CGBuiltin.cpp

Go to the documentation of this file.

//===---- CGBuiltin.cpp - Emit LLVM Code for builtins ---------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This contains code to emit Builtin calls as LLVM code.
//
//===----------------------------------------------------------------------===//
 
#include "ABIInfo.h"
#include "CGCUDARuntime.h"
#include "CGCXXABI.h"
#include "CGObjCRuntime.h"
#include "CGOpenCLRuntime.h"
#include "CGRecordLayout.h"
#include "CodeGenFunction.h"
#include "CodeGenModule.h"
#include "ConstantEmitter.h"
#include "PatternInit.h"
#include "TargetInfo.h"
#include "clang/AST/ASTContext.h"
#include "clang/AST/Attr.h"
#include "clang/AST/Decl.h"
#include "clang/AST/OSLog.h"
#include "clang/Basic/TargetBuiltins.h"
#include "clang/Basic/TargetInfo.h"
#include "clang/CodeGen/CGFunctionInfo.h"
#include "clang/Frontend/FrontendDiagnostic.h"
#include "llvm/ADT/APFloat.h"
#include "llvm/ADT/APInt.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/InlineAsm.h"
#include "llvm/IR/Intrinsics.h"
#include "llvm/IR/IntrinsicsAArch64.h"
#include "llvm/IR/IntrinsicsAMDGPU.h"
#include "llvm/IR/IntrinsicsARM.h"
#include "llvm/IR/IntrinsicsBPF.h"
#include "llvm/IR/IntrinsicsHexagon.h"
#include "llvm/IR/IntrinsicsLoongArch.h"
#include "llvm/IR/IntrinsicsNVPTX.h"
#include "llvm/IR/IntrinsicsPowerPC.h"
#include "llvm/IR/IntrinsicsR600.h"
#include "llvm/IR/IntrinsicsRISCV.h"
#include "llvm/IR/IntrinsicsS390.h"
#include "llvm/IR/IntrinsicsVE.h"
#include "llvm/IR/IntrinsicsWebAssembly.h"
#include "llvm/IR/IntrinsicsX86.h"
#include "llvm/IR/MDBuilder.h"
#include "llvm/IR/MatrixBuilder.h"
#include "llvm/Support/ConvertUTF.h"
#include "llvm/Support/ScopedPrinter.h"
#include "llvm/TargetParser/AArch64TargetParser.h"
#include "llvm/TargetParser/X86TargetParser.h"
#include <optional>
#include <sstream>
 
using namespace clang;
using namespace CodeGen;
using namespace llvm;
 
static void initializeAlloca(CodeGenFunction &CGF, AllocaInst *AI, Value *Size,
                             Align AlignmentInBytes) {
  ConstantInt *Byte;
  switch (CGF.getLangOpts().getTrivialAutoVarInit()) {
  case LangOptions::TrivialAutoVarInitKind::Uninitialized:
    // Nothing to initialize.
    return;
  case LangOptions::TrivialAutoVarInitKind::Zero:
    Byte = CGF.Builder.getInt8(0x00);
    break;
  case LangOptions::TrivialAutoVarInitKind::Pattern: {
    llvm::Type *Int8 = llvm::IntegerType::getInt8Ty(CGF.CGM.getLLVMContext());
    Byte = llvm::dyn_cast<llvm::ConstantInt>(
        initializationPatternFor(CGF.CGM, Int8));
    break;
  }
  }
  if (CGF.CGM.stopAutoInit())
    return;
  auto *I = CGF.Builder.CreateMemSet(AI, Byte, Size, AlignmentInBytes);
  I->addAnnotationMetadata("auto-init");
}
 
/// getBuiltinLibFunction - Given a builtin id for a function like
/// "__builtin_fabsf", return a Function* for "fabsf".
llvm::Constant *CodeGenModule::getBuiltinLibFunction(const FunctionDecl *FD,
                                                     unsigned BuiltinID) {
  assert(Context.BuiltinInfo.isLibFunction(BuiltinID));
 
  // Get the name, skip over the __builtin_ prefix (if necessary).
  StringRef Name;
  GlobalDecl D(FD);
 
  // TODO: This list should be expanded or refactored after all GCC-compatible
  // std libcall builtins are implemented.
  static SmallDenseMap<unsigned, StringRef, 8> F128Builtins{
      {Builtin::BI__builtin_printf, "__printfieee128"},
      {Builtin::BI__builtin_vsnprintf, "__vsnprintfieee128"},
      {Builtin::BI__builtin_vsprintf, "__vsprintfieee128"},
      {Builtin::BI__builtin_sprintf, "__sprintfieee128"},
      {Builtin::BI__builtin_snprintf, "__snprintfieee128"},
      {Builtin::BI__builtin_fprintf, "__fprintfieee128"},
      {Builtin::BI__builtin_nexttowardf128, "__nexttowardieee128"},
  };
 
  // The AIX library functions frexpl, ldexpl, and modfl are for 128-bit
  // IBM 'long double' (i.e. __ibm128). Map to the 'double' versions
  // if it is 64-bit 'long double' mode.
  static SmallDenseMap<unsigned, StringRef, 4> AIXLongDouble64Builtins{
      {Builtin::BI__builtin_frexpl, "frexp"},
      {Builtin::BI__builtin_ldexpl, "ldexp"},
      {Builtin::BI__builtin_modfl, "modf"},
  };
 
  // If the builtin has been declared explicitly with an assembler label,
  // use the mangled name. This differs from the plain label on platforms
  // that prefix labels.
  if (FD->hasAttr<AsmLabelAttr>())
    Name = getMangledName(D);
  else {
    // TODO: This mutation should also be applied to other targets other than
    // PPC, after backend supports IEEE 128-bit style libcalls.
    if (getTriple().isPPC64() &&
        &getTarget().getLongDoubleFormat() == &llvm::APFloat::IEEEquad() &&
        F128Builtins.find(BuiltinID) != F128Builtins.end())
      Name = F128Builtins[BuiltinID];
    else if (getTriple().isOSAIX() &&
             &getTarget().getLongDoubleFormat() ==
                 &llvm::APFloat::IEEEdouble() &&
             AIXLongDouble64Builtins.find(BuiltinID) !=
                 AIXLongDouble64Builtins.end())
      Name = AIXLongDouble64Builtins[BuiltinID];
    else
      Name = Context.BuiltinInfo.getName(BuiltinID).substr(10);
  }
 
  llvm::FunctionType *Ty =
    cast<llvm::FunctionType>(getTypes().ConvertType(FD->getType()));
 
  return GetOrCreateLLVMFunction(Name, Ty, D, /*ForVTable=*/false);
}
 
/// Emit the conversions required to turn the given value into an
/// integer of the given size.
static Value *EmitToInt(CodeGenFunction &CGF, llvm::Value *V,
                        QualType T, llvm::IntegerType *IntType) {
  V = CGF.EmitToMemory(V, T);
 
  if (V->getType()->isPointerTy())
    return CGF.Builder.CreatePtrToInt(V, IntType);
 
  assert(V->getType() == IntType);
  return V;
}
 
static Value *EmitFromInt(CodeGenFunction &CGF, llvm::Value *V,
                          QualType T, llvm::Type *ResultType) {
  V = CGF.EmitFromMemory(V, T);
 
  if (ResultType->isPointerTy())
    return CGF.Builder.CreateIntToPtr(V, ResultType);
 
  assert(V->getType() == ResultType);
  return V;
}
 
/// Utility to insert an atomic instruction based on Intrinsic::ID
/// and the expression node.
static Value *MakeBinaryAtomicValue(
    CodeGenFunction &CGF, llvm::AtomicRMWInst::BinOp Kind, const CallExpr *E,
    AtomicOrdering Ordering = AtomicOrdering::SequentiallyConsistent) {
 
  QualType T = E->getType();
  assert(E->getArg(0)->getType()->isPointerType());
  assert(CGF.getContext().hasSameUnqualifiedType(T,
                                  E->getArg(0)->getType()->getPointeeType()));
  assert(CGF.getContext().hasSameUnqualifiedType(T, E->getArg(1)->getType()));
 
  llvm::Value *DestPtr = CGF.EmitScalarExpr(E->getArg(0));
  unsigned AddrSpace = DestPtr->getType()->getPointerAddressSpace();
 
  llvm::IntegerType *IntType =
    llvm::IntegerType::get(CGF.getLLVMContext(),
                           CGF.getContext().getTypeSize(T));
  llvm::Type *IntPtrType = IntType->getPointerTo(AddrSpace);
 
  llvm::Value *Args[2];
  Args[0] = CGF.Builder.CreateBitCast(DestPtr, IntPtrType);
  Args[1] = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Type *ValueType = Args[1]->getType();
  Args[1] = EmitToInt(CGF, Args[1], T, IntType);
 
  llvm::Value *Result = CGF.Builder.CreateAtomicRMW(
      Kind, Args[0], Args[1], Ordering);
  return EmitFromInt(CGF, Result, T, ValueType);
}
 
static Value *EmitNontemporalStore(CodeGenFunction &CGF, const CallExpr *E) {
  Value *Val = CGF.EmitScalarExpr(E->getArg(0));
  Value *Address = CGF.EmitScalarExpr(E->getArg(1));
 
  // Convert the type of the pointer to a pointer to the stored type.
  Val = CGF.EmitToMemory(Val, E->getArg(0)->getType());
  unsigned SrcAddrSpace = Address->getType()->getPointerAddressSpace();
  Value *BC = CGF.Builder.CreateBitCast(
      Address, llvm::PointerType::get(Val->getType(), SrcAddrSpace), "cast");
  LValue LV = CGF.MakeNaturalAlignAddrLValue(BC, E->getArg(0)->getType());
  LV.setNontemporal(true);
  CGF.EmitStoreOfScalar(Val, LV, false);
  return nullptr;
}
 
static Value *EmitNontemporalLoad(CodeGenFunction &CGF, const CallExpr *E) {
  Value *Address = CGF.EmitScalarExpr(E->getArg(0));
 
  LValue LV = CGF.MakeNaturalAlignAddrLValue(Address, E->getType());
  LV.setNontemporal(true);
  return CGF.EmitLoadOfScalar(LV, E->getExprLoc());
}
 
static void CheckAtomicAlignment(CodeGenFunction &CGF, const CallExpr *E) {
  ASTContext &Ctx = CGF.getContext();
  Address Ptr = CGF.EmitPointerWithAlignment(E->getArg(0));
  unsigned Bytes = Ptr.getElementType()->isPointerTy()
                       ? Ctx.getTypeSizeInChars(Ctx.VoidPtrTy).getQuantity()
                       : Ptr.getElementType()->getScalarSizeInBits() / 8;
  unsigned Align = Ptr.getAlignment().getQuantity();
  if (Align % Bytes != 0) {
    DiagnosticsEngine &Diags = CGF.CGM.getDiags();
    Diags.Report(E->getBeginLoc(), diag::warn_sync_op_misaligned);
  }
}
 
static RValue EmitBinaryAtomic(CodeGenFunction &CGF,
                               llvm::AtomicRMWInst::BinOp Kind,
                               const CallExpr *E) {
  CheckAtomicAlignment(CGF, E);
  return RValue::get(MakeBinaryAtomicValue(CGF, Kind, E));
}
 
/// Utility to insert an atomic instruction based Intrinsic::ID and
/// the expression node, where the return value is the result of the
/// operation.
static RValue EmitBinaryAtomicPost(CodeGenFunction &CGF,
                                   llvm::AtomicRMWInst::BinOp Kind,
                                   const CallExpr *E,
                                   Instruction::BinaryOps Op,
                                   bool Invert = false) {
  CheckAtomicAlignment(CGF, E);
  QualType T = E->getType();
  assert(E->getArg(0)->getType()->isPointerType());
  assert(CGF.getContext().hasSameUnqualifiedType(T,
                                  E->getArg(0)->getType()->getPointeeType()));
  assert(CGF.getContext().hasSameUnqualifiedType(T, E->getArg(1)->getType()));
 
  llvm::Value *DestPtr = CGF.EmitScalarExpr(E->getArg(0));
  unsigned AddrSpace = DestPtr->getType()->getPointerAddressSpace();
 
  llvm::IntegerType *IntType =
    llvm::IntegerType::get(CGF.getLLVMContext(),
                           CGF.getContext().getTypeSize(T));
  llvm::Type *IntPtrType = IntType->getPointerTo(AddrSpace);
 
  llvm::Value *Args[2];
  Args[1] = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Type *ValueType = Args[1]->getType();
  Args[1] = EmitToInt(CGF, Args[1], T, IntType);
  Args[0] = CGF.Builder.CreateBitCast(DestPtr, IntPtrType);
 
  llvm::Value *Result = CGF.Builder.CreateAtomicRMW(
      Kind, Args[0], Args[1], llvm::AtomicOrdering::SequentiallyConsistent);
  Result = CGF.Builder.CreateBinOp(Op, Result, Args[1]);
  if (Invert)
    Result =
        CGF.Builder.CreateBinOp(llvm::Instruction::Xor, Result,
                                llvm::ConstantInt::getAllOnesValue(IntType));
  Result = EmitFromInt(CGF, Result, T, ValueType);
  return RValue::get(Result);
}
 
/// Utility to insert an atomic cmpxchg instruction.
///
/// @param CGF The current codegen function.
/// @param E   Builtin call expression to convert to cmpxchg.
///            arg0 - address to operate on
///            arg1 - value to compare with
///            arg2 - new value
/// @param ReturnBool Specifies whether to return success flag of
///                   cmpxchg result or the old value.
///
/// @returns result of cmpxchg, according to ReturnBool
///
/// Note: In order to lower Microsoft's _InterlockedCompareExchange* intrinsics
/// invoke the function EmitAtomicCmpXchgForMSIntrin.
static Value *MakeAtomicCmpXchgValue(CodeGenFunction &CGF, const CallExpr *E,
                                     bool ReturnBool) {
  CheckAtomicAlignment(CGF, E);
  QualType T = ReturnBool ? E->getArg(1)->getType() : E->getType();
  llvm::Value *DestPtr = CGF.EmitScalarExpr(E->getArg(0));
  unsigned AddrSpace = DestPtr->getType()->getPointerAddressSpace();
 
  llvm::IntegerType *IntType = llvm::IntegerType::get(
      CGF.getLLVMContext(), CGF.getContext().getTypeSize(T));
  llvm::Type *IntPtrType = IntType->getPointerTo(AddrSpace);
 
  Value *Args[3];
  Args[0] = CGF.Builder.CreateBitCast(DestPtr, IntPtrType);
  Args[1] = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Type *ValueType = Args[1]->getType();
  Args[1] = EmitToInt(CGF, Args[1], T, IntType);
  Args[2] = EmitToInt(CGF, CGF.EmitScalarExpr(E->getArg(2)), T, IntType);
 
  Value *Pair = CGF.Builder.CreateAtomicCmpXchg(
      Args[0], Args[1], Args[2], llvm::AtomicOrdering::SequentiallyConsistent,
      llvm::AtomicOrdering::SequentiallyConsistent);
  if (ReturnBool)
    // Extract boolean success flag and zext it to int.
    return CGF.Builder.CreateZExt(CGF.Builder.CreateExtractValue(Pair, 1),
                                  CGF.ConvertType(E->getType()));
  else
    // Extract old value and emit it using the same type as compare value.
    return EmitFromInt(CGF, CGF.Builder.CreateExtractValue(Pair, 0), T,
                       ValueType);
}
 
/// This function should be invoked to emit atomic cmpxchg for Microsoft's
/// _InterlockedCompareExchange* intrinsics which have the following signature:
/// T _InterlockedCompareExchange(T volatile *Destination,
///                               T Exchange,
///                               T Comparand);
///
/// Whereas the llvm 'cmpxchg' instruction has the following syntax:
/// cmpxchg *Destination, Comparand, Exchange.
/// So we need to swap Comparand and Exchange when invoking
/// CreateAtomicCmpXchg. That is the reason we could not use the above utility
/// function MakeAtomicCmpXchgValue since it expects the arguments to be
/// already swapped.
 
static
Value *EmitAtomicCmpXchgForMSIntrin(CodeGenFunction &CGF, const CallExpr *E,
    AtomicOrdering SuccessOrdering = AtomicOrdering::SequentiallyConsistent) {
  assert(E->getArg(0)->getType()->isPointerType());
  assert(CGF.getContext().hasSameUnqualifiedType(
      E->getType(), E->getArg(0)->getType()->getPointeeType()));
  assert(CGF.getContext().hasSameUnqualifiedType(E->getType(),
                                                 E->getArg(1)->getType()));
  assert(CGF.getContext().hasSameUnqualifiedType(E->getType(),
                                                 E->getArg(2)->getType()));
 
  auto *Destination = CGF.EmitScalarExpr(E->getArg(0));
  auto *Comparand = CGF.EmitScalarExpr(E->getArg(2));
  auto *Exchange = CGF.EmitScalarExpr(E->getArg(1));
 
  // For Release ordering, the failure ordering should be Monotonic.
  auto FailureOrdering = SuccessOrdering == AtomicOrdering::Release ?
                         AtomicOrdering::Monotonic :
                         SuccessOrdering;
 
  // The atomic instruction is marked volatile for consistency with MSVC. This
  // blocks the few atomics optimizations that LLVM has. If we want to optimize
  // _Interlocked* operations in the future, we will have to remove the volatile
  // marker.
  auto *Result = CGF.Builder.CreateAtomicCmpXchg(
                   Destination, Comparand, Exchange,
                   SuccessOrdering, FailureOrdering);
  Result->setVolatile(true);
  return CGF.Builder.CreateExtractValue(Result, 0);
}
 
// 64-bit Microsoft platforms support 128 bit cmpxchg operations. They are
// prototyped like this:
//
// unsigned char _InterlockedCompareExchange128...(
//     __int64 volatile * _Destination,
//     __int64 _ExchangeHigh,
//     __int64 _ExchangeLow,
//     __int64 * _ComparandResult);
static Value *EmitAtomicCmpXchg128ForMSIntrin(CodeGenFunction &CGF,
                                              const CallExpr *E,
                                              AtomicOrdering SuccessOrdering) {
  assert(E->getNumArgs() == 4);
  llvm::Value *Destination = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *ExchangeHigh = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Value *ExchangeLow = CGF.EmitScalarExpr(E->getArg(2));
  llvm::Value *ComparandPtr = CGF.EmitScalarExpr(E->getArg(3));
 
  assert(Destination->getType()->isPointerTy());
  assert(!ExchangeHigh->getType()->isPointerTy());
  assert(!ExchangeLow->getType()->isPointerTy());
  assert(ComparandPtr->getType()->isPointerTy());
 
  // For Release ordering, the failure ordering should be Monotonic.
  auto FailureOrdering = SuccessOrdering == AtomicOrdering::Release
                             ? AtomicOrdering::Monotonic
                             : SuccessOrdering;
 
  // Convert to i128 pointers and values.
  llvm::Type *Int128Ty = llvm::IntegerType::get(CGF.getLLVMContext(), 128);
  llvm::Type *Int128PtrTy = Int128Ty->getPointerTo();
  Destination = CGF.Builder.CreateBitCast(Destination, Int128PtrTy);
  Address ComparandResult(CGF.Builder.CreateBitCast(ComparandPtr, Int128PtrTy),
                          Int128Ty, CGF.getContext().toCharUnitsFromBits(128));
 
  // (((i128)hi) << 64) | ((i128)lo)
  ExchangeHigh = CGF.Builder.CreateZExt(ExchangeHigh, Int128Ty);
  ExchangeLow = CGF.Builder.CreateZExt(ExchangeLow, Int128Ty);
  ExchangeHigh =
      CGF.Builder.CreateShl(ExchangeHigh, llvm::ConstantInt::get(Int128Ty, 64));
  llvm::Value *Exchange = CGF.Builder.CreateOr(ExchangeHigh, ExchangeLow);
 
  // Load the comparand for the instruction.
  llvm::Value *Comparand = CGF.Builder.CreateLoad(ComparandResult);
 
  auto *CXI = CGF.Builder.CreateAtomicCmpXchg(Destination, Comparand, Exchange,
                                              SuccessOrdering, FailureOrdering);
 
  // The atomic instruction is marked volatile for consistency with MSVC. This
  // blocks the few atomics optimizations that LLVM has. If we want to optimize
  // _Interlocked* operations in the future, we will have to remove the volatile
  // marker.
  CXI->setVolatile(true);
 
  // Store the result as an outparameter.
  CGF.Builder.CreateStore(CGF.Builder.CreateExtractValue(CXI, 0),
                          ComparandResult);
 
  // Get the success boolean and zero extend it to i8.
  Value *Success = CGF.Builder.CreateExtractValue(CXI, 1);
  return CGF.Builder.CreateZExt(Success, CGF.Int8Ty);
}
 
static Value *EmitAtomicIncrementValue(CodeGenFunction &CGF, const CallExpr *E,
    AtomicOrdering Ordering = AtomicOrdering::SequentiallyConsistent) {
  assert(E->getArg(0)->getType()->isPointerType());
 
  auto *IntTy = CGF.ConvertType(E->getType());
  auto *Result = CGF.Builder.CreateAtomicRMW(
                   AtomicRMWInst::Add,
                   CGF.EmitScalarExpr(E->getArg(0)),
                   ConstantInt::get(IntTy, 1),
                   Ordering);
  return CGF.Builder.CreateAdd(Result, ConstantInt::get(IntTy, 1));
}
 
static Value *EmitAtomicDecrementValue(CodeGenFunction &CGF, const CallExpr *E,
    AtomicOrdering Ordering = AtomicOrdering::SequentiallyConsistent) {
  assert(E->getArg(0)->getType()->isPointerType());
 
  auto *IntTy = CGF.ConvertType(E->getType());
  auto *Result = CGF.Builder.CreateAtomicRMW(
                   AtomicRMWInst::Sub,
                   CGF.EmitScalarExpr(E->getArg(0)),
                   ConstantInt::get(IntTy, 1),
                   Ordering);
  return CGF.Builder.CreateSub(Result, ConstantInt::get(IntTy, 1));
}
 
// Build a plain volatile load.
static Value *EmitISOVolatileLoad(CodeGenFunction &CGF, const CallExpr *E) {
  Value *Ptr = CGF.EmitScalarExpr(E->getArg(0));
  QualType ElTy = E->getArg(0)->getType()->getPointeeType();
  CharUnits LoadSize = CGF.getContext().getTypeSizeInChars(ElTy);
  llvm::Type *ITy =
      llvm::IntegerType::get(CGF.getLLVMContext(), LoadSize.getQuantity() * 8);
  Ptr = CGF.Builder.CreateBitCast(Ptr, ITy->getPointerTo());
  llvm::LoadInst *Load = CGF.Builder.CreateAlignedLoad(ITy, Ptr, LoadSize);
  Load->setVolatile(true);
  return Load;
}
 
// Build a plain volatile store.
static Value *EmitISOVolatileStore(CodeGenFunction &CGF, const CallExpr *E) {
  Value *Ptr = CGF.EmitScalarExpr(E->getArg(0));
  Value *Value = CGF.EmitScalarExpr(E->getArg(1));
  QualType ElTy = E->getArg(0)->getType()->getPointeeType();
  CharUnits StoreSize = CGF.getContext().getTypeSizeInChars(ElTy);
  llvm::Type *ITy =
      llvm::IntegerType::get(CGF.getLLVMContext(), StoreSize.getQuantity() * 8);
  Ptr = CGF.Builder.CreateBitCast(Ptr, ITy->getPointerTo());
  llvm::StoreInst *Store =
      CGF.Builder.CreateAlignedStore(Value, Ptr, StoreSize);
  Store->setVolatile(true);
  return Store;
}
 
// Emit a simple mangled intrinsic that has 1 argument and a return type
// matching the argument type. Depending on mode, this may be a constrained
// floating-point intrinsic.
static Value *emitUnaryMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
                                const CallExpr *E, unsigned IntrinsicID,
                                unsigned ConstrainedIntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
 
  if (CGF.Builder.getIsFPConstrained()) {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
    Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Src0->getType());
    return CGF.Builder.CreateConstrainedFPCall(F, { Src0 });
  } else {
    Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
    return CGF.Builder.CreateCall(F, Src0);
  }
}
 
// Emit an intrinsic that has 2 operands of the same type as its result.
// Depending on mode, this may be a constrained floating-point intrinsic.
static Value *emitBinaryMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
                                const CallExpr *E, unsigned IntrinsicID,
                                unsigned ConstrainedIntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
 
  if (CGF.Builder.getIsFPConstrained()) {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
    Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Src0->getType());
    return CGF.Builder.CreateConstrainedFPCall(F, { Src0, Src1 });
  } else {
    Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
    return CGF.Builder.CreateCall(F, { Src0, Src1 });
  }
}
 
// Emit an intrinsic that has 3 operands of the same type as its result.
// Depending on mode, this may be a constrained floating-point intrinsic.
static Value *emitTernaryMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
                                 const CallExpr *E, unsigned IntrinsicID,
                                 unsigned ConstrainedIntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Value *Src2 = CGF.EmitScalarExpr(E->getArg(2));
 
  if (CGF.Builder.getIsFPConstrained()) {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
    Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Src0->getType());
    return CGF.Builder.CreateConstrainedFPCall(F, { Src0, Src1, Src2 });
  } else {
    Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
    return CGF.Builder.CreateCall(F, { Src0, Src1, Src2 });
  }
}
 
// Emit an intrinsic where all operands are of the same type as the result.
// Depending on mode, this may be a constrained floating-point intrinsic.
static Value *emitCallMaybeConstrainedFPBuiltin(CodeGenFunction &CGF,
                                                unsigned IntrinsicID,
                                                unsigned ConstrainedIntrinsicID,
                                                llvm::Type *Ty,
                                                ArrayRef<Value *> Args) {
  Function *F;
  if (CGF.Builder.getIsFPConstrained())
    F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID, Ty);
  else
    F = CGF.CGM.getIntrinsic(IntrinsicID, Ty);
 
  if (CGF.Builder.getIsFPConstrained())
    return CGF.Builder.CreateConstrainedFPCall(F, Args);
  else
    return CGF.Builder.CreateCall(F, Args);
}
 
