[PATCH] D109821: [ScalarEvolution] Infer loop max trip count from array accesses

[Liren.Peng via Phabricator via llvm-commits]

Peakulorain added inline comments.


================
Comment at: llvm/lib/Analysis/ScalarEvolution.cpp:7342
+      // And index of GEP wraps arround.
+      if (Step->getAPInt().getActiveBits() > 32 ||
+          Step->getAPInt().getZExtValue() !=
----------------
mkazantsev wrote:
> I don't quite get where this is coming from. Did we ever check the something regarding type width?
This check is to guard against huge step so that we  have not to do some extreme calculations.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D109821/new/

https://reviews.llvm.org/D109821