Index: llvm/lib/CodeGen/MachineLICM.cpp
===================================================================
--- llvm/lib/CodeGen/MachineLICM.cpp
+++ llvm/lib/CodeGen/MachineLICM.cpp
@@ -327,6 +327,21 @@
   return true;
 }
 
+static void addSubLoopsToWorkList(MachineLoop *Loop,
+                                  SmallVectorImpl<MachineLoop *> &Worklist,
+                                  bool PreRA) {
+  // Add loop to worklist
+  Worklist.push_back(Loop);
+
+  // If it is pre-ra LICM, add sub loops to worklist.
+  if (PreRA && !Loop->isInnermost()) {
+    MachineLoop::iterator MLI = Loop->begin();
+    MachineLoop::iterator MLE = Loop->end();
+    for (; MLI != MLE; ++MLI)
+      addSubLoopsToWorkList(*MLI, Worklist, PreRA);
+  }
+}
+
 bool MachineLICMBase::runOnMachineFunction(MachineFunction &MF) {
   if (skipFunction(MF.getFunction()))
     return false;
@@ -366,19 +381,18 @@
   DT  = &getAnalysis<MachineDominatorTree>();
   AA = &getAnalysis<AAResultsWrapperPass>().getAAResults();
 
-  SmallVector<MachineLoop *, 8> Worklist(MLI->begin(), MLI->end());
+  SmallVector<MachineLoop *, 8> Worklist;
+
+  MachineLoopInfo::iterator MLII = MLI->begin();
+  MachineLoopInfo::iterator MLIE = MLI->end();
+  for (; MLII != MLIE; ++MLII)
+    addSubLoopsToWorkList(*MLII, Worklist, PreRegAlloc);
+
   while (!Worklist.empty()) {
     CurLoop = Worklist.pop_back_val();
     CurPreheader = nullptr;
     ExitBlocks.clear();
 
-    // If this is done before regalloc, only visit outer-most preheader-sporting
-    // loops.
-    if (PreRegAlloc && !LoopIsOuterMostWithPredecessor(CurLoop)) {
-      Worklist.append(CurLoop->begin(), CurLoop->end());
-      continue;
-    }
-
     CurLoop->getExitBlocks(ExitBlocks);
 
     if (!PreRegAlloc)
Index: llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll
===================================================================
--- llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll
+++ llvm/test/CodeGen/AArch64/machine-licm-sub-loop.ll
@@ -38,6 +38,7 @@
 ; CHECK-NEXT:    .p2align 5, , 16
 ; CHECK-NEXT:  .LBB0_5: // %vector.ph
 ; CHECK-NEXT:    // in Loop: Header=BB0_3 Depth=1
+; CHECK-NEXT:    dup v0.8h, w15
 ; CHECK-NEXT:    mov x16, x14
 ; CHECK-NEXT:    mov x17, x12
 ; CHECK-NEXT:    mov x18, x11
@@ -45,9 +46,8 @@
 ; CHECK-NEXT:  .LBB0_6: // %vector.body
 ; CHECK-NEXT:    // Parent Loop BB0_3 Depth=1
 ; CHECK-NEXT:    // => This Inner Loop Header: Depth=2
-; CHECK-NEXT:    dup v0.8h, w15
-; CHECK-NEXT:    subs x18, x18, #16
 ; CHECK-NEXT:    ldp q1, q2, [x16, #-16]
+; CHECK-NEXT:    subs x18, x18, #16
 ; CHECK-NEXT:    add x16, x16, #32
 ; CHECK-NEXT:    ldp q4, q3, [x17, #-32]
 ; CHECK-NEXT:    smlal v4.4s, v0.4h, v1.4h