llvm/lib/Target/AArch64/AArch64InstructionSelector.cpp

//===- AArch64InstructionSelector.cpp ----------------------------*- C++ -*-==//
//
// 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
//
//===----------------------------------------------------------------------===//
/// \file
/// This file implements the targeting of the InstructionSelector class for
/// AArch64.
/// \todo This should be generated by TableGen.
//===----------------------------------------------------------------------===//

#include "AArch64InstrInfo.h"
#include "AArch64MachineFunctionInfo.h"
#include "AArch64RegisterBankInfo.h"
#include "AArch64RegisterInfo.h"
#include "AArch64Subtarget.h"
#include "AArch64TargetMachine.h"
#include "MCTargetDesc/AArch64AddressingModes.h"
#include "llvm/ADT/Optional.h"
#include "llvm/CodeGen/GlobalISel/InstructionSelector.h"
#include "llvm/CodeGen/GlobalISel/InstructionSelectorImpl.h"
#include "llvm/CodeGen/GlobalISel/MachineIRBuilder.h"
#include "llvm/CodeGen/GlobalISel/Utils.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineConstantPool.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/IR/Type.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"

#define DEBUG_TYPE "aarch64-isel"

using namespace llvm;

namespace {

#define GET_GLOBALISEL_PREDICATE_BITSET
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_PREDICATE_BITSET

class AArch64InstructionSelector : public InstructionSelector {
public:
  AArch64InstructionSelector(const AArch64TargetMachine &TM,
                             const AArch64Subtarget &STI,
                             const AArch64RegisterBankInfo &RBI);

  bool select(MachineInstr &I, CodeGenCoverage &CoverageInfo) const override;
  static const char *getName() { return DEBUG_TYPE; }

private:
  /// tblgen-erated 'select' implementation, used as the initial selector for
  /// the patterns that don't require complex C++.
  bool selectImpl(MachineInstr &I, CodeGenCoverage &CoverageInfo) const;

  bool selectVaStartAAPCS(MachineInstr &I, MachineFunction &MF,
                          MachineRegisterInfo &MRI) const;
  bool selectVaStartDarwin(MachineInstr &I, MachineFunction &MF,
                           MachineRegisterInfo &MRI) const;

  bool selectCompareBranch(MachineInstr &I, MachineFunction &MF,
                           MachineRegisterInfo &MRI) const;

  // Helper to generate an equivalent of scalar_to_vector into a new register,
  // returned via 'Dst'.
  MachineInstr *emitScalarToVector(unsigned EltSize,
                                   const TargetRegisterClass *DstRC,
                                   unsigned Scalar,
                                   MachineIRBuilder &MIRBuilder) const;

  /// Emit a lane insert into \p DstReg, or a new vector register if None is
  /// provided.
  ///
  /// The lane inserted into is defined by \p LaneIdx. The vector source
  /// register is given by \p SrcReg. The register containing the element is
  /// given by \p EltReg.
  MachineInstr *emitLaneInsert(Optional<unsigned> DstReg, unsigned SrcReg,
                               unsigned EltReg, unsigned LaneIdx,
                               const RegisterBank &RB,
                               MachineIRBuilder &MIRBuilder) const;
  bool selectInsertElt(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectBuildVector(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectMergeValues(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectUnmergeValues(MachineInstr &I, MachineRegisterInfo &MRI) const;

  void collectShuffleMaskIndices(MachineInstr &I, MachineRegisterInfo &MRI,
                                 SmallVectorImpl<int> &Idxs) const;
  bool selectShuffleVector(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectExtractElt(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectConcatVectors(MachineInstr &I, MachineRegisterInfo &MRI) const;
  bool selectSplitVectorUnmerge(MachineInstr &I,
                                MachineRegisterInfo &MRI) const;

  unsigned emitConstantPoolEntry(Constant *CPVal, MachineFunction &MF) const;
  MachineInstr *emitLoadFromConstantPool(Constant *CPVal,
                                         MachineIRBuilder &MIRBuilder) const;

  // Emit a vector concat operation.
  MachineInstr *emitVectorConcat(Optional<unsigned> Dst, unsigned Op1,
                                 unsigned Op2,
                                 MachineIRBuilder &MIRBuilder) const;
  MachineInstr *emitExtractVectorElt(Optional<unsigned> DstReg,
                                     const RegisterBank &DstRB, LLT ScalarTy,
                                     unsigned VecReg, unsigned LaneIdx,
                                     MachineIRBuilder &MIRBuilder) const;

  ComplexRendererFns selectArithImmed(MachineOperand &Root) const;

  ComplexRendererFns selectAddrModeUnscaled(MachineOperand &Root,
                                            unsigned Size) const;

  ComplexRendererFns selectAddrModeUnscaled8(MachineOperand &Root) const {
    return selectAddrModeUnscaled(Root, 1);
  }
  ComplexRendererFns selectAddrModeUnscaled16(MachineOperand &Root) const {
    return selectAddrModeUnscaled(Root, 2);
  }
  ComplexRendererFns selectAddrModeUnscaled32(MachineOperand &Root) const {
    return selectAddrModeUnscaled(Root, 4);
  }
  ComplexRendererFns selectAddrModeUnscaled64(MachineOperand &Root) const {
    return selectAddrModeUnscaled(Root, 8);
  }
  ComplexRendererFns selectAddrModeUnscaled128(MachineOperand &Root) const {
    return selectAddrModeUnscaled(Root, 16);
  }

