--- a/lib/Target/X86/X86FrameLowering.cpp	Fri Nov 25 19:14:25 2016 +0900
+++ b/lib/Target/X86/X86FrameLowering.cpp	Fri Oct 27 17:07:41 2017 +0900
@@ -29,8 +29,8 @@
 #include "llvm/IR/Function.h"
 #include "llvm/MC/MCAsmInfo.h"
 #include "llvm/MC/MCSymbol.h"
+#include "llvm/Support/Debug.h"
 #include "llvm/Target/TargetOptions.h"
-#include "llvm/Support/Debug.h"
 #include <cstdlib>
 
 using namespace llvm;
@@ -83,14 +83,12 @@
 /// or if frame pointer elimination is disabled.
 bool X86FrameLowering::hasFP(const MachineFunction &MF) const {
   const MachineFrameInfo &MFI = MF.getFrameInfo();
-  const MachineModuleInfo &MMI = MF.getMMI();
-
   return (MF.getTarget().Options.DisableFramePointerElim(MF) ||
           TRI->needsStackRealignment(MF) ||
           MFI.hasVarSizedObjects() ||
           MFI.isFrameAddressTaken() || MFI.hasOpaqueSPAdjustment() ||
           MF.getInfo<X86MachineFunctionInfo>()->getForceFramePointer() ||
-          MMI.callsUnwindInit() || MMI.hasEHFunclets() || MMI.callsEHReturn() ||
+          MF.callsUnwindInit() || MF.hasEHFunclets() || MF.callsEHReturn() ||
           MFI.hasStackMap() || MFI.hasPatchPoint() ||
           MFI.hasCopyImplyingStackAdjustment());
 }
@@ -151,7 +149,7 @@
                                        bool Is64Bit) {
   const MachineFunction *MF = MBB.getParent();
   const Function *F = MF->getFunction();
-  if (!F || MF->getMMI().callsEHReturn())
+  if (!F || MF->callsEHReturn())
     return 0;
 
   const TargetRegisterClass &AvailableRegs = *TRI->getGPRsForTailCall(*MF);
@@ -254,40 +252,76 @@
                                     int64_t NumBytes, bool InEpilogue) const {
   bool isSub = NumBytes < 0;
   uint64_t Offset = isSub ? -NumBytes : NumBytes;
+  MachineInstr::MIFlag Flag =
+      isSub ? MachineInstr::FrameSetup : MachineInstr::FrameDestroy;
 
   uint64_t Chunk = (1LL << 31) - 1;
   DebugLoc DL = MBB.findDebugLoc(MBBI);
 