// Emit a simple mangled intrinsic that has 1 argument and a return type
// matching the argument type.
static Value *emitUnaryBuiltin(CodeGenFunction &CGF, const CallExpr *E,
                               unsigned IntrinsicID,
                               llvm::StringRef Name = "") {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
 
  Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
  return CGF.Builder.CreateCall(F, Src0, Name);
}
 
// Emit an intrinsic that has 2 operands of the same type as its result.
static Value *emitBinaryBuiltin(CodeGenFunction &CGF,
                                const CallExpr *E,
                                unsigned IntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
 
  Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
  return CGF.Builder.CreateCall(F, { Src0, Src1 });
}
 
// Emit an intrinsic that has 3 operands of the same type as its result.
static Value *emitTernaryBuiltin(CodeGenFunction &CGF,
                                 const CallExpr *E,
                                 unsigned IntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
  llvm::Value *Src2 = CGF.EmitScalarExpr(E->getArg(2));
 
  Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
  return CGF.Builder.CreateCall(F, { Src0, Src1, Src2 });
}
 
// Emit an intrinsic that has 1 float or double operand, and 1 integer.
static Value *emitFPIntBuiltin(CodeGenFunction &CGF,
                               const CallExpr *E,
                               unsigned IntrinsicID) {
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
  llvm::Value *Src1 = CGF.EmitScalarExpr(E->getArg(1));
 
  Function *F = CGF.CGM.getIntrinsic(IntrinsicID, Src0->getType());
  return CGF.Builder.CreateCall(F, {Src0, Src1});
}
 
// Emit an intrinsic that has overloaded integer result and fp operand.
static Value *
emitMaybeConstrainedFPToIntRoundBuiltin(CodeGenFunction &CGF, const CallExpr *E,
                                        unsigned IntrinsicID,
                                        unsigned ConstrainedIntrinsicID) {
  llvm::Type *ResultType = CGF.ConvertType(E->getType());
  llvm::Value *Src0 = CGF.EmitScalarExpr(E->getArg(0));
 
  if (CGF.Builder.getIsFPConstrained()) {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(CGF, E);
    Function *F = CGF.CGM.getIntrinsic(ConstrainedIntrinsicID,
                                       {ResultType, Src0->getType()});
    return CGF.Builder.CreateConstrainedFPCall(F, {Src0});
  } else {
    Function *F =
        CGF.CGM.getIntrinsic(IntrinsicID, {ResultType, Src0->getType()});
    return CGF.Builder.CreateCall(F, Src0);
  }
}
 
/// EmitFAbs - Emit a call to @llvm.fabs().
static Value *EmitFAbs(CodeGenFunction &CGF, Value *V) {
  Function *F = CGF.CGM.getIntrinsic(Intrinsic::fabs, V->getType());
  llvm::CallInst *Call = CGF.Builder.CreateCall(F, V);
  Call->setDoesNotAccessMemory();
  return Call;
}
 
/// Emit the computation of the sign bit for a floating point value. Returns
/// the i1 sign bit value.
static Value *EmitSignBit(CodeGenFunction &CGF, Value *V) {
  LLVMContext &C = CGF.CGM.getLLVMContext();
 
  llvm::Type *Ty = V->getType();
  int Width = Ty->getPrimitiveSizeInBits();
  llvm::Type *IntTy = llvm::IntegerType::get(C, Width);
  V = CGF.Builder.CreateBitCast(V, IntTy);
  if (Ty->isPPC_FP128Ty()) {
    // We want the sign bit of the higher-order double. The bitcast we just
    // did works as if the double-double was stored to memory and then
    // read as an i128. The "store" will put the higher-order double in the
    // lower address in both little- and big-Endian modes, but the "load"
    // will treat those bits as a different part of the i128: the low bits in
    // little-Endian, the high bits in big-Endian. Therefore, on big-Endian
    // we need to shift the high bits down to the low before truncating.
    Width >>= 1;
    if (CGF.getTarget().isBigEndian()) {
      Value *ShiftCst = llvm::ConstantInt::get(IntTy, Width);
      V = CGF.Builder.CreateLShr(V, ShiftCst);
    }
    // We are truncating value in order to extract the higher-order
    // double, which we will be using to extract the sign from.
    IntTy = llvm::IntegerType::get(C, Width);
    V = CGF.Builder.CreateTrunc(V, IntTy);
  }
  Value *Zero = llvm::Constant::getNullValue(IntTy);
  return CGF.Builder.CreateICmpSLT(V, Zero);
}
 
static RValue emitLibraryCall(CodeGenFunction &CGF, const FunctionDecl *FD,
                              const CallExpr *E, llvm::Constant *calleeValue) {
  CGCallee callee = CGCallee::forDirect(calleeValue, GlobalDecl(FD));
  return CGF.EmitCall(E->getCallee()->getType(), callee, E, ReturnValueSlot());
}
 
/// Emit a call to llvm.{sadd,uadd,ssub,usub,smul,umul}.with.overflow.*
/// depending on IntrinsicID.
///
/// \arg CGF The current codegen function.
/// \arg IntrinsicID The ID for the Intrinsic we wish to generate.
/// \arg X The first argument to the llvm.*.with.overflow.*.
/// \arg Y The second argument to the llvm.*.with.overflow.*.
/// \arg Carry The carry returned by the llvm.*.with.overflow.*.
/// \returns The result (i.e. sum/product) returned by the intrinsic.
static llvm::Value *EmitOverflowIntrinsic(CodeGenFunction &CGF,
                                          const llvm::Intrinsic::ID IntrinsicID,
                                          llvm::Value *X, llvm::Value *Y,
                                          llvm::Value *&Carry) {
  // Make sure we have integers of the same width.
  assert(X->getType() == Y->getType() &&
         "Arguments must be the same type. (Did you forget to make sure both "
         "arguments have the same integer width?)");
 
  Function *Callee = CGF.CGM.getIntrinsic(IntrinsicID, X->getType());
  llvm::Value *Tmp = CGF.Builder.CreateCall(Callee, {X, Y});
  Carry = CGF.Builder.CreateExtractValue(Tmp, 1);
  return CGF.Builder.CreateExtractValue(Tmp, 0);
}
 
static Value *emitRangedBuiltin(CodeGenFunction &CGF,
                                unsigned IntrinsicID,
                                int low, int high) {
    llvm::MDBuilder MDHelper(CGF.getLLVMContext());
    llvm::MDNode *RNode = MDHelper.createRange(APInt(32, low), APInt(32, high));
    Function *F = CGF.CGM.getIntrinsic(IntrinsicID, {});
    llvm::Instruction *Call = CGF.Builder.CreateCall(F);
    Call->setMetadata(llvm::LLVMContext::MD_range, RNode);
    Call->setMetadata(llvm::LLVMContext::MD_noundef,
                      llvm::MDNode::get(CGF.getLLVMContext(), std::nullopt));
    return Call;
}
 
namespace {
  struct WidthAndSignedness {
    unsigned Width;
    bool Signed;
  };
}
 
static WidthAndSignedness
getIntegerWidthAndSignedness(const clang::ASTContext &context,
                             const clang::QualType Type) {
  assert(Type->isIntegerType() && "Given type is not an integer.");
  unsigned Width = Type->isBooleanType()  ? 1
                   : Type->isBitIntType() ? context.getIntWidth(Type)
                                          : context.getTypeInfo(Type).Width;
  bool Signed = Type->isSignedIntegerType();
  return {Width, Signed};
}
 
// Given one or more integer types, this function produces an integer type that
// encompasses them: any value in one of the given types could be expressed in
// the encompassing type.
static struct WidthAndSignedness
EncompassingIntegerType(ArrayRef<struct WidthAndSignedness> Types) {
  assert(Types.size() > 0 && "Empty list of types.");
 
  // If any of the given types is signed, we must return a signed type.
  bool Signed = false;
  for (const auto &Type : Types) {
    Signed |= Type.Signed;
  }
 
  // The encompassing type must have a width greater than or equal to the width
  // of the specified types.  Additionally, if the encompassing type is signed,
  // its width must be strictly greater than the width of any unsigned types
  // given.
  unsigned Width = 0;
  for (const auto &Type : Types) {
    unsigned MinWidth = Type.Width + (Signed && !Type.Signed);
    if (Width < MinWidth) {
      Width = MinWidth;
    }
  }
 
  return {Width, Signed};
}
 
Value *CodeGenFunction::EmitVAStartEnd(Value *ArgValue, bool IsStart) {
  llvm::Type *DestType = Int8PtrTy;
  if (ArgValue->getType() != DestType)
    ArgValue =
        Builder.CreateBitCast(ArgValue, DestType, ArgValue->getName().data());
 
  Intrinsic::ID inst = IsStart ? Intrinsic::vastart : Intrinsic::vaend;
  return Builder.CreateCall(CGM.getIntrinsic(inst), ArgValue);
}
 
/// Checks if using the result of __builtin_object_size(p, @p From) in place of
/// __builtin_object_size(p, @p To) is correct
static bool areBOSTypesCompatible(int From, int To) {
  // Note: Our __builtin_object_size implementation currently treats Type=0 and
  // Type=2 identically. Encoding this implementation detail here may make
  // improving __builtin_object_size difficult in the future, so it's omitted.
  return From == To || (From == 0 && To == 1) || (From == 3 && To == 2);
}
 
static llvm::Value *
getDefaultBuiltinObjectSizeResult(unsigned Type, llvm::IntegerType *ResType) {
  return ConstantInt::get(ResType, (Type & 2) ? 0 : -1, /*isSigned=*/true);
}
 
llvm::Value *
CodeGenFunction::evaluateOrEmitBuiltinObjectSize(const Expr *E, unsigned Type,
                                                 llvm::IntegerType *ResType,
                                                 llvm::Value *EmittedE,
                                                 bool IsDynamic) {
  uint64_t ObjectSize;
  if (!E->tryEvaluateObjectSize(ObjectSize, getContext(), Type))
    return emitBuiltinObjectSize(E, Type, ResType, EmittedE, IsDynamic);
  return ConstantInt::get(ResType, ObjectSize, /*isSigned=*/true);
}
 
/// Returns a Value corresponding to the size of the given expression.
/// This Value may be either of the following:
///   - A llvm::Argument (if E is a param with the pass_object_size attribute on
///     it)
///   - A call to the @llvm.objectsize intrinsic
///
/// EmittedE is the result of emitting `E` as a scalar expr. If it's non-null
/// and we wouldn't otherwise try to reference a pass_object_size parameter,
/// we'll call @llvm.objectsize on EmittedE, rather than emitting E.
llvm::Value *
CodeGenFunction::emitBuiltinObjectSize(const Expr *E, unsigned Type,
                                       llvm::IntegerType *ResType,
                                       llvm::Value *EmittedE, bool IsDynamic) {
  // We need to reference an argument if the pointer is a parameter with the
  // pass_object_size attribute.
  if (auto *D = dyn_cast<DeclRefExpr>(E->IgnoreParenImpCasts())) {
    auto *Param = dyn_cast<ParmVarDecl>(D->getDecl());
    auto *PS = D->getDecl()->getAttr<PassObjectSizeAttr>();
    if (Param != nullptr && PS != nullptr &&
        areBOSTypesCompatible(PS->getType(), Type)) {
      auto Iter = SizeArguments.find(Param);
      assert(Iter != SizeArguments.end());
 
      const ImplicitParamDecl *D = Iter->second;
      auto DIter = LocalDeclMap.find(D);
      assert(DIter != LocalDeclMap.end());
 
      return EmitLoadOfScalar(DIter->second, /*Volatile=*/false,
                              getContext().getSizeType(), E->getBeginLoc());
    }
  }
 
  // LLVM can't handle Type=3 appropriately, and __builtin_object_size shouldn't
  // evaluate E for side-effects. In either case, we shouldn't lower to
  // @llvm.objectsize.
  if (Type == 3 || (!EmittedE && E->HasSideEffects(getContext())))
    return getDefaultBuiltinObjectSizeResult(Type, ResType);
 
  Value *Ptr = EmittedE ? EmittedE : EmitScalarExpr(E);
  assert(Ptr->getType()->isPointerTy() &&
         "Non-pointer passed to __builtin_object_size?");
 
  Function *F =
      CGM.getIntrinsic(Intrinsic::objectsize, {ResType, Ptr->getType()});
 
  // LLVM only supports 0 and 2, make sure that we pass along that as a boolean.
  Value *Min = Builder.getInt1((Type & 2) != 0);
  // For GCC compatibility, __builtin_object_size treat NULL as unknown size.
  Value *NullIsUnknown = Builder.getTrue();
  Value *Dynamic = Builder.getInt1(IsDynamic);
  return Builder.CreateCall(F, {Ptr, Min, NullIsUnknown, Dynamic});
}
 
namespace {
/// A struct to generically describe a bit test intrinsic.
struct BitTest {
  enum ActionKind : uint8_t { TestOnly, Complement, Reset, Set };
  enum InterlockingKind : uint8_t {
    Unlocked,
    Sequential,
    Acquire,
    Release,
    NoFence
  };
 
  ActionKind Action;
  InterlockingKind Interlocking;
  bool Is64Bit;
 
  static BitTest decodeBitTestBuiltin(unsigned BuiltinID);
};
} // namespace
 
BitTest BitTest::decodeBitTestBuiltin(unsigned BuiltinID) {
  switch (BuiltinID) {
    // Main portable variants.
  case Builtin::BI_bittest:
    return {TestOnly, Unlocked, false};
  case Builtin::BI_bittestandcomplement:
    return {Complement, Unlocked, false};
  case Builtin::BI_bittestandreset:
    return {Reset, Unlocked, false};
  case Builtin::BI_bittestandset:
    return {Set, Unlocked, false};
  case Builtin::BI_interlockedbittestandreset:
    return {Reset, Sequential, false};
  case Builtin::BI_interlockedbittestandset:
    return {Set, Sequential, false};
 
    // X86-specific 64-bit variants.
  case Builtin::BI_bittest64:
    return {TestOnly, Unlocked, true};
  case Builtin::BI_bittestandcomplement64:
    return {Complement, Unlocked, true};
  case Builtin::BI_bittestandreset64:
    return {Reset, Unlocked, true};
  case Builtin::BI_bittestandset64:
    return {Set, Unlocked, true};
  case Builtin::BI_interlockedbittestandreset64:
    return {Reset, Sequential, true};
  case Builtin::BI_interlockedbittestandset64:
    return {Set, Sequential, true};
 
    // ARM/AArch64-specific ordering variants.
  case Builtin::BI_interlockedbittestandset_acq:
    return {Set, Acquire, false};
  case Builtin::BI_interlockedbittestandset_rel:
    return {Set, Release, false};
  case Builtin::BI_interlockedbittestandset_nf:
    return {Set, NoFence, false};
  case Builtin::BI_interlockedbittestandreset_acq:
    return {Reset, Acquire, false};
  case Builtin::BI_interlockedbittestandreset_rel:
    return {Reset, Release, false};
  case Builtin::BI_interlockedbittestandreset_nf:
    return {Reset, NoFence, false};
  }
  llvm_unreachable("expected only bittest intrinsics");
}
 
static char bitActionToX86BTCode(BitTest::ActionKind A) {
  switch (A) {
  case BitTest::TestOnly:   return '\0';
  case BitTest::Complement: return 'c';
  case BitTest::Reset:      return 'r';
  case BitTest::Set:        return 's';
  }
  llvm_unreachable("invalid action");
}
 
static llvm::Value *EmitX86BitTestIntrinsic(CodeGenFunction &CGF,
                                            BitTest BT,
                                            const CallExpr *E, Value *BitBase,
                                            Value *BitPos) {
  char Action = bitActionToX86BTCode(BT.Action);
  char SizeSuffix = BT.Is64Bit ? 'q' : 'l';
 
  // Build the assembly.
  SmallString<64> Asm;
  raw_svector_ostream AsmOS(Asm);
  if (BT.Interlocking != BitTest::Unlocked)
    AsmOS << "lock ";
  AsmOS << "bt";
  if (Action)
    AsmOS << Action;
  AsmOS << SizeSuffix << " $2, ($1)";
 
  // Build the constraints. FIXME: We should support immediates when possible.
  std::string Constraints = "={@ccc},r,r,~{cc},~{memory}";
  std::string MachineClobbers = CGF.getTarget().getClobbers();
  if (!MachineClobbers.empty()) {
    Constraints += ',';
    Constraints += MachineClobbers;
  }
  llvm::IntegerType *IntType = llvm::IntegerType::get(
      CGF.getLLVMContext(),
      CGF.getContext().getTypeSize(E->getArg(1)->getType()));
  llvm::Type *IntPtrType = IntType->getPointerTo();
  llvm::FunctionType *FTy =
      llvm::FunctionType::get(CGF.Int8Ty, {IntPtrType, IntType}, false);
 
  llvm::InlineAsm *IA =
      llvm::InlineAsm::get(FTy, Asm, Constraints, /*hasSideEffects=*/true);
  return CGF.Builder.CreateCall(IA, {BitBase, BitPos});
}
 
static llvm::AtomicOrdering
getBitTestAtomicOrdering(BitTest::InterlockingKind I) {
  switch (I) {
  case BitTest::Unlocked:   return llvm::AtomicOrdering::NotAtomic;
  case BitTest::Sequential: return llvm::AtomicOrdering::SequentiallyConsistent;
  case BitTest::Acquire:    return llvm::AtomicOrdering::Acquire;
  case BitTest::Release:    return llvm::AtomicOrdering::Release;
  case BitTest::NoFence:    return llvm::AtomicOrdering::Monotonic;
  }
  llvm_unreachable("invalid interlocking");
}
 
/// Emit a _bittest* intrinsic. These intrinsics take a pointer to an array of
/// bits and a bit position and read and optionally modify the bit at that
/// position. The position index can be arbitrarily large, i.e. it can be larger
/// than 31 or 63, so we need an indexed load in the general case.
static llvm::Value *EmitBitTestIntrinsic(CodeGenFunction &CGF,
                                         unsigned BuiltinID,
                                         const CallExpr *E) {
  Value *BitBase = CGF.EmitScalarExpr(E->getArg(0));
  Value *BitPos = CGF.EmitScalarExpr(E->getArg(1));
 
  BitTest BT = BitTest::decodeBitTestBuiltin(BuiltinID);
 
  // X86 has special BT, BTC, BTR, and BTS instructions that handle the array
  // indexing operation internally. Use them if possible.
  if (CGF.getTarget().getTriple().isX86())
    return EmitX86BitTestIntrinsic(CGF, BT, E, BitBase, BitPos);
 
  // Otherwise, use generic code to load one byte and test the bit. Use all but
  // the bottom three bits as the array index, and the bottom three bits to form
  // a mask.
  // Bit = BitBaseI8[BitPos >> 3] & (1 << (BitPos & 0x7)) != 0;
  Value *ByteIndex = CGF.Builder.CreateAShr(
      BitPos, llvm::ConstantInt::get(BitPos->getType(), 3), "bittest.byteidx");
  Value *BitBaseI8 = CGF.Builder.CreatePointerCast(BitBase, CGF.Int8PtrTy);
  Address ByteAddr(CGF.Builder.CreateInBoundsGEP(CGF.Int8Ty, BitBaseI8,
                                                 ByteIndex, "bittest.byteaddr"),
                   CGF.Int8Ty, CharUnits::One());
  Value *PosLow =
      CGF.Builder.CreateAnd(CGF.Builder.CreateTrunc(BitPos, CGF.Int8Ty),
                            llvm::ConstantInt::get(CGF.Int8Ty, 0x7));
 
  // The updating instructions will need a mask.
  Value *Mask = nullptr;
  if (BT.Action != BitTest::TestOnly) {
    Mask = CGF.Builder.CreateShl(llvm::ConstantInt::get(CGF.Int8Ty, 1), PosLow,
                                 "bittest.mask");
  }
 
  // Check the action and ordering of the interlocked intrinsics.
  llvm::AtomicOrdering Ordering = getBitTestAtomicOrdering(BT.Interlocking);
 
  Value *OldByte = nullptr;
  if (Ordering != llvm::AtomicOrdering::NotAtomic) {
    // Emit a combined atomicrmw load/store operation for the interlocked
    // intrinsics.
    llvm::AtomicRMWInst::BinOp RMWOp = llvm::AtomicRMWInst::Or;
    if (BT.Action == BitTest::Reset) {
      Mask = CGF.Builder.CreateNot(Mask);
      RMWOp = llvm::AtomicRMWInst::And;
    }
    OldByte = CGF.Builder.CreateAtomicRMW(RMWOp, ByteAddr.getPointer(), Mask,
                                          Ordering);
  } else {
    // Emit a plain load for the non-interlocked intrinsics.
    OldByte = CGF.Builder.CreateLoad(ByteAddr, "bittest.byte");
    Value *NewByte = nullptr;
    switch (BT.Action) {
    case BitTest::TestOnly:
      // Don't store anything.
      break;
    case BitTest::Complement:
      NewByte = CGF.Builder.CreateXor(OldByte, Mask);
      break;
    case BitTest::Reset:
      NewByte = CGF.Builder.CreateAnd(OldByte, CGF.Builder.CreateNot(Mask));
      break;
    case BitTest::Set:
      NewByte = CGF.Builder.CreateOr(OldByte, Mask);
      break;
    }
    if (NewByte)
      CGF.Builder.CreateStore(NewByte, ByteAddr);
  }
 
  // However we loaded the old byte, either by plain load or atomicrmw, shift
  // the bit into the low position and mask it to 0 or 1.
  Value *ShiftedByte = CGF.Builder.CreateLShr(OldByte, PosLow, "bittest.shr");
  return CGF.Builder.CreateAnd(
      ShiftedByte, llvm::ConstantInt::get(CGF.Int8Ty, 1), "bittest.res");
}
 
static llvm::Value *emitPPCLoadReserveIntrinsic(CodeGenFunction &CGF,
                                                unsigned BuiltinID,
                                                const CallExpr *E) {
  Value *Addr = CGF.EmitScalarExpr(E->getArg(0));
 
  SmallString<64> Asm;
  raw_svector_ostream AsmOS(Asm);
  llvm::IntegerType *RetType = CGF.Int32Ty;
 
  switch (BuiltinID) {
  case clang::PPC::BI__builtin_ppc_ldarx:
    AsmOS << "ldarx ";
    RetType = CGF.Int64Ty;
    break;
  case clang::PPC::BI__builtin_ppc_lwarx:
    AsmOS << "lwarx ";
    RetType = CGF.Int32Ty;
    break;
  case clang::PPC::BI__builtin_ppc_lharx:
    AsmOS << "lharx ";
    RetType = CGF.Int16Ty;
    break;
  case clang::PPC::BI__builtin_ppc_lbarx:
    AsmOS << "lbarx ";
    RetType = CGF.Int8Ty;
    break;
  default:
    llvm_unreachable("Expected only PowerPC load reserve intrinsics");
  }
 
  AsmOS << "$0, ${1:y}";
 
  std::string Constraints = "=r,*Z,~{memory}";
  std::string MachineClobbers = CGF.getTarget().getClobbers();
  if (!MachineClobbers.empty()) {
    Constraints += ',';
    Constraints += MachineClobbers;
  }
 
  llvm::Type *IntPtrType = RetType->getPointerTo();
  llvm::FunctionType *FTy =
      llvm::FunctionType::get(RetType, {IntPtrType}, false);
 
  llvm::InlineAsm *IA =
      llvm::InlineAsm::get(FTy, Asm, Constraints, /*hasSideEffects=*/true);
  llvm::CallInst *CI = CGF.Builder.CreateCall(IA, {Addr});
  CI->addParamAttr(
      0, Attribute::get(CGF.getLLVMContext(), Attribute::ElementType, RetType));
  return CI;
}
 
namespace {
enum class MSVCSetJmpKind {
  _setjmpex,
  _setjmp3,
  _setjmp
};
}
 
/// MSVC handles setjmp a bit differently on different platforms. On every
/// architecture except 32-bit x86, the frame address is passed. On x86, extra
/// parameters can be passed as variadic arguments, but we always pass none.
static RValue EmitMSVCRTSetJmp(CodeGenFunction &CGF, MSVCSetJmpKind SJKind,
                               const CallExpr *E) {
  llvm::Value *Arg1 = nullptr;
  llvm::Type *Arg1Ty = nullptr;
  StringRef Name;
  bool IsVarArg = false;
  if (SJKind == MSVCSetJmpKind::_setjmp3) {
    Name = "_setjmp3";
    Arg1Ty = CGF.Int32Ty;
    Arg1 = llvm::ConstantInt::get(CGF.IntTy, 0);
    IsVarArg = true;
  } else {
    Name = SJKind == MSVCSetJmpKind::_setjmp ? "_setjmp" : "_setjmpex";
    Arg1Ty = CGF.Int8PtrTy;
    if (CGF.getTarget().getTriple().getArch() == llvm::Triple::aarch64) {
      Arg1 = CGF.Builder.CreateCall(
          CGF.CGM.getIntrinsic(Intrinsic::sponentry, CGF.AllocaInt8PtrTy));
    } else
      Arg1 = CGF.Builder.CreateCall(
          CGF.CGM.getIntrinsic(Intrinsic::frameaddress, CGF.AllocaInt8PtrTy),
          llvm::ConstantInt::get(CGF.Int32Ty, 0));
  }
 
  // Mark the call site and declaration with ReturnsTwice.
  llvm::Type *ArgTypes[2] = {CGF.Int8PtrTy, Arg1Ty};
  llvm::AttributeList ReturnsTwiceAttr = llvm::AttributeList::get(
      CGF.getLLVMContext(), llvm::AttributeList::FunctionIndex,
      llvm::Attribute::ReturnsTwice);
  llvm::FunctionCallee SetJmpFn = CGF.CGM.CreateRuntimeFunction(
      llvm::FunctionType::get(CGF.IntTy, ArgTypes, IsVarArg), Name,
      ReturnsTwiceAttr, /*Local=*/true);
 
  llvm::Value *Buf = CGF.Builder.CreateBitOrPointerCast(
      CGF.EmitScalarExpr(E->getArg(0)), CGF.Int8PtrTy);
  llvm::Value *Args[] = {Buf, Arg1};
  llvm::CallBase *CB = CGF.EmitRuntimeCallOrInvoke(SetJmpFn, Args);
  CB->setAttributes(ReturnsTwiceAttr);
  return RValue::get(CB);
}
 
