[LoopFlatten] Replace unchecked dyn_cast with cast.

[craig.topper]

Summary

Spotted while reading through the code.

[fhahn]

llvm/lib/Transforms/Scalar/LoopFlatten.cpp

414

I *think* this must by a binary op at the moment, is maybe it's possible to just use cast. If it triggers, we will at least have a test case :)

[craig.topper]

Use cast instead of dyn_cast

[SjoerdMeijer]

I also think this must be a binary op.

llvm/lib/Transforms/Scalar/LoopFlatten.cpp

 //===- LoopFlatten.cpp - Loop flattening pass------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
static bool findLoopComponents(
   IterationInstructions.insert(Compare);
   LLVM_DEBUG(dbgs() << "Found comparison: "; Compare->dump());
 
   // Find increment and trip count.
   // There are exactly 2 incoming values to the induction phi; one from the
   // pre-header and one from the latch. The incoming latch value is the
   // increment variable.
   Increment =
       dyn_cast<BinaryOperator>(InductionPHI->getIncomingValueForBlock(Latch));
-  if (Increment->hasNUsesOrMore(3)) {
+  if (!Increment || Increment->hasNUsesOrMore(3)) {
     LLVM_DEBUG(dbgs() << "Could not find valid increment\n");
     return false;
   }
   // The trip count is the RHS of the compare. If this doesn't match the trip
   // count computed by SCEV then this is because the trip count variable
   // has been widened so the types don't match, or because it is a constant and
   // another transformation has changed the compare (e.g. icmp ult %inc,
   // tripcount -> icmp ult %j, tripcount-1), or both.