-  while (Offset) {
-    if (Offset > Chunk) {
-      // Rather than emit a long series of instructions for large offsets,
-      // load the offset into a register and do one sub/add
-      unsigned Reg = 0;
-
-      if (isSub && !isEAXLiveIn(MBB))
-        Reg = (unsigned)(Is64Bit ? X86::RAX : X86::EAX);
+  if (Offset > Chunk) {
+    // Rather than emit a long series of instructions for large offsets,
+    // load the offset into a register and do one sub/add
+    unsigned Reg = 0;
+    unsigned Rax = (unsigned)(Is64Bit ? X86::RAX : X86::EAX);
+
+    if (isSub && !isEAXLiveIn(MBB))
+      Reg = Rax;
+    else
+      Reg = findDeadCallerSavedReg(MBB, MBBI, TRI, Is64Bit);
+
+    unsigned MovRIOpc = Is64Bit ? X86::MOV64ri : X86::MOV32ri;
+    unsigned AddSubRROpc =
+        isSub ? getSUBrrOpcode(Is64Bit) : getADDrrOpcode(Is64Bit);
+    if (Reg) {
+      BuildMI(MBB, MBBI, DL, TII.get(MovRIOpc), Reg)
+          .addImm(Offset)
+          .setMIFlag(Flag);
+      MachineInstr *MI = BuildMI(MBB, MBBI, DL, TII.get(AddSubRROpc), StackPtr)
+                             .addReg(StackPtr)
+                             .addReg(Reg);
+      MI->getOperand(3).setIsDead(); // The EFLAGS implicit def is dead.
+      return;
+    } else if (Offset > 8 * Chunk) {
+      // If we would need more than 8 add or sub instructions (a >16GB stack
+      // frame), it's worth spilling RAX to materialize this immediate.
+      //   pushq %rax
+      //   movabsq +-$Offset+-SlotSize, %rax
+      //   addq %rsp, %rax
+      //   xchg %rax, (%rsp)
+      //   movq (%rsp), %rsp
+      assert(Is64Bit && "can't have 32-bit 16GB stack frame");
+      BuildMI(MBB, MBBI, DL, TII.get(X86::PUSH64r))
+          .addReg(Rax, RegState::Kill)
+          .setMIFlag(Flag);
+      // Subtract is not commutative, so negate the offset and always use add.
+      // Subtract 8 less and add 8 more to account for the PUSH we just did.
+      if (isSub)
+        Offset = -(Offset - SlotSize);
       else
-        Reg = findDeadCallerSavedReg(MBB, MBBI, TRI, Is64Bit);
-
-      if (Reg) {
-        unsigned Opc = Is64Bit ? X86::MOV64ri : X86::MOV32ri;
-        BuildMI(MBB, MBBI, DL, TII.get(Opc), Reg)
-          .addImm(Offset);
-        Opc = isSub
-          ? getSUBrrOpcode(Is64Bit)
-          : getADDrrOpcode(Is64Bit);
-        MachineInstr *MI = BuildMI(MBB, MBBI, DL, TII.get(Opc), StackPtr)
-          .addReg(StackPtr)
-          .addReg(Reg);
-        MI->getOperand(3).setIsDead(); // The EFLAGS implicit def is dead.
-        Offset = 0;
-        continue;
-      }
+        Offset = Offset + SlotSize;
+      BuildMI(MBB, MBBI, DL, TII.get(MovRIOpc), Rax)
+          .addImm(Offset)
+          .setMIFlag(Flag);
+      MachineInstr *MI = BuildMI(MBB, MBBI, DL, TII.get(X86::ADD64rr), Rax)
+                             .addReg(Rax)
+                             .addReg(StackPtr);
+      MI->getOperand(3).setIsDead(); // The EFLAGS implicit def is dead.
+      // Exchange the new SP in RAX with the top of the stack.
+      addRegOffset(
+          BuildMI(MBB, MBBI, DL, TII.get(X86::XCHG64rm), Rax).addReg(Rax),
+          StackPtr, false, 0);
+      // Load new SP from the top of the stack into RSP.
+      addRegOffset(BuildMI(MBB, MBBI, DL, TII.get(X86::MOV64rm), StackPtr),
+                   StackPtr, false, 0);
+      return;
     }
-
+  }
+
+  while (Offset) {
     uint64_t ThisVal = std::min(Offset, Chunk);
-    if (ThisVal == (Is64Bit ? 8 : 4)) {
-      // Use push / pop instead.
+    if (ThisVal == SlotSize) {
+      // Use push / pop for slot sized adjustments as a size optimization. We
+      // need to find a dead register when using pop.
       unsigned Reg = isSub
         ? (unsigned)(Is64Bit ? X86::RAX : X86::EAX)
         : findDeadCallerSavedReg(MBB, MBBI, TRI, Is64Bit);
@@ -295,23 +329,16 @@
         unsigned Opc = isSub
           ? (Is64Bit ? X86::PUSH64r : X86::PUSH32r)
           : (Is64Bit ? X86::POP64r  : X86::POP32r);
-        MachineInstr *MI = BuildMI(MBB, MBBI, DL, TII.get(Opc))
-          .addReg(Reg, getDefRegState(!isSub) | getUndefRegState(isSub));
-        if (isSub)
-          MI->setFlag(MachineInstr::FrameSetup);
-        else
-          MI->setFlag(MachineInstr::FrameDestroy);
+        BuildMI(MBB, MBBI, DL, TII.get(Opc))
+            .addReg(Reg, getDefRegState(!isSub) | getUndefRegState(isSub))
+            .setMIFlag(Flag);
         Offset -= ThisVal;
         continue;
       }
     }
 
-    MachineInstrBuilder MI = BuildStackAdjustment(
-        MBB, MBBI, DL, isSub ? -ThisVal : ThisVal, InEpilogue);
-    if (isSub)
-      MI.setMIFlag(MachineInstr::FrameSetup);
-    else
-      MI.setMIFlag(MachineInstr::FrameDestroy);
+    BuildStackAdjustment(MBB, MBBI, DL, isSub ? -ThisVal : ThisVal, InEpilogue)
+        .setMIFlag(Flag);
 
     Offset -= ThisVal;
   }
@@ -375,6 +402,10 @@
   MachineBasicBlock::iterator PI = doMergeWithPrevious ? std::prev(MBBI) : MBBI;
   MachineBasicBlock::iterator NI = doMergeWithPrevious ? nullptr
                                                        : std::next(MBBI);
+  PI = skipDebugInstructionsBackward(PI, MBB.begin());
+  if (NI != nullptr)
+    NI = skipDebugInstructionsForward(NI, MBB.end());
+
   unsigned Opc = PI->getOpcode();
   int Offset = 0;
 