// Many of MSVC builtins are on x64, ARM and AArch64; to avoid repeating code,
// we handle them here.
enum class CodeGenFunction::MSVCIntrin {
  _BitScanForward,
  _BitScanReverse,
  _InterlockedAnd,
  _InterlockedDecrement,
  _InterlockedExchange,
  _InterlockedExchangeAdd,
  _InterlockedExchangeSub,
  _InterlockedIncrement,
  _InterlockedOr,
  _InterlockedXor,
  _InterlockedExchangeAdd_acq,
  _InterlockedExchangeAdd_rel,
  _InterlockedExchangeAdd_nf,
  _InterlockedExchange_acq,
  _InterlockedExchange_rel,
  _InterlockedExchange_nf,
  _InterlockedCompareExchange_acq,
  _InterlockedCompareExchange_rel,
  _InterlockedCompareExchange_nf,
  _InterlockedCompareExchange128,
  _InterlockedCompareExchange128_acq,
  _InterlockedCompareExchange128_rel,
  _InterlockedCompareExchange128_nf,
  _InterlockedOr_acq,
  _InterlockedOr_rel,
  _InterlockedOr_nf,
  _InterlockedXor_acq,
  _InterlockedXor_rel,
  _InterlockedXor_nf,
  _InterlockedAnd_acq,
  _InterlockedAnd_rel,
  _InterlockedAnd_nf,
  _InterlockedIncrement_acq,
  _InterlockedIncrement_rel,
  _InterlockedIncrement_nf,
  _InterlockedDecrement_acq,
  _InterlockedDecrement_rel,
  _InterlockedDecrement_nf,
  __fastfail,
};
 
static std::optional<CodeGenFunction::MSVCIntrin>
translateArmToMsvcIntrin(unsigned BuiltinID) {
  using MSVCIntrin = CodeGenFunction::MSVCIntrin;
  switch (BuiltinID) {
  default:
    return std::nullopt;
  case clang::ARM::BI_BitScanForward:
  case clang::ARM::BI_BitScanForward64:
    return MSVCIntrin::_BitScanForward;
  case clang::ARM::BI_BitScanReverse:
  case clang::ARM::BI_BitScanReverse64:
    return MSVCIntrin::_BitScanReverse;
  case clang::ARM::BI_InterlockedAnd64:
    return MSVCIntrin::_InterlockedAnd;
  case clang::ARM::BI_InterlockedExchange64:
    return MSVCIntrin::_InterlockedExchange;
  case clang::ARM::BI_InterlockedExchangeAdd64:
    return MSVCIntrin::_InterlockedExchangeAdd;
  case clang::ARM::BI_InterlockedExchangeSub64:
    return MSVCIntrin::_InterlockedExchangeSub;
  case clang::ARM::BI_InterlockedOr64:
    return MSVCIntrin::_InterlockedOr;
  case clang::ARM::BI_InterlockedXor64:
    return MSVCIntrin::_InterlockedXor;
  case clang::ARM::BI_InterlockedDecrement64:
    return MSVCIntrin::_InterlockedDecrement;
  case clang::ARM::BI_InterlockedIncrement64:
    return MSVCIntrin::_InterlockedIncrement;
  case clang::ARM::BI_InterlockedExchangeAdd8_acq:
  case clang::ARM::BI_InterlockedExchangeAdd16_acq:
  case clang::ARM::BI_InterlockedExchangeAdd_acq:
  case clang::ARM::BI_InterlockedExchangeAdd64_acq:
    return MSVCIntrin::_InterlockedExchangeAdd_acq;
  case clang::ARM::BI_InterlockedExchangeAdd8_rel:
  case clang::ARM::BI_InterlockedExchangeAdd16_rel:
  case clang::ARM::BI_InterlockedExchangeAdd_rel:
  case clang::ARM::BI_InterlockedExchangeAdd64_rel:
    return MSVCIntrin::_InterlockedExchangeAdd_rel;
  case clang::ARM::BI_InterlockedExchangeAdd8_nf:
  case clang::ARM::BI_InterlockedExchangeAdd16_nf:
  case clang::ARM::BI_InterlockedExchangeAdd_nf:
  case clang::ARM::BI_InterlockedExchangeAdd64_nf:
    return MSVCIntrin::_InterlockedExchangeAdd_nf;
  case clang::ARM::BI_InterlockedExchange8_acq:
  case clang::ARM::BI_InterlockedExchange16_acq:
  case clang::ARM::BI_InterlockedExchange_acq:
  case clang::ARM::BI_InterlockedExchange64_acq:
    return MSVCIntrin::_InterlockedExchange_acq;
  case clang::ARM::BI_InterlockedExchange8_rel:
  case clang::ARM::BI_InterlockedExchange16_rel:
  case clang::ARM::BI_InterlockedExchange_rel:
  case clang::ARM::BI_InterlockedExchange64_rel:
    return MSVCIntrin::_InterlockedExchange_rel;
  case clang::ARM::BI_InterlockedExchange8_nf:
  case clang::ARM::BI_InterlockedExchange16_nf:
  case clang::ARM::BI_InterlockedExchange_nf:
  case clang::ARM::BI_InterlockedExchange64_nf:
    return MSVCIntrin::_InterlockedExchange_nf;
  case clang::ARM::BI_InterlockedCompareExchange8_acq:
  case clang::ARM::BI_InterlockedCompareExchange16_acq:
  case clang::ARM::BI_InterlockedCompareExchange_acq:
  case clang::ARM::BI_InterlockedCompareExchange64_acq:
    return MSVCIntrin::_InterlockedCompareExchange_acq;
  case clang::ARM::BI_InterlockedCompareExchange8_rel:
  case clang::ARM::BI_InterlockedCompareExchange16_rel:
  case clang::ARM::BI_InterlockedCompareExchange_rel:
  case clang::ARM::BI_InterlockedCompareExchange64_rel:
    return MSVCIntrin::_InterlockedCompareExchange_rel;
  case clang::ARM::BI_InterlockedCompareExchange8_nf:
  case clang::ARM::BI_InterlockedCompareExchange16_nf:
  case clang::ARM::BI_InterlockedCompareExchange_nf:
  case clang::ARM::BI_InterlockedCompareExchange64_nf:
    return MSVCIntrin::_InterlockedCompareExchange_nf;
  case clang::ARM::BI_InterlockedOr8_acq:
  case clang::ARM::BI_InterlockedOr16_acq:
  case clang::ARM::BI_InterlockedOr_acq:
  case clang::ARM::BI_InterlockedOr64_acq:
    return MSVCIntrin::_InterlockedOr_acq;
  case clang::ARM::BI_InterlockedOr8_rel:
  case clang::ARM::BI_InterlockedOr16_rel:
  case clang::ARM::BI_InterlockedOr_rel:
  case clang::ARM::BI_InterlockedOr64_rel:
    return MSVCIntrin::_InterlockedOr_rel;
  case clang::ARM::BI_InterlockedOr8_nf:
  case clang::ARM::BI_InterlockedOr16_nf:
  case clang::ARM::BI_InterlockedOr_nf:
  case clang::ARM::BI_InterlockedOr64_nf:
    return MSVCIntrin::_InterlockedOr_nf;
  case clang::ARM::BI_InterlockedXor8_acq:
  case clang::ARM::BI_InterlockedXor16_acq:
  case clang::ARM::BI_InterlockedXor_acq:
  case clang::ARM::BI_InterlockedXor64_acq:
    return MSVCIntrin::_InterlockedXor_acq;
  case clang::ARM::BI_InterlockedXor8_rel:
  case clang::ARM::BI_InterlockedXor16_rel:
  case clang::ARM::BI_InterlockedXor_rel:
  case clang::ARM::BI_InterlockedXor64_rel:
    return MSVCIntrin::_InterlockedXor_rel;
  case clang::ARM::BI_InterlockedXor8_nf:
  case clang::ARM::BI_InterlockedXor16_nf:
  case clang::ARM::BI_InterlockedXor_nf:
  case clang::ARM::BI_InterlockedXor64_nf:
    return MSVCIntrin::_InterlockedXor_nf;
  case clang::ARM::BI_InterlockedAnd8_acq:
  case clang::ARM::BI_InterlockedAnd16_acq:
  case clang::ARM::BI_InterlockedAnd_acq:
  case clang::ARM::BI_InterlockedAnd64_acq:
    return MSVCIntrin::_InterlockedAnd_acq;
  case clang::ARM::BI_InterlockedAnd8_rel:
  case clang::ARM::BI_InterlockedAnd16_rel:
  case clang::ARM::BI_InterlockedAnd_rel:
  case clang::ARM::BI_InterlockedAnd64_rel:
    return MSVCIntrin::_InterlockedAnd_rel;
  case clang::ARM::BI_InterlockedAnd8_nf:
  case clang::ARM::BI_InterlockedAnd16_nf:
  case clang::ARM::BI_InterlockedAnd_nf:
  case clang::ARM::BI_InterlockedAnd64_nf:
    return MSVCIntrin::_InterlockedAnd_nf;
  case clang::ARM::BI_InterlockedIncrement16_acq:
  case clang::ARM::BI_InterlockedIncrement_acq:
  case clang::ARM::BI_InterlockedIncrement64_acq:
    return MSVCIntrin::_InterlockedIncrement_acq;
  case clang::ARM::BI_InterlockedIncrement16_rel:
  case clang::ARM::BI_InterlockedIncrement_rel:
  case clang::ARM::BI_InterlockedIncrement64_rel:
    return MSVCIntrin::_InterlockedIncrement_rel;
  case clang::ARM::BI_InterlockedIncrement16_nf:
  case clang::ARM::BI_InterlockedIncrement_nf:
  case clang::ARM::BI_InterlockedIncrement64_nf:
    return MSVCIntrin::_InterlockedIncrement_nf;
  case clang::ARM::BI_InterlockedDecrement16_acq:
  case clang::ARM::BI_InterlockedDecrement_acq:
  case clang::ARM::BI_InterlockedDecrement64_acq:
    return MSVCIntrin::_InterlockedDecrement_acq;
  case clang::ARM::BI_InterlockedDecrement16_rel:
  case clang::ARM::BI_InterlockedDecrement_rel:
  case clang::ARM::BI_InterlockedDecrement64_rel:
    return MSVCIntrin::_InterlockedDecrement_rel;
  case clang::ARM::BI_InterlockedDecrement16_nf:
  case clang::ARM::BI_InterlockedDecrement_nf:
  case clang::ARM::BI_InterlockedDecrement64_nf:
    return MSVCIntrin::_InterlockedDecrement_nf;
  }
  llvm_unreachable("must return from switch");
}
 
static std::optional<CodeGenFunction::MSVCIntrin>
translateAarch64ToMsvcIntrin(unsigned BuiltinID) {
  using MSVCIntrin = CodeGenFunction::MSVCIntrin;
  switch (BuiltinID) {
  default:
    return std::nullopt;
  case clang::AArch64::BI_BitScanForward:
  case clang::AArch64::BI_BitScanForward64:
    return MSVCIntrin::_BitScanForward;
  case clang::AArch64::BI_BitScanReverse:
  case clang::AArch64::BI_BitScanReverse64:
    return MSVCIntrin::_BitScanReverse;
  case clang::AArch64::BI_InterlockedAnd64:
    return MSVCIntrin::_InterlockedAnd;
  case clang::AArch64::BI_InterlockedExchange64:
    return MSVCIntrin::_InterlockedExchange;
  case clang::AArch64::BI_InterlockedExchangeAdd64:
    return MSVCIntrin::_InterlockedExchangeAdd;
  case clang::AArch64::BI_InterlockedExchangeSub64:
    return MSVCIntrin::_InterlockedExchangeSub;
  case clang::AArch64::BI_InterlockedOr64:
    return MSVCIntrin::_InterlockedOr;
  case clang::AArch64::BI_InterlockedXor64:
    return MSVCIntrin::_InterlockedXor;
  case clang::AArch64::BI_InterlockedDecrement64:
    return MSVCIntrin::_InterlockedDecrement;
  case clang::AArch64::BI_InterlockedIncrement64:
    return MSVCIntrin::_InterlockedIncrement;
  case clang::AArch64::BI_InterlockedExchangeAdd8_acq:
  case clang::AArch64::BI_InterlockedExchangeAdd16_acq:
  case clang::AArch64::BI_InterlockedExchangeAdd_acq:
  case clang::AArch64::BI_InterlockedExchangeAdd64_acq:
    return MSVCIntrin::_InterlockedExchangeAdd_acq;
  case clang::AArch64::BI_InterlockedExchangeAdd8_rel:
  case clang::AArch64::BI_InterlockedExchangeAdd16_rel:
  case clang::AArch64::BI_InterlockedExchangeAdd_rel:
  case clang::AArch64::BI_InterlockedExchangeAdd64_rel:
    return MSVCIntrin::_InterlockedExchangeAdd_rel;
  case clang::AArch64::BI_InterlockedExchangeAdd8_nf:
  case clang::AArch64::BI_InterlockedExchangeAdd16_nf:
  case clang::AArch64::BI_InterlockedExchangeAdd_nf:
  case clang::AArch64::BI_InterlockedExchangeAdd64_nf:
    return MSVCIntrin::_InterlockedExchangeAdd_nf;
  case clang::AArch64::BI_InterlockedExchange8_acq:
  case clang::AArch64::BI_InterlockedExchange16_acq:
  case clang::AArch64::BI_InterlockedExchange_acq:
  case clang::AArch64::BI_InterlockedExchange64_acq:
    return MSVCIntrin::_InterlockedExchange_acq;
  case clang::AArch64::BI_InterlockedExchange8_rel:
  case clang::AArch64::BI_InterlockedExchange16_rel:
  case clang::AArch64::BI_InterlockedExchange_rel:
  case clang::AArch64::BI_InterlockedExchange64_rel:
    return MSVCIntrin::_InterlockedExchange_rel;
  case clang::AArch64::BI_InterlockedExchange8_nf:
  case clang::AArch64::BI_InterlockedExchange16_nf:
  case clang::AArch64::BI_InterlockedExchange_nf:
  case clang::AArch64::BI_InterlockedExchange64_nf:
    return MSVCIntrin::_InterlockedExchange_nf;
  case clang::AArch64::BI_InterlockedCompareExchange8_acq:
  case clang::AArch64::BI_InterlockedCompareExchange16_acq:
  case clang::AArch64::BI_InterlockedCompareExchange_acq:
  case clang::AArch64::BI_InterlockedCompareExchange64_acq:
    return MSVCIntrin::_InterlockedCompareExchange_acq;
  case clang::AArch64::BI_InterlockedCompareExchange8_rel:
  case clang::AArch64::BI_InterlockedCompareExchange16_rel:
  case clang::AArch64::BI_InterlockedCompareExchange_rel:
  case clang::AArch64::BI_InterlockedCompareExchange64_rel:
    return MSVCIntrin::_InterlockedCompareExchange_rel;
  case clang::AArch64::BI_InterlockedCompareExchange8_nf:
  case clang::AArch64::BI_InterlockedCompareExchange16_nf:
  case clang::AArch64::BI_InterlockedCompareExchange_nf:
  case clang::AArch64::BI_InterlockedCompareExchange64_nf:
    return MSVCIntrin::_InterlockedCompareExchange_nf;
  case clang::AArch64::BI_InterlockedCompareExchange128:
    return MSVCIntrin::_InterlockedCompareExchange128;
  case clang::AArch64::BI_InterlockedCompareExchange128_acq:
    return MSVCIntrin::_InterlockedCompareExchange128_acq;
  case clang::AArch64::BI_InterlockedCompareExchange128_nf:
    return MSVCIntrin::_InterlockedCompareExchange128_nf;
  case clang::AArch64::BI_InterlockedCompareExchange128_rel:
    return MSVCIntrin::_InterlockedCompareExchange128_rel;
  case clang::AArch64::BI_InterlockedOr8_acq:
  case clang::AArch64::BI_InterlockedOr16_acq:
  case clang::AArch64::BI_InterlockedOr_acq:
  case clang::AArch64::BI_InterlockedOr64_acq:
    return MSVCIntrin::_InterlockedOr_acq;
  case clang::AArch64::BI_InterlockedOr8_rel:
  case clang::AArch64::BI_InterlockedOr16_rel:
  case clang::AArch64::BI_InterlockedOr_rel:
  case clang::AArch64::BI_InterlockedOr64_rel:
    return MSVCIntrin::_InterlockedOr_rel;
  case clang::AArch64::BI_InterlockedOr8_nf:
  case clang::AArch64::BI_InterlockedOr16_nf:
  case clang::AArch64::BI_InterlockedOr_nf:
  case clang::AArch64::BI_InterlockedOr64_nf:
    return MSVCIntrin::_InterlockedOr_nf;
  case clang::AArch64::BI_InterlockedXor8_acq:
  case clang::AArch64::BI_InterlockedXor16_acq:
  case clang::AArch64::BI_InterlockedXor_acq:
  case clang::AArch64::BI_InterlockedXor64_acq:
    return MSVCIntrin::_InterlockedXor_acq;
  case clang::AArch64::BI_InterlockedXor8_rel:
  case clang::AArch64::BI_InterlockedXor16_rel:
  case clang::AArch64::BI_InterlockedXor_rel:
  case clang::AArch64::BI_InterlockedXor64_rel:
    return MSVCIntrin::_InterlockedXor_rel;
  case clang::AArch64::BI_InterlockedXor8_nf:
  case clang::AArch64::BI_InterlockedXor16_nf:
  case clang::AArch64::BI_InterlockedXor_nf:
  case clang::AArch64::BI_InterlockedXor64_nf:
    return MSVCIntrin::_InterlockedXor_nf;
  case clang::AArch64::BI_InterlockedAnd8_acq:
  case clang::AArch64::BI_InterlockedAnd16_acq:
  case clang::AArch64::BI_InterlockedAnd_acq:
  case clang::AArch64::BI_InterlockedAnd64_acq:
    return MSVCIntrin::_InterlockedAnd_acq;
  case clang::AArch64::BI_InterlockedAnd8_rel:
  case clang::AArch64::BI_InterlockedAnd16_rel:
  case clang::AArch64::BI_InterlockedAnd_rel:
  case clang::AArch64::BI_InterlockedAnd64_rel:
    return MSVCIntrin::_InterlockedAnd_rel;
  case clang::AArch64::BI_InterlockedAnd8_nf:
  case clang::AArch64::BI_InterlockedAnd16_nf:
  case clang::AArch64::BI_InterlockedAnd_nf:
  case clang::AArch64::BI_InterlockedAnd64_nf:
    return MSVCIntrin::_InterlockedAnd_nf;
  case clang::AArch64::BI_InterlockedIncrement16_acq:
  case clang::AArch64::BI_InterlockedIncrement_acq:
  case clang::AArch64::BI_InterlockedIncrement64_acq:
    return MSVCIntrin::_InterlockedIncrement_acq;
  case clang::AArch64::BI_InterlockedIncrement16_rel:
  case clang::AArch64::BI_InterlockedIncrement_rel:
  case clang::AArch64::BI_InterlockedIncrement64_rel:
    return MSVCIntrin::_InterlockedIncrement_rel;
  case clang::AArch64::BI_InterlockedIncrement16_nf:
  case clang::AArch64::BI_InterlockedIncrement_nf:
  case clang::AArch64::BI_InterlockedIncrement64_nf:
    return MSVCIntrin::_InterlockedIncrement_nf;
  case clang::AArch64::BI_InterlockedDecrement16_acq:
  case clang::AArch64::BI_InterlockedDecrement_acq:
  case clang::AArch64::BI_InterlockedDecrement64_acq:
    return MSVCIntrin::_InterlockedDecrement_acq;
  case clang::AArch64::BI_InterlockedDecrement16_rel:
  case clang::AArch64::BI_InterlockedDecrement_rel:
  case clang::AArch64::BI_InterlockedDecrement64_rel:
    return MSVCIntrin::_InterlockedDecrement_rel;
  case clang::AArch64::BI_InterlockedDecrement16_nf:
  case clang::AArch64::BI_InterlockedDecrement_nf:
  case clang::AArch64::BI_InterlockedDecrement64_nf:
    return MSVCIntrin::_InterlockedDecrement_nf;
  }
  llvm_unreachable("must return from switch");
}
 
static std::optional<CodeGenFunction::MSVCIntrin>
translateX86ToMsvcIntrin(unsigned BuiltinID) {
  using MSVCIntrin = CodeGenFunction::MSVCIntrin;
  switch (BuiltinID) {
  default:
    return std::nullopt;
  case clang::X86::BI_BitScanForward:
  case clang::X86::BI_BitScanForward64:
    return MSVCIntrin::_BitScanForward;
  case clang::X86::BI_BitScanReverse:
  case clang::X86::BI_BitScanReverse64:
    return MSVCIntrin::_BitScanReverse;
  case clang::X86::BI_InterlockedAnd64:
    return MSVCIntrin::_InterlockedAnd;
  case clang::X86::BI_InterlockedCompareExchange128:
    return MSVCIntrin::_InterlockedCompareExchange128;
  case clang::X86::BI_InterlockedExchange64:
    return MSVCIntrin::_InterlockedExchange;
  case clang::X86::BI_InterlockedExchangeAdd64:
    return MSVCIntrin::_InterlockedExchangeAdd;
  case clang::X86::BI_InterlockedExchangeSub64:
    return MSVCIntrin::_InterlockedExchangeSub;
  case clang::X86::BI_InterlockedOr64:
    return MSVCIntrin::_InterlockedOr;
  case clang::X86::BI_InterlockedXor64:
    return MSVCIntrin::_InterlockedXor;
  case clang::X86::BI_InterlockedDecrement64:
    return MSVCIntrin::_InterlockedDecrement;
  case clang::X86::BI_InterlockedIncrement64:
    return MSVCIntrin::_InterlockedIncrement;
  }
  llvm_unreachable("must return from switch");
}
 
// Emit an MSVC intrinsic. Assumes that arguments have *not* been evaluated.
Value *CodeGenFunction::EmitMSVCBuiltinExpr(MSVCIntrin BuiltinID,
                                            const CallExpr *E) {
  switch (BuiltinID) {
  case MSVCIntrin::_BitScanForward:
  case MSVCIntrin::_BitScanReverse: {
    Address IndexAddress(EmitPointerWithAlignment(E->getArg(0)));
    Value *ArgValue = EmitScalarExpr(E->getArg(1));
 
    llvm::Type *ArgType = ArgValue->getType();
    llvm::Type *IndexType = IndexAddress.getElementType();
    llvm::Type *ResultType = ConvertType(E->getType());
 
    Value *ArgZero = llvm::Constant::getNullValue(ArgType);
    Value *ResZero = llvm::Constant::getNullValue(ResultType);
    Value *ResOne = llvm::ConstantInt::get(ResultType, 1);
 
    BasicBlock *Begin = Builder.GetInsertBlock();
    BasicBlock *End = createBasicBlock("bitscan_end", this->CurFn);
    Builder.SetInsertPoint(End);
    PHINode *Result = Builder.CreatePHI(ResultType, 2, "bitscan_result");
 
    Builder.SetInsertPoint(Begin);
    Value *IsZero = Builder.CreateICmpEQ(ArgValue, ArgZero);
    BasicBlock *NotZero = createBasicBlock("bitscan_not_zero", this->CurFn);
    Builder.CreateCondBr(IsZero, End, NotZero);
    Result->addIncoming(ResZero, Begin);
 
    Builder.SetInsertPoint(NotZero);
 
    if (BuiltinID == MSVCIntrin::_BitScanForward) {
      Function *F = CGM.getIntrinsic(Intrinsic::cttz, ArgType);
      Value *ZeroCount = Builder.CreateCall(F, {ArgValue, Builder.getTrue()});
      ZeroCount = Builder.CreateIntCast(ZeroCount, IndexType, false);
      Builder.CreateStore(ZeroCount, IndexAddress, false);
    } else {
      unsigned ArgWidth = cast<llvm::IntegerType>(ArgType)->getBitWidth();
      Value *ArgTypeLastIndex = llvm::ConstantInt::get(IndexType, ArgWidth - 1);
 
      Function *F = CGM.getIntrinsic(Intrinsic::ctlz, ArgType);
      Value *ZeroCount = Builder.CreateCall(F, {ArgValue, Builder.getTrue()});
      ZeroCount = Builder.CreateIntCast(ZeroCount, IndexType, false);
      Value *Index = Builder.CreateNSWSub(ArgTypeLastIndex, ZeroCount);
      Builder.CreateStore(Index, IndexAddress, false);
    }
    Builder.CreateBr(End);
    Result->addIncoming(ResOne, NotZero);
 
