Index: llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
===================================================================
--- llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
+++ llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
@@ -579,6 +579,12 @@
          !F.hasFnAttribute("no-stack-arg-probe");
 }
 
+static bool needsWinCFI(const MachineFunction &MF) {
+  const Function &F = MF.getFunction();
+  return MF.getTarget().getMCAsmInfo()->usesWindowsCFI() &&
+         F.needsUnwindTableEntry();
+}
+
 bool AArch64FrameLowering::shouldCombineCSRLocalStackBump(
     MachineFunction &MF, uint64_t StackBumpBytes) const {
   AArch64FunctionInfo *AFI = MF.getInfo<AArch64FunctionInfo>();
@@ -589,6 +595,12 @@
   if (AFI->getLocalStackSize() == 0)
     return false;
 
+  // For WinCFI, prefer to not combine the stack bump (to force a stp with
+  // predecrement) to match the packed unwind format, provided that there
+  // actually are any callee saved registers to merge the decrement with.
+  if (needsWinCFI(MF) && AFI->getCalleeSavedStackSize() > 0)
+    return false;
+
   // 512 is the maximum immediate for stp/ldp that will be used for
   // callee-save save/restores
   if (StackBumpBytes >= 512 || windowsRequiresStackProbe(MF, StackBumpBytes))
@@ -982,12 +994,6 @@
   //
 }
 
-static bool needsWinCFI(const MachineFunction &MF) {
-  const Function &F = MF.getFunction();
-  return MF.getTarget().getMCAsmInfo()->usesWindowsCFI() &&
-         F.needsUnwindTableEntry();
-}
-
 static bool isTargetWindows(const MachineFunction &MF) {
   return MF.getSubtarget<AArch64Subtarget>().isTargetWindows();
 }
Index: llvm/test/CodeGen/AArch64/addr-of-ret-addr.ll
===================================================================
--- llvm/test/CodeGen/AArch64/addr-of-ret-addr.ll
+++ llvm/test/CodeGen/AArch64/addr-of-ret-addr.ll
@@ -41,11 +41,11 @@
   ret i32 %add
 
 ; CHECK-LABEL: bar
-; CHECK: sub sp, sp, #96
-; CHECK: stp x29, x30, [sp, #16]
-; CHECK: add x29, sp, #16
+; CHECK: stp x29, x30, [sp, #-80]!
+; CHECK: mov x29, sp
+; CHECK: sub sp, sp, #16
 ; CHECK: stp x1, x2, [x29, #24]
 ; CHECK: add x1, x29, #8
-; CHECK: ldp x29, x30, [sp, #16]
-; CHECK: add sp, sp, #96
+; CHECK: add sp, sp, #16
+; CHECK: ldp x29, x30, [sp], #80
 }
Index: llvm/test/CodeGen/AArch64/seh-finally.ll
===================================================================
--- llvm/test/CodeGen/AArch64/seh-finally.ll
+++ llvm/test/CodeGen/AArch64/seh-finally.ll
@@ -35,7 +35,8 @@
 define void @simple_seh() #0 personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*) {
 entry:
 ; CHECK-LABEL: simple_seh
-; CHECK: add     x29, sp, #16
+; CHECK: mov     x29, sp
+; CHECK: sub     sp, sp, #16
 ; CHECK: mov     x0, #-2
 ; CHECK: stur    x0, [x29, #16]
 ; CHECK: .set .Lsimple_seh$frame_escape_0, -8
@@ -140,7 +141,8 @@
 define void @vla_present(i32 %n) #0 personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*) {
 entry:
 ; CHECK-LABEL: vla_present
-; CHECK: add     x29, sp, #32
+; CHECK: mov     x29, sp
+; CHECK: sub     sp, sp, #32
 ; CHECK: mov     x1, #-2
 ; CHECK: stur    x1, [x29, #16]
 ; CHECK: .set .Lvla_present$frame_escape_0, -4
Index: llvm/test/CodeGen/AArch64/sponentry.ll
===================================================================
--- llvm/test/CodeGen/AArch64/sponentry.ll
+++ llvm/test/CodeGen/AArch64/sponentry.ll
@@ -37,12 +37,14 @@
 }
 
 ; CHECK: foo:
-; CHECK: sub     sp, sp, #448
-; CHECK: add     x29, sp, #424
+; CHECK: str     x28, [sp, #-32]!
+; CHECK: add     x29, sp, #8
+; CHECK: sub     sp, sp, #416
 ; CHECK: add     x1, x29, #24
 ; CHECK: bl      _setjmpex
 
-; NOFP: sub     sp, sp, #432
+; NOFP: stp     x29, x30, [sp, #-16]!
+; NOFP: sub     sp, sp, #416
 ; NOFP: add     x1, sp, #432
 
 define dso_local void @var_args(i8*, ...) {
@@ -67,12 +69,14 @@
 }
 
 ; CHECK: var_args:
-; CHECK: sub     sp, sp, #96
-; CHECK: add     x29, sp, #16
+; CHECK: stp     x29, x30, [sp, #-80]!
+; CHECK: mov     x29, sp
+; CHECK: sub     sp, sp, #16
 ; CHECK: add     x1, x29, #80
 ; CHECK: bl      _setjmpex
 