@@ -418,7 +449,7 @@
                                 const DebugLoc &DL,
                                 const MCCFIInstruction &CFIInst) const {
   MachineFunction &MF = *MBB.getParent();
-  unsigned CFIIndex = MF.getMMI().addFrameInst(CFIInst);
+  unsigned CFIIndex = MF.addFrameInst(CFIInst);
   BuildMI(MBB, MBBI, DL, TII.get(TargetOpcode::CFI_INSTRUCTION))
       .addCFIIndex(CFIIndex);
 }
@@ -717,17 +748,7 @@
   else
     CallOp = X86::CALLpcrel32;
 
-  const char *Symbol;
-  if (Is64Bit) {
-    if (STI.isTargetCygMing()) {
-      Symbol = "___chkstk_ms";
-    } else {
-      Symbol = "__chkstk";
-    }
-  } else if (STI.isTargetCygMing())
-    Symbol = "_alloca";
-  else
-    Symbol = "_chkstk";
+  StringRef Symbol = STI.getTargetLowering()->getStackProbeSymbolName(MF);
 
   MachineInstrBuilder CI;
   MachineBasicBlock::iterator ExpansionMBBI = std::prev(MBBI);
@@ -738,10 +759,11 @@
     // For the large code model, we have to call through a register. Use R11,
     // as it is scratch in all supported calling conventions.
     BuildMI(MBB, MBBI, DL, TII.get(X86::MOV64ri), X86::R11)
-        .addExternalSymbol(Symbol);
+        .addExternalSymbol(MF.createExternalSymbolName(Symbol));
     CI = BuildMI(MBB, MBBI, DL, TII.get(CallOp)).addReg(X86::R11);
   } else {
-    CI = BuildMI(MBB, MBBI, DL, TII.get(CallOp)).addExternalSymbol(Symbol);
+    CI = BuildMI(MBB, MBBI, DL, TII.get(CallOp))
+        .addExternalSymbol(MF.createExternalSymbolName(Symbol));
   }
 
   unsigned AX = Is64Bit ? X86::RAX : X86::EAX;
@@ -752,13 +774,16 @@
       .addReg(SP, RegState::Define | RegState::Implicit)
       .addReg(X86::EFLAGS, RegState::Define | RegState::Implicit);
 
-  if (Is64Bit) {
+  if (STI.isTargetWin64() || !STI.isOSWindows()) {
+    // MSVC x32's _chkstk and cygwin/mingw's _alloca adjust %esp themselves.
     // MSVC x64's __chkstk and cygwin/mingw's ___chkstk_ms do not adjust %rsp
-    // themselves. It also does not clobber %rax so we can reuse it when
+    // themselves. They also does not clobber %rax so we can reuse it when
     // adjusting %rsp.
-    BuildMI(MBB, MBBI, DL, TII.get(X86::SUB64rr), X86::RSP)
-        .addReg(X86::RSP)
-        .addReg(X86::RAX);
+    // All other platforms do not specify a particular ABI for the stack probe
+    // function, so we arbitrarily define it to not adjust %esp/%rsp itself.
+    BuildMI(MBB, MBBI, DL, TII.get(getSUBrrOpcode(Is64Bit)), SP)
+        .addReg(SP)
+        .addReg(AX);
   }
 
   if (InProlog) {
@@ -899,6 +924,7 @@
 
   Notes:
   - .seh directives are emitted only for Windows 64 ABI
+  - .cv_fpo directives are emitted on win32 when emitting CodeView
   - .cfi directives are emitted for all other ABIs
   - for 32-bit code, substitute %e?? registers for %r??
 */
@@ -919,12 +945,16 @@
   if (Fn->hasPersonalityFn())
     Personality = classifyEHPersonality(Fn->getPersonalityFn());
   bool FnHasClrFunclet =
-      MMI.hasEHFunclets() && Personality == EHPersonality::CoreCLR;
+      MF.hasEHFunclets() && Personality == EHPersonality::CoreCLR;
   bool IsClrFunclet = IsFunclet && FnHasClrFunclet;
   bool HasFP = hasFP(MF);
   bool IsWin64CC = STI.isCallingConvWin64(Fn->getCallingConv());
   bool IsWin64Prologue = MF.getTarget().getMCAsmInfo()->usesWindowsCFI();
-  bool NeedsWinCFI = IsWin64Prologue && Fn->needsUnwindTableEntry();
+  bool NeedsWin64CFI = IsWin64Prologue && Fn->needsUnwindTableEntry();
+  // FIXME: Emit FPO data for EH funclets.
+  bool NeedsWinFPO =
+      !IsFunclet && STI.isTargetWin32() && MMI.getModule()->getCodeViewFlag();
+  bool NeedsWinCFI = NeedsWin64CFI || NeedsWinFPO;
   bool NeedsDwarfCFI =
       !IsWin64Prologue && (MMI.hasDebugInfo() || Fn->needsUnwindTableEntry());
   unsigned FramePtr = TRI->getFrameRegister(MF);
@@ -933,7 +963,7 @@
           ? getX86SubSuperRegister(FramePtr, 64) : FramePtr;
   unsigned BasePtr = TRI->getBaseRegister();
   bool HasWinCFI = false;
-  
+
   // Debug location must be unknown since the first debug location is used
   // to determine the end of the prologue.
   DebugLoc DL;
@@ -947,7 +977,7 @@
     X86FI->setCalleeSavedFrameSize(
       X86FI->getCalleeSavedFrameSize() - TailCallReturnAddrDelta);
 
