diff --git a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index cc6eac70ba7..c88aad66934 100644
--- a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -1814,12 +1814,14 @@ void SelectionDAGBuilder::visitSPDescriptorParent(StackProtectorDescriptor &SPD,
 
   SDValue Guard;
 
-  // If useLoadStackGuardNode returns true, retrieve the guard value from
-  // the virtual register holding the value. Otherwise, emit a volatile load
-  // to retrieve the stack guard value.
-  if (TLI->useLoadStackGuardNode())
-    Guard = DAG.getCopyFromReg(DAG.getEntryNode(), getCurSDLoc(),
-                               SPD.getGuardReg(), PtrTy);
+  // If GuardReg is set and useLoadStackGuardNode returns true, retrieve the
+  // guard value from the virtual register holding the value. Otherwise, emit a
+  // volatile load to retrieve the stack guard value.
+  unsigned GuardReg = SPD.getGuardReg();
+
+  if (GuardReg && TLI->useLoadStackGuardNode())
+    Guard = DAG.getCopyFromReg(DAG.getEntryNode(), getCurSDLoc(), GuardReg,
+                               PtrTy);
   else
     Guard = DAG.getLoad(PtrTy, getCurSDLoc(), DAG.getEntryNode(),
                         GuardPtr, MachinePointerInfo(IRGuard, 0),
@@ -5263,13 +5265,21 @@ SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I, unsigned Intrinsic) {
     MachineFrameInfo *MFI = MF.getFrameInfo();
     EVT PtrTy = TLI->getPointerTy();
     SDValue Src, Chain = getRoot();
+    const Value *Ptr = cast<LoadInst>(I.getArgOperand(0))->getPointerOperand();
+    const GlobalVariable *GV = dyn_cast<GlobalVariable>(Ptr);
 
-    if (TLI->useLoadStackGuardNode()) {
+    // See if Ptr is a bitcast. If it is, look through it and see if we can get
+    // global variable __stack_chk_guard.
+    if (!GV)
+      if (const Operator *BC = dyn_cast<Operator>(Ptr))
+        if (BC->getOpcode() == Instruction::BitCast)
+          GV = dyn_cast<GlobalVariable>(BC->getOperand(0));
+
+    if (GV && TLI->useLoadStackGuardNode()) {
       // Emit a LOAD_STACK_GUARD node.
       MachineSDNode *Node = DAG.getMachineNode(TargetOpcode::LOAD_STACK_GUARD,
                                                sdl, PtrTy, Chain);
-      LoadInst *LI = cast<LoadInst>(I.getArgOperand(0));
-      MachinePointerInfo MPInfo(LI->getPointerOperand());
+      MachinePointerInfo MPInfo(GV);
       MachineInstr::mmo_iterator MemRefs = MF.allocateMemRefsArray(1);
       unsigned Flags = MachineMemOperand::MOLoad |
                        MachineMemOperand::MOInvariant;
diff --git a/test/CodeGen/AArch64/stack-guard-remat-bitcast.ll b/test/CodeGen/AArch64/stack-guard-remat-bitcast.ll
new file mode 100644
index 00000000000..bedbf5facbb
--- /dev/null
+++ b/test/CodeGen/AArch64/stack-guard-remat-bitcast.ll
@@ -0,0 +1,26 @@
+; RUN: llc < %s -mtriple=arm64-apple-ios -relocation-model=pic | FileCheck %s
+
+@__stack_chk_guard = external global i64*
+
+; PR20558
+
+; CHECK: adrp [[R0:x[0-9]+]], ___stack_chk_guard@GOTPAGE
+; CHECK: ldr  [[R1:x[0-9]+]], {{\[}}[[R0]], ___stack_chk_guard@GOTPAGEOFF{{\]}}
+; CHECK: ldr  [[R2:x[0-9]+]], {{\[}}[[R1]]{{\]}}
+; CHECK: stur [[R2]], {{\[}}x29, [[SLOT0:[0-9#\-]+]]{{\]}}
+; CHECK: ldur [[R3:x[0-9]+]], {{\[}}x29, [[SLOT0]]{{\]}}
+; CHECK: sub  [[R4:x[0-9]+]], [[R2]], [[R3]]
+; CHECK: cbnz [[R4]], LBB
+
+define i32 @test_stack_guard_remat2() {
+entry:
+  %StackGuardSlot = alloca i8*
+  %StackGuard = load i8** bitcast (i64** @__stack_chk_guard to i8**)
+  call void @llvm.stackprotector(i8* %StackGuard, i8** %StackGuardSlot)
+  %container = alloca [32 x i8], align 1
+  call void @llvm.stackprotectorcheck(i8** bitcast (i64** @__stack_chk_guard to i8**))
+  ret i32 -1
+}
+
+declare void @llvm.stackprotector(i8*, i8**)
+declare void @llvm.stackprotectorcheck(i8**)