  ComplexRendererFns selectAddrModeIndexed(MachineOperand &Root,
                                           unsigned Size) const;
  template <int Width>
  ComplexRendererFns selectAddrModeIndexed(MachineOperand &Root) const {
    return selectAddrModeIndexed(Root, Width / 8);
  }

  void renderTruncImm(MachineInstrBuilder &MIB, const MachineInstr &MI) const;

  // Materialize a GlobalValue or BlockAddress using a movz+movk sequence.
  void materializeLargeCMVal(MachineInstr &I, const Value *V,
                             unsigned char OpFlags) const;

  const AArch64TargetMachine &TM;
  const AArch64Subtarget &STI;
  const AArch64InstrInfo &TII;
  const AArch64RegisterInfo &TRI;
  const AArch64RegisterBankInfo &RBI;

#define GET_GLOBALISEL_PREDICATES_DECL
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_PREDICATES_DECL

// We declare the temporaries used by selectImpl() in the class to minimize the
// cost of constructing placeholder values.
#define GET_GLOBALISEL_TEMPORARIES_DECL
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_TEMPORARIES_DECL
};

} // end anonymous namespace

#define GET_GLOBALISEL_IMPL
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_IMPL

AArch64InstructionSelector::AArch64InstructionSelector(
    const AArch64TargetMachine &TM, const AArch64Subtarget &STI,
    const AArch64RegisterBankInfo &RBI)
    : InstructionSelector(), TM(TM), STI(STI), TII(*STI.getInstrInfo()),
      TRI(*STI.getRegisterInfo()), RBI(RBI),
#define GET_GLOBALISEL_PREDICATES_INIT
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_PREDICATES_INIT
#define GET_GLOBALISEL_TEMPORARIES_INIT
#include "AArch64GenGlobalISel.inc"
#undef GET_GLOBALISEL_TEMPORARIES_INIT
{
}

// FIXME: This should be target-independent, inferred from the types declared
// for each class in the bank.
static const TargetRegisterClass *
getRegClassForTypeOnBank(LLT Ty, const RegisterBank &RB,
                         const RegisterBankInfo &RBI,
                         bool GetAllRegSet = false) {
  if (RB.getID() == AArch64::GPRRegBankID) {
    if (Ty.getSizeInBits() <= 32)
      return GetAllRegSet ? &AArch64::GPR32allRegClass
                          : &AArch64::GPR32RegClass;
    if (Ty.getSizeInBits() == 64)
      return GetAllRegSet ? &AArch64::GPR64allRegClass
                          : &AArch64::GPR64RegClass;
    return nullptr;
  }

  if (RB.getID() == AArch64::FPRRegBankID) {
    if (Ty.getSizeInBits() <= 16)
      return &AArch64::FPR16RegClass;
    if (Ty.getSizeInBits() == 32)
      return &AArch64::FPR32RegClass;
    if (Ty.getSizeInBits() == 64)
      return &AArch64::FPR64RegClass;
    if (Ty.getSizeInBits() == 128)
      return &AArch64::FPR128RegClass;
    return nullptr;
  }

  return nullptr;
}

/// Given a register bank, and size in bits, return the smallest register class
/// that can represent that combination.
static const TargetRegisterClass *
getMinClassForRegBank(const RegisterBank &RB, unsigned SizeInBits,
                      bool GetAllRegSet = false) {
  unsigned RegBankID = RB.getID();

  if (RegBankID == AArch64::GPRRegBankID) {
    if (SizeInBits <= 32)
      return GetAllRegSet ? &AArch64::GPR32allRegClass
                          : &AArch64::GPR32RegClass;
    if (SizeInBits == 64)
      return GetAllRegSet ? &AArch64::GPR64allRegClass
                          : &AArch64::GPR64RegClass;
  }

  if (RegBankID == AArch64::FPRRegBankID) {
    switch (SizeInBits) {
    default:
      return nullptr;
    case 8:
      return &AArch64::FPR8RegClass;
    case 16:
      return &AArch64::FPR16RegClass;
    case 32:
      return &AArch64::FPR32RegClass;
    case 64:
      return &AArch64::FPR64RegClass;
    case 128:
      return &AArch64::FPR128RegClass;
    }
  }

  return nullptr;
}

/// Returns the correct subregister to use for a given register class.
static bool getSubRegForClass(const TargetRegisterClass *RC,
                              const TargetRegisterInfo &TRI, unsigned &SubReg) {
  switch (TRI.getRegSizeInBits(*RC)) {
  case 8:
    SubReg = AArch64::bsub;
    break;
  case 16:
    SubReg = AArch64::hsub;
    break;
  case 32:
    if (RC == &AArch64::GPR32RegClass)
      SubReg = AArch64::sub_32;
    else
      SubReg = AArch64::ssub;
    break;
  case 64:
    SubReg = AArch64::dsub;
    break;
  default:
    LLVM_DEBUG(
        dbgs() << "Couldn't find appropriate subregister for register class.");
    return false;
  }