-  bool UseStackProbe = (STI.isOSWindows() && !STI.isTargetMachO());
+  bool UseStackProbe = !STI.getTargetLowering()->getStackProbeSymbolName(MF).empty();
 
   // The default stack probe size is 4096 if the function has no stackprobesize
   // attribute.
@@ -957,6 +987,16 @@
         .getValueAsString()
         .getAsInteger(0, StackProbeSize);
 
+  // Re-align the stack on 64-bit if the x86-interrupt calling convention is
+  // used and an error code was pushed, since the x86-64 ABI requires a 16-byte
+  // stack alignment.
+  if (Fn->getCallingConv() == CallingConv::X86_INTR && Is64Bit &&
+      Fn->arg_size() == 2) {
+    StackSize += 8;
+    MFI.setStackSize(StackSize);
+    emitSPUpdate(MBB, MBBI, -8, /*InEpilogue=*/false);
+  }
+
   // If this is x86-64 and the Red Zone is not disabled, if we are a leaf
   // function, and use up to 128 bytes of stack space, don't have a frame
   // pointer, calls, or dynamic alloca then we do not need to adjust the
@@ -966,6 +1006,7 @@
       !TRI->needsStackRealignment(MF) &&
       !MFI.hasVarSizedObjects() &&             // No dynamic alloca.
       !MFI.adjustsStack() &&                   // No calls.
+      !UseStackProbe &&                        // No stack probes.
       !IsWin64CC &&                            // Win64 has no Red Zone
       !MFI.hasCopyImplyingStackAdjustment() && // Don't push and pop.
       !MF.shouldSplitStack()) {                // Regular stack
@@ -1021,6 +1062,8 @@
   }
 
   if (HasFP) {
+    assert(MF.getRegInfo().isReserved(MachineFramePtr) && "FP reserved");
+
     // Calculate required stack adjustment.
     uint64_t FrameSize = StackSize - SlotSize;
     // If required, include space for extra hidden slot for stashing base pointer.
@@ -1082,13 +1125,15 @@
         BuildCFI(MBB, MBBI, DL, MCCFIInstruction::createDefCfaRegister(
                                     nullptr, DwarfFramePtr));
       }
-    }
-
-    // Mark the FramePtr as live-in in every block. Don't do this again for
-    // funclet prologues.
-    if (!IsFunclet) {
-      for (MachineBasicBlock &EveryMBB : MF)
-        EveryMBB.addLiveIn(MachineFramePtr);
+
+      if (NeedsWinFPO) {
+        // .cv_fpo_setframe $FramePtr
+        HasWinCFI = true;
+        BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_SetFrame))
+            .addImm(FramePtr)
+            .addImm(0)
+            .setMIFlag(MachineInstr::FrameSetup);
+      }
     }
   } else {
     assert(!IsFunclet && "funclets without FPs not yet implemented");
@@ -1124,8 +1169,9 @@
 
     if (NeedsWinCFI) {
       HasWinCFI = true;
-      BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_PushReg)).addImm(Reg).setMIFlag(
-          MachineInstr::FrameSetup);
+      BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_PushReg))
+          .addImm(Reg)
+          .setMIFlag(MachineInstr::FrameSetup);
     }
   }
 
@@ -1156,6 +1202,9 @@
   if (IsWin64Prologue && !IsFunclet && TRI->needsStackRealignment(MF))
     AlignedNumBytes = alignTo(AlignedNumBytes, MaxAlign);
   if (AlignedNumBytes >= StackProbeSize && UseStackProbe) {
+    assert(!X86FI->getUsesRedZone() &&
+           "The Red Zone is not accounted for in stack probes");
+
     // Check whether EAX is livein for this block.
     bool isEAXAlive = isEAXLiveIn(MBB);
 
@@ -1261,6 +1310,7 @@
 
     // If this is not a funclet, emit the CFI describing our frame pointer.
     if (NeedsWinCFI && !IsFunclet) {
+      assert(!NeedsWinFPO && "this setframe incompatible with FPO data");
       HasWinCFI = true;
       BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_SetFrame))
           .addImm(FramePtr)
@@ -1299,6 +1349,7 @@
           Offset += SEHFrameOffset;
 