    Builder.SetInsertPoint(End);
    return Result;
  }
  case MSVCIntrin::_InterlockedAnd:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::And, E);
  case MSVCIntrin::_InterlockedExchange:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xchg, E);
  case MSVCIntrin::_InterlockedExchangeAdd:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E);
  case MSVCIntrin::_InterlockedExchangeSub:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Sub, E);
  case MSVCIntrin::_InterlockedOr:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Or, E);
  case MSVCIntrin::_InterlockedXor:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xor, E);
  case MSVCIntrin::_InterlockedExchangeAdd_acq:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
                                 AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedExchangeAdd_rel:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
                                 AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedExchangeAdd_nf:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Add, E,
                                 AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedExchange_acq:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xchg, E,
                                 AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedExchange_rel:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xchg, E,
                                 AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedExchange_nf:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xchg, E,
                                 AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedCompareExchange_acq:
    return EmitAtomicCmpXchgForMSIntrin(*this, E, AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedCompareExchange_rel:
    return EmitAtomicCmpXchgForMSIntrin(*this, E, AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedCompareExchange_nf:
    return EmitAtomicCmpXchgForMSIntrin(*this, E, AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedCompareExchange128:
    return EmitAtomicCmpXchg128ForMSIntrin(
        *this, E, AtomicOrdering::SequentiallyConsistent);
  case MSVCIntrin::_InterlockedCompareExchange128_acq:
    return EmitAtomicCmpXchg128ForMSIntrin(*this, E, AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedCompareExchange128_rel:
    return EmitAtomicCmpXchg128ForMSIntrin(*this, E, AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedCompareExchange128_nf:
    return EmitAtomicCmpXchg128ForMSIntrin(*this, E, AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedOr_acq:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Or, E,
                                 AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedOr_rel:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Or, E,
                                 AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedOr_nf:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Or, E,
                                 AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedXor_acq:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xor, E,
                                 AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedXor_rel:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xor, E,
                                 AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedXor_nf:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::Xor, E,
                                 AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedAnd_acq:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::And, E,
                                 AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedAnd_rel:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::And, E,
                                 AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedAnd_nf:
    return MakeBinaryAtomicValue(*this, AtomicRMWInst::And, E,
                                 AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedIncrement_acq:
    return EmitAtomicIncrementValue(*this, E, AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedIncrement_rel:
    return EmitAtomicIncrementValue(*this, E, AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedIncrement_nf:
    return EmitAtomicIncrementValue(*this, E, AtomicOrdering::Monotonic);
  case MSVCIntrin::_InterlockedDecrement_acq:
    return EmitAtomicDecrementValue(*this, E, AtomicOrdering::Acquire);
  case MSVCIntrin::_InterlockedDecrement_rel:
    return EmitAtomicDecrementValue(*this, E, AtomicOrdering::Release);
  case MSVCIntrin::_InterlockedDecrement_nf:
    return EmitAtomicDecrementValue(*this, E, AtomicOrdering::Monotonic);
 
  case MSVCIntrin::_InterlockedDecrement:
    return EmitAtomicDecrementValue(*this, E);
  case MSVCIntrin::_InterlockedIncrement:
    return EmitAtomicIncrementValue(*this, E);
 
  case MSVCIntrin::__fastfail: {
    // Request immediate process termination from the kernel. The instruction
    // sequences to do this are documented on MSDN:
    // https://msdn.microsoft.com/en-us/library/dn774154.aspx
    llvm::Triple::ArchType ISA = getTarget().getTriple().getArch();
    StringRef Asm, Constraints;
    switch (ISA) {
    default:
      ErrorUnsupported(E, "__fastfail call for this architecture");
      break;
    case llvm::Triple::x86:
    case llvm::Triple::x86_64:
      Asm = "int $$0x29";
      Constraints = "{cx}";
      break;
    case llvm::Triple::thumb:
      Asm = "udf #251";
      Constraints = "{r0}";
      break;
    case llvm::Triple::aarch64:
      Asm = "brk #0xF003";
      Constraints = "{w0}";
    }
    llvm::FunctionType *FTy = llvm::FunctionType::get(VoidTy, {Int32Ty}, false);
    llvm::InlineAsm *IA =
        llvm::InlineAsm::get(FTy, Asm, Constraints, /*hasSideEffects=*/true);
    llvm::AttributeList NoReturnAttr = llvm::AttributeList::get(
        getLLVMContext(), llvm::AttributeList::FunctionIndex,
        llvm::Attribute::NoReturn);
    llvm::CallInst *CI = Builder.CreateCall(IA, EmitScalarExpr(E->getArg(0)));
    CI->setAttributes(NoReturnAttr);
    return CI;
  }
  }
  llvm_unreachable("Incorrect MSVC intrinsic!");
}
 
namespace {
// ARC cleanup for __builtin_os_log_format
struct CallObjCArcUse final : EHScopeStack::Cleanup {
  CallObjCArcUse(llvm::Value *object) : object(object) {}
  llvm::Value *object;
 
  void Emit(CodeGenFunction &CGF, Flags flags) override {
    CGF.EmitARCIntrinsicUse(object);
  }
};
}
 
Value *CodeGenFunction::EmitCheckedArgForBuiltin(const Expr *E,
                                                 BuiltinCheckKind Kind) {
  assert((Kind == BCK_CLZPassedZero || Kind == BCK_CTZPassedZero)
          && "Unsupported builtin check kind");
 
  Value *ArgValue = EmitScalarExpr(E);
  if (!SanOpts.has(SanitizerKind::Builtin) || !getTarget().isCLZForZeroUndef())
    return ArgValue;
 
  SanitizerScope SanScope(this);
  Value *Cond = Builder.CreateICmpNE(
      ArgValue, llvm::Constant::getNullValue(ArgValue->getType()));
  EmitCheck(std::make_pair(Cond, SanitizerKind::Builtin),
            SanitizerHandler::InvalidBuiltin,
            {EmitCheckSourceLocation(E->getExprLoc()),
             llvm::ConstantInt::get(Builder.getInt8Ty(), Kind)},
            std::nullopt);
  return ArgValue;
}
 
/// Get the argument type for arguments to os_log_helper.
static CanQualType getOSLogArgType(ASTContext &C, int Size) {
  QualType UnsignedTy = C.getIntTypeForBitwidth(Size * 8, /*Signed=*/false);
  return C.getCanonicalType(UnsignedTy);
}
 
llvm::Function *CodeGenFunction::generateBuiltinOSLogHelperFunction(
    const analyze_os_log::OSLogBufferLayout &Layout,
    CharUnits BufferAlignment) {
  ASTContext &Ctx = getContext();
 
  llvm::SmallString<64> Name;
  {
    raw_svector_ostream OS(Name);
    OS << "__os_log_helper";
    OS << "_" << BufferAlignment.getQuantity();
    OS << "_" << int(Layout.getSummaryByte());
    OS << "_" << int(Layout.getNumArgsByte());
    for (const auto &Item : Layout.Items)
      OS << "_" << int(Item.getSizeByte()) << "_"
         << int(Item.getDescriptorByte());
  }
 
  if (llvm::Function *F = CGM.getModule().getFunction(Name))
    return F;
 
  llvm::SmallVector<QualType, 4> ArgTys;
  FunctionArgList Args;
  Args.push_back(ImplicitParamDecl::Create(
      Ctx, nullptr, SourceLocation(), &Ctx.Idents.get("buffer"), Ctx.VoidPtrTy,
      ImplicitParamDecl::Other));
  ArgTys.emplace_back(Ctx.VoidPtrTy);
 
  for (unsigned int I = 0, E = Layout.Items.size(); I < E; ++I) {
    char Size = Layout.Items[I].getSizeByte();
    if (!Size)
      continue;
 
    QualType ArgTy = getOSLogArgType(Ctx, Size);
    Args.push_back(ImplicitParamDecl::Create(
        Ctx, nullptr, SourceLocation(),
        &Ctx.Idents.get(std::string("arg") + llvm::to_string(I)), ArgTy,
        ImplicitParamDecl::Other));
    ArgTys.emplace_back(ArgTy);
  }
 
  QualType ReturnTy = Ctx.VoidTy;
 
  // The helper function has linkonce_odr linkage to enable the linker to merge
  // identical functions. To ensure the merging always happens, 'noinline' is
  // attached to the function when compiling with -Oz.
  const CGFunctionInfo &FI =
      CGM.getTypes().arrangeBuiltinFunctionDeclaration(ReturnTy, Args);
  llvm::FunctionType *FuncTy = CGM.getTypes().GetFunctionType(FI);
  llvm::Function *Fn = llvm::Function::Create(
      FuncTy, llvm::GlobalValue::LinkOnceODRLinkage, Name, &CGM.getModule());
  Fn->setVisibility(llvm::GlobalValue::HiddenVisibility);
  CGM.SetLLVMFunctionAttributes(GlobalDecl(), FI, Fn, /*IsThunk=*/false);
  CGM.SetLLVMFunctionAttributesForDefinition(nullptr, Fn);
  Fn->setDoesNotThrow();
 
  // Attach 'noinline' at -Oz.
  if (CGM.getCodeGenOpts().OptimizeSize == 2)
    Fn->addFnAttr(llvm::Attribute::NoInline);
 
  auto NL = ApplyDebugLocation::CreateEmpty(*this);
  StartFunction(GlobalDecl(), ReturnTy, Fn, FI, Args);
 
  // Create a scope with an artificial location for the body of this function.
  auto AL = ApplyDebugLocation::CreateArtificial(*this);
 
  CharUnits Offset;
  Address BufAddr =
      Address(Builder.CreateLoad(GetAddrOfLocalVar(Args[0]), "buf"), Int8Ty,
              BufferAlignment);
  Builder.CreateStore(Builder.getInt8(Layout.getSummaryByte()),
                      Builder.CreateConstByteGEP(BufAddr, Offset++, "summary"));
  Builder.CreateStore(Builder.getInt8(Layout.getNumArgsByte()),
                      Builder.CreateConstByteGEP(BufAddr, Offset++, "numArgs"));
 
  unsigned I = 1;
  for (const auto &Item : Layout.Items) {
    Builder.CreateStore(
        Builder.getInt8(Item.getDescriptorByte()),
        Builder.CreateConstByteGEP(BufAddr, Offset++, "argDescriptor"));
    Builder.CreateStore(
        Builder.getInt8(Item.getSizeByte()),
        Builder.CreateConstByteGEP(BufAddr, Offset++, "argSize"));
 
    CharUnits Size = Item.size();
    if (!Size.getQuantity())
      continue;
 
    Address Arg = GetAddrOfLocalVar(Args[I]);
    Address Addr = Builder.CreateConstByteGEP(BufAddr, Offset, "argData");
    Addr =
        Builder.CreateElementBitCast(Addr, Arg.getElementType(), "argDataCast");
    Builder.CreateStore(Builder.CreateLoad(Arg), Addr);
    Offset += Size;
    ++I;
  }
 
  FinishFunction();
 
  return Fn;
}
 
RValue CodeGenFunction::emitBuiltinOSLogFormat(const CallExpr &E) {
  assert(E.getNumArgs() >= 2 &&
         "__builtin_os_log_format takes at least 2 arguments");
  ASTContext &Ctx = getContext();
  analyze_os_log::OSLogBufferLayout Layout;
  analyze_os_log::computeOSLogBufferLayout(Ctx, &E, Layout);
  Address BufAddr = EmitPointerWithAlignment(E.getArg(0));
  llvm::SmallVector<llvm::Value *, 4> RetainableOperands;
 
  // Ignore argument 1, the format string. It is not currently used.
  CallArgList Args;
  Args.add(RValue::get(BufAddr.getPointer()), Ctx.VoidPtrTy);
 
  for (const auto &Item : Layout.Items) {
    int Size = Item.getSizeByte();
    if (!Size)
      continue;
 
    llvm::Value *ArgVal;
 
    if (Item.getKind() == analyze_os_log::OSLogBufferItem::MaskKind) {
      uint64_t Val = 0;
      for (unsigned I = 0, E = Item.getMaskType().size(); I < E; ++I)
        Val |= ((uint64_t)Item.getMaskType()[I]) << I * 8;
      ArgVal = llvm::Constant::getIntegerValue(Int64Ty, llvm::APInt(64, Val));
    } else if (const Expr *TheExpr = Item.getExpr()) {
      ArgVal = EmitScalarExpr(TheExpr, /*Ignore*/ false);
 
      // If a temporary object that requires destruction after the full
      // expression is passed, push a lifetime-extended cleanup to extend its
      // lifetime to the end of the enclosing block scope.
      auto LifetimeExtendObject = [&](const Expr *E) {
        E = E->IgnoreParenCasts();
        // Extend lifetimes of objects returned by function calls and message
        // sends.
 
        // FIXME: We should do this in other cases in which temporaries are
        //        created including arguments of non-ARC types (e.g., C++
        //        temporaries).
        if (isa<CallExpr>(E) || isa<ObjCMessageExpr>(E))
          return true;
        return false;
      };
 
      if (TheExpr->getType()->isObjCRetainableType() &&
          getLangOpts().ObjCAutoRefCount && LifetimeExtendObject(TheExpr)) {
        assert(getEvaluationKind(TheExpr->getType()) == TEK_Scalar &&
               "Only scalar can be a ObjC retainable type");
        if (!isa<Constant>(ArgVal)) {
          CleanupKind Cleanup = getARCCleanupKind();
          QualType Ty = TheExpr->getType();
          Address Alloca = Address::invalid();
          Address Addr = CreateMemTemp(Ty, "os.log.arg", &Alloca);
          ArgVal = EmitARCRetain(Ty, ArgVal);
          Builder.CreateStore(ArgVal, Addr);
          pushLifetimeExtendedDestroy(Cleanup, Alloca, Ty,
                                      CodeGenFunction::destroyARCStrongPrecise,
                                      Cleanup & EHCleanup);
 
          // Push a clang.arc.use call to ensure ARC optimizer knows that the
          // argument has to be alive.
          if (CGM.getCodeGenOpts().OptimizationLevel != 0)
            pushCleanupAfterFullExpr<CallObjCArcUse>(Cleanup, ArgVal);
        }
      }
    } else {
      ArgVal = Builder.getInt32(Item.getConstValue().getQuantity());
    }
 
    unsigned ArgValSize =
        CGM.getDataLayout().getTypeSizeInBits(ArgVal->getType());
    llvm::IntegerType *IntTy = llvm::Type::getIntNTy(getLLVMContext(),
                                                     ArgValSize);
    ArgVal = Builder.CreateBitOrPointerCast(ArgVal, IntTy);
    CanQualType ArgTy = getOSLogArgType(Ctx, Size);
    // If ArgVal has type x86_fp80, zero-extend ArgVal.
    ArgVal = Builder.CreateZExtOrBitCast(ArgVal, ConvertType(ArgTy));
    Args.add(RValue::get(ArgVal), ArgTy);
  }
 
  const CGFunctionInfo &FI =
      CGM.getTypes().arrangeBuiltinFunctionCall(Ctx.VoidTy, Args);
  llvm::Function *F = CodeGenFunction(CGM).generateBuiltinOSLogHelperFunction(
      Layout, BufAddr.getAlignment());
  EmitCall(FI, CGCallee::forDirect(F), ReturnValueSlot(), Args);
  return RValue::get(BufAddr.getPointer());
}
 
static bool isSpecialUnsignedMultiplySignedResult(
    unsigned BuiltinID, WidthAndSignedness Op1Info, WidthAndSignedness Op2Info,
    WidthAndSignedness ResultInfo) {
  return BuiltinID == Builtin::BI__builtin_mul_overflow &&
         Op1Info.Width == Op2Info.Width && Op2Info.Width == ResultInfo.Width &&
         !Op1Info.Signed && !Op2Info.Signed && ResultInfo.Signed;
}
 
static RValue EmitCheckedUnsignedMultiplySignedResult(
    CodeGenFunction &CGF, const clang::Expr *Op1, WidthAndSignedness Op1Info,
    const clang::Expr *Op2, WidthAndSignedness Op2Info,
    const clang::Expr *ResultArg, QualType ResultQTy,
    WidthAndSignedness ResultInfo) {
  assert(isSpecialUnsignedMultiplySignedResult(
             Builtin::BI__builtin_mul_overflow, Op1Info, Op2Info, ResultInfo) &&
         "Cannot specialize this multiply");
 
  llvm::Value *V1 = CGF.EmitScalarExpr(Op1);
  llvm::Value *V2 = CGF.EmitScalarExpr(Op2);
 
  llvm::Value *HasOverflow;
  llvm::Value *Result = EmitOverflowIntrinsic(
      CGF, llvm::Intrinsic::umul_with_overflow, V1, V2, HasOverflow);
 
  // The intrinsic call will detect overflow when the value is > UINT_MAX,
  // however, since the original builtin had a signed result, we need to report
  // an overflow when the result is greater than INT_MAX.
  auto IntMax = llvm::APInt::getSignedMaxValue(ResultInfo.Width);
  llvm::Value *IntMaxValue = llvm::ConstantInt::get(Result->getType(), IntMax);
 
  llvm::Value *IntMaxOverflow = CGF.Builder.CreateICmpUGT(Result, IntMaxValue);
  HasOverflow = CGF.Builder.CreateOr(HasOverflow, IntMaxOverflow);
 
  bool isVolatile =
      ResultArg->getType()->getPointeeType().isVolatileQualified();
  Address ResultPtr = CGF.EmitPointerWithAlignment(ResultArg);
  CGF.Builder.CreateStore(CGF.EmitToMemory(Result, ResultQTy), ResultPtr,
                          isVolatile);
  return RValue::get(HasOverflow);
}
 
/// Determine if a binop is a checked mixed-sign multiply we can specialize.
static bool isSpecialMixedSignMultiply(unsigned BuiltinID,
                                       WidthAndSignedness Op1Info,
                                       WidthAndSignedness Op2Info,
                                       WidthAndSignedness ResultInfo) {
  return BuiltinID == Builtin::BI__builtin_mul_overflow &&
         std::max(Op1Info.Width, Op2Info.Width) >= ResultInfo.Width &&
         Op1Info.Signed != Op2Info.Signed;
}
 
/// Emit a checked mixed-sign multiply. This is a cheaper specialization of
/// the generic checked-binop irgen.
static RValue
EmitCheckedMixedSignMultiply(CodeGenFunction &CGF, const clang::Expr *Op1,
                             WidthAndSignedness Op1Info, const clang::Expr *Op2,
                             WidthAndSignedness Op2Info,
                             const clang::Expr *ResultArg, QualType ResultQTy,
                             WidthAndSignedness ResultInfo) {
  assert(isSpecialMixedSignMultiply(Builtin::BI__builtin_mul_overflow, Op1Info,
                                    Op2Info, ResultInfo) &&
         "Not a mixed-sign multipliction we can specialize");
 
  // Emit the signed and unsigned operands.
  const clang::Expr *SignedOp = Op1Info.Signed ? Op1 : Op2;
  const clang::Expr *UnsignedOp = Op1Info.Signed ? Op2 : Op1;
  llvm::Value *Signed = CGF.EmitScalarExpr(SignedOp);
  llvm::Value *Unsigned = CGF.EmitScalarExpr(UnsignedOp);
  unsigned SignedOpWidth = Op1Info.Signed ? Op1Info.Width : Op2Info.Width;
  unsigned UnsignedOpWidth = Op1Info.Signed ? Op2Info.Width : Op1Info.Width;
 
  // One of the operands may be smaller than the other. If so, [s|z]ext it.
  if (SignedOpWidth < UnsignedOpWidth)
    Signed = CGF.Builder.CreateSExt(Signed, Unsigned->getType(), "op.sext");
  if (UnsignedOpWidth < SignedOpWidth)
    Unsigned = CGF.Builder.CreateZExt(Unsigned, Signed->getType(), "op.zext");
 
  llvm::Type *OpTy = Signed->getType();
  llvm::Value *Zero = llvm::Constant::getNullValue(OpTy);
  Address ResultPtr = CGF.EmitPointerWithAlignment(ResultArg);
  llvm::Type *ResTy = ResultPtr.getElementType();
  unsigned OpWidth = std::max(Op1Info.Width, Op2Info.Width);
 
  // Take the absolute value of the signed operand.
  llvm::Value *IsNegative = CGF.Builder.CreateICmpSLT(Signed, Zero);
  llvm::Value *AbsOfNegative = CGF.Builder.CreateSub(Zero, Signed);
  llvm::Value *AbsSigned =
      CGF.Builder.CreateSelect(IsNegative, AbsOfNegative, Signed);
 
  // Perform a checked unsigned multiplication.
  llvm::Value *UnsignedOverflow;
  llvm::Value *UnsignedResult =
      EmitOverflowIntrinsic(CGF, llvm::Intrinsic::umul_with_overflow, AbsSigned,
                            Unsigned, UnsignedOverflow);
 
  llvm::Value *Overflow, *Result;
  if (ResultInfo.Signed) {
    // Signed overflow occurs if the result is greater than INT_MAX or lesser
    // than INT_MIN, i.e when |Result| > (INT_MAX + IsNegative).
    auto IntMax =
        llvm::APInt::getSignedMaxValue(ResultInfo.Width).zext(OpWidth);
    llvm::Value *MaxResult =
        CGF.Builder.CreateAdd(llvm::ConstantInt::get(OpTy, IntMax),
                              CGF.Builder.CreateZExt(IsNegative, OpTy));
    llvm::Value *SignedOverflow =
        CGF.Builder.CreateICmpUGT(UnsignedResult, MaxResult);
    Overflow = CGF.Builder.CreateOr(UnsignedOverflow, SignedOverflow);
 
    // Prepare the signed result (possibly by negating it).
    llvm::Value *NegativeResult = CGF.Builder.CreateNeg(UnsignedResult);
    llvm::Value *SignedResult =
        CGF.Builder.CreateSelect(IsNegative, NegativeResult, UnsignedResult);
    Result = CGF.Builder.CreateTrunc(SignedResult, ResTy);
  } else {
    // Unsigned overflow occurs if the result is < 0 or greater than UINT_MAX.
    llvm::Value *Underflow = CGF.Builder.CreateAnd(
        IsNegative, CGF.Builder.CreateIsNotNull(UnsignedResult));
    Overflow = CGF.Builder.CreateOr(UnsignedOverflow, Underflow);
    if (ResultInfo.Width < OpWidth) {
      auto IntMax =
          llvm::APInt::getMaxValue(ResultInfo.Width).zext(OpWidth);
      llvm::Value *TruncOverflow = CGF.Builder.CreateICmpUGT(
          UnsignedResult, llvm::ConstantInt::get(OpTy, IntMax));
      Overflow = CGF.Builder.CreateOr(Overflow, TruncOverflow);
    }
 
    // Negate the product if it would be negative in infinite precision.
    Result = CGF.Builder.CreateSelect(
        IsNegative, CGF.Builder.CreateNeg(UnsignedResult), UnsignedResult);
 
    Result = CGF.Builder.CreateTrunc(Result, ResTy);
  }
  assert(Overflow && Result && "Missing overflow or result");
 
  bool isVolatile =
      ResultArg->getType()->getPointeeType().isVolatileQualified();
  CGF.Builder.CreateStore(CGF.EmitToMemory(Result, ResultQTy), ResultPtr,
                          isVolatile);
  return RValue::get(Overflow);
}
 
static bool
TypeRequiresBuiltinLaunderImp(const ASTContext &Ctx, QualType Ty,
                              llvm::SmallPtrSetImpl<const Decl *> &Seen) {
  if (const auto *Arr = Ctx.getAsArrayType(Ty))
    Ty = Ctx.getBaseElementType(Arr);
 
  const auto *Record = Ty->getAsCXXRecordDecl();
  if (!Record)
    return false;
 
  // We've already checked this type, or are in the process of checking it.
  if (!Seen.insert(Record).second)
    return false;
 
  assert(Record->hasDefinition() &&
         "Incomplete types should already be diagnosed");
 
  if (Record->isDynamicClass())
    return true;
 
  for (FieldDecl *F : Record->fields()) {
    if (TypeRequiresBuiltinLaunderImp(Ctx, F->getType(), Seen))
      return true;
  }
  return false;
}
 
/// Determine if the specified type requires laundering by checking if it is a
/// dynamic class type or contains a subobject which is a dynamic class type.
static bool TypeRequiresBuiltinLaunder(CodeGenModule &CGM, QualType Ty) {
  if (!CGM.getCodeGenOpts().StrictVTablePointers)
    return false;
  llvm::SmallPtrSet<const Decl *, 16> Seen;
  return TypeRequiresBuiltinLaunderImp(CGM.getContext(), Ty, Seen);
}
 
RValue CodeGenFunction::emitRotate(const CallExpr *E, bool IsRotateRight) {
  llvm::Value *Src = EmitScalarExpr(E->getArg(0));
  llvm::Value *ShiftAmt = EmitScalarExpr(E->getArg(1));
 
  // The builtin's shift arg may have a different type than the source arg and
  // result, but the LLVM intrinsic uses the same type for all values.
  llvm::Type *Ty = Src->getType();
  ShiftAmt = Builder.CreateIntCast(ShiftAmt, Ty, false);
 
  // Rotate is a special case of LLVM funnel shift - 1st 2 args are the same.
  unsigned IID = IsRotateRight ? Intrinsic::fshr : Intrinsic::fshl;
  Function *F = CGM.getIntrinsic(IID, Ty);
  return RValue::get(Builder.CreateCall(F, { Src, Src, ShiftAmt }));
}
 
// Map math builtins for long-double to f128 version.
static unsigned mutateLongDoubleBuiltin(unsigned BuiltinID) {
  switch (BuiltinID) {
#define MUTATE_LDBL(func) \
  case Builtin::BI__builtin_##func##l: \
    return Builtin::BI__builtin_##func##f128;
  MUTATE_LDBL(sqrt)
  MUTATE_LDBL(cbrt)
  MUTATE_LDBL(fabs)
  MUTATE_LDBL(log)
  MUTATE_LDBL(log2)
  MUTATE_LDBL(log10)
  MUTATE_LDBL(log1p)
  MUTATE_LDBL(logb)
  MUTATE_LDBL(exp)
  MUTATE_LDBL(exp2)
  MUTATE_LDBL(expm1)
  MUTATE_LDBL(fdim)
  MUTATE_LDBL(hypot)
  MUTATE_LDBL(ilogb)
  MUTATE_LDBL(pow)
  MUTATE_LDBL(fmin)
  MUTATE_LDBL(fmax)
  MUTATE_LDBL(ceil)
  MUTATE_LDBL(trunc)
  MUTATE_LDBL(rint)
  MUTATE_LDBL(nearbyint)
  MUTATE_LDBL(round)
  MUTATE_LDBL(floor)
  MUTATE_LDBL(lround)
  MUTATE_LDBL(llround)
  MUTATE_LDBL(lrint)
  MUTATE_LDBL(llrint)
  MUTATE_LDBL(fmod)
  MUTATE_LDBL(modf)
  MUTATE_LDBL(nan)
  MUTATE_LDBL(nans)
  MUTATE_LDBL(inf)
  MUTATE_LDBL(fma)
  MUTATE_LDBL(sin)
  MUTATE_LDBL(cos)
  MUTATE_LDBL(tan)
  MUTATE_LDBL(sinh)
  MUTATE_LDBL(cosh)
  MUTATE_LDBL(tanh)
  MUTATE_LDBL(asin)
  MUTATE_LDBL(acos)
  MUTATE_LDBL(atan)
  MUTATE_LDBL(asinh)
  MUTATE_LDBL(acosh)
  MUTATE_LDBL(atanh)
  MUTATE_LDBL(atan2)
  MUTATE_LDBL(erf)
  MUTATE_LDBL(erfc)
  MUTATE_LDBL(ldexp)
  MUTATE_LDBL(frexp)
  MUTATE_LDBL(huge_val)
  MUTATE_LDBL(copysign)
  MUTATE_LDBL(nextafter)
  MUTATE_LDBL(nexttoward)
  MUTATE_LDBL(remainder)
  MUTATE_LDBL(remquo)
  MUTATE_LDBL(scalbln)
  MUTATE_LDBL(scalbn)
  MUTATE_LDBL(tgamma)
  MUTATE_LDBL(lgamma)
#undef MUTATE_LDBL
  default:
    return BuiltinID;
  }
}
 
RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
                                        const CallExpr *E,
                                        ReturnValueSlot ReturnValue) {
  const FunctionDecl *FD = GD.getDecl()->getAsFunction();
  // See if we can constant fold this builtin.  If so, don't emit it at all.
  // TODO: Extend this handling to all builtin calls that we can constant-fold.
  Expr::EvalResult Result;
  if (E->isPRValue() && E->EvaluateAsRValue(Result, CGM.getContext()) &&
      !Result.hasSideEffects()) {
    if (Result.Val.isInt())
      return RValue::get(llvm::ConstantInt::get(getLLVMContext(),
                                                Result.Val.getInt()));
    if (Result.Val.isFloat())
      return RValue::get(llvm::ConstantFP::get(getLLVMContext(),
                                               Result.Val.getFloat()));
  }
 
  // If current long-double semantics is IEEE 128-bit, replace math builtins
  // of long-double with f128 equivalent.
  // TODO: This mutation should also be applied to other targets other than PPC,
  // after backend supports IEEE 128-bit style libcalls.
  if (getTarget().getTriple().isPPC64() &&
      &getTarget().getLongDoubleFormat() == &llvm::APFloat::IEEEquad())
    BuiltinID = mutateLongDoubleBuiltin(BuiltinID);
 
  // If the builtin has been declared explicitly with an assembler label,
  // disable the specialized emitting below. Ideally we should communicate the
  // rename in IR, or at least avoid generating the intrinsic calls that are
  // likely to get lowered to the renamed library functions.
  const unsigned BuiltinIDIfNoAsmLabel =
      FD->hasAttr<AsmLabelAttr>() ? 0 : BuiltinID;
 
  // There are LLVM math intrinsics/instructions corresponding to math library
  // functions except the LLVM op will never set errno while the math library
  // might. Also, math builtins have the same semantics as their math library
  // twins. Thus, we can transform math library and builtin calls to their
  // LLVM counterparts if the call is marked 'const' (known to never set errno).
  // In case FP exceptions are enabled, the experimental versions of the
  // intrinsics model those.
  bool ConstWithoutErrnoAndExceptions =
      getContext().BuiltinInfo.isConstWithoutErrnoAndExceptions(BuiltinID);
  bool ConstWithoutExceptions =
      getContext().BuiltinInfo.isConstWithoutExceptions(BuiltinID);
  if (FD->hasAttr<ConstAttr>() ||
      ((ConstWithoutErrnoAndExceptions || ConstWithoutExceptions) &&
       (!ConstWithoutErrnoAndExceptions || (!getLangOpts().MathErrno)))) {
    switch (BuiltinIDIfNoAsmLabel) {
    case Builtin::BIceil:
    case Builtin::BIceilf:
    case Builtin::BIceill:
    case Builtin::BI__builtin_ceil:
    case Builtin::BI__builtin_ceilf:
    case Builtin::BI__builtin_ceilf16:
    case Builtin::BI__builtin_ceill:
    case Builtin::BI__builtin_ceilf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::ceil,
                                   Intrinsic::experimental_constrained_ceil));
 
    case Builtin::BIcopysign:
    case Builtin::BIcopysignf:
    case Builtin::BIcopysignl:
    case Builtin::BI__builtin_copysign:
    case Builtin::BI__builtin_copysignf:
    case Builtin::BI__builtin_copysignf16:
    case Builtin::BI__builtin_copysignl:
    case Builtin::BI__builtin_copysignf128:
      return RValue::get(emitBinaryBuiltin(*this, E, Intrinsic::copysign));
 
    case Builtin::BIcos:
    case Builtin::BIcosf:
    case Builtin::BIcosl:
    case Builtin::BI__builtin_cos:
    case Builtin::BI__builtin_cosf:
    case Builtin::BI__builtin_cosf16:
    case Builtin::BI__builtin_cosl:
    case Builtin::BI__builtin_cosf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::cos,
                                   Intrinsic::experimental_constrained_cos));
 
    case Builtin::BIexp:
    case Builtin::BIexpf:
    case Builtin::BIexpl:
    case Builtin::BI__builtin_exp:
    case Builtin::BI__builtin_expf:
    case Builtin::BI__builtin_expf16:
    case Builtin::BI__builtin_expl:
    case Builtin::BI__builtin_expf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::exp,
                                   Intrinsic::experimental_constrained_exp));
 
    case Builtin::BIexp2:
    case Builtin::BIexp2f:
    case Builtin::BIexp2l:
    case Builtin::BI__builtin_exp2:
    case Builtin::BI__builtin_exp2f:
    case Builtin::BI__builtin_exp2f16:
    case Builtin::BI__builtin_exp2l:
    case Builtin::BI__builtin_exp2f128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::exp2,
                                   Intrinsic::experimental_constrained_exp2));
 
    case Builtin::BIfabs:
    case Builtin::BIfabsf:
    case Builtin::BIfabsl:
    case Builtin::BI__builtin_fabs:
    case Builtin::BI__builtin_fabsf:
    case Builtin::BI__builtin_fabsf16:
    case Builtin::BI__builtin_fabsl:
    case Builtin::BI__builtin_fabsf128:
      return RValue::get(emitUnaryBuiltin(*this, E, Intrinsic::fabs));
 
    case Builtin::BIfloor:
    case Builtin::BIfloorf:
    case Builtin::BIfloorl:
    case Builtin::BI__builtin_floor:
    case Builtin::BI__builtin_floorf:
    case Builtin::BI__builtin_floorf16:
    case Builtin::BI__builtin_floorl:
    case Builtin::BI__builtin_floorf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::floor,
                                   Intrinsic::experimental_constrained_floor));
 
    case Builtin::BIfma:
    case Builtin::BIfmaf:
    case Builtin::BIfmal:
    case Builtin::BI__builtin_fma:
    case Builtin::BI__builtin_fmaf:
    case Builtin::BI__builtin_fmaf16:
    case Builtin::BI__builtin_fmal:
    case Builtin::BI__builtin_fmaf128:
      return RValue::get(emitTernaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::fma,
                                   Intrinsic::experimental_constrained_fma));
 
    case Builtin::BIfmax:
    case Builtin::BIfmaxf:
    case Builtin::BIfmaxl:
    case Builtin::BI__builtin_fmax:
    case Builtin::BI__builtin_fmaxf:
    case Builtin::BI__builtin_fmaxf16:
    case Builtin::BI__builtin_fmaxl:
    case Builtin::BI__builtin_fmaxf128:
      return RValue::get(emitBinaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::maxnum,
                                   Intrinsic::experimental_constrained_maxnum));
 
    case Builtin::BIfmin:
    case Builtin::BIfminf:
    case Builtin::BIfminl:
    case Builtin::BI__builtin_fmin:
    case Builtin::BI__builtin_fminf:
    case Builtin::BI__builtin_fminf16:
    case Builtin::BI__builtin_fminl:
    case Builtin::BI__builtin_fminf128:
      return RValue::get(emitBinaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::minnum,
                                   Intrinsic::experimental_constrained_minnum));
 
    // fmod() is a special-case. It maps to the frem instruction rather than an
    // LLVM intrinsic.
    case Builtin::BIfmod:
    case Builtin::BIfmodf:
    case Builtin::BIfmodl:
    case Builtin::BI__builtin_fmod:
    case Builtin::BI__builtin_fmodf:
    case Builtin::BI__builtin_fmodf16:
    case Builtin::BI__builtin_fmodl:
    case Builtin::BI__builtin_fmodf128: {
      CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
      Value *Arg1 = EmitScalarExpr(E->getArg(0));
      Value *Arg2 = EmitScalarExpr(E->getArg(1));
      return RValue::get(Builder.CreateFRem(Arg1, Arg2, "fmod"));
    }
 
    case Builtin::BIlog:
    case Builtin::BIlogf:
    case Builtin::BIlogl:
    case Builtin::BI__builtin_log:
    case Builtin::BI__builtin_logf:
    case Builtin::BI__builtin_logf16:
    case Builtin::BI__builtin_logl:
    case Builtin::BI__builtin_logf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::log,
                                   Intrinsic::experimental_constrained_log));
 
    case Builtin::BIlog10:
    case Builtin::BIlog10f:
    case Builtin::BIlog10l:
    case Builtin::BI__builtin_log10:
    case Builtin::BI__builtin_log10f:
    case Builtin::BI__builtin_log10f16:
    case Builtin::BI__builtin_log10l:
    case Builtin::BI__builtin_log10f128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::log10,
                                   Intrinsic::experimental_constrained_log10));
 
    case Builtin::BIlog2:
    case Builtin::BIlog2f:
    case Builtin::BIlog2l:
    case Builtin::BI__builtin_log2:
    case Builtin::BI__builtin_log2f:
    case Builtin::BI__builtin_log2f16:
    case Builtin::BI__builtin_log2l:
    case Builtin::BI__builtin_log2f128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::log2,
                                   Intrinsic::experimental_constrained_log2));
 
    case Builtin::BInearbyint:
    case Builtin::BInearbyintf:
    case Builtin::BInearbyintl:
    case Builtin::BI__builtin_nearbyint:
    case Builtin::BI__builtin_nearbyintf:
    case Builtin::BI__builtin_nearbyintl:
    case Builtin::BI__builtin_nearbyintf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                Intrinsic::nearbyint,
                                Intrinsic::experimental_constrained_nearbyint));
 
    case Builtin::BIpow:
    case Builtin::BIpowf:
    case Builtin::BIpowl:
    case Builtin::BI__builtin_pow:
    case Builtin::BI__builtin_powf:
    case Builtin::BI__builtin_powf16:
    case Builtin::BI__builtin_powl:
    case Builtin::BI__builtin_powf128:
      return RValue::get(emitBinaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::pow,
                                   Intrinsic::experimental_constrained_pow));
 
    case Builtin::BIrint:
    case Builtin::BIrintf:
    case Builtin::BIrintl:
    case Builtin::BI__builtin_rint:
    case Builtin::BI__builtin_rintf:
    case Builtin::BI__builtin_rintf16:
    case Builtin::BI__builtin_rintl:
    case Builtin::BI__builtin_rintf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::rint,
                                   Intrinsic::experimental_constrained_rint));
 
    case Builtin::BIround:
    case Builtin::BIroundf:
    case Builtin::BIroundl:
    case Builtin::BI__builtin_round:
    case Builtin::BI__builtin_roundf:
    case Builtin::BI__builtin_roundf16:
    case Builtin::BI__builtin_roundl:
    case Builtin::BI__builtin_roundf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::round,
                                   Intrinsic::experimental_constrained_round));
 
    case Builtin::BIroundeven:
    case Builtin::BIroundevenf:
    case Builtin::BIroundevenl:
    case Builtin::BI__builtin_roundeven:
    case Builtin::BI__builtin_roundevenf:
    case Builtin::BI__builtin_roundevenf16:
    case Builtin::BI__builtin_roundevenl:
    case Builtin::BI__builtin_roundevenf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::roundeven,
                                   Intrinsic::experimental_constrained_roundeven));
                                   
    case Builtin::BIsin:
    case Builtin::BIsinf:
    case Builtin::BIsinl:
    case Builtin::BI__builtin_sin:
    case Builtin::BI__builtin_sinf:
    case Builtin::BI__builtin_sinf16:
    case Builtin::BI__builtin_sinl:
    case Builtin::BI__builtin_sinf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::sin,
                                   Intrinsic::experimental_constrained_sin));
 
    case Builtin::BIsqrt:
    case Builtin::BIsqrtf:
    case Builtin::BIsqrtl:
    case Builtin::BI__builtin_sqrt:
    case Builtin::BI__builtin_sqrtf:
    case Builtin::BI__builtin_sqrtf16:
    case Builtin::BI__builtin_sqrtl:
    case Builtin::BI__builtin_sqrtf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::sqrt,
                                   Intrinsic::experimental_constrained_sqrt));
 
    case Builtin::BItrunc:
    case Builtin::BItruncf:
    case Builtin::BItruncl:
    case Builtin::BI__builtin_trunc:
    case Builtin::BI__builtin_truncf:
    case Builtin::BI__builtin_truncf16:
    case Builtin::BI__builtin_truncl:
    case Builtin::BI__builtin_truncf128:
      return RValue::get(emitUnaryMaybeConstrainedFPBuiltin(*this, E,
                                   Intrinsic::trunc,
                                   Intrinsic::experimental_constrained_trunc));
 
    case Builtin::BIlround:
    case Builtin::BIlroundf:
    case Builtin::BIlroundl:
    case Builtin::BI__builtin_lround:
    case Builtin::BI__builtin_lroundf:
    case Builtin::BI__builtin_lroundl:
    case Builtin::BI__builtin_lroundf128:
      return RValue::get(emitMaybeConstrainedFPToIntRoundBuiltin(
          *this, E, Intrinsic::lround,
          Intrinsic::experimental_constrained_lround));
 
    case Builtin::BIllround:
    case Builtin::BIllroundf:
    case Builtin::BIllroundl:
    case Builtin::BI__builtin_llround:
    case Builtin::BI__builtin_llroundf:
    case Builtin::BI__builtin_llroundl:
    case Builtin::BI__builtin_llroundf128:
      return RValue::get(emitMaybeConstrainedFPToIntRoundBuiltin(
          *this, E, Intrinsic::llround,
          Intrinsic::experimental_constrained_llround));
 
    case Builtin::BIlrint:
    case Builtin::BIlrintf:
    case Builtin::BIlrintl:
    case Builtin::BI__builtin_lrint:
    case Builtin::BI__builtin_lrintf:
    case Builtin::BI__builtin_lrintl:
    case Builtin::BI__builtin_lrintf128:
      return RValue::get(emitMaybeConstrainedFPToIntRoundBuiltin(
          *this, E, Intrinsic::lrint,
          Intrinsic::experimental_constrained_lrint));
 
    case Builtin::BIllrint:
    case Builtin::BIllrintf:
    case Builtin::BIllrintl:
    case Builtin::BI__builtin_llrint:
    case Builtin::BI__builtin_llrintf:
    case Builtin::BI__builtin_llrintl:
    case Builtin::BI__builtin_llrintf128:
      return RValue::get(emitMaybeConstrainedFPToIntRoundBuiltin(
          *this, E, Intrinsic::llrint,
          Intrinsic::experimental_constrained_llrint));
 
    default:
      break;
    }
  }
 
  switch (BuiltinIDIfNoAsmLabel) {
  default: break;
  case Builtin::BI__builtin___CFStringMakeConstantString:
  case Builtin::BI__builtin___NSStringMakeConstantString:
    return RValue::get(ConstantEmitter(*this).emitAbstract(E, E->getType()));
  case Builtin::BI__builtin_stdarg_start:
  case Builtin::BI__builtin_va_start:
  case Builtin::BI__va_start:
  case Builtin::BI__builtin_va_end:
    EmitVAStartEnd(BuiltinID == Builtin::BI__va_start
                       ? EmitScalarExpr(E->getArg(0))
                       : EmitVAListRef(E->getArg(0)).getPointer(),
                   BuiltinID != Builtin::BI__builtin_va_end);
    return RValue::get(nullptr);
  case Builtin::BI__builtin_va_copy: {
    Value *DstPtr = EmitVAListRef(E->getArg(0)).getPointer();
    Value *SrcPtr = EmitVAListRef(E->getArg(1)).getPointer();
 
    llvm::Type *Type = Int8PtrTy;
 
    DstPtr = Builder.CreateBitCast(DstPtr, Type);
    SrcPtr = Builder.CreateBitCast(SrcPtr, Type);
    Builder.CreateCall(CGM.getIntrinsic(Intrinsic::vacopy), {DstPtr, SrcPtr});
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_abs:
  case Builtin::BI__builtin_labs:
  case Builtin::BI__builtin_llabs: {
    // X < 0 ? -X : X
    // The negation has 'nsw' because abs of INT_MIN is undefined.
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
    Value *NegOp = Builder.CreateNSWNeg(ArgValue, "neg");
    Constant *Zero = llvm::Constant::getNullValue(ArgValue->getType());
    Value *CmpResult = Builder.CreateICmpSLT(ArgValue, Zero, "abscond");
    Value *Result = Builder.CreateSelect(CmpResult, NegOp, ArgValue, "abs");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_complex: {
    Value *Real = EmitScalarExpr(E->getArg(0));
    Value *Imag = EmitScalarExpr(E->getArg(1));
    return RValue::getComplex({Real, Imag});
  }
  case Builtin::BI__builtin_conj:
  case Builtin::BI__builtin_conjf:
  case Builtin::BI__builtin_conjl:
  case Builtin::BIconj:
  case Builtin::BIconjf:
  case Builtin::BIconjl: {
    ComplexPairTy ComplexVal = EmitComplexExpr(E->getArg(0));
    Value *Real = ComplexVal.first;
    Value *Imag = ComplexVal.second;
    Imag = Builder.CreateFNeg(Imag, "neg");
    return RValue::getComplex(std::make_pair(Real, Imag));
  }
  case Builtin::BI__builtin_creal:
  case Builtin::BI__builtin_crealf:
  case Builtin::BI__builtin_creall:
  case Builtin::BIcreal:
  case Builtin::BIcrealf:
  case Builtin::BIcreall: {
    ComplexPairTy ComplexVal = EmitComplexExpr(E->getArg(0));
    return RValue::get(ComplexVal.first);
  }
 
  case Builtin::BI__builtin_preserve_access_index: {
    // Only enabled preserved access index region when debuginfo
    // is available as debuginfo is needed to preserve user-level
    // access pattern.
    if (!getDebugInfo()) {
      CGM.Error(E->getExprLoc(), "using builtin_preserve_access_index() without -g");
      return RValue::get(EmitScalarExpr(E->getArg(0)));
    }
 
    // Nested builtin_preserve_access_index() not supported
    if (IsInPreservedAIRegion) {
      CGM.Error(E->getExprLoc(), "nested builtin_preserve_access_index() not supported");
      return RValue::get(EmitScalarExpr(E->getArg(0)));
    }
 
    IsInPreservedAIRegion = true;
    Value *Res = EmitScalarExpr(E->getArg(0));
    IsInPreservedAIRegion = false;
    return RValue::get(Res);
  }
 
  case Builtin::BI__builtin_cimag:
  case Builtin::BI__builtin_cimagf:
  case Builtin::BI__builtin_cimagl:
  case Builtin::BIcimag:
  case Builtin::BIcimagf:
  case Builtin::BIcimagl: {
    ComplexPairTy ComplexVal = EmitComplexExpr(E->getArg(0));
    return RValue::get(ComplexVal.second);
  }
 
  case Builtin::BI__builtin_clrsb:
  case Builtin::BI__builtin_clrsbl:
  case Builtin::BI__builtin_clrsbll: {
    // clrsb(x) -> clz(x < 0 ? ~x : x) - 1 or
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::ctlz, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Zero = llvm::Constant::getNullValue(ArgType);
    Value *IsNeg = Builder.CreateICmpSLT(ArgValue, Zero, "isneg");
    Value *Inverse = Builder.CreateNot(ArgValue, "not");
    Value *Tmp = Builder.CreateSelect(IsNeg, Inverse, ArgValue);
    Value *Ctlz = Builder.CreateCall(F, {Tmp, Builder.getFalse()});
    Value *Result = Builder.CreateSub(Ctlz, llvm::ConstantInt::get(ArgType, 1));
    Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                   "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_ctzs:
  case Builtin::BI__builtin_ctz:
  case Builtin::BI__builtin_ctzl:
  case Builtin::BI__builtin_ctzll: {
    Value *ArgValue = EmitCheckedArgForBuiltin(E->getArg(0), BCK_CTZPassedZero);
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::cttz, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *ZeroUndef = Builder.getInt1(getTarget().isCLZForZeroUndef());
    Value *Result = Builder.CreateCall(F, {ArgValue, ZeroUndef});
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_clzs:
  case Builtin::BI__builtin_clz:
  case Builtin::BI__builtin_clzl:
  case Builtin::BI__builtin_clzll: {
    Value *ArgValue = EmitCheckedArgForBuiltin(E->getArg(0), BCK_CLZPassedZero);
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::ctlz, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *ZeroUndef = Builder.getInt1(getTarget().isCLZForZeroUndef());
    Value *Result = Builder.CreateCall(F, {ArgValue, ZeroUndef});
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_ffs:
  case Builtin::BI__builtin_ffsl:
  case Builtin::BI__builtin_ffsll: {
    // ffs(x) -> x ? cttz(x) + 1 : 0
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::cttz, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Tmp =
        Builder.CreateAdd(Builder.CreateCall(F, {ArgValue, Builder.getTrue()}),
                          llvm::ConstantInt::get(ArgType, 1));
    Value *Zero = llvm::Constant::getNullValue(ArgType);
    Value *IsZero = Builder.CreateICmpEQ(ArgValue, Zero, "iszero");
    Value *Result = Builder.CreateSelect(IsZero, Zero, Tmp, "ffs");
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_parity:
  case Builtin::BI__builtin_parityl:
  case Builtin::BI__builtin_parityll: {
    // parity(x) -> ctpop(x) & 1
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::ctpop, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Tmp = Builder.CreateCall(F, ArgValue);
    Value *Result = Builder.CreateAnd(Tmp, llvm::ConstantInt::get(ArgType, 1));
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__lzcnt16:
  case Builtin::BI__lzcnt:
  case Builtin::BI__lzcnt64: {
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::ctlz, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Result = Builder.CreateCall(F, {ArgValue, Builder.getFalse()});
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__popcnt16:
  case Builtin::BI__popcnt:
  case Builtin::BI__popcnt64:
  case Builtin::BI__builtin_popcount:
  case Builtin::BI__builtin_popcountl:
  case Builtin::BI__builtin_popcountll: {
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
    llvm::Type *ArgType = ArgValue->getType();
    Function *F = CGM.getIntrinsic(Intrinsic::ctpop, ArgType);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Result = Builder.CreateCall(F, ArgValue);
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_unpredictable: {
    // Always return the argument of __builtin_unpredictable. LLVM does not
    // handle this builtin. Metadata for this builtin should be added directly
    // to instructions such as branches or switches that use it.
    return RValue::get(EmitScalarExpr(E->getArg(0)));
  }
  case Builtin::BI__builtin_expect: {
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
    llvm::Type *ArgType = ArgValue->getType();
 
    Value *ExpectedValue = EmitScalarExpr(E->getArg(1));
    // Don't generate llvm.expect on -O0 as the backend won't use it for
    // anything.
    // Note, we still IRGen ExpectedValue because it could have side-effects.
    if (CGM.getCodeGenOpts().OptimizationLevel == 0)
      return RValue::get(ArgValue);
 
    Function *FnExpect = CGM.getIntrinsic(Intrinsic::expect, ArgType);
    Value *Result =
        Builder.CreateCall(FnExpect, {ArgValue, ExpectedValue}, "expval");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_expect_with_probability: {
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
    llvm::Type *ArgType = ArgValue->getType();
 
    Value *ExpectedValue = EmitScalarExpr(E->getArg(1));
    llvm::APFloat Probability(0.0);
    const Expr *ProbArg = E->getArg(2);
    bool EvalSucceed = ProbArg->EvaluateAsFloat(Probability, CGM.getContext());
    assert(EvalSucceed && "probability should be able to evaluate as float");
    (void)EvalSucceed;
    bool LoseInfo = false;
    Probability.convert(llvm::APFloat::IEEEdouble(),
                        llvm::RoundingMode::Dynamic, &LoseInfo);
    llvm::Type *Ty = ConvertType(ProbArg->getType());
    Constant *Confidence = ConstantFP::get(Ty, Probability);
    // Don't generate llvm.expect.with.probability on -O0 as the backend
    // won't use it for anything.
    // Note, we still IRGen ExpectedValue because it could have side-effects.
    if (CGM.getCodeGenOpts().OptimizationLevel == 0)
      return RValue::get(ArgValue);
 
    Function *FnExpect =
        CGM.getIntrinsic(Intrinsic::expect_with_probability, ArgType);
    Value *Result = Builder.CreateCall(
        FnExpect, {ArgValue, ExpectedValue, Confidence}, "expval");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_assume_aligned: {
    const Expr *Ptr = E->getArg(0);
    Value *PtrValue = EmitScalarExpr(Ptr);
    if (PtrValue->getType() != VoidPtrTy)
      PtrValue = EmitCastToVoidPtr(PtrValue);
    Value *OffsetValue =
      (E->getNumArgs() > 2) ? EmitScalarExpr(E->getArg(2)) : nullptr;
 
    Value *AlignmentValue = EmitScalarExpr(E->getArg(1));
    ConstantInt *AlignmentCI = cast<ConstantInt>(AlignmentValue);
    if (AlignmentCI->getValue().ugt(llvm::Value::MaximumAlignment))
      AlignmentCI = ConstantInt::get(AlignmentCI->getType(),
                                     llvm::Value::MaximumAlignment);
 
    emitAlignmentAssumption(PtrValue, Ptr,
                            /*The expr loc is sufficient.*/ SourceLocation(),
                            AlignmentCI, OffsetValue);
    return RValue::get(PtrValue);
  }
  case Builtin::BI__assume:
  case Builtin::BI__builtin_assume: {
    if (E->getArg(0)->HasSideEffects(getContext()))
      return RValue::get(nullptr);
 
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
    Function *FnAssume = CGM.getIntrinsic(Intrinsic::assume);
    Builder.CreateCall(FnAssume, ArgValue);
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_assume_separate_storage: {
    const Expr *Arg0 = E->getArg(0);
    const Expr *Arg1 = E->getArg(1);
 
    Value *Value0 = EmitScalarExpr(Arg0);
    Value *Value1 = EmitScalarExpr(Arg1);
 
