[PATCH] D104148: [LoopUtils] Fix incorrect runtime checks

[Mindong Chen via Phabricator via llvm-commits]

mdchen added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/pr50686.ll:19
 ; CHECK-NEXT:    [[SCEVGEP2:%.*]] = bitcast i32* [[SCEVGEP]] to i8*
-; CHECK-NEXT:    [[SCEVGEP3:%.*]] = getelementptr i32, i32* [[P2]], i64 2
-; CHECK-NEXT:    [[SCEVGEP34:%.*]] = bitcast i32* [[SCEVGEP3]] to i8*
-; CHECK-NEXT:    [[UGLYGEP:%.*]] = getelementptr i8, i8* [[SCEVGEP34]], i64 1
-; CHECK-NEXT:    [[BC:%.*]] = bitcast i32* [[ARRAYIDX9_2]] to i8*
-; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ult i8* [[P1]], [[UGLYGEP]]
-; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ult i8* [[BC]], [[SCEVGEP2]]
+; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ult i8* [[P1]], [[ARRAYIDX9_24]]
+; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ult i8* [[P23]], [[SCEVGEP2]]
----------------
Ahh..I think the range is still incorrect now, which should be [0, 3) here rather than [0, 2). 


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D104148/new/

https://reviews.llvm.org/D104148