           HasWinCFI = true;
+          assert(!NeedsWinFPO && "SEH_SaveXMM incompatible with FPO data");
           BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_SaveXMM))
               .addImm(Reg)
               .addImm(Offset)
@@ -1488,10 +1539,8 @@
                                     MachineBasicBlock &MBB) const {
   const MachineFrameInfo &MFI = MF.getFrameInfo();
   X86MachineFunctionInfo *X86FI = MF.getInfo<X86MachineFunctionInfo>();
-  MachineBasicBlock::iterator MBBI = MBB.getFirstTerminator();
-  Optional<unsigned> RetOpcode;
-  if (MBBI != MBB.end())
-    RetOpcode = MBBI->getOpcode();
+  MachineBasicBlock::iterator Terminator = MBB.getFirstTerminator();
+  MachineBasicBlock::iterator MBBI = Terminator;
   DebugLoc DL;
   if (MBBI != MBB.end())
     DL = MBBI->getDebugLoc();
@@ -1502,38 +1551,21 @@
       Is64BitILP32 ? getX86SubSuperRegister(FramePtr, 64) : FramePtr;
 
   bool IsWin64Prologue = MF.getTarget().getMCAsmInfo()->usesWindowsCFI();
-  bool NeedsWinCFI =
+  bool NeedsWin64CFI =
       IsWin64Prologue && MF.getFunction()->needsUnwindTableEntry();
   bool IsFunclet = MBBI == MBB.end() ? false : isFuncletReturnInstr(*MBBI);
-  MachineBasicBlock *TargetMBB = nullptr;
 
   // Get the number of bytes to allocate from the FrameInfo.
   uint64_t StackSize = MFI.getStackSize();
   uint64_t MaxAlign = calculateMaxStackAlign(MF);
   unsigned CSSize = X86FI->getCalleeSavedFrameSize();
+  bool HasFP = hasFP(MF);
   uint64_t NumBytes = 0;
 
-  if (RetOpcode && *RetOpcode == X86::CATCHRET) {
-    // SEH shouldn't use catchret.
-    assert(!isAsynchronousEHPersonality(
-               classifyEHPersonality(MF.getFunction()->getPersonalityFn())) &&
-           "SEH should not use CATCHRET");
-
+  if (IsFunclet) {
+    assert(HasFP && "EH funclets without FP not yet implemented");
     NumBytes = getWinEHFuncletFrameSize(MF);
-    assert(hasFP(MF) && "EH funclets without FP not yet implemented");
-    TargetMBB = MBBI->getOperand(0).getMBB();
-
-    // Pop EBP.
-    BuildMI(MBB, MBBI, DL, TII.get(Is64Bit ? X86::POP64r : X86::POP32r),
-            MachineFramePtr)
-        .setMIFlag(MachineInstr::FrameDestroy);
-  } else if (RetOpcode && *RetOpcode == X86::CLEANUPRET) {
-    NumBytes = getWinEHFuncletFrameSize(MF);
-    assert(hasFP(MF) && "EH funclets without FP not yet implemented");
-    BuildMI(MBB, MBBI, DL, TII.get(Is64Bit ? X86::POP64r : X86::POP32r),
-            MachineFramePtr)
-        .setMIFlag(MachineInstr::FrameDestroy);
-  } else if (hasFP(MF)) {
+  } else if (HasFP) {
     // Calculate required stack adjustment.
     uint64_t FrameSize = StackSize - SlotSize;
     NumBytes = FrameSize - CSSize;
@@ -1542,50 +1574,37 @@
     // realigned.
     if (TRI->needsStackRealignment(MF) && !IsWin64Prologue)
       NumBytes = alignTo(FrameSize, MaxAlign);
-
-    // Pop EBP.
-    BuildMI(MBB, MBBI, DL,
-            TII.get(Is64Bit ? X86::POP64r : X86::POP32r), MachineFramePtr)
-        .setMIFlag(MachineInstr::FrameDestroy);
   } else {
     NumBytes = StackSize - CSSize;
   }
   uint64_t SEHStackAllocAmt = NumBytes;
 
+  if (HasFP) {
+    // Pop EBP.
+    BuildMI(MBB, MBBI, DL, TII.get(Is64Bit ? X86::POP64r : X86::POP32r),
+            MachineFramePtr)
+        .setMIFlag(MachineInstr::FrameDestroy);
+  }
+
+  MachineBasicBlock::iterator FirstCSPop = MBBI;
   // Skip the callee-saved pop instructions.
   while (MBBI != MBB.begin()) {
     MachineBasicBlock::iterator PI = std::prev(MBBI);
     unsigned Opc = PI->getOpcode();
 