  return true;
}

/// Check whether \p I is a currently unsupported binary operation:
/// - it has an unsized type
/// - an operand is not a vreg
/// - all operands are not in the same bank
/// These are checks that should someday live in the verifier, but right now,
/// these are mostly limitations of the aarch64 selector.
static bool unsupportedBinOp(const MachineInstr &I,
                             const AArch64RegisterBankInfo &RBI,
                             const MachineRegisterInfo &MRI,
                             const AArch64RegisterInfo &TRI) {
  LLT Ty = MRI.getType(I.getOperand(0).getReg());
  if (!Ty.isValid()) {
    LLVM_DEBUG(dbgs() << "Generic binop register should be typed\n");
    return true;
  }

  const RegisterBank *PrevOpBank = nullptr;
  for (auto &MO : I.operands()) {
    // FIXME: Support non-register operands.
    if (!MO.isReg()) {
      LLVM_DEBUG(dbgs() << "Generic inst non-reg operands are unsupported\n");
      return true;
    }

    // FIXME: Can generic operations have physical registers operands? If
    // so, this will need to be taught about that, and we'll need to get the
    // bank out of the minimal class for the register.
    // Either way, this needs to be documented (and possibly verified).
    if (!TargetRegisterInfo::isVirtualRegister(MO.getReg())) {
      LLVM_DEBUG(dbgs() << "Generic inst has physical register operand\n");
      return true;
    }

    const RegisterBank *OpBank = RBI.getRegBank(MO.getReg(), MRI, TRI);
    if (!OpBank) {
      LLVM_DEBUG(dbgs() << "Generic register has no bank or class\n");
      return true;
    }

    if (PrevOpBank && OpBank != PrevOpBank) {
      LLVM_DEBUG(dbgs() << "Generic inst operands have different banks\n");
      return true;
    }
    PrevOpBank = OpBank;
  }
  return false;
}

/// Select the AArch64 opcode for the basic binary operation \p GenericOpc
/// (such as G_OR or G_SDIV), appropriate for the register bank \p RegBankID
/// and of size \p OpSize.
/// \returns \p GenericOpc if the combination is unsupported.
static unsigned selectBinaryOp(unsigned GenericOpc, unsigned RegBankID,
                               unsigned OpSize) {
  switch (RegBankID) {
  case AArch64::GPRRegBankID:
    if (OpSize == 32) {
      switch (GenericOpc) {
      case TargetOpcode::G_SHL:
        return AArch64::LSLVWr;
      case TargetOpcode::G_LSHR:
        return AArch64::LSRVWr;
      case TargetOpcode::G_ASHR:
        return AArch64::ASRVWr;
      default:
        return GenericOpc;
      }
    } else if (OpSize == 64) {
      switch (GenericOpc) {
      case TargetOpcode::G_GEP:
        return AArch64::ADDXrr;
      case TargetOpcode::G_SHL:
        return AArch64::LSLVXr;
      case TargetOpcode::G_LSHR:
        return AArch64::LSRVXr;
      case TargetOpcode::G_ASHR:
        return AArch64::ASRVXr;
      default:
        return GenericOpc;
      }
    }
    break;
  case AArch64::FPRRegBankID:
    switch (OpSize) {
    case 32:
      switch (GenericOpc) {
      case TargetOpcode::G_FADD:
        return AArch64::FADDSrr;
      case TargetOpcode::G_FSUB:
        return AArch64::FSUBSrr;
      case TargetOpcode::G_FMUL:
        return AArch64::FMULSrr;
      case TargetOpcode::G_FDIV:
        return AArch64::FDIVSrr;
      default:
        return GenericOpc;
      }
    case 64:
      switch (GenericOpc) {
      case TargetOpcode::G_FADD:
        return AArch64::FADDDrr;
      case TargetOpcode::G_FSUB:
        return AArch64::FSUBDrr;
      case TargetOpcode::G_FMUL:
        return AArch64::FMULDrr;
      case TargetOpcode::G_FDIV:
        return AArch64::FDIVDrr;
      case TargetOpcode::G_OR:
        return AArch64::ORRv8i8;
      default:
        return GenericOpc;
      }
    }
    break;
  }
  return GenericOpc;
}

/// Select the AArch64 opcode for the G_LOAD or G_STORE operation \p GenericOpc,
/// appropriate for the (value) register bank \p RegBankID and of memory access
/// size \p OpSize.  This returns the variant with the base+unsigned-immediate
/// addressing mode (e.g., LDRXui).
/// \returns \p GenericOpc if the combination is unsupported.
static unsigned selectLoadStoreUIOp(unsigned GenericOpc, unsigned RegBankID,
                                    unsigned OpSize) {
  const bool isStore = GenericOpc == TargetOpcode::G_STORE;
  switch (RegBankID) {
  case AArch64::GPRRegBankID:
    switch (OpSize) {
    case 8:
      return isStore ? AArch64::STRBBui : AArch64::LDRBBui;
    case 16:
      return isStore ? AArch64::STRHHui : AArch64::LDRHHui;
    case 32:
      return isStore ? AArch64::STRWui : AArch64::LDRWui;
    case 64:
      return isStore ? AArch64::STRXui : AArch64::LDRXui;
    }
    break;
  case AArch64::FPRRegBankID:
    switch (OpSize) {
    case 8:
      return isStore ? AArch64::STRBui : AArch64::LDRBui;
    case 16:
      return isStore ? AArch64::STRHui : AArch64::LDRHui;
    case 32:
      return isStore ? AArch64::STRSui : AArch64::LDRSui;
    case 64:
      return isStore ? AArch64::STRDui : AArch64::LDRDui;
    }
    break;
  }
  return GenericOpc;
}

