Index: lib/Transforms/Vectorize/LoadStoreVectorizer.cpp
===================================================================
--- lib/Transforms/Vectorize/LoadStoreVectorizer.cpp
+++ lib/Transforms/Vectorize/LoadStoreVectorizer.cpp
@@ -49,6 +49,7 @@
 #include "llvm/ADT/Statistic.h"
 #include "llvm/ADT/iterator_range.h"
 #include "llvm/Analysis/AliasAnalysis.h"
+#include "llvm/Analysis/InstructionSimplify.h"
 #include "llvm/Analysis/MemoryLocation.h"
 #include "llvm/Analysis/OrderedBasicBlock.h"
 #include "llvm/Analysis/ScalarEvolution.h"
@@ -473,6 +474,7 @@
   OrderedBasicBlock OBB(I->getParent());
   SmallPtrSet<Instruction *, 16> InstructionsToMove;
   SmallVector<Instruction *, 16> Worklist;
+  SimplifyQuery SQ(DL, nullptr, &DT);
 
   Worklist.push_back(I);
   while (!Worklist.empty()) {
@@ -489,6 +491,11 @@
         continue;
 
       if (!OBB.dominates(IM, I)) {
+        if (auto *S = SimplifyInstruction(IM, SQ.getWithInstruction(IM))) {
+          IW->setOperand(i, S);
+          i--;
+          continue;
+        }
         InstructionsToMove.insert(IM);
         Worklist.push_back(IM);
       }
Index: test/Transforms/LoadStoreVectorizer/AArch64/reorder-infinite-loop.ll
===================================================================
--- /dev/null
+++ test/Transforms/LoadStoreVectorizer/AArch64/reorder-infinite-loop.ll
@@ -0,0 +1,27 @@
+; RUN: opt -load-store-vectorizer -S -o - %s | FileCheck %s
+target triple = "aarch64"
+
+@a = dso_local global i32 0, align 1
+@b = dso_local global [1 x [3 x i32]] zeroinitializer, align 1
+
+; arrayidx4 should have the last index replaced with 0 due to simplification,
+; to avoid the infinite loop in reorder.
+; CHECK-LABEL: @main()
+; CHECK: %arrayidx4 = getelementptr inbounds [1 x [3 x i32]], [1 x [3 x i32]]* @b, i32 0, i32 0
+
+; Function Attrs: noinline nounwind
+define dso_local i16 @main() #0 {
+entry:
+  %0 = load i32, i32* @a, align 1
+  %rem = urem i32 %0, 1
+  %arrayidx = getelementptr inbounds [1 x [3 x i32]], [1 x [3 x i32]]* @b, i32 0, i32 %rem
+  %arrayidx1 = getelementptr inbounds [3 x i32], [3 x i32]* %arrayidx, i32 0, i32 2
+  %1 = load i32, i32* %arrayidx1, align 1
+  %rem2 = urem i32 %1, 1
+  %arrayidx4 = getelementptr inbounds [1 x [3 x i32]], [1 x [3 x i32]]* @b, i32 0, i32 %rem2
+  %arrayidx5 = getelementptr inbounds [3 x i32], [3 x i32]* %arrayidx4, i32 0, i32 1
+  %2 = load i32, i32* %arrayidx5, align 1
+  ret i16 0
+}
+
+attributes #0 = { noinline nounwind }