-    if ((Opc != X86::POP32r || !PI->getFlag(MachineInstr::FrameDestroy)) &&
-        (Opc != X86::POP64r || !PI->getFlag(MachineInstr::FrameDestroy)) &&
-        Opc != X86::DBG_VALUE && !PI->isTerminator())
-      break;
+    if (Opc != X86::DBG_VALUE && !PI->isTerminator()) {
+      if ((Opc != X86::POP32r || !PI->getFlag(MachineInstr::FrameDestroy)) &&
+          (Opc != X86::POP64r || !PI->getFlag(MachineInstr::FrameDestroy)))
+        break;
+      FirstCSPop = PI;
+    }
 
     --MBBI;
   }
-  MachineBasicBlock::iterator FirstCSPop = MBBI;
-
-  if (TargetMBB) {
-    // Fill EAX/RAX with the address of the target block.
-    unsigned ReturnReg = STI.is64Bit() ? X86::RAX : X86::EAX;
-    if (STI.is64Bit()) {
-      // LEA64r TargetMBB(%rip), %rax
-      BuildMI(MBB, FirstCSPop, DL, TII.get(X86::LEA64r), ReturnReg)
-          .addReg(X86::RIP)
-          .addImm(0)
-          .addReg(0)
-          .addMBB(TargetMBB)
-          .addReg(0);
-    } else {
-      // MOV32ri $TargetMBB, %eax
-      BuildMI(MBB, FirstCSPop, DL, TII.get(X86::MOV32ri), ReturnReg)
-          .addMBB(TargetMBB);
-    }
-    // Record that we've taken the address of TargetMBB and no longer just
-    // reference it in a terminator.
-    TargetMBB->setHasAddressTaken();
-  }
+  MBBI = FirstCSPop;
+
+  if (IsFunclet && Terminator->getOpcode() == X86::CATCHRET)
+    emitCatchRetReturnValue(MBB, FirstCSPop, &*Terminator);
 
   if (MBBI != MBB.end())
     DL = MBBI->getDebugLoc();
@@ -1637,38 +1656,33 @@
   // into the epilogue.  To cope with that, we insert an epilogue marker here,
   // then replace it with a 'nop' if it ends up immediately after a CALL in the
   // final emitted code.
-  if (NeedsWinCFI && MF.hasWinCFI())
+  if (NeedsWin64CFI && MF.hasWinCFI())
     BuildMI(MBB, MBBI, DL, TII.get(X86::SEH_Epilogue));
 
-  if (!RetOpcode || !isTailCallOpcode(*RetOpcode)) {
+  if (Terminator == MBB.end() || !isTailCallOpcode(Terminator->getOpcode())) {
     // Add the return addr area delta back since we are not tail calling.
     int Offset = -1 * X86FI->getTCReturnAddrDelta();
     assert(Offset >= 0 && "TCDelta should never be positive");
     if (Offset) {
-      MBBI = MBB.getFirstTerminator();
-
       // Check for possible merge with preceding ADD instruction.
-      Offset += mergeSPUpdates(MBB, MBBI, true);
-      emitSPUpdate(MBB, MBBI, Offset, /*InEpilogue=*/true);
+      Offset += mergeSPUpdates(MBB, Terminator, true);
+      emitSPUpdate(MBB, Terminator, Offset, /*InEpilogue=*/true);
     }
   }
 }
 
-// NOTE: this only has a subset of the full frame index logic. In
-// particular, the FI < 0 and AfterFPPop logic is handled in
-// X86RegisterInfo::eliminateFrameIndex, but not here. Possibly
-// (probably?) it should be moved into here.
 int X86FrameLowering::getFrameIndexReference(const MachineFunction &MF, int FI,
                                              unsigned &FrameReg) const {
   const MachineFrameInfo &MFI = MF.getFrameInfo();
 
+  bool IsFixed = MFI.isFixedObjectIndex(FI);
   // We can't calculate offset from frame pointer if the stack is realigned,
   // so enforce usage of stack/base pointer.  The base pointer is used when we
   // have dynamic allocas in addition to dynamic realignment.
   if (TRI->hasBasePointer(MF))
-    FrameReg = TRI->getBaseRegister();
+    FrameReg = IsFixed ? TRI->getFramePtr() : TRI->getBaseRegister();
   else if (TRI->needsStackRealignment(MF))
-    FrameReg = TRI->getStackRegister();
+    FrameReg = IsFixed ? TRI->getFramePtr() : TRI->getStackRegister();
   else
     FrameReg = TRI->getFrameRegister(MF);
 
