Index: llvm/lib/Transforms/Scalar/LoopRerollPass.cpp
===================================================================
--- llvm/lib/Transforms/Scalar/LoopRerollPass.cpp
+++ llvm/lib/Transforms/Scalar/LoopRerollPass.cpp
@@ -1080,6 +1080,12 @@
   DenseSet<Instruction*> V;
   collectInLoopUserSet(LoopIncs, Exclude, PossibleRedSet, V);
   for (auto *I : V) {
+    if (I->mayHaveSideEffects()) {
+      LLVM_DEBUG(dbgs() << "LRR: Aborting - An instruction which does not "
+                        << "belongs to any root sets may have side effects: "
+                        << *I);
+      return false;
+    }
     Uses[I].set(IL_All);
   }
 
Index: llvm/test/Transforms/LoopReroll/extra_instr.ll
===================================================================
--- /dev/null
+++ llvm/test/Transforms/LoopReroll/extra_instr.ll
@@ -0,0 +1,135 @@
+; RUN: opt -S  -loop-reroll   %s | FileCheck %s
+target triple = "aarch64--linux-gnu"
+
+define void @rerollable1([2 x i32]* nocapture %a) {
+entry:
+  br label %loop
+
+loop:
+
+; CHECK-LABEL: loop:
+; CHECK-NEXT:    %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+; CHECK-NEXT:    %scevgep
+
+  ; base instruction
+  %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+
+  ; NO unrerollable instructions
+
+  ; extra simple arithmetic operations, used by root instructions
+  %plus20 = add nuw nsw i64 %indvars, 20
+  %plus10 = add nuw nsw i64 %indvars, 10
+
+  ; root instruction 0
+  %ldptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 0
+  %value0 = load i32, i32* %ldptr0, align 4
+  %stptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 0
+  store i32 %value0, i32* %stptr0, align 4
+
+  ; root instruction 1
+  %ldptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 1
+  %value1 = load i32, i32* %ldptr1, align 4
+  %stptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 1
+  store i32 %value1, i32* %stptr1, align 4
+
+  ; loop-increment
+  %indvars.next = add nuw nsw i64 %indvars, 1
+
+  ; latch
+  %exitcond = icmp eq i64 %indvars.next, 5
+  br i1 %exitcond, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @unrerollable1([2 x i32]* nocapture %a) {
+entry:
+  br label %loop
+
+loop:
+
+; CHECK-LABEL: loop:
+; CHECK-NEXT:    %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+; CHECK-NEXT:    %stptrx = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %indvars, i64 0
+; CHECK-NEXT:    store i32 999, i32* %stptrx, align 4
+
+  ; base instruction
+  %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+
+  ; unrerollable instructions using %indvars
+  %stptrx = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %indvars, i64 0
+  store i32 999, i32* %stptrx, align 4
+
+  ; extra simple arithmetic operations, used by root instructions
+  %plus20 = add nuw nsw i64 %indvars, 20
+  %plus10 = add nuw nsw i64 %indvars, 10
+
+  ; root instruction 0
+  %ldptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 0
+  %value0 = load i32, i32* %ldptr0, align 4
+  %stptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 0
+  store i32 %value0, i32* %stptr0, align 4
+
+  ; root instruction 1
+  %ldptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 1
+  %value1 = load i32, i32* %ldptr1, align 4
+  %stptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 1
+  store i32 %value1, i32* %stptr1, align 4
+
+  ; loop-increment
+  %indvars.next = add nuw nsw i64 %indvars, 1
+
+  ; latch
+  %exitcond = icmp eq i64 %indvars.next, 5
+  br i1 %exitcond, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @unrerollable2([2 x i32]* nocapture %a) {
+entry:
+  br label %loop
+
+loop:
+
+; CHECK-LABEL: loop:
+; CHECK-NEXT:    %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+; CHECK-NEXT:    %indvars.next = add nuw nsw i64 %indvars, 1
+; CHECK-NEXT:    %stptrx = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %indvars.next, i64 0
+; CHECK-NEXT:    store i32 999, i32* %stptrx, align 4
+
+  ; base instruction
+  %indvars = phi i64 [ 0, %entry ], [ %indvars.next, %loop ]
+
+  ; loop-increment
+  %indvars.next = add nuw nsw i64 %indvars, 1
+
+  ; unrerollable instructions using %indvars.next
+  %stptrx = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %indvars.next, i64 0
+  store i32 999, i32* %stptrx, align 4
+
+  ; extra simple arithmetic operations, used by root instructions
+  %plus20 = add nuw nsw i64 %indvars, 20
+  %plus10 = add nuw nsw i64 %indvars, 10
+
+  ; root instruction 0
+  %ldptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 0
+  %value0 = load i32, i32* %ldptr0, align 4
+  %stptr0 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 0
+  store i32 %value0, i32* %stptr0, align 4
+
+  ; root instruction 1
+  %ldptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus20, i64 1
+  %value1 = load i32, i32* %ldptr1, align 4
+  %stptr1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 %plus10, i64 1
+  store i32 %value1, i32* %stptr1, align 4
+
+  ; latch
+  %exitcond = icmp eq i64 %indvars.next, 5
+  br i1 %exitcond, label %exit, label %loop
+
+exit:
+  ret void
+}