    Value *Values[] = {Value0, Value1};
    OperandBundleDefT<Value *> OBD("separate_storage", Values);
    Builder.CreateAssumption(ConstantInt::getTrue(getLLVMContext()), {OBD});
    return RValue::get(nullptr);
  }
  case Builtin::BI__arithmetic_fence: {
    // Create the builtin call if FastMath is selected, and the target
    // supports the builtin, otherwise just return the argument.
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    llvm::FastMathFlags FMF = Builder.getFastMathFlags();
    bool isArithmeticFenceEnabled =
        FMF.allowReassoc() &&
        getContext().getTargetInfo().checkArithmeticFenceSupported();
    QualType ArgType = E->getArg(0)->getType();
    if (ArgType->isComplexType()) {
      if (isArithmeticFenceEnabled) {
        QualType ElementType = ArgType->castAs<ComplexType>()->getElementType();
        ComplexPairTy ComplexVal = EmitComplexExpr(E->getArg(0));
        Value *Real = Builder.CreateArithmeticFence(ComplexVal.first,
                                                    ConvertType(ElementType));
        Value *Imag = Builder.CreateArithmeticFence(ComplexVal.second,
                                                    ConvertType(ElementType));
        return RValue::getComplex(std::make_pair(Real, Imag));
      }
      ComplexPairTy ComplexVal = EmitComplexExpr(E->getArg(0));
      Value *Real = ComplexVal.first;
      Value *Imag = ComplexVal.second;
      return RValue::getComplex(std::make_pair(Real, Imag));
    }
    Value *ArgValue = EmitScalarExpr(E->getArg(0));
    if (isArithmeticFenceEnabled)
      return RValue::get(
          Builder.CreateArithmeticFence(ArgValue, ConvertType(ArgType)));
    return RValue::get(ArgValue);
  }
  case Builtin::BI__builtin_bswap16:
  case Builtin::BI__builtin_bswap32:
  case Builtin::BI__builtin_bswap64:
  case Builtin::BI_byteswap_ushort:
  case Builtin::BI_byteswap_ulong:
  case Builtin::BI_byteswap_uint64: {
    return RValue::get(emitUnaryBuiltin(*this, E, Intrinsic::bswap));
  }
  case Builtin::BI__builtin_bitreverse8:
  case Builtin::BI__builtin_bitreverse16:
  case Builtin::BI__builtin_bitreverse32:
  case Builtin::BI__builtin_bitreverse64: {
    return RValue::get(emitUnaryBuiltin(*this, E, Intrinsic::bitreverse));
  }
  case Builtin::BI__builtin_rotateleft8:
  case Builtin::BI__builtin_rotateleft16:
  case Builtin::BI__builtin_rotateleft32:
  case Builtin::BI__builtin_rotateleft64:
  case Builtin::BI_rotl8: // Microsoft variants of rotate left
  case Builtin::BI_rotl16:
  case Builtin::BI_rotl:
  case Builtin::BI_lrotl:
  case Builtin::BI_rotl64:
    return emitRotate(E, false);
 
  case Builtin::BI__builtin_rotateright8:
  case Builtin::BI__builtin_rotateright16:
  case Builtin::BI__builtin_rotateright32:
  case Builtin::BI__builtin_rotateright64:
  case Builtin::BI_rotr8: // Microsoft variants of rotate right
  case Builtin::BI_rotr16:
  case Builtin::BI_rotr:
  case Builtin::BI_lrotr:
  case Builtin::BI_rotr64:
    return emitRotate(E, true);
 
  case Builtin::BI__builtin_constant_p: {
    llvm::Type *ResultType = ConvertType(E->getType());
 
    const Expr *Arg = E->getArg(0);
    QualType ArgType = Arg->getType();
    // FIXME: The allowance for Obj-C pointers and block pointers is historical
    // and likely a mistake.
    if (!ArgType->isIntegralOrEnumerationType() && !ArgType->isFloatingType() &&
        !ArgType->isObjCObjectPointerType() && !ArgType->isBlockPointerType())
      // Per the GCC documentation, only numeric constants are recognized after
      // inlining.
      return RValue::get(ConstantInt::get(ResultType, 0));
 
    if (Arg->HasSideEffects(getContext()))
      // The argument is unevaluated, so be conservative if it might have
      // side-effects.
      return RValue::get(ConstantInt::get(ResultType, 0));
 
    Value *ArgValue = EmitScalarExpr(Arg);
    if (ArgType->isObjCObjectPointerType()) {
      // Convert Objective-C objects to id because we cannot distinguish between
      // LLVM types for Obj-C classes as they are opaque.
      ArgType = CGM.getContext().getObjCIdType();
      ArgValue = Builder.CreateBitCast(ArgValue, ConvertType(ArgType));
    }
    Function *F =
        CGM.getIntrinsic(Intrinsic::is_constant, ConvertType(ArgType));
    Value *Result = Builder.CreateCall(F, ArgValue);
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/false);
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_dynamic_object_size:
  case Builtin::BI__builtin_object_size: {
    unsigned Type =
        E->getArg(1)->EvaluateKnownConstInt(getContext()).getZExtValue();
    auto *ResType = cast<llvm::IntegerType>(ConvertType(E->getType()));
 
    // We pass this builtin onto the optimizer so that it can figure out the
    // object size in more complex cases.
    bool IsDynamic = BuiltinID == Builtin::BI__builtin_dynamic_object_size;
    return RValue::get(emitBuiltinObjectSize(E->getArg(0), Type, ResType,
                                             /*EmittedE=*/nullptr, IsDynamic));
  }
  case Builtin::BI__builtin_prefetch: {
    Value *Locality, *RW, *Address = EmitScalarExpr(E->getArg(0));
    // FIXME: Technically these constants should of type 'int', yes?
    RW = (E->getNumArgs() > 1) ? EmitScalarExpr(E->getArg(1)) :
      llvm::ConstantInt::get(Int32Ty, 0);
    Locality = (E->getNumArgs() > 2) ? EmitScalarExpr(E->getArg(2)) :
      llvm::ConstantInt::get(Int32Ty, 3);
    Value *Data = llvm::ConstantInt::get(Int32Ty, 1);
    Function *F = CGM.getIntrinsic(Intrinsic::prefetch, Address->getType());
    Builder.CreateCall(F, {Address, RW, Locality, Data});
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_readcyclecounter: {
    Function *F = CGM.getIntrinsic(Intrinsic::readcyclecounter);
    return RValue::get(Builder.CreateCall(F));
  }
  case Builtin::BI__builtin___clear_cache: {
    Value *Begin = EmitScalarExpr(E->getArg(0));
    Value *End = EmitScalarExpr(E->getArg(1));
    Function *F = CGM.getIntrinsic(Intrinsic::clear_cache);
    return RValue::get(Builder.CreateCall(F, {Begin, End}));
  }
  case Builtin::BI__builtin_trap:
    EmitTrapCall(Intrinsic::trap);
    return RValue::get(nullptr);
  case Builtin::BI__debugbreak:
    EmitTrapCall(Intrinsic::debugtrap);
    return RValue::get(nullptr);
  case Builtin::BI__builtin_unreachable: {
    EmitUnreachable(E->getExprLoc());
 
    // We do need to preserve an insertion point.
    EmitBlock(createBasicBlock("unreachable.cont"));
 
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__builtin_powi:
  case Builtin::BI__builtin_powif:
  case Builtin::BI__builtin_powil: {
    llvm::Value *Src0 = EmitScalarExpr(E->getArg(0));
    llvm::Value *Src1 = EmitScalarExpr(E->getArg(1));
 
    if (Builder.getIsFPConstrained()) {
      CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
      Function *F = CGM.getIntrinsic(Intrinsic::experimental_constrained_powi,
                                     Src0->getType());
      return RValue::get(Builder.CreateConstrainedFPCall(F, { Src0, Src1 }));
    }
 
    Function *F = CGM.getIntrinsic(Intrinsic::powi,
                                   { Src0->getType(), Src1->getType() });
    return RValue::get(Builder.CreateCall(F, { Src0, Src1 }));
  }
  case Builtin::BI__builtin_isgreater:
  case Builtin::BI__builtin_isgreaterequal:
  case Builtin::BI__builtin_isless:
  case Builtin::BI__builtin_islessequal:
  case Builtin::BI__builtin_islessgreater:
  case Builtin::BI__builtin_isunordered: {
    // Ordered comparisons: we know the arguments to these are matching scalar
    // floating point values.
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    Value *LHS = EmitScalarExpr(E->getArg(0));
    Value *RHS = EmitScalarExpr(E->getArg(1));
 
    switch (BuiltinID) {
    default: llvm_unreachable("Unknown ordered comparison");
    case Builtin::BI__builtin_isgreater:
      LHS = Builder.CreateFCmpOGT(LHS, RHS, "cmp");
      break;
    case Builtin::BI__builtin_isgreaterequal:
      LHS = Builder.CreateFCmpOGE(LHS, RHS, "cmp");
      break;
    case Builtin::BI__builtin_isless:
      LHS = Builder.CreateFCmpOLT(LHS, RHS, "cmp");
      break;
    case Builtin::BI__builtin_islessequal:
      LHS = Builder.CreateFCmpOLE(LHS, RHS, "cmp");
      break;
    case Builtin::BI__builtin_islessgreater:
      LHS = Builder.CreateFCmpONE(LHS, RHS, "cmp");
      break;
    case Builtin::BI__builtin_isunordered:
      LHS = Builder.CreateFCmpUNO(LHS, RHS, "cmp");
      break;
    }
    // ZExt bool to int type.
    return RValue::get(Builder.CreateZExt(LHS, ConvertType(E->getType())));
  }
  case Builtin::BI__builtin_isnan: {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    Value *V = EmitScalarExpr(E->getArg(0));
    llvm::Type *Ty = V->getType();
    const llvm::fltSemantics &Semantics = Ty->getFltSemantics();
    if (!Builder.getIsFPConstrained() ||
        Builder.getDefaultConstrainedExcept() == fp::ebIgnore ||
        !Ty->isIEEE()) {
      V = Builder.CreateFCmpUNO(V, V, "cmp");
      return RValue::get(Builder.CreateZExt(V, ConvertType(E->getType())));
    }
 
    if (Value *Result = getTargetHooks().testFPKind(V, BuiltinID, Builder, CGM))
      return RValue::get(Result);
 
    // NaN has all exp bits set and a non zero significand. Therefore:
    // isnan(V) == ((exp mask - (abs(V) & exp mask)) < 0)
    unsigned bitsize = Ty->getScalarSizeInBits();
    llvm::IntegerType *IntTy = Builder.getIntNTy(bitsize);
    Value *IntV = Builder.CreateBitCast(V, IntTy);
    APInt AndMask = APInt::getSignedMaxValue(bitsize);
    Value *AbsV =
        Builder.CreateAnd(IntV, llvm::ConstantInt::get(IntTy, AndMask));
    APInt ExpMask = APFloat::getInf(Semantics).bitcastToAPInt();
    Value *Sub =
        Builder.CreateSub(llvm::ConstantInt::get(IntTy, ExpMask), AbsV);
    // V = sign bit (Sub) <=> V = (Sub < 0)
    V = Builder.CreateLShr(Sub, llvm::ConstantInt::get(IntTy, bitsize - 1));
    if (bitsize > 32)
      V = Builder.CreateTrunc(V, ConvertType(E->getType()));
    return RValue::get(V);
  }
 
  case Builtin::BI__builtin_nondeterministic_value: {
    llvm::Type *Ty = ConvertType(E->getArg(0)->getType());
 
    Value *Result = PoisonValue::get(Ty);
    Result = Builder.CreateFreeze(Result);
 
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_elementwise_abs: {
    Value *Result;
    QualType QT = E->getArg(0)->getType();
 
    if (auto *VecTy = QT->getAs<VectorType>())
      QT = VecTy->getElementType();
    if (QT->isIntegerType())
      Result = Builder.CreateBinaryIntrinsic(
          llvm::Intrinsic::abs, EmitScalarExpr(E->getArg(0)),
          Builder.getFalse(), nullptr, "elt.abs");
    else
      Result = emitUnaryBuiltin(*this, E, llvm::Intrinsic::fabs, "elt.abs");
 
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_elementwise_ceil:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::ceil, "elt.ceil"));
  case Builtin::BI__builtin_elementwise_exp:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::exp, "elt.exp"));
  case Builtin::BI__builtin_elementwise_exp2:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::exp2, "elt.exp2"));
  case Builtin::BI__builtin_elementwise_log:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::log, "elt.log"));
  case Builtin::BI__builtin_elementwise_log2:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::log2, "elt.log2"));
  case Builtin::BI__builtin_elementwise_log10:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::log10, "elt.log10"));
  case Builtin::BI__builtin_elementwise_cos:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::cos, "elt.cos"));
  case Builtin::BI__builtin_elementwise_floor:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::floor, "elt.floor"));
  case Builtin::BI__builtin_elementwise_roundeven:
    return RValue::get(emitUnaryBuiltin(*this, E, llvm::Intrinsic::roundeven,
                                        "elt.roundeven"));
  case Builtin::BI__builtin_elementwise_sin:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::sin, "elt.sin"));
 
  case Builtin::BI__builtin_elementwise_trunc:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::trunc, "elt.trunc"));
  case Builtin::BI__builtin_elementwise_canonicalize:
    return RValue::get(
        emitUnaryBuiltin(*this, E, llvm::Intrinsic::canonicalize, "elt.trunc"));
  case Builtin::BI__builtin_elementwise_copysign:
    return RValue::get(emitBinaryBuiltin(*this, E, llvm::Intrinsic::copysign));
  case Builtin::BI__builtin_elementwise_fma:
    return RValue::get(emitTernaryBuiltin(*this, E, llvm::Intrinsic::fma));
  case Builtin::BI__builtin_elementwise_add_sat:
  case Builtin::BI__builtin_elementwise_sub_sat: {
    Value *Op0 = EmitScalarExpr(E->getArg(0));
    Value *Op1 = EmitScalarExpr(E->getArg(1));
    Value *Result;
    assert(Op0->getType()->isIntOrIntVectorTy() && "integer type expected");
    QualType Ty = E->getArg(0)->getType();
    if (auto *VecTy = Ty->getAs<VectorType>())
      Ty = VecTy->getElementType();
    bool IsSigned = Ty->isSignedIntegerType();
    unsigned Opc;
    if (BuiltinIDIfNoAsmLabel == Builtin::BI__builtin_elementwise_add_sat)
      Opc = IsSigned ? llvm::Intrinsic::sadd_sat : llvm::Intrinsic::uadd_sat;
    else
      Opc = IsSigned ? llvm::Intrinsic::ssub_sat : llvm::Intrinsic::usub_sat;
    Result = Builder.CreateBinaryIntrinsic(Opc, Op0, Op1, nullptr, "elt.sat");
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_elementwise_max: {
    Value *Op0 = EmitScalarExpr(E->getArg(0));
    Value *Op1 = EmitScalarExpr(E->getArg(1));
    Value *Result;
    if (Op0->getType()->isIntOrIntVectorTy()) {
      QualType Ty = E->getArg(0)->getType();
      if (auto *VecTy = Ty->getAs<VectorType>())
        Ty = VecTy->getElementType();
      Result = Builder.CreateBinaryIntrinsic(Ty->isSignedIntegerType()
                                                 ? llvm::Intrinsic::smax
                                                 : llvm::Intrinsic::umax,
                                             Op0, Op1, nullptr, "elt.max");
    } else
      Result = Builder.CreateMaxNum(Op0, Op1, "elt.max");
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_elementwise_min: {
    Value *Op0 = EmitScalarExpr(E->getArg(0));
    Value *Op1 = EmitScalarExpr(E->getArg(1));
    Value *Result;
    if (Op0->getType()->isIntOrIntVectorTy()) {
      QualType Ty = E->getArg(0)->getType();
      if (auto *VecTy = Ty->getAs<VectorType>())
        Ty = VecTy->getElementType();
      Result = Builder.CreateBinaryIntrinsic(Ty->isSignedIntegerType()
                                                 ? llvm::Intrinsic::smin
                                                 : llvm::Intrinsic::umin,
                                             Op0, Op1, nullptr, "elt.min");
    } else
      Result = Builder.CreateMinNum(Op0, Op1, "elt.min");
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_reduce_max: {
    auto GetIntrinsicID = [](QualType QT) {
      if (auto *VecTy = QT->getAs<VectorType>())
        QT = VecTy->getElementType();
      if (QT->isSignedIntegerType())
        return llvm::Intrinsic::vector_reduce_smax;
      if (QT->isUnsignedIntegerType())
        return llvm::Intrinsic::vector_reduce_umax;
      assert(QT->isFloatingType() && "must have a float here");
      return llvm::Intrinsic::vector_reduce_fmax;
    };
    return RValue::get(emitUnaryBuiltin(
        *this, E, GetIntrinsicID(E->getArg(0)->getType()), "rdx.min"));
  }
 
  case Builtin::BI__builtin_reduce_min: {
    auto GetIntrinsicID = [](QualType QT) {
      if (auto *VecTy = QT->getAs<VectorType>())
        QT = VecTy->getElementType();
      if (QT->isSignedIntegerType())
        return llvm::Intrinsic::vector_reduce_smin;
      if (QT->isUnsignedIntegerType())
        return llvm::Intrinsic::vector_reduce_umin;
      assert(QT->isFloatingType() && "must have a float here");
      return llvm::Intrinsic::vector_reduce_fmin;
    };
 
    return RValue::get(emitUnaryBuiltin(
        *this, E, GetIntrinsicID(E->getArg(0)->getType()), "rdx.min"));
  }
 
  case Builtin::BI__builtin_reduce_add:
    return RValue::get(emitUnaryBuiltin(
        *this, E, llvm::Intrinsic::vector_reduce_add, "rdx.add"));
  case Builtin::BI__builtin_reduce_mul:
    return RValue::get(emitUnaryBuiltin(
        *this, E, llvm::Intrinsic::vector_reduce_mul, "rdx.mul"));
  case Builtin::BI__builtin_reduce_xor:
    return RValue::get(emitUnaryBuiltin(
        *this, E, llvm::Intrinsic::vector_reduce_xor, "rdx.xor"));
  case Builtin::BI__builtin_reduce_or:
    return RValue::get(emitUnaryBuiltin(
        *this, E, llvm::Intrinsic::vector_reduce_or, "rdx.or"));
  case Builtin::BI__builtin_reduce_and:
    return RValue::get(emitUnaryBuiltin(
        *this, E, llvm::Intrinsic::vector_reduce_and, "rdx.and"));
 
  case Builtin::BI__builtin_matrix_transpose: {
    auto *MatrixTy = E->getArg(0)->getType()->castAs<ConstantMatrixType>();
    Value *MatValue = EmitScalarExpr(E->getArg(0));
    MatrixBuilder MB(Builder);
    Value *Result = MB.CreateMatrixTranspose(MatValue, MatrixTy->getNumRows(),
                                             MatrixTy->getNumColumns());
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_matrix_column_major_load: {
    MatrixBuilder MB(Builder);
    // Emit everything that isn't dependent on the first parameter type
    Value *Stride = EmitScalarExpr(E->getArg(3));
    const auto *ResultTy = E->getType()->getAs<ConstantMatrixType>();
    auto *PtrTy = E->getArg(0)->getType()->getAs<PointerType>();
    assert(PtrTy && "arg0 must be of pointer type");
    bool IsVolatile = PtrTy->getPointeeType().isVolatileQualified();
 
    Address Src = EmitPointerWithAlignment(E->getArg(0));
    EmitNonNullArgCheck(RValue::get(Src.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    Value *Result = MB.CreateColumnMajorLoad(
        Src.getElementType(), Src.getPointer(),
        Align(Src.getAlignment().getQuantity()), Stride, IsVolatile,
        ResultTy->getNumRows(), ResultTy->getNumColumns(),
        "matrix");
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_matrix_column_major_store: {
    MatrixBuilder MB(Builder);
    Value *Matrix = EmitScalarExpr(E->getArg(0));
    Address Dst = EmitPointerWithAlignment(E->getArg(1));
    Value *Stride = EmitScalarExpr(E->getArg(2));
 
    const auto *MatrixTy = E->getArg(0)->getType()->getAs<ConstantMatrixType>();
    auto *PtrTy = E->getArg(1)->getType()->getAs<PointerType>();
    assert(PtrTy && "arg1 must be of pointer type");
    bool IsVolatile = PtrTy->getPointeeType().isVolatileQualified();
 
    EmitNonNullArgCheck(RValue::get(Dst.getPointer()), E->getArg(1)->getType(),
                        E->getArg(1)->getExprLoc(), FD, 0);
    Value *Result = MB.CreateColumnMajorStore(
        Matrix, Dst.getPointer(), Align(Dst.getAlignment().getQuantity()),
        Stride, IsVolatile, MatrixTy->getNumRows(), MatrixTy->getNumColumns());
    return RValue::get(Result);
  }
 
  case Builtin::BIfinite:
  case Builtin::BI__finite:
  case Builtin::BIfinitef:
  case Builtin::BI__finitef:
  case Builtin::BIfinitel:
  case Builtin::BI__finitel:
  case Builtin::BI__builtin_isinf:
  case Builtin::BI__builtin_isfinite: {
    // isinf(x)    --> fabs(x) == infinity
    // isfinite(x) --> fabs(x) != infinity
    // x != NaN via the ordered compare in either case.
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    Value *V = EmitScalarExpr(E->getArg(0));
    llvm::Type *Ty = V->getType();
    if (!Builder.getIsFPConstrained() ||
        Builder.getDefaultConstrainedExcept() == fp::ebIgnore ||
        !Ty->isIEEE()) {
      Value *Fabs = EmitFAbs(*this, V);
      Constant *Infinity = ConstantFP::getInfinity(V->getType());
      CmpInst::Predicate Pred = (BuiltinID == Builtin::BI__builtin_isinf)
                                    ? CmpInst::FCMP_OEQ
                                    : CmpInst::FCMP_ONE;
      Value *FCmp = Builder.CreateFCmp(Pred, Fabs, Infinity, "cmpinf");
      return RValue::get(Builder.CreateZExt(FCmp, ConvertType(E->getType())));
    }
 
    if (Value *Result = getTargetHooks().testFPKind(V, BuiltinID, Builder, CGM))
      return RValue::get(Result);
 
    // Inf values have all exp bits set and a zero significand. Therefore:
    // isinf(V) == ((V << 1) == ((exp mask) << 1))
    // isfinite(V) == ((V << 1) < ((exp mask) << 1)) using unsigned comparison
    unsigned bitsize = Ty->getScalarSizeInBits();
    llvm::IntegerType *IntTy = Builder.getIntNTy(bitsize);
    Value *IntV = Builder.CreateBitCast(V, IntTy);
    Value *Shl1 = Builder.CreateShl(IntV, 1);
    const llvm::fltSemantics &Semantics = Ty->getFltSemantics();
    APInt ExpMask = APFloat::getInf(Semantics).bitcastToAPInt();
    Value *ExpMaskShl1 = llvm::ConstantInt::get(IntTy, ExpMask.shl(1));
    if (BuiltinID == Builtin::BI__builtin_isinf)
      V = Builder.CreateICmpEQ(Shl1, ExpMaskShl1);
    else
      V = Builder.CreateICmpULT(Shl1, ExpMaskShl1);
    return RValue::get(Builder.CreateZExt(V, ConvertType(E->getType())));
  }
 
  case Builtin::BI__builtin_isinf_sign: {
    // isinf_sign(x) -> fabs(x) == infinity ? (signbit(x) ? -1 : 1) : 0
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    // FIXME: for strictfp/IEEE-754 we need to not trap on SNaN here.
    Value *Arg = EmitScalarExpr(E->getArg(0));
    Value *AbsArg = EmitFAbs(*this, Arg);
    Value *IsInf = Builder.CreateFCmpOEQ(
        AbsArg, ConstantFP::getInfinity(Arg->getType()), "isinf");
    Value *IsNeg = EmitSignBit(*this, Arg);
 
    llvm::Type *IntTy = ConvertType(E->getType());
    Value *Zero = Constant::getNullValue(IntTy);
    Value *One = ConstantInt::get(IntTy, 1);
    Value *NegativeOne = ConstantInt::get(IntTy, -1);
    Value *SignResult = Builder.CreateSelect(IsNeg, NegativeOne, One);
    Value *Result = Builder.CreateSelect(IsInf, SignResult, Zero);
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_isnormal: {
    // isnormal(x) --> x == x && fabsf(x) < infinity && fabsf(x) >= float_min
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    // FIXME: for strictfp/IEEE-754 we need to not trap on SNaN here.
    Value *V = EmitScalarExpr(E->getArg(0));
    Value *Eq = Builder.CreateFCmpOEQ(V, V, "iseq");
 
    Value *Abs = EmitFAbs(*this, V);
    Value *IsLessThanInf =
      Builder.CreateFCmpULT(Abs, ConstantFP::getInfinity(V->getType()),"isinf");
    APFloat Smallest = APFloat::getSmallestNormalized(
                   getContext().getFloatTypeSemantics(E->getArg(0)->getType()));
    Value *IsNormal =
      Builder.CreateFCmpUGE(Abs, ConstantFP::get(V->getContext(), Smallest),
                            "isnormal");
    V = Builder.CreateAnd(Eq, IsLessThanInf, "and");
    V = Builder.CreateAnd(V, IsNormal, "and");
    return RValue::get(Builder.CreateZExt(V, ConvertType(E->getType())));
  }
 
  case Builtin::BI__builtin_flt_rounds: {
    Function *F = CGM.getIntrinsic(Intrinsic::get_rounding);
 
    llvm::Type *ResultType = ConvertType(E->getType());
    Value *Result = Builder.CreateCall(F);
    if (Result->getType() != ResultType)
      Result = Builder.CreateIntCast(Result, ResultType, /*isSigned*/true,
                                     "cast");
    return RValue::get(Result);
  }
 
  case Builtin::BI__builtin_set_flt_rounds: {
    Function *F = CGM.getIntrinsic(Intrinsic::set_rounding);
 
    Value *V = EmitScalarExpr(E->getArg(0));
    Builder.CreateCall(F, V);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__builtin_fpclassify: {
    CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
    // FIXME: for strictfp/IEEE-754 we need to not trap on SNaN here.
    Value *V = EmitScalarExpr(E->getArg(5));
    llvm::Type *Ty = ConvertType(E->getArg(5)->getType());
 