@@ -1742,6 +1756,14 @@
   return Offset + FPDelta;
 }
 
+int X86FrameLowering::getFrameIndexReferenceSP(const MachineFunction &MF,
+                                               int FI, unsigned &FrameReg,
+                                               int Adjustment) const {
+  const MachineFrameInfo &MFI = MF.getFrameInfo();
+  FrameReg = TRI->getStackRegister();
+  return MFI.getObjectOffset(FI) - getOffsetOfLocalArea() + Adjustment;
+}
+
 int
 X86FrameLowering::getFrameIndexReferencePreferSP(const MachineFunction &MF,
                                                  int FI, unsigned &FrameReg,
@@ -1798,9 +1820,6 @@
   assert(MF.getInfo<X86MachineFunctionInfo>()->getTCReturnAddrDelta() >= 0 &&
          "we don't handle this case!");
 
-  // Fill in FrameReg output argument.
-  FrameReg = TRI->getStackRegister();
-
   // This is how the math works out:
   //
   //  %rsp grows (i.e. gets lower) left to right. Each box below is
@@ -1825,12 +1844,8 @@
   // (C - E) == (C - A) - (B - A) + (B - E)
   //            { Using [1], [2] and [3] above }
   //         == getObjectOffset - LocalAreaOffset + StackSize
-  //
-
-  // Get the Offset from the StackPointer
-  int Offset = MFI.getObjectOffset(FI) - getOffsetOfLocalArea();
-
-  return Offset + StackSize;
+
+  return getFrameIndexReferenceSP(MF, FI, FrameReg, StackSize);
 }
 
 bool X86FrameLowering::assignCalleeSavedSpillSlots(
@@ -1882,14 +1897,15 @@
       continue;
 
     const TargetRegisterClass *RC = TRI->getMinimalPhysRegClass(Reg);
+    unsigned Size = TRI->getSpillSize(*RC);
+    unsigned Align = TRI->getSpillAlignment(*RC);
     // ensure alignment
-    SpillSlotOffset -= std::abs(SpillSlotOffset) % RC->getAlignment();
+    SpillSlotOffset -= std::abs(SpillSlotOffset) % Align;
     // spill into slot
-    SpillSlotOffset -= RC->getSize();
-    int SlotIndex =
-        MFI.CreateFixedSpillStackObject(RC->getSize(), SpillSlotOffset);
+    SpillSlotOffset -= Size;
+    int SlotIndex = MFI.CreateFixedSpillStackObject(Size, SpillSlotOffset);
     CSI[i - 1].setFrameIdx(SlotIndex);
-    MFI.ensureMaxAlignment(RC->getAlignment());
+    MFI.ensureMaxAlignment(Align);
   }
 
   return true;
@@ -1961,9 +1977,39 @@
   return true;
 }
 
+void X86FrameLowering::emitCatchRetReturnValue(MachineBasicBlock &MBB,
+                                               MachineBasicBlock::iterator MBBI,
+                                               MachineInstr *CatchRet) const {
+  // SEH shouldn't use catchret.
+  assert(!isAsynchronousEHPersonality(classifyEHPersonality(
+             MBB.getParent()->getFunction()->getPersonalityFn())) &&
+         "SEH should not use CATCHRET");
+  DebugLoc DL = CatchRet->getDebugLoc();
+  MachineBasicBlock *CatchRetTarget = CatchRet->getOperand(0).getMBB();
+
+  // Fill EAX/RAX with the address of the target block.
+  if (STI.is64Bit()) {
+    // LEA64r CatchRetTarget(%rip), %rax
+    BuildMI(MBB, MBBI, DL, TII.get(X86::LEA64r), X86::RAX)
+        .addReg(X86::RIP)
+        .addImm(0)
+        .addReg(0)
+        .addMBB(CatchRetTarget)
+        .addReg(0);
+  } else {
+    // MOV32ri $CatchRetTarget, %eax
+    BuildMI(MBB, MBBI, DL, TII.get(X86::MOV32ri), X86::EAX)
+        .addMBB(CatchRetTarget);
+  }
+
+  // Record that we've taken the address of CatchRetTarget and no longer just
+  // reference it in a terminator.
+  CatchRetTarget->setHasAddressTaken();
+}
+
 bool X86FrameLowering::restoreCalleeSavedRegisters(MachineBasicBlock &MBB,
                                                MachineBasicBlock::iterator MI,
-                                        const std::vector<CalleeSavedInfo> &CSI,
+                                          std::vector<CalleeSavedInfo> &CSI,
                                           const TargetRegisterInfo *TRI) const {
   if (CSI.empty())
     return false;
@@ -2040,7 +2086,7 @@
     SavedRegs.set(TRI->getBaseRegister());
 