#ifndef NDEBUG
/// Helper function that verifies that we have a valid copy at the end of
/// selectCopy. Verifies that the source and dest have the expected sizes and
/// then returns true.
static bool isValidCopy(const MachineInstr &I, const RegisterBank &DstBank,
                        const MachineRegisterInfo &MRI,
                        const TargetRegisterInfo &TRI,
                        const RegisterBankInfo &RBI) {
  const unsigned DstReg = I.getOperand(0).getReg();
  const unsigned SrcReg = I.getOperand(1).getReg();
  const unsigned DstSize = RBI.getSizeInBits(DstReg, MRI, TRI);
  const unsigned SrcSize = RBI.getSizeInBits(SrcReg, MRI, TRI);

  // Make sure the size of the source and dest line up.
  assert(
      (DstSize == SrcSize ||
       // Copies are a mean to setup initial types, the number of
       // bits may not exactly match.
       (TargetRegisterInfo::isPhysicalRegister(SrcReg) && DstSize <= SrcSize) ||
       // Copies are a mean to copy bits around, as long as we are
       // on the same register class, that's fine. Otherwise, that
       // means we need some SUBREG_TO_REG or AND & co.
       (((DstSize + 31) / 32 == (SrcSize + 31) / 32) && DstSize > SrcSize)) &&
      "Copy with different width?!");

  // Check the size of the destination.
  assert((DstSize <= 64 || DstBank.getID() == AArch64::FPRRegBankID) &&
         "GPRs cannot get more than 64-bit width values");

  return true;
}
#endif

/// Helper function for selectCopy. Inserts a subregister copy from
/// \p *From to \p *To, linking it up to \p I.
///
/// e.g, given I = "Dst = COPY SrcReg", we'll transform that into
///
/// CopyReg (From class) = COPY SrcReg
/// SubRegCopy (To class) = COPY CopyReg:SubReg
/// Dst = COPY SubRegCopy
static bool selectSubregisterCopy(MachineInstr &I, MachineRegisterInfo &MRI,
                                  const RegisterBankInfo &RBI, unsigned SrcReg,
                                  const TargetRegisterClass *From,
                                  const TargetRegisterClass *To,
                                  unsigned SubReg) {
  MachineIRBuilder MIB(I);
  auto Copy = MIB.buildCopy({From}, {SrcReg});
  auto SubRegCopy = MIB.buildInstr(TargetOpcode::COPY, {To}, {})
                        .addReg(Copy.getReg(0), 0, SubReg);
  MachineOperand &RegOp = I.getOperand(1);
  RegOp.setReg(SubRegCopy.getReg(0));

  // It's possible that the destination register won't be constrained. Make
  // sure that happens.
  if (!TargetRegisterInfo::isPhysicalRegister(I.getOperand(0).getReg()))
    RBI.constrainGenericRegister(I.getOperand(0).getReg(), *To, MRI);

  return true;
}

static bool selectCopy(MachineInstr &I, const TargetInstrInfo &TII,
                       MachineRegisterInfo &MRI, const TargetRegisterInfo &TRI,
                       const RegisterBankInfo &RBI) {

  unsigned DstReg = I.getOperand(0).getReg();
  unsigned SrcReg = I.getOperand(1).getReg();
  const RegisterBank &DstRegBank = *RBI.getRegBank(DstReg, MRI, TRI);
  const RegisterBank &SrcRegBank = *RBI.getRegBank(SrcReg, MRI, TRI);
  const TargetRegisterClass *DstRC = getMinClassForRegBank(
      DstRegBank, RBI.getSizeInBits(DstReg, MRI, TRI), true);
  if (!DstRC) {
    LLVM_DEBUG(dbgs() << "Unexpected dest size "
                      << RBI.getSizeInBits(DstReg, MRI, TRI) << '\n');
    return false;
  }

  // A couple helpers below, for making sure that the copy we produce is valid.

  // Set to true if we insert a SUBREG_TO_REG. If we do this, then we don't want
  // to verify that the src and dst are the same size, since that's handled by
  // the SUBREG_TO_REG.
  bool KnownValid = false;

  // Returns true, or asserts if something we don't expect happens. Instead of
  // returning true, we return isValidCopy() to ensure that we verify the
  // result.
  auto CheckCopy = [&]() {
    // If we have a bitcast or something, we can't have physical registers.
    assert(
        (I.isCopy() ||
         (!TargetRegisterInfo::isPhysicalRegister(I.getOperand(0).getReg()) &&
          !TargetRegisterInfo::isPhysicalRegister(I.getOperand(1).getReg()))) &&
        "No phys reg on generic operator!");
    assert(KnownValid || isValidCopy(I, DstRegBank, MRI, TRI, RBI));
    (void)KnownValid;
    return true;
  };

  // Is this a copy? If so, then we may need to insert a subregister copy, or
  // a SUBREG_TO_REG.
  if (I.isCopy()) {
    // Yes. Check if there's anything to fix up.
    const TargetRegisterClass *SrcRC = getMinClassForRegBank(
        SrcRegBank, RBI.getSizeInBits(SrcReg, MRI, TRI), true);
    if (!SrcRC) {
      LLVM_DEBUG(dbgs() << "Couldn't determine source register class\n");
      return false;
    }