    // Create Result
    BasicBlock *Begin = Builder.GetInsertBlock();
    BasicBlock *End = createBasicBlock("fpclassify_end", this->CurFn);
    Builder.SetInsertPoint(End);
    PHINode *Result =
      Builder.CreatePHI(ConvertType(E->getArg(0)->getType()), 4,
                        "fpclassify_result");
 
    // if (V==0) return FP_ZERO
    Builder.SetInsertPoint(Begin);
    Value *IsZero = Builder.CreateFCmpOEQ(V, Constant::getNullValue(Ty),
                                          "iszero");
    Value *ZeroLiteral = EmitScalarExpr(E->getArg(4));
    BasicBlock *NotZero = createBasicBlock("fpclassify_not_zero", this->CurFn);
    Builder.CreateCondBr(IsZero, End, NotZero);
    Result->addIncoming(ZeroLiteral, Begin);
 
    // if (V != V) return FP_NAN
    Builder.SetInsertPoint(NotZero);
    Value *IsNan = Builder.CreateFCmpUNO(V, V, "cmp");
    Value *NanLiteral = EmitScalarExpr(E->getArg(0));
    BasicBlock *NotNan = createBasicBlock("fpclassify_not_nan", this->CurFn);
    Builder.CreateCondBr(IsNan, End, NotNan);
    Result->addIncoming(NanLiteral, NotZero);
 
    // if (fabs(V) == infinity) return FP_INFINITY
    Builder.SetInsertPoint(NotNan);
    Value *VAbs = EmitFAbs(*this, V);
    Value *IsInf =
      Builder.CreateFCmpOEQ(VAbs, ConstantFP::getInfinity(V->getType()),
                            "isinf");
    Value *InfLiteral = EmitScalarExpr(E->getArg(1));
    BasicBlock *NotInf = createBasicBlock("fpclassify_not_inf", this->CurFn);
    Builder.CreateCondBr(IsInf, End, NotInf);
    Result->addIncoming(InfLiteral, NotNan);
 
    // if (fabs(V) >= MIN_NORMAL) return FP_NORMAL else FP_SUBNORMAL
    Builder.SetInsertPoint(NotInf);
    APFloat Smallest = APFloat::getSmallestNormalized(
        getContext().getFloatTypeSemantics(E->getArg(5)->getType()));
    Value *IsNormal =
      Builder.CreateFCmpUGE(VAbs, ConstantFP::get(V->getContext(), Smallest),
                            "isnormal");
    Value *NormalResult =
      Builder.CreateSelect(IsNormal, EmitScalarExpr(E->getArg(2)),
                           EmitScalarExpr(E->getArg(3)));
    Builder.CreateBr(End);
    Result->addIncoming(NormalResult, NotInf);
 
    // return Result
    Builder.SetInsertPoint(End);
    return RValue::get(Result);
  }
 
  case Builtin::BIalloca:
  case Builtin::BI_alloca:
  case Builtin::BI__builtin_alloca_uninitialized:
  case Builtin::BI__builtin_alloca: {
    Value *Size = EmitScalarExpr(E->getArg(0));
    const TargetInfo &TI = getContext().getTargetInfo();
    // The alignment of the alloca should correspond to __BIGGEST_ALIGNMENT__.
    const Align SuitableAlignmentInBytes =
        CGM.getContext()
            .toCharUnitsFromBits(TI.getSuitableAlign())
            .getAsAlign();
    AllocaInst *AI = Builder.CreateAlloca(Builder.getInt8Ty(), Size);
    AI->setAlignment(SuitableAlignmentInBytes);
    if (BuiltinID != Builtin::BI__builtin_alloca_uninitialized)
      initializeAlloca(*this, AI, Size, SuitableAlignmentInBytes);
    return RValue::get(AI);
  }
 
  case Builtin::BI__builtin_alloca_with_align_uninitialized:
  case Builtin::BI__builtin_alloca_with_align: {
    Value *Size = EmitScalarExpr(E->getArg(0));
    Value *AlignmentInBitsValue = EmitScalarExpr(E->getArg(1));
    auto *AlignmentInBitsCI = cast<ConstantInt>(AlignmentInBitsValue);
    unsigned AlignmentInBits = AlignmentInBitsCI->getZExtValue();
    const Align AlignmentInBytes =
        CGM.getContext().toCharUnitsFromBits(AlignmentInBits).getAsAlign();
    AllocaInst *AI = Builder.CreateAlloca(Builder.getInt8Ty(), Size);
    AI->setAlignment(AlignmentInBytes);
    if (BuiltinID != Builtin::BI__builtin_alloca_with_align_uninitialized)
      initializeAlloca(*this, AI, Size, AlignmentInBytes);
    return RValue::get(AI);
  }
 
  case Builtin::BIbzero:
  case Builtin::BI__builtin_bzero: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Value *SizeVal = EmitScalarExpr(E->getArg(1));
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    Builder.CreateMemSet(Dest, Builder.getInt8(0), SizeVal, false);
    return RValue::get(nullptr);
  }
  case Builtin::BImemcpy:
  case Builtin::BI__builtin_memcpy:
  case Builtin::BImempcpy:
  case Builtin::BI__builtin_mempcpy: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Address Src = EmitPointerWithAlignment(E->getArg(1));
    Value *SizeVal = EmitScalarExpr(E->getArg(2));
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    EmitNonNullArgCheck(RValue::get(Src.getPointer()), E->getArg(1)->getType(),
                        E->getArg(1)->getExprLoc(), FD, 1);
    Builder.CreateMemCpy(Dest, Src, SizeVal, false);
    if (BuiltinID == Builtin::BImempcpy ||
        BuiltinID == Builtin::BI__builtin_mempcpy)
      return RValue::get(Builder.CreateInBoundsGEP(Dest.getElementType(),
                                                   Dest.getPointer(), SizeVal));
    else
      return RValue::get(Dest.getPointer());
  }
 
  case Builtin::BI__builtin_memcpy_inline: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Address Src = EmitPointerWithAlignment(E->getArg(1));
    uint64_t Size =
        E->getArg(2)->EvaluateKnownConstInt(getContext()).getZExtValue();
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    EmitNonNullArgCheck(RValue::get(Src.getPointer()), E->getArg(1)->getType(),
                        E->getArg(1)->getExprLoc(), FD, 1);
    Builder.CreateMemCpyInline(Dest, Src, Size);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__builtin_char_memchr:
    BuiltinID = Builtin::BI__builtin_memchr;
    break;
 
  case Builtin::BI__builtin___memcpy_chk: {
    // fold __builtin_memcpy_chk(x, y, cst1, cst2) to memcpy iff cst1<=cst2.
    Expr::EvalResult SizeResult, DstSizeResult;
    if (!E->getArg(2)->EvaluateAsInt(SizeResult, CGM.getContext()) ||
        !E->getArg(3)->EvaluateAsInt(DstSizeResult, CGM.getContext()))
      break;
    llvm::APSInt Size = SizeResult.Val.getInt();
    llvm::APSInt DstSize = DstSizeResult.Val.getInt();
    if (Size.ugt(DstSize))
      break;
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Address Src = EmitPointerWithAlignment(E->getArg(1));
    Value *SizeVal = llvm::ConstantInt::get(Builder.getContext(), Size);
    Builder.CreateMemCpy(Dest, Src, SizeVal, false);
    return RValue::get(Dest.getPointer());
  }
 
  case Builtin::BI__builtin_objc_memmove_collectable: {
    Address DestAddr = EmitPointerWithAlignment(E->getArg(0));
    Address SrcAddr = EmitPointerWithAlignment(E->getArg(1));
    Value *SizeVal = EmitScalarExpr(E->getArg(2));
    CGM.getObjCRuntime().EmitGCMemmoveCollectable(*this,
                                                  DestAddr, SrcAddr, SizeVal);
    return RValue::get(DestAddr.getPointer());
  }
 
  case Builtin::BI__builtin___memmove_chk: {
    // fold __builtin_memmove_chk(x, y, cst1, cst2) to memmove iff cst1<=cst2.
    Expr::EvalResult SizeResult, DstSizeResult;
    if (!E->getArg(2)->EvaluateAsInt(SizeResult, CGM.getContext()) ||
        !E->getArg(3)->EvaluateAsInt(DstSizeResult, CGM.getContext()))
      break;
    llvm::APSInt Size = SizeResult.Val.getInt();
    llvm::APSInt DstSize = DstSizeResult.Val.getInt();
    if (Size.ugt(DstSize))
      break;
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Address Src = EmitPointerWithAlignment(E->getArg(1));
    Value *SizeVal = llvm::ConstantInt::get(Builder.getContext(), Size);
    Builder.CreateMemMove(Dest, Src, SizeVal, false);
    return RValue::get(Dest.getPointer());
  }
 
  case Builtin::BImemmove:
  case Builtin::BI__builtin_memmove: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Address Src = EmitPointerWithAlignment(E->getArg(1));
    Value *SizeVal = EmitScalarExpr(E->getArg(2));
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    EmitNonNullArgCheck(RValue::get(Src.getPointer()), E->getArg(1)->getType(),
                        E->getArg(1)->getExprLoc(), FD, 1);
    Builder.CreateMemMove(Dest, Src, SizeVal, false);
    return RValue::get(Dest.getPointer());
  }
  case Builtin::BImemset:
  case Builtin::BI__builtin_memset: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Value *ByteVal = Builder.CreateTrunc(EmitScalarExpr(E->getArg(1)),
                                         Builder.getInt8Ty());
    Value *SizeVal = EmitScalarExpr(E->getArg(2));
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    Builder.CreateMemSet(Dest, ByteVal, SizeVal, false);
    return RValue::get(Dest.getPointer());
  }
  case Builtin::BI__builtin_memset_inline: {
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Value *ByteVal =
        Builder.CreateTrunc(EmitScalarExpr(E->getArg(1)), Builder.getInt8Ty());
    uint64_t Size =
        E->getArg(2)->EvaluateKnownConstInt(getContext()).getZExtValue();
    EmitNonNullArgCheck(RValue::get(Dest.getPointer()), E->getArg(0)->getType(),
                        E->getArg(0)->getExprLoc(), FD, 0);
    Builder.CreateMemSetInline(Dest, ByteVal, Size);
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin___memset_chk: {
    // fold __builtin_memset_chk(x, y, cst1, cst2) to memset iff cst1<=cst2.
    Expr::EvalResult SizeResult, DstSizeResult;
    if (!E->getArg(2)->EvaluateAsInt(SizeResult, CGM.getContext()) ||
        !E->getArg(3)->EvaluateAsInt(DstSizeResult, CGM.getContext()))
      break;
    llvm::APSInt Size = SizeResult.Val.getInt();
    llvm::APSInt DstSize = DstSizeResult.Val.getInt();
    if (Size.ugt(DstSize))
      break;
    Address Dest = EmitPointerWithAlignment(E->getArg(0));
    Value *ByteVal = Builder.CreateTrunc(EmitScalarExpr(E->getArg(1)),
                                         Builder.getInt8Ty());
    Value *SizeVal = llvm::ConstantInt::get(Builder.getContext(), Size);
    Builder.CreateMemSet(Dest, ByteVal, SizeVal, false);
    return RValue::get(Dest.getPointer());
  }
  case Builtin::BI__builtin_wmemchr: {
    // The MSVC runtime library does not provide a definition of wmemchr, so we
    // need an inline implementation.
    if (!getTarget().getTriple().isOSMSVCRT())
      break;
 
    llvm::Type *WCharTy = ConvertType(getContext().WCharTy);
    Value *Str = EmitScalarExpr(E->getArg(0));
    Value *Chr = EmitScalarExpr(E->getArg(1));
    Value *Size = EmitScalarExpr(E->getArg(2));
 
    BasicBlock *Entry = Builder.GetInsertBlock();
    BasicBlock *CmpEq = createBasicBlock("wmemchr.eq");
    BasicBlock *Next = createBasicBlock("wmemchr.next");
    BasicBlock *Exit = createBasicBlock("wmemchr.exit");
    Value *SizeEq0 = Builder.CreateICmpEQ(Size, ConstantInt::get(SizeTy, 0));
    Builder.CreateCondBr(SizeEq0, Exit, CmpEq);
 
    EmitBlock(CmpEq);
    PHINode *StrPhi = Builder.CreatePHI(Str->getType(), 2);
    StrPhi->addIncoming(Str, Entry);
    PHINode *SizePhi = Builder.CreatePHI(SizeTy, 2);
    SizePhi->addIncoming(Size, Entry);
    CharUnits WCharAlign =
        getContext().getTypeAlignInChars(getContext().WCharTy);
    Value *StrCh = Builder.CreateAlignedLoad(WCharTy, StrPhi, WCharAlign);
    Value *FoundChr = Builder.CreateConstInBoundsGEP1_32(WCharTy, StrPhi, 0);
    Value *StrEqChr = Builder.CreateICmpEQ(StrCh, Chr);
    Builder.CreateCondBr(StrEqChr, Exit, Next);
 
    EmitBlock(Next);
    Value *NextStr = Builder.CreateConstInBoundsGEP1_32(WCharTy, StrPhi, 1);
    Value *NextSize = Builder.CreateSub(SizePhi, ConstantInt::get(SizeTy, 1));
    Value *NextSizeEq0 =
        Builder.CreateICmpEQ(NextSize, ConstantInt::get(SizeTy, 0));
    Builder.CreateCondBr(NextSizeEq0, Exit, CmpEq);
    StrPhi->addIncoming(NextStr, Next);
    SizePhi->addIncoming(NextSize, Next);
 
    EmitBlock(Exit);
    PHINode *Ret = Builder.CreatePHI(Str->getType(), 3);
    Ret->addIncoming(llvm::Constant::getNullValue(Str->getType()), Entry);
    Ret->addIncoming(llvm::Constant::getNullValue(Str->getType()), Next);
    Ret->addIncoming(FoundChr, CmpEq);
    return RValue::get(Ret);
  }
  case Builtin::BI__builtin_wmemcmp: {
    // The MSVC runtime library does not provide a definition of wmemcmp, so we
    // need an inline implementation.
    if (!getTarget().getTriple().isOSMSVCRT())
      break;
 
    llvm::Type *WCharTy = ConvertType(getContext().WCharTy);
 
    Value *Dst = EmitScalarExpr(E->getArg(0));
    Value *Src = EmitScalarExpr(E->getArg(1));
    Value *Size = EmitScalarExpr(E->getArg(2));
 
    BasicBlock *Entry = Builder.GetInsertBlock();
    BasicBlock *CmpGT = createBasicBlock("wmemcmp.gt");
    BasicBlock *CmpLT = createBasicBlock("wmemcmp.lt");
    BasicBlock *Next = createBasicBlock("wmemcmp.next");
    BasicBlock *Exit = createBasicBlock("wmemcmp.exit");
    Value *SizeEq0 = Builder.CreateICmpEQ(Size, ConstantInt::get(SizeTy, 0));
    Builder.CreateCondBr(SizeEq0, Exit, CmpGT);
 
    EmitBlock(CmpGT);
    PHINode *DstPhi = Builder.CreatePHI(Dst->getType(), 2);
    DstPhi->addIncoming(Dst, Entry);
    PHINode *SrcPhi = Builder.CreatePHI(Src->getType(), 2);
    SrcPhi->addIncoming(Src, Entry);
    PHINode *SizePhi = Builder.CreatePHI(SizeTy, 2);
    SizePhi->addIncoming(Size, Entry);
    CharUnits WCharAlign =
        getContext().getTypeAlignInChars(getContext().WCharTy);
    Value *DstCh = Builder.CreateAlignedLoad(WCharTy, DstPhi, WCharAlign);
    Value *SrcCh = Builder.CreateAlignedLoad(WCharTy, SrcPhi, WCharAlign);
    Value *DstGtSrc = Builder.CreateICmpUGT(DstCh, SrcCh);
    Builder.CreateCondBr(DstGtSrc, Exit, CmpLT);
 
    EmitBlock(CmpLT);
    Value *DstLtSrc = Builder.CreateICmpULT(DstCh, SrcCh);
    Builder.CreateCondBr(DstLtSrc, Exit, Next);
 
    EmitBlock(Next);
    Value *NextDst = Builder.CreateConstInBoundsGEP1_32(WCharTy, DstPhi, 1);
    Value *NextSrc = Builder.CreateConstInBoundsGEP1_32(WCharTy, SrcPhi, 1);
    Value *NextSize = Builder.CreateSub(SizePhi, ConstantInt::get(SizeTy, 1));
    Value *NextSizeEq0 =
        Builder.CreateICmpEQ(NextSize, ConstantInt::get(SizeTy, 0));
    Builder.CreateCondBr(NextSizeEq0, Exit, CmpGT);
    DstPhi->addIncoming(NextDst, Next);
    SrcPhi->addIncoming(NextSrc, Next);
    SizePhi->addIncoming(NextSize, Next);
 
    EmitBlock(Exit);
    PHINode *Ret = Builder.CreatePHI(IntTy, 4);
    Ret->addIncoming(ConstantInt::get(IntTy, 0), Entry);
    Ret->addIncoming(ConstantInt::get(IntTy, 1), CmpGT);
    Ret->addIncoming(ConstantInt::get(IntTy, -1), CmpLT);
    Ret->addIncoming(ConstantInt::get(IntTy, 0), Next);
    return RValue::get(Ret);
  }
  case Builtin::BI__builtin_dwarf_cfa: {
    // The offset in bytes from the first argument to the CFA.
    //
    // Why on earth is this in the frontend?  Is there any reason at
    // all that the backend can't reasonably determine this while
    // lowering llvm.eh.dwarf.cfa()?
    //
    // TODO: If there's a satisfactory reason, add a target hook for
    // this instead of hard-coding 0, which is correct for most targets.
    int32_t Offset = 0;
 
    Function *F = CGM.getIntrinsic(Intrinsic::eh_dwarf_cfa);
    return RValue::get(Builder.CreateCall(F,
                                      llvm::ConstantInt::get(Int32Ty, Offset)));
  }
  case Builtin::BI__builtin_return_address: {
    Value *Depth = ConstantEmitter(*this).emitAbstract(E->getArg(0),
                                                   getContext().UnsignedIntTy);
    Function *F = CGM.getIntrinsic(Intrinsic::returnaddress);
    return RValue::get(Builder.CreateCall(F, Depth));
  }
  case Builtin::BI_ReturnAddress: {
    Function *F = CGM.getIntrinsic(Intrinsic::returnaddress);
    return RValue::get(Builder.CreateCall(F, Builder.getInt32(0)));
  }
  case Builtin::BI__builtin_frame_address: {
    Value *Depth = ConstantEmitter(*this).emitAbstract(E->getArg(0),
                                                   getContext().UnsignedIntTy);
    Function *F = CGM.getIntrinsic(Intrinsic::frameaddress, AllocaInt8PtrTy);
    return RValue::get(Builder.CreateCall(F, Depth));
  }
  case Builtin::BI__builtin_extract_return_addr: {
    Value *Address = EmitScalarExpr(E->getArg(0));
    Value *Result = getTargetHooks().decodeReturnAddress(*this, Address);
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_frob_return_addr: {
    Value *Address = EmitScalarExpr(E->getArg(0));
    Value *Result = getTargetHooks().encodeReturnAddress(*this, Address);
    return RValue::get(Result);
  }
  case Builtin::BI__builtin_dwarf_sp_column: {
    llvm::IntegerType *Ty
      = cast<llvm::IntegerType>(ConvertType(E->getType()));
    int Column = getTargetHooks().getDwarfEHStackPointer(CGM);
    if (Column == -1) {
      CGM.ErrorUnsupported(E, "__builtin_dwarf_sp_column");
      return RValue::get(llvm::UndefValue::get(Ty));
    }
    return RValue::get(llvm::ConstantInt::get(Ty, Column, true));
  }
  case Builtin::BI__builtin_init_dwarf_reg_size_table: {
    Value *Address = EmitScalarExpr(E->getArg(0));
    if (getTargetHooks().initDwarfEHRegSizeTable(*this, Address))
      CGM.ErrorUnsupported(E, "__builtin_init_dwarf_reg_size_table");
    return RValue::get(llvm::UndefValue::get(ConvertType(E->getType())));
  }
  case Builtin::BI__builtin_eh_return: {
    Value *Int = EmitScalarExpr(E->getArg(0));
    Value *Ptr = EmitScalarExpr(E->getArg(1));
 
    llvm::IntegerType *IntTy = cast<llvm::IntegerType>(Int->getType());
    assert((IntTy->getBitWidth() == 32 || IntTy->getBitWidth() == 64) &&
           "LLVM's __builtin_eh_return only supports 32- and 64-bit variants");
    Function *F =
        CGM.getIntrinsic(IntTy->getBitWidth() == 32 ? Intrinsic::eh_return_i32
                                                    : Intrinsic::eh_return_i64);
    Builder.CreateCall(F, {Int, Ptr});
    Builder.CreateUnreachable();
 
    // We do need to preserve an insertion point.
    EmitBlock(createBasicBlock("builtin_eh_return.cont"));
 
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_unwind_init: {
    Function *F = CGM.getIntrinsic(Intrinsic::eh_unwind_init);
    Builder.CreateCall(F);
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_extend_pointer: {
    // Extends a pointer to the size of an _Unwind_Word, which is
    // uint64_t on all platforms.  Generally this gets poked into a
    // register and eventually used as an address, so if the
    // addressing registers are wider than pointers and the platform
    // doesn't implicitly ignore high-order bits when doing
    // addressing, we need to make sure we zext / sext based on
    // the platform's expectations.
    //
    // See: http://gcc.gnu.org/ml/gcc-bugs/2002-02/msg00237.html
 
    // Cast the pointer to intptr_t.
    Value *Ptr = EmitScalarExpr(E->getArg(0));
    Value *Result = Builder.CreatePtrToInt(Ptr, IntPtrTy, "extend.cast");
 
    // If that's 64 bits, we're done.
    if (IntPtrTy->getBitWidth() == 64)
      return RValue::get(Result);
 
    // Otherwise, ask the codegen data what to do.
    if (getTargetHooks().extendPointerWithSExt())
      return RValue::get(Builder.CreateSExt(Result, Int64Ty, "extend.sext"));
    else
      return RValue::get(Builder.CreateZExt(Result, Int64Ty, "extend.zext"));
  }
  case Builtin::BI__builtin_setjmp: {
    // Buffer is a void**.
    Address Buf = EmitPointerWithAlignment(E->getArg(0));
 
    // Store the frame pointer to the setjmp buffer.
    Value *FrameAddr = Builder.CreateCall(
        CGM.getIntrinsic(Intrinsic::frameaddress, AllocaInt8PtrTy),
        ConstantInt::get(Int32Ty, 0));
    Builder.CreateStore(FrameAddr, Buf);
 
    // Store the stack pointer to the setjmp buffer.
    Value *StackAddr =
        Builder.CreateCall(CGM.getIntrinsic(Intrinsic::stacksave));
    Address StackSaveSlot = Builder.CreateConstInBoundsGEP(Buf, 2);
    Builder.CreateStore(StackAddr, StackSaveSlot);
 
    // Call LLVM's EH setjmp, which is lightweight.
    Function *F = CGM.getIntrinsic(Intrinsic::eh_sjlj_setjmp);
    Buf = Builder.CreateElementBitCast(Buf, Int8Ty);
    return RValue::get(Builder.CreateCall(F, Buf.getPointer()));
  }
  case Builtin::BI__builtin_longjmp: {
    Value *Buf = EmitScalarExpr(E->getArg(0));
    Buf = Builder.CreateBitCast(Buf, Int8PtrTy);
 
    // Call LLVM's EH longjmp, which is lightweight.
    Builder.CreateCall(CGM.getIntrinsic(Intrinsic::eh_sjlj_longjmp), Buf);
 
    // longjmp doesn't return; mark this as unreachable.
    Builder.CreateUnreachable();
 
    // We do need to preserve an insertion point.
    EmitBlock(createBasicBlock("longjmp.cont"));
 
    return RValue::get(nullptr);
  }
  case Builtin::BI__builtin_launder: {
    const Expr *Arg = E->getArg(0);
    QualType ArgTy = Arg->getType()->getPointeeType();
    Value *Ptr = EmitScalarExpr(Arg);
    if (TypeRequiresBuiltinLaunder(CGM, ArgTy))
      Ptr = Builder.CreateLaunderInvariantGroup(Ptr);
 
    return RValue::get(Ptr);
  }
  case Builtin::BI__sync_fetch_and_add:
  case Builtin::BI__sync_fetch_and_sub:
  case Builtin::BI__sync_fetch_and_or:
  case Builtin::BI__sync_fetch_and_and:
  case Builtin::BI__sync_fetch_and_xor:
  case Builtin::BI__sync_fetch_and_nand:
  case Builtin::BI__sync_add_and_fetch:
  case Builtin::BI__sync_sub_and_fetch:
  case Builtin::BI__sync_and_and_fetch:
  case Builtin::BI__sync_or_and_fetch:
  case Builtin::BI__sync_xor_and_fetch:
  case Builtin::BI__sync_nand_and_fetch:
  case Builtin::BI__sync_val_compare_and_swap:
  case Builtin::BI__sync_bool_compare_and_swap:
  case Builtin::BI__sync_lock_test_and_set:
  case Builtin::BI__sync_lock_release:
  case Builtin::BI__sync_swap:
    llvm_unreachable("Shouldn't make it through sema");
  case Builtin::BI__sync_fetch_and_add_1:
  case Builtin::BI__sync_fetch_and_add_2:
  case Builtin::BI__sync_fetch_and_add_4:
  case Builtin::BI__sync_fetch_and_add_8:
  case Builtin::BI__sync_fetch_and_add_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Add, E);
  case Builtin::BI__sync_fetch_and_sub_1:
  case Builtin::BI__sync_fetch_and_sub_2:
  case Builtin::BI__sync_fetch_and_sub_4:
  case Builtin::BI__sync_fetch_and_sub_8:
  case Builtin::BI__sync_fetch_and_sub_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Sub, E);
  case Builtin::BI__sync_fetch_and_or_1:
  case Builtin::BI__sync_fetch_and_or_2:
  case Builtin::BI__sync_fetch_and_or_4:
  case Builtin::BI__sync_fetch_and_or_8:
  case Builtin::BI__sync_fetch_and_or_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Or, E);
  case Builtin::BI__sync_fetch_and_and_1:
  case Builtin::BI__sync_fetch_and_and_2:
  case Builtin::BI__sync_fetch_and_and_4:
  case Builtin::BI__sync_fetch_and_and_8:
  case Builtin::BI__sync_fetch_and_and_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::And, E);
  case Builtin::BI__sync_fetch_and_xor_1:
  case Builtin::BI__sync_fetch_and_xor_2:
  case Builtin::BI__sync_fetch_and_xor_4:
  case Builtin::BI__sync_fetch_and_xor_8:
  case Builtin::BI__sync_fetch_and_xor_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Xor, E);
  case Builtin::BI__sync_fetch_and_nand_1:
  case Builtin::BI__sync_fetch_and_nand_2:
  case Builtin::BI__sync_fetch_and_nand_4:
  case Builtin::BI__sync_fetch_and_nand_8:
  case Builtin::BI__sync_fetch_and_nand_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Nand, E);
 