-; NOFP: sub     sp, sp, #96
+; NOFP: str     x30, [sp, #-80]!
+; NOFP: sub     sp, sp, #16
 ; NOFP: add     x1, sp, #96
 
 define dso_local void @manyargs(i64 %x1, i64 %x2, i64 %x3, i64 %x4, i64 %x5, i64 %x6, i64 %x7, i64 %x8, i64 %x9, i64 %x10) {
Index: llvm/test/CodeGen/AArch64/win64_vararg.ll
===================================================================
--- llvm/test/CodeGen/AArch64/win64_vararg.ll
+++ llvm/test/CodeGen/AArch64/win64_vararg.ll
@@ -211,9 +211,9 @@
 declare void @llvm.stackrestore(i8*)
 
 ; CHECK-LABEL: snprintf
-; CHECK-DAG: sub     sp,  sp, #96
-; CHECK-DAG: stp     x19, x20, [sp, #16]
-; CHECK-DAG: stp     x21, x30, [sp, #32]
+; CHECK-DAG: stp     x19, x20, [sp, #-80]!
+; CHECK-DAG: stp     x21, x30, [sp, #16]
+; CHECK-DAG: sub     sp, sp, #16
 ; CHECK-DAG: add     x8, sp, #56
 ; CHECK-DAG: mov     x19, x2
 ; CHECK-DAG: mov     x20, x1
@@ -231,11 +231,11 @@
 ; CHECK-DAG: mov     x3, x19
 ; CHECK-DAG: mov     x4, xzr
 ; CHECK-DAG: bl      __stdio_common_vsprintf
-; CHECK-DAG: ldp     x21, x30, [sp, #32]
-; CHECK-DAG: ldp     x19, x20, [sp, #16]
 ; CHECK-DAG: cmp     w0, #0
 ; CHECK-DAG: csinv   w0, w0, wzr, ge
-; CHECK-DAG: add     sp, sp, #96
+; CHECK-DAG: add     sp, sp, #16
+; CHECK-DAG: ldp     x21, x30, [sp, #16]
+; CHECK-DAG: ldp     x19, x20, [sp], #80
 ; CHECK-DAG: ret
 define i32 @snprintf(i8*, i64, i8*, ...) local_unnamed_addr #5 {
   %4 = alloca i8*, align 8
Index: llvm/test/CodeGen/AArch64/wineh-frame-predecrement.mir
===================================================================
--- /dev/null
+++ llvm/test/CodeGen/AArch64/wineh-frame-predecrement.mir
@@ -0,0 +1,65 @@
+# RUN: llc -o - %s -mtriple=aarch64-windows -start-before=prologepilog \
+# RUN:   -stop-after=prologepilog | FileCheck %s
+
+# Check that the callee-saved registers are saved starting with a STP
+# with predecrement, followed by a separate stack adjustment later.
+
+# CHECK:      early-clobber $sp = frame-setup STPXpre killed $x19, killed $x20, $sp, -2
+# CHECK-NEXT: frame-setup SEH_SaveRegP_X 19, 20, -16
+# CHECK-NEXT: $sp = frame-setup SUBXri $sp, 16, 0
+# CHECK-NEXT: frame-setup SEH_StackAlloc 16
+# CHECK-NEXT: frame-setup SEH_PrologEnd
+
+...
+---
+name:            func
+alignment:       4
+exposesReturnsTwice: false
+legalized:       false
+regBankSelected: false
+selected:        false
+failedISel:      false
+tracksRegLiveness: true
+hasWinCFI:       false
+registers:       []
+liveins:         []
+frameInfo:
+  isFrameAddressTaken: false
+  isReturnAddressTaken: false
+  hasStackMap:     false
+  hasPatchPoint:   false
+  stackSize:       0
+  offsetAdjustment: 0
+  maxAlignment:    4
+  adjustsStack:    false
+  hasCalls:        false
+  stackProtector:  ''
+  maxCallFrameSize: 0
+  cvBytesOfCalleeSavedRegisters: 0
+  hasOpaqueSPAdjustment: false
+  hasVAStart:      false
+  hasMustTailInVarArgFunc: false
+  localFrameSize:  4
+  savePoint:       ''
+  restorePoint:    ''
+fixedStack:      []
+stack:
+  - { id: 0, name: '', type: default, offset: 0, size: 4, alignment: 4,
+      stack-id: default, callee-saved-register: '', callee-saved-restored: true,
+      local-offset: -4, debug-info-variable: '', debug-info-expression: '',
+      debug-info-location: '' }
+callSites:       []
+constants:       []
+machineFunctionInfo: {}
+body:             |
+  bb.0.entry:
+    liveins: $x0, $x19, $x20
+
+    renamable $x8 = ADDXri %stack.0, 0, 0
+    $x19 = ADDXrr $x0, $x8
+    $x20 = ADDXrr $x19, $x0
+    $x0 = ADDXrr $x0, killed $x20
+
+    RET_ReallyLR
+
+...