    // Is this a cross-bank copy?
    if (DstRegBank.getID() != SrcRegBank.getID()) {
      // If we're doing a cross-bank copy on different-sized registers, we need
      // to do a bit more work.
      unsigned SrcSize = TRI.getRegSizeInBits(*SrcRC);
      unsigned DstSize = TRI.getRegSizeInBits(*DstRC);

      if (SrcSize > DstSize) {
        // We're doing a cross-bank copy into a smaller register. We need a
        // subregister copy. First, get a register class that's on the same bank
        // as the destination, but the same size as the source.
        const TargetRegisterClass *SubregRC =
            getMinClassForRegBank(DstRegBank, SrcSize, true);
        assert(SubregRC && "Didn't get a register class for subreg?");

        // Get the appropriate subregister for the destination.
        unsigned SubReg = 0;
        if (!getSubRegForClass(DstRC, TRI, SubReg)) {
          LLVM_DEBUG(dbgs() << "Couldn't determine subregister for copy.\n");
          return false;
        }

        // Now, insert a subregister copy using the new register class.
        selectSubregisterCopy(I, MRI, RBI, SrcReg, SubregRC, DstRC, SubReg);
        return CheckCopy();
      }

      else if (DstRegBank.getID() == AArch64::GPRRegBankID && DstSize == 32 &&
               SrcSize == 16) {
        // Special case for FPR16 to GPR32.
        // FIXME: This can probably be generalized like the above case.
        unsigned PromoteReg =
            MRI.createVirtualRegister(&AArch64::FPR32RegClass);
        BuildMI(*I.getParent(), I, I.getDebugLoc(),
                TII.get(AArch64::SUBREG_TO_REG), PromoteReg)
            .addImm(0)
            .addUse(SrcReg)
            .addImm(AArch64::hsub);
        MachineOperand &RegOp = I.getOperand(1);
        RegOp.setReg(PromoteReg);

        // Promise that the copy is implicitly validated by the SUBREG_TO_REG.
        KnownValid = true;
      }
    }

    // If the destination is a physical register, then there's nothing to
    // change, so we're done.
    if (TargetRegisterInfo::isPhysicalRegister(DstReg))
      return CheckCopy();
  }

  // No need to constrain SrcReg. It will get constrained when we hit another
  // of its use or its defs. Copies do not have constraints.
  if (!RBI.constrainGenericRegister(DstReg, *DstRC, MRI)) {
    LLVM_DEBUG(dbgs() << "Failed to constrain " << TII.getName(I.getOpcode())
                      << " operand\n");
    return false;
  }
  I.setDesc(TII.get(AArch64::COPY));
  return CheckCopy();
}

static unsigned selectFPConvOpc(unsigned GenericOpc, LLT DstTy, LLT SrcTy) {
  if (!DstTy.isScalar() || !SrcTy.isScalar())
    return GenericOpc;

  const unsigned DstSize = DstTy.getSizeInBits();
  const unsigned SrcSize = SrcTy.getSizeInBits();

  switch (DstSize) {
  case 32:
    switch (SrcSize) {
    case 32:
      switch (GenericOpc) {
      case TargetOpcode::G_SITOFP:
        return AArch64::SCVTFUWSri;
      case TargetOpcode::G_UITOFP:
        return AArch64::UCVTFUWSri;
      case TargetOpcode::G_FPTOSI:
        return AArch64::FCVTZSUWSr;
      case TargetOpcode::G_FPTOUI:
        return AArch64::FCVTZUUWSr;
      default:
        return GenericOpc;
      }
    case 64:
      switch (GenericOpc) {
      case TargetOpcode::G_SITOFP:
        return AArch64::SCVTFUXSri;
      case TargetOpcode::G_UITOFP:
        return AArch64::UCVTFUXSri;
      case TargetOpcode::G_FPTOSI:
        return AArch64::FCVTZSUWDr;
      case TargetOpcode::G_FPTOUI:
        return AArch64::FCVTZUUWDr;
      default:
        return GenericOpc;
      }
    default:
      return GenericOpc;
    }
  case 64:
    switch (SrcSize) {
    case 32:
      switch (GenericOpc) {
      case TargetOpcode::G_SITOFP:
        return AArch64::SCVTFUWDri;
      case TargetOpcode::G_UITOFP:
        return AArch64::UCVTFUWDri;
      case TargetOpcode::G_FPTOSI:
        return AArch64::FCVTZSUXSr;
      case TargetOpcode::G_FPTOUI:
        return AArch64::FCVTZUUXSr;
      default:
        return GenericOpc;
      }
    case 64:
      switch (GenericOpc) {
      case TargetOpcode::G_SITOFP:
        return AArch64::SCVTFUXDri;
      case TargetOpcode::G_UITOFP:
        return AArch64::UCVTFUXDri;
      case TargetOpcode::G_FPTOSI:
        return AArch64::FCVTZSUXDr;
      case TargetOpcode::G_FPTOUI:
        return AArch64::FCVTZUUXDr;
      default:
        return GenericOpc;
      }
    default:
      return GenericOpc;
    }
  default:
    return GenericOpc;
  };
  return GenericOpc;
}

