 insertions, 0 deletions

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
index 3ba56bbe53e..446e6758096 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
@@ -2370,6 +2370,12 @@ Instruction *InstCombiner::optimizeBitCastFromPhi(CastInst &CI, PHINode *PN) {
         llvm_unreachable("all uses should be handled");
       }
     }
+
+    // At this point the old phi has either no users, or is only used
+    // in other old phis. Replace with undef to break circles and remove
+    // the instruction.
+    replaceInstUsesWith(*OldPN, UndefValue::get(OldPN->getType()));
+    eraseInstFromFunction(*OldPN);
   }
 
   return RetVal;