[PATCH] D148633: [SCEV] Clarify inference in isAddRecNeverPoison()

[Max Kazantsev via Phabricator via llvm-commits]

mkazantsev accepted this revision.
mkazantsev added inline comments.
This revision is now accepted and ready to land.


================
Comment at: llvm/lib/Analysis/ScalarEvolution.cpp:7292
     for (const Use &U : Poison->uses()) {
-      const User *PoisonUser = U.getUser();
-      if (propagatesPoison(U)) {
-        if (Pushed.insert(cast<Instruction>(PoisonUser)).second)
-          PoisonStack.push_back(cast<Instruction>(PoisonUser));
-      } else if (auto *BI = dyn_cast<BranchInst>(PoisonUser)) {
-        assert(BI->isConditional() && "Only possibility!");
-        if (BI->getParent() == LatchBB) {
-          LatchControlDependentOnPoison = true;
-          break;
-        }
-      }
+      const Instruction *PoisonUser = cast<Instruction>(U.getUser());
+      if (mustTriggerUB(PoisonUser, KnownPoison) &&
----------------
Skip users that are not in the same loop just to save some time?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148633/new/

https://reviews.llvm.org/D148633