static AArch64CC::CondCode changeICMPPredToAArch64CC(CmpInst::Predicate P) {
  switch (P) {
  default:
    llvm_unreachable("Unknown condition code!");
  case CmpInst::ICMP_NE:
    return AArch64CC::NE;
  case CmpInst::ICMP_EQ:
    return AArch64CC::EQ;
  case CmpInst::ICMP_SGT:
    return AArch64CC::GT;
  case CmpInst::ICMP_SGE:
    return AArch64CC::GE;
  case CmpInst::ICMP_SLT:
    return AArch64CC::LT;
  case CmpInst::ICMP_SLE:
    return AArch64CC::LE;
  case CmpInst::ICMP_UGT:
    return AArch64CC::HI;
  case CmpInst::ICMP_UGE:
    return AArch64CC::HS;
  case CmpInst::ICMP_ULT:
    return AArch64CC::LO;
  case CmpInst::ICMP_ULE:
    return AArch64CC::LS;
  }
}

static void changeFCMPPredToAArch64CC(CmpInst::Predicate P,
                                      AArch64CC::CondCode &CondCode,
                                      AArch64CC::CondCode &CondCode2) {
  CondCode2 = AArch64CC::AL;
  switch (P) {
  default:
    llvm_unreachable("Unknown FP condition!");
  case CmpInst::FCMP_OEQ:
    CondCode = AArch64CC::EQ;
    break;
  case CmpInst::FCMP_OGT:
    CondCode = AArch64CC::GT;
    break;
  case CmpInst::FCMP_OGE:
    CondCode = AArch64CC::GE;
    break;
  case CmpInst::FCMP_OLT:
    CondCode = AArch64CC::MI;
    break;
  case CmpInst::FCMP_OLE:
    CondCode = AArch64CC::LS;
    break;
  case CmpInst::FCMP_ONE:
    CondCode = AArch64CC::MI;
    CondCode2 = AArch64CC::GT;
    break;
  case CmpInst::FCMP_ORD:
    CondCode = AArch64CC::VC;
    break;
  case CmpInst::FCMP_UNO:
    CondCode = AArch64CC::VS;
    break;
  case CmpInst::FCMP_UEQ:
    CondCode = AArch64CC::EQ;
    CondCode2 = AArch64CC::VS;
    break;
  case CmpInst::FCMP_UGT:
    CondCode = AArch64CC::HI;
    break;
  case CmpInst::FCMP_UGE:
    CondCode = AArch64CC::PL;
    break;
  case CmpInst::FCMP_ULT:
    CondCode = AArch64CC::LT;
    break;
  case CmpInst::FCMP_ULE:
    CondCode = AArch64CC::LE;
    break;
  case CmpInst::FCMP_UNE:
    CondCode = AArch64CC::NE;
    break;
  }
}

bool AArch64InstructionSelector::selectCompareBranch(
    MachineInstr &I, MachineFunction &MF, MachineRegisterInfo &MRI) const {

  const unsigned CondReg = I.getOperand(0).getReg();
  MachineBasicBlock *DestMBB = I.getOperand(1).getMBB();
  MachineInstr *CCMI = MRI.getVRegDef(CondReg);
  if (CCMI->getOpcode() == TargetOpcode::G_TRUNC)
    CCMI = MRI.getVRegDef(CCMI->getOperand(1).getReg());
  if (CCMI->getOpcode() != TargetOpcode::G_ICMP)
    return false;

  unsigned LHS = CCMI->getOperand(2).getReg();
  unsigned RHS = CCMI->getOperand(3).getReg();
  if (!getConstantVRegVal(RHS, MRI))
    std::swap(RHS, LHS);

  const auto RHSImm = getConstantVRegVal(RHS, MRI);
  if (!RHSImm || *RHSImm != 0)
    return false;

  const RegisterBank &RB = *RBI.getRegBank(LHS, MRI, TRI);
  if (RB.getID() != AArch64::GPRRegBankID)
    return false;

  const auto Pred = (CmpInst::Predicate)CCMI->getOperand(1).getPredicate();
  if (Pred != CmpInst::ICMP_NE && Pred != CmpInst::ICMP_EQ)
    return false;

  const unsigned CmpWidth = MRI.getType(LHS).getSizeInBits();
  unsigned CBOpc = 0;
  if (CmpWidth <= 32)
    CBOpc = (Pred == CmpInst::ICMP_EQ ? AArch64::CBZW : AArch64::CBNZW);
  else if (CmpWidth == 64)
    CBOpc = (Pred == CmpInst::ICMP_EQ ? AArch64::CBZX : AArch64::CBNZX);
  else
    return false;

  BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(CBOpc))
      .addUse(LHS)
      .addMBB(DestMBB)
      .constrainAllUses(TII, TRI, RBI);