     // Allocate a spill slot for EBP if we have a base pointer and EH funclets.
-    if (MF.getMMI().hasEHFunclets()) {
+    if (MF.hasEHFunclets()) {
       int FI = MFI.CreateSpillStackObject(SlotSize, SlotSize);
       X86FI->setHasSEHFramePtrSave(true);
       X86FI->setSEHFramePtrSaveIndex(FI);
@@ -2582,9 +2628,10 @@
   unsigned Opcode = I->getOpcode();
   bool isDestroy = Opcode == TII.getCallFrameDestroyOpcode();
   DebugLoc DL = I->getDebugLoc();
-  uint64_t Amount = !reserveCallFrame ? I->getOperand(0).getImm() : 0;
-  uint64_t InternalAmt = (isDestroy || Amount) ? I->getOperand(1).getImm() : 0;
+  uint64_t Amount = !reserveCallFrame ? TII.getFrameSize(*I) : 0;
+  uint64_t InternalAmt = (isDestroy || Amount) ? TII.getFrameAdjustment(*I) : 0;
   I = MBB.erase(I);
+  auto InsertPos = skipDebugInstructionsForward(I, MBB.end());
 
   if (!reserveCallFrame) {
     // If the stack pointer can be changed after prologue, turn the
@@ -2610,12 +2657,11 @@
     // GNU_ARGS_SIZE.
     // TODO: We don't need to reset this between subsequent functions,
     // if it didn't change.
-    bool HasDwarfEHHandlers = !WindowsCFI &&
-                              !MF.getMMI().getLandingPads().empty();
+    bool HasDwarfEHHandlers = !WindowsCFI && !MF.getLandingPads().empty();
 
     if (HasDwarfEHHandlers && !isDestroy &&
         MF.getInfo<X86MachineFunctionInfo>()->getHasPushSequences())
-      BuildCFI(MBB, I, DL,
+      BuildCFI(MBB, InsertPos, DL,
                MCCFIInstruction::createGnuArgsSize(nullptr, Amount));
 
     if (Amount == 0)
@@ -2629,7 +2675,7 @@
     // If this is a callee-pop calling convention, emit a CFA adjust for
     // the amount the callee popped.
     if (isDestroy && InternalAmt && DwarfCFI && !hasFP(MF))
-      BuildCFI(MBB, I, DL, 
+      BuildCFI(MBB, InsertPos, DL,
                MCCFIInstruction::createAdjustCfaOffset(nullptr, -InternalAmt));
 
     // Add Amount to SP to destroy a frame, or subtract to setup.
@@ -2640,13 +2686,13 @@
       // Merge with any previous or following adjustment instruction. Note: the
       // instructions merged with here do not have CFI, so their stack
       // adjustments do not feed into CfaAdjustment.
-      StackAdjustment += mergeSPUpdates(MBB, I, true);
-      StackAdjustment += mergeSPUpdates(MBB, I, false);
+      StackAdjustment += mergeSPUpdates(MBB, InsertPos, true);
+      StackAdjustment += mergeSPUpdates(MBB, InsertPos, false);
 
       if (StackAdjustment) {
         if (!(Fn->optForMinSize() &&
-              adjustStackWithPops(MBB, I, DL, StackAdjustment)))
-          BuildStackAdjustment(MBB, I, DL, StackAdjustment,
+              adjustStackWithPops(MBB, InsertPos, DL, StackAdjustment)))
+          BuildStackAdjustment(MBB, InsertPos, DL, StackAdjustment,
                                /*InEpilogue=*/false);
       }
     }
@@ -2662,8 +2708,9 @@
       // TODO: When not using precise CFA, we also need to adjust for the
       // InternalAmt here.
       if (CfaAdjustment) {
-        BuildCFI(MBB, I, DL, MCCFIInstruction::createAdjustCfaOffset(
-                                 nullptr, CfaAdjustment));
+        BuildCFI(MBB, InsertPos, DL,
+                 MCCFIInstruction::createAdjustCfaOffset(nullptr,
+                                                         CfaAdjustment));
       }
     }
 
@@ -2946,10 +2993,14 @@
 
 void X86FrameLowering::processFunctionBeforeFrameFinalized(
     MachineFunction &MF, RegScavenger *RS) const {
+  // Mark the function as not having WinCFI. We will set it back to true in
+  // emitPrologue if it gets called and emits CFI.
+  MF.setHasWinCFI(false);
+
   // If this function isn't doing Win64-style C++ EH, we don't need to do
   // anything.
   const Function *Fn = MF.getFunction();
-  if (!STI.is64Bit() || !MF.getMMI().hasEHFunclets() ||
+  if (!STI.is64Bit() || !MF.hasEHFunclets() ||
       classifyEHPersonality(Fn->getPersonalityFn()) != EHPersonality::MSVC_CXX)
     return;