  // Clang extensions: not overloaded yet.
  case Builtin::BI__sync_fetch_and_min:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Min, E);
  case Builtin::BI__sync_fetch_and_max:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Max, E);
  case Builtin::BI__sync_fetch_and_umin:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::UMin, E);
  case Builtin::BI__sync_fetch_and_umax:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::UMax, E);
 
  case Builtin::BI__sync_add_and_fetch_1:
  case Builtin::BI__sync_add_and_fetch_2:
  case Builtin::BI__sync_add_and_fetch_4:
  case Builtin::BI__sync_add_and_fetch_8:
  case Builtin::BI__sync_add_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::Add, E,
                                llvm::Instruction::Add);
  case Builtin::BI__sync_sub_and_fetch_1:
  case Builtin::BI__sync_sub_and_fetch_2:
  case Builtin::BI__sync_sub_and_fetch_4:
  case Builtin::BI__sync_sub_and_fetch_8:
  case Builtin::BI__sync_sub_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::Sub, E,
                                llvm::Instruction::Sub);
  case Builtin::BI__sync_and_and_fetch_1:
  case Builtin::BI__sync_and_and_fetch_2:
  case Builtin::BI__sync_and_and_fetch_4:
  case Builtin::BI__sync_and_and_fetch_8:
  case Builtin::BI__sync_and_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::And, E,
                                llvm::Instruction::And);
  case Builtin::BI__sync_or_and_fetch_1:
  case Builtin::BI__sync_or_and_fetch_2:
  case Builtin::BI__sync_or_and_fetch_4:
  case Builtin::BI__sync_or_and_fetch_8:
  case Builtin::BI__sync_or_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::Or, E,
                                llvm::Instruction::Or);
  case Builtin::BI__sync_xor_and_fetch_1:
  case Builtin::BI__sync_xor_and_fetch_2:
  case Builtin::BI__sync_xor_and_fetch_4:
  case Builtin::BI__sync_xor_and_fetch_8:
  case Builtin::BI__sync_xor_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::Xor, E,
                                llvm::Instruction::Xor);
  case Builtin::BI__sync_nand_and_fetch_1:
  case Builtin::BI__sync_nand_and_fetch_2:
  case Builtin::BI__sync_nand_and_fetch_4:
  case Builtin::BI__sync_nand_and_fetch_8:
  case Builtin::BI__sync_nand_and_fetch_16:
    return EmitBinaryAtomicPost(*this, llvm::AtomicRMWInst::Nand, E,
                                llvm::Instruction::And, true);
 
  case Builtin::BI__sync_val_compare_and_swap_1:
  case Builtin::BI__sync_val_compare_and_swap_2:
  case Builtin::BI__sync_val_compare_and_swap_4:
  case Builtin::BI__sync_val_compare_and_swap_8:
  case Builtin::BI__sync_val_compare_and_swap_16:
    return RValue::get(MakeAtomicCmpXchgValue(*this, E, false));
 
  case Builtin::BI__sync_bool_compare_and_swap_1:
  case Builtin::BI__sync_bool_compare_and_swap_2:
  case Builtin::BI__sync_bool_compare_and_swap_4:
  case Builtin::BI__sync_bool_compare_and_swap_8:
  case Builtin::BI__sync_bool_compare_and_swap_16:
    return RValue::get(MakeAtomicCmpXchgValue(*this, E, true));
 
  case Builtin::BI__sync_swap_1:
  case Builtin::BI__sync_swap_2:
  case Builtin::BI__sync_swap_4:
  case Builtin::BI__sync_swap_8:
  case Builtin::BI__sync_swap_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Xchg, E);
 
  case Builtin::BI__sync_lock_test_and_set_1:
  case Builtin::BI__sync_lock_test_and_set_2:
  case Builtin::BI__sync_lock_test_and_set_4:
  case Builtin::BI__sync_lock_test_and_set_8:
  case Builtin::BI__sync_lock_test_and_set_16:
    return EmitBinaryAtomic(*this, llvm::AtomicRMWInst::Xchg, E);
 
  case Builtin::BI__sync_lock_release_1:
  case Builtin::BI__sync_lock_release_2:
  case Builtin::BI__sync_lock_release_4:
  case Builtin::BI__sync_lock_release_8:
  case Builtin::BI__sync_lock_release_16: {
    CheckAtomicAlignment(*this, E);
    Value *Ptr = EmitScalarExpr(E->getArg(0));
    QualType ElTy = E->getArg(0)->getType()->getPointeeType();
    CharUnits StoreSize = getContext().getTypeSizeInChars(ElTy);
    llvm::Type *ITy = llvm::IntegerType::get(getLLVMContext(),
                                             StoreSize.getQuantity() * 8);
    Ptr = Builder.CreateBitCast(Ptr, ITy->getPointerTo());
    llvm::StoreInst *Store =
      Builder.CreateAlignedStore(llvm::Constant::getNullValue(ITy), Ptr,
                                 StoreSize);
    Store->setAtomic(llvm::AtomicOrdering::Release);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__sync_synchronize: {
    // We assume this is supposed to correspond to a C++0x-style
    // sequentially-consistent fence (i.e. this is only usable for
    // synchronization, not device I/O or anything like that). This intrinsic
    // is really badly designed in the sense that in theory, there isn't
    // any way to safely use it... but in practice, it mostly works
    // to use it with non-atomic loads and stores to get acquire/release
    // semantics.
    Builder.CreateFence(llvm::AtomicOrdering::SequentiallyConsistent);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__builtin_nontemporal_load:
    return RValue::get(EmitNontemporalLoad(*this, E));
  case Builtin::BI__builtin_nontemporal_store:
    return RValue::get(EmitNontemporalStore(*this, E));
  case Builtin::BI__c11_atomic_is_lock_free:
  case Builtin::BI__atomic_is_lock_free: {
    // Call "bool __atomic_is_lock_free(size_t size, void *ptr)". For the
    // __c11 builtin, ptr is 0 (indicating a properly-aligned object), since
    // _Atomic(T) is always properly-aligned.
    const char *LibCallName = "__atomic_is_lock_free";
    CallArgList Args;
    Args.add(RValue::get(EmitScalarExpr(E->getArg(0))),
             getContext().getSizeType());
    if (BuiltinID == Builtin::BI__atomic_is_lock_free)
      Args.add(RValue::get(EmitScalarExpr(E->getArg(1))),
               getContext().VoidPtrTy);
    else
      Args.add(RValue::get(llvm::Constant::getNullValue(VoidPtrTy)),
               getContext().VoidPtrTy);
    const CGFunctionInfo &FuncInfo =
        CGM.getTypes().arrangeBuiltinFunctionCall(E->getType(), Args);
    llvm::FunctionType *FTy = CGM.getTypes().GetFunctionType(FuncInfo);
    llvm::FunctionCallee Func = CGM.CreateRuntimeFunction(FTy, LibCallName);
    return EmitCall(FuncInfo, CGCallee::forDirect(Func),
                    ReturnValueSlot(), Args);
  }
 
  case Builtin::BI__atomic_test_and_set: {
    // Look at the argument type to determine whether this is a volatile
    // operation. The parameter type is always volatile.
    QualType PtrTy = E->getArg(0)->IgnoreImpCasts()->getType();
    bool Volatile =
        PtrTy->castAs<PointerType>()->getPointeeType().isVolatileQualified();
 
    Value *Ptr = EmitScalarExpr(E->getArg(0));
    unsigned AddrSpace = Ptr->getType()->getPointerAddressSpace();
    Ptr = Builder.CreateBitCast(Ptr, Int8Ty->getPointerTo(AddrSpace));
    Value *NewVal = Builder.getInt8(1);
    Value *Order = EmitScalarExpr(E->getArg(1));
    if (isa<llvm::ConstantInt>(Order)) {
      int ord = cast<llvm::ConstantInt>(Order)->getZExtValue();
      AtomicRMWInst *Result = nullptr;
      switch (ord) {
      case 0:  // memory_order_relaxed
      default: // invalid order
        Result = Builder.CreateAtomicRMW(llvm::AtomicRMWInst::Xchg, Ptr, NewVal,
                                         llvm::AtomicOrdering::Monotonic);
        break;
      case 1: // memory_order_consume
      case 2: // memory_order_acquire
        Result = Builder.CreateAtomicRMW(llvm::AtomicRMWInst::Xchg, Ptr, NewVal,
                                         llvm::AtomicOrdering::Acquire);
        break;
      case 3: // memory_order_release
        Result = Builder.CreateAtomicRMW(llvm::AtomicRMWInst::Xchg, Ptr, NewVal,
                                         llvm::AtomicOrdering::Release);
        break;
      case 4: // memory_order_acq_rel
 
        Result = Builder.CreateAtomicRMW(llvm::AtomicRMWInst::Xchg, Ptr, NewVal,
                                         llvm::AtomicOrdering::AcquireRelease);
        break;
      case 5: // memory_order_seq_cst
        Result = Builder.CreateAtomicRMW(
            llvm::AtomicRMWInst::Xchg, Ptr, NewVal,
            llvm::AtomicOrdering::SequentiallyConsistent);
        break;
      }
      Result->setVolatile(Volatile);
      return RValue::get(Builder.CreateIsNotNull(Result, "tobool"));
    }
 
    llvm::BasicBlock *ContBB = createBasicBlock("atomic.continue", CurFn);
 
    llvm::BasicBlock *BBs[5] = {
      createBasicBlock("monotonic", CurFn),
      createBasicBlock("acquire", CurFn),
      createBasicBlock("release", CurFn),
      createBasicBlock("acqrel", CurFn),
      createBasicBlock("seqcst", CurFn)
    };
    llvm::AtomicOrdering Orders[5] = {
        llvm::AtomicOrdering::Monotonic, llvm::AtomicOrdering::Acquire,
        llvm::AtomicOrdering::Release, llvm::AtomicOrdering::AcquireRelease,
        llvm::AtomicOrdering::SequentiallyConsistent};
 
    Order = Builder.CreateIntCast(Order, Builder.getInt32Ty(), false);
    llvm::SwitchInst *SI = Builder.CreateSwitch(Order, BBs[0]);
 
    Builder.SetInsertPoint(ContBB);
    PHINode *Result = Builder.CreatePHI(Int8Ty, 5, "was_set");
 
    for (unsigned i = 0; i < 5; ++i) {
      Builder.SetInsertPoint(BBs[i]);
      AtomicRMWInst *RMW = Builder.CreateAtomicRMW(llvm::AtomicRMWInst::Xchg,
                                                   Ptr, NewVal, Orders[i]);
      RMW->setVolatile(Volatile);
      Result->addIncoming(RMW, BBs[i]);
      Builder.CreateBr(ContBB);
    }
 
    SI->addCase(Builder.getInt32(0), BBs[0]);
    SI->addCase(Builder.getInt32(1), BBs[1]);
    SI->addCase(Builder.getInt32(2), BBs[1]);
    SI->addCase(Builder.getInt32(3), BBs[2]);
    SI->addCase(Builder.getInt32(4), BBs[3]);
    SI->addCase(Builder.getInt32(5), BBs[4]);
 
    Builder.SetInsertPoint(ContBB);
    return RValue::get(Builder.CreateIsNotNull(Result, "tobool"));
  }
 
  case Builtin::BI__atomic_clear: {
    QualType PtrTy = E->getArg(0)->IgnoreImpCasts()->getType();
    bool Volatile =
        PtrTy->castAs<PointerType>()->getPointeeType().isVolatileQualified();
 
    Address Ptr = EmitPointerWithAlignment(E->getArg(0));
    Ptr = Builder.CreateElementBitCast(Ptr, Int8Ty);
    Value *NewVal = Builder.getInt8(0);
    Value *Order = EmitScalarExpr(E->getArg(1));
    if (isa<llvm::ConstantInt>(Order)) {
      int ord = cast<llvm::ConstantInt>(Order)->getZExtValue();
      StoreInst *Store = Builder.CreateStore(NewVal, Ptr, Volatile);
      switch (ord) {
      case 0:  // memory_order_relaxed
      default: // invalid order
        Store->setOrdering(llvm::AtomicOrdering::Monotonic);
        break;
      case 3:  // memory_order_release
        Store->setOrdering(llvm::AtomicOrdering::Release);
        break;
      case 5:  // memory_order_seq_cst
        Store->setOrdering(llvm::AtomicOrdering::SequentiallyConsistent);
        break;
      }
      return RValue::get(nullptr);
    }
 
    llvm::BasicBlock *ContBB = createBasicBlock("atomic.continue", CurFn);
 
    llvm::BasicBlock *BBs[3] = {
      createBasicBlock("monotonic", CurFn),
      createBasicBlock("release", CurFn),
      createBasicBlock("seqcst", CurFn)
    };
    llvm::AtomicOrdering Orders[3] = {
        llvm::AtomicOrdering::Monotonic, llvm::AtomicOrdering::Release,
        llvm::AtomicOrdering::SequentiallyConsistent};
 
    Order = Builder.CreateIntCast(Order, Builder.getInt32Ty(), false);
    llvm::SwitchInst *SI = Builder.CreateSwitch(Order, BBs[0]);
 
    for (unsigned i = 0; i < 3; ++i) {
      Builder.SetInsertPoint(BBs[i]);
      StoreInst *Store = Builder.CreateStore(NewVal, Ptr, Volatile);
      Store->setOrdering(Orders[i]);
      Builder.CreateBr(ContBB);
    }
 
    SI->addCase(Builder.getInt32(0), BBs[0]);
    SI->addCase(Builder.getInt32(3), BBs[1]);
    SI->addCase(Builder.getInt32(5), BBs[2]);
 
    Builder.SetInsertPoint(ContBB);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__atomic_thread_fence:
  case Builtin::BI__atomic_signal_fence:
  case Builtin::BI__c11_atomic_thread_fence:
  case Builtin::BI__c11_atomic_signal_fence: {
    llvm::SyncScope::ID SSID;
    if (BuiltinID == Builtin::BI__atomic_signal_fence ||
        BuiltinID == Builtin::BI__c11_atomic_signal_fence)
      SSID = llvm::SyncScope::SingleThread;
    else
      SSID = llvm::SyncScope::System;
    Value *Order = EmitScalarExpr(E->getArg(0));
    if (isa<llvm::ConstantInt>(Order)) {
      int ord = cast<llvm::ConstantInt>(Order)->getZExtValue();
      switch (ord) {
      case 0:  // memory_order_relaxed
      default: // invalid order
        break;
      case 1:  // memory_order_consume
      case 2:  // memory_order_acquire
        Builder.CreateFence(llvm::AtomicOrdering::Acquire, SSID);
        break;
      case 3:  // memory_order_release
        Builder.CreateFence(llvm::AtomicOrdering::Release, SSID);
        break;
      case 4:  // memory_order_acq_rel
        Builder.CreateFence(llvm::AtomicOrdering::AcquireRelease, SSID);
        break;
      case 5:  // memory_order_seq_cst
        Builder.CreateFence(llvm::AtomicOrdering::SequentiallyConsistent, SSID);
        break;
      }
      return RValue::get(nullptr);
    }
 
    llvm::BasicBlock *AcquireBB, *ReleaseBB, *AcqRelBB, *SeqCstBB;
    AcquireBB = createBasicBlock("acquire", CurFn);
    ReleaseBB = createBasicBlock("release", CurFn);
    AcqRelBB = createBasicBlock("acqrel", CurFn);
    SeqCstBB = createBasicBlock("seqcst", CurFn);
    llvm::BasicBlock *ContBB = createBasicBlock("atomic.continue", CurFn);
 
    Order = Builder.CreateIntCast(Order, Builder.getInt32Ty(), false);
    llvm::SwitchInst *SI = Builder.CreateSwitch(Order, ContBB);
 
    Builder.SetInsertPoint(AcquireBB);
    Builder.CreateFence(llvm::AtomicOrdering::Acquire, SSID);
    Builder.CreateBr(ContBB);
    SI->addCase(Builder.getInt32(1), AcquireBB);
    SI->addCase(Builder.getInt32(2), AcquireBB);
 
    Builder.SetInsertPoint(ReleaseBB);
    Builder.CreateFence(llvm::AtomicOrdering::Release, SSID);
    Builder.CreateBr(ContBB);
    SI->addCase(Builder.getInt32(3), ReleaseBB);
 
    Builder.SetInsertPoint(AcqRelBB);
    Builder.CreateFence(llvm::AtomicOrdering::AcquireRelease, SSID);
    Builder.CreateBr(ContBB);
    SI->addCase(Builder.getInt32(4), AcqRelBB);
 
    Builder.SetInsertPoint(SeqCstBB);
    Builder.CreateFence(llvm::AtomicOrdering::SequentiallyConsistent, SSID);
    Builder.CreateBr(ContBB);
    SI->addCase(Builder.getInt32(5), SeqCstBB);
 
    Builder.SetInsertPoint(ContBB);
    return RValue::get(nullptr);
  }
 
  case Builtin::BI__builtin_signbit:
  case Builtin::BI__builtin_signbitf:
  case Builtin::BI__builtin_signbitl: {
    return RValue::get(
        Builder.CreateZExt(EmitSignBit(*this, EmitScalarExpr(E->getArg(0))),
                           ConvertType(E->getType())));
  }
  case Builtin::BI__warn_memset_zero_len:
    return RValue::getIgnored();
  case Builtin::BI__annotation: {
    // Re-encode each wide string to UTF8 and make an MDString.
    SmallVector<Metadata *, 1> Strings;
    for (const Expr *Arg : E->arguments()) {
      const auto *Str = cast<StringLiteral>(Arg->IgnoreParenCasts());
      assert(Str->getCharByteWidth() == 2);
      StringRef WideBytes = Str->getBytes();
      std::string StrUtf8;
      if (!convertUTF16ToUTF8String(
              ArrayRef(WideBytes.data(), WideBytes.size()), StrUtf8)) {
        CGM.ErrorUnsupported(E, "non-UTF16 __annotation argument");
        continue;
      }
      Strings.push_back(llvm::MDString::get(getLLVMContext(), StrUtf8));
    }
 
    // Build and MDTuple of MDStrings and emit the intrinsic call.
    llvm::Function *F =
        CGM.getIntrinsic(llvm::Intrinsic::codeview_annotation, {});
    MDTuple *StrTuple = MDTuple::get(getLLVMContext(), Strings);
    Builder.CreateCall(F, MetadataAsValue::get(getLLVMContext(), StrTuple));
    return RValue::getIgnored();
  }
  case Builtin::BI__builtin_annotation: {
    llvm::Value *AnnVal = EmitScalarExpr(E->getArg(0));
    llvm::Function *F =
        CGM.getIntrinsic(llvm::Intrinsic::annotation,
                         {AnnVal->getType(), CGM.ConstGlobalsPtrTy});
 
    // Get the annotation string, go through casts. Sema requires this to be a
    // non-wide string literal, potentially casted, so the cast<> is safe.
    const Expr *AnnotationStrExpr = E->getArg(1)->IgnoreParenCasts();
    StringRef Str = cast<StringLiteral>(AnnotationStrExpr)->getString();
    return RValue::get(
        EmitAnnotationCall(F, AnnVal, Str, E->getExprLoc(), nullptr));
  }
  case Builtin::BI__builtin_addcb:
  case Builtin::BI__builtin_addcs:
  case Builtin::BI__builtin_addc:
  case Builtin::BI__builtin_addcl:
  case Builtin::BI__builtin_addcll:
  case Builtin::BI__builtin_subcb:
  case Builtin::BI__builtin_subcs:
  case Builtin::BI__builtin_subc:
  case Builtin::BI__builtin_subcl:
  case Builtin::BI__builtin_subcll: {
 
    // We translate all of these builtins from expressions of the form:
    //   int x = ..., y = ..., carryin = ..., carryout, result;
    //   result = __builtin_addc(x, y, carryin, &carryout);
    //
    // to LLVM IR of the form:
    //
    //   %tmp1 = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %x, i32 %y)
    //   %tmpsum1 = extractvalue {i32, i1} %tmp1, 0
    //   %carry1 = extractvalue {i32, i1} %tmp1, 1
    //   %tmp2 = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %tmpsum1,
    //                                                       i32 %carryin)
    //   %result = extractvalue {i32, i1} %tmp2, 0
    //   %carry2 = extractvalue {i32, i1} %tmp2, 1
    //   %tmp3 = or i1 %carry1, %carry2
    //   %tmp4 = zext i1 %tmp3 to i32
    //   store i32 %tmp4, i32* %carryout
 
    // Scalarize our inputs.
    llvm::Value *X = EmitScalarExpr(E->getArg(0));
    llvm::Value *Y = EmitScalarExpr(E->getArg(1));
    llvm::Value *Carryin = EmitScalarExpr(E->getArg(2));
    Address CarryOutPtr = EmitPointerWithAlignment(E->getArg(3));
 
    // Decide if we are lowering to a uadd.with.overflow or usub.with.overflow.
    llvm::Intrinsic::ID IntrinsicId;
    switch (BuiltinID) {
    default: llvm_unreachable("Unknown multiprecision builtin id.");
    case Builtin::BI__builtin_addcb:
    case Builtin::BI__builtin_addcs:
    case Builtin::BI__builtin_addc:
    case Builtin::BI__builtin_addcl:
    case Builtin::BI__builtin_addcll:
      IntrinsicId = llvm::Intrinsic::uadd_with_overflow;
      break;
    case Builtin::BI__builtin_subcb:
    case Builtin::BI__builtin_subcs:
    case Builtin::BI__builtin_subc:
    case Builtin::BI__builtin_subcl:
    case Builtin::BI__builtin_subcll:
      IntrinsicId = llvm::Intrinsic::usub_with_overflow;
      break;
    }
 
    // Construct our resulting LLVM IR expression.
    llvm::Value *Carry1;
    llvm::Value *Sum1 = EmitOverflowIntrinsic(*this, IntrinsicId,
                                              X, Y, Carry1);
    llvm::Value *Carry2;
    llvm::Value *Sum2 = EmitOverflowIntrinsic(*this, IntrinsicId,
                                              Sum1, Carryin, Carry2);
    llvm::Value *CarryOut = Builder.CreateZExt(Builder.CreateOr(Carry1, Carry2),
                                               X->getType());
    Builder.CreateStore(CarryOut, CarryOutPtr);
    return RValue::get(Sum2);
  }
 
  case Builtin::BI__builtin_add_overflow:
  case Builtin::BI__builtin_sub_overflow:
  case Builtin::BI__builtin_mul_overflow: {
    const clang::Expr *LeftArg = E->getArg(0);
    const clang::Expr *RightArg = E->getArg(1);
    const clang::Expr *ResultArg = E->getArg(2);
 
    clang::QualType ResultQTy =
        ResultArg->getType()->castAs<PointerType>()->getPointeeType();
 
    WidthAndSignedness LeftInfo =
        getIntegerWidthAndSignedness(CGM.getContext(), LeftArg->getType());
    WidthAndSignedness RightInfo =
        getIntegerWidthAndSignedness(CGM.getContext(), RightArg->getType());
    WidthAndSignedness ResultInfo =
        getIntegerWidthAndSignedness(CGM.getContext(), ResultQTy);
 
    // Handle mixed-sign multiplication as a special case, because adding
    // runtime or backend support for our generic irgen would be too expensive.
    if (isSpecialMixedSignMultiply(BuiltinID, LeftInfo, RightInfo, ResultInfo))
      return EmitCheckedMixedSignMultiply(*this, LeftArg, LeftInfo, RightArg,
                                          RightInfo, ResultArg, ResultQTy,
                                          ResultInfo);
 
    if (isSpecialUnsignedMultiplySignedResult(BuiltinID, LeftInfo, RightInfo,
                                              ResultInfo))
      return EmitCheckedUnsignedMultiplySignedResult(
          *this, LeftArg, LeftInfo, RightArg, RightInfo, ResultArg, ResultQTy,
          ResultInfo);
 
    WidthAndSignedness EncompassingInfo =
        EncompassingIntegerType({LeftInfo, RightInfo, ResultInfo});
 
    llvm::Type *EncompassingLLVMTy =
        llvm::IntegerType::get(CGM.getLLVMContext(), EncompassingInfo.Width);
 
    llvm::Type *ResultLLVMTy = CGM.getTypes().ConvertType(ResultQTy);
 
    llvm::Intrinsic::ID IntrinsicId;
    switch (BuiltinID) {
    default:
      llvm_unreachable("Unknown overflow builtin id.");
    case Builtin::BI__builtin_add_overflow:
      IntrinsicId = EncompassingInfo.Signed
                        ? llvm::Intrinsic::sadd_with_overflow
                        : llvm::Intrinsic::uadd_with_overflow;
      break;
    case Builtin::BI__builtin_sub_overflow:
      IntrinsicId = EncompassingInfo.Signed
                        ? llvm::Intrinsic::ssub_with_overflow
                        : llvm::Intrinsic::usub_with_overflow;
      break;
    case Builtin::BI__builtin_mul_overflow:
      IntrinsicId = EncompassingInfo.Signed
                        ? llvm::Intrinsic::smul_with_overflow
                        : llvm::Intrinsic::umul_with_overflow;
      break;
    }
 
    llvm::Value *Left = EmitScalarExpr(LeftArg);
    llvm::Value *Right = EmitScalarExpr(RightArg);
    Address ResultPtr = EmitPointerWithAlignment(ResultArg);
 
    <