  I.eraseFromParent();
  return true;
}

bool AArch64InstructionSelector::selectVaStartAAPCS(
    MachineInstr &I, MachineFunction &MF, MachineRegisterInfo &MRI) const {
  return false;
}

bool AArch64InstructionSelector::selectVaStartDarwin(
    MachineInstr &I, MachineFunction &MF, MachineRegisterInfo &MRI) const {
  AArch64FunctionInfo *FuncInfo = MF.getInfo<AArch64FunctionInfo>();
  unsigned ListReg = I.getOperand(0).getReg();

  unsigned ArgsAddrReg = MRI.createVirtualRegister(&AArch64::GPR64RegClass);

  auto MIB =
      BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(AArch64::ADDXri))
          .addDef(ArgsAddrReg)
          .addFrameIndex(FuncInfo->getVarArgsStackIndex())
          .addImm(0)
          .addImm(0);

  constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);

  MIB = BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(AArch64::STRXui))
            .addUse(ArgsAddrReg)
            .addUse(ListReg)
            .addImm(0)
            .addMemOperand(*I.memoperands_begin());

  constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
  I.eraseFromParent();
  return true;
}

void AArch64InstructionSelector::materializeLargeCMVal(
    MachineInstr &I, const Value *V, unsigned char OpFlags) const {
  MachineBasicBlock &MBB = *I.getParent();
  MachineFunction &MF = *MBB.getParent();
  MachineRegisterInfo &MRI = MF.getRegInfo();
  MachineIRBuilder MIB(I);

  auto MovZ = MIB.buildInstr(AArch64::MOVZXi, {&AArch64::GPR64RegClass}, {});
  MovZ->addOperand(MF, I.getOperand(1));
  MovZ->getOperand(1).setTargetFlags(OpFlags | AArch64II::MO_G0 |
                                     AArch64II::MO_NC);
  MovZ->addOperand(MF, MachineOperand::CreateImm(0));
  constrainSelectedInstRegOperands(*MovZ, TII, TRI, RBI);

  auto BuildMovK = [&](unsigned SrcReg, unsigned char Flags, unsigned Offset,
                       unsigned ForceDstReg) {
    unsigned DstReg = ForceDstReg
                          ? ForceDstReg
                          : MRI.createVirtualRegister(&AArch64::GPR64RegClass);
    auto MovI = MIB.buildInstr(AArch64::MOVKXi).addDef(DstReg).addUse(SrcReg);
    if (auto *GV = dyn_cast<GlobalValue>(V)) {
      MovI->addOperand(MF, MachineOperand::CreateGA(
                               GV, MovZ->getOperand(1).getOffset(), Flags));
    } else {
      MovI->addOperand(
          MF, MachineOperand::CreateBA(cast<BlockAddress>(V),
                                       MovZ->getOperand(1).getOffset(), Flags));
    }
    MovI->addOperand(MF, MachineOperand::CreateImm(Offset));
    constrainSelectedInstRegOperands(*MovI, TII, TRI, RBI);
    return DstReg;
  };
  unsigned DstReg = BuildMovK(MovZ.getReg(0),
                              AArch64II::MO_G1 | AArch64II::MO_NC, 16, 0);
  DstReg = BuildMovK(DstReg, AArch64II::MO_G2 | AArch64II::MO_NC, 32, 0);
  BuildMovK(DstReg, AArch64II::MO_G3, 48, I.getOperand(0).getReg());
  return;
}

bool AArch64InstructionSelector::select(MachineInstr &I,
                                        CodeGenCoverage &CoverageInfo) const {
  assert(I.getParent() && "Instruction should be in a basic block!");
  assert(I.getParent()->getParent() && "Instruction should be in a function!");

  MachineBasicBlock &MBB = *I.getParent();
  MachineFunction &MF = *MBB.getParent();
  MachineRegisterInfo &MRI = MF.getRegInfo();

  unsigned Opcode = I.getOpcode();
  // G_PHI requires same handling as PHI
  if (!isPreISelGenericOpcode(Opcode) || Opcode == TargetOpcode::G_PHI) {
    // Certain non-generic instructions also need some special handling.

    if (Opcode ==  TargetOpcode::LOAD_STACK_GUARD)
      return constrainSelectedInstRegOperands(I, TII, TRI, RBI);

    if (Opcode == TargetOpcode::PHI || Opcode == TargetOpcode::G_PHI) {
      const unsigned DefReg = I.getOperand(0).getReg();
      const LLT DefTy = MRI.getType(DefReg);

      const TargetRegisterClass *DefRC = nullptr;
      if (TargetRegisterInfo::isPhysicalRegister(DefReg)) {
        DefRC = TRI.getRegClass(DefReg);
      } else {
        const RegClassOrRegBank &RegClassOrBank =
            MRI.getRegClassOrRegBank(DefReg);

        DefRC = RegClassOrBank.dyn_cast<const TargetRegisterClass *>();
        if (!DefRC) {
          if (!DefTy.isValid()) {
            LLVM_DEBUG(dbgs() << "PHI operand has no type, not a gvreg?\n");
            return false;
          }
          const RegisterBank &RB = *RegClassOrBank.get<const RegisterBank *>();
          DefRC = getRegClassForTypeOnBank(DefTy, RB, RBI);
          if (!DefRC) {
            LLVM_DEBUG(dbgs() << "PHI operand has unexpected size/bank\n");
            return false;
          }
        }
      }
      I.setDesc(TII.get(TargetOpcode::PHI));

      return RBI.constrainGenericRegister(DefReg, *DefRC, MRI);
    }

    if (I.isCopy())
      return selectCopy(I, TII, MRI, TRI, RBI);

    return true;
  }


  if (I.getNumOperands() != I.getNumExplicitOperands()) {
    LLVM_DEBUG(
        dbgs() << "Generic instruction has unexpected implicit operands\n");
    return false;
  }

  if (selectImpl(I, CoverageInfo))
    return true;

  LLT Ty =
      I.getOperand(0).isReg() ? MRI.getType(I.getOperand(0).getReg()) : LLT{};

  MachineIRBuilder MIB(I);

  switch (Opcode) {
  case TargetOpcode::G_BRCOND: {
    if (Ty.getSizeInBits() > 32) {
      // We shouldn't need this on AArch64, but it would be implemented as an
      // EXTRACT_SUBREG followed by a TBNZW because TBNZX has no encoding if the
      // bit being tested is < 32.
      LLVM_DEBUG(dbgs() << "G_BRCOND has type: " << Ty
                        << ", expected at most 32-bits");
      return false;
    }

    const unsigned CondReg = I.getOperand(0).getReg();
    MachineBasicBlock *DestMBB = I.getOperand(1).getMBB();

    // Speculation tracking/SLH assumes that optimized TB(N)Z/CB(N)Z
    // instructions will not be produced, as they are conditional branch
    // instructions that do not set flags.
    bool ProduceNonFlagSettingCondBr =
        !MF.getFunction().hasFnAttribute(Attribute::SpeculativeLoadHardening);
    if (ProduceNonFlagSettingCondBr && selectCompareBranch(I, MF, MRI))
      return true;

    if (ProduceNonFlagSettingCondBr) {
      auto MIB = BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::TBNZW))
                     .addUse(CondReg)
                     .addImm(/*bit offset=*/0)
                     .addMBB(DestMBB);

      I.eraseFromParent();
      return constrainSelectedInstRegOperands(*MIB.getInstr(), TII, TRI, RBI);
    } else {
      auto CMP = BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::ANDSWri))
                     .addDef(AArch64::WZR)
                     .addUse(CondReg)
                     .addImm(1);
      constrainSelectedInstRegOperands(*CMP.getInstr(), TII, TRI, RBI);
      auto Bcc =
          BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::Bcc))
              .addImm(AArch64CC::EQ)
              .addMBB(DestMBB);

      I.eraseFromParent();
      return constrainSelectedInstRegOperands(*Bcc.getInstr(), TII, TRI, RBI);
    }
  }

  case TargetOpcode::G_BRINDIRECT: {
    I.setDesc(TII.get(AArch64::BR));
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_FCONSTANT:
  case TargetOpcode::G_CONSTANT: {
    const bool isFP = Opcode == TargetOpcode::G_FCONSTANT;

    const LLT s32 = LLT::scalar(32);
    const LLT s64 = LLT::scalar(64);
    const LLT p0 = LLT::pointer(0, 64);

    const unsigned DefReg = I.getOperand(0).getReg();
    const LLT DefTy = MRI.getType(DefReg);
    const unsigned DefSize = DefTy.getSizeInBits();
    const RegisterBank &RB = *RBI.getRegBank(DefReg, MRI, TRI);

    // FIXME: Redundant check, but even less readable when factored out.
    if (isFP) {
      if (Ty != s32 && Ty != s64) {
        LLVM_DEBUG(dbgs() << "Unable to materialize FP " << Ty
                          << " constant, expected: " << s32 << " or " << s64
                          << '\n');
        return false;
      }

      if (RB.getID() != AArch64::FPRRegBankID) {
        LLVM_DEBUG(dbgs() << "Unable to materialize FP " << Ty
                          << " constant on bank: " << RB
                          << ", expected: FPR\n");
        return false;
      }

      // The case when we have 0.0 is covered by tablegen. Reject it here so we
      // can be sure tablegen works correctly and isn't rescued by this code.
      if (I.getOperand(1).getFPImm()->getValueAPF().isExactlyValue(0.0))
        return false;
    } else {
      // s32 and s64 are covered by tablegen.
      if (Ty != p0) {
        LLVM_DEBUG(dbgs() << "Unable to materialize integer " << Ty
                          << " constant, expected: " << s32 << ", " << s64
                          << ", or " << p0 << '\n');
        return false;
      }

      if (RB.getID() != AArch64::GPRRegBankID) {
        LLVM_DEBUG(dbgs() << "Unable to materialize integer " << Ty
                          << " constant on bank: " << RB
                          << ", expected: GPR\n");
        return false;
      }
    }

    const unsigned MovOpc =
        DefSize == 32 ? AArch64::MOVi32imm : AArch64::MOVi64imm;

    I.setDesc(TII.get(MovOpc));

    if (isFP) {
      const TargetRegisterClass &GPRRC =
          DefSize == 32 ? AArch64::GPR32RegClass : AArch64::GPR64RegClass;
      const TargetRegisterClass &FPRRC =
          DefSize == 32 ? AArch64::FPR32RegClass : AArch64::FPR64RegClass;

      const unsigned DefGPRReg = MRI.createVirtualRegister(&GPRRC);
      MachineOperand &RegOp = I.getOperand(0);
      RegOp.setReg(DefGPRReg);
      MIB.setInsertPt(MIB.getMBB(), std::next(I.getIterator()));
      MIB.buildCopy({DefReg}, {DefGPRReg});

      if (!RBI.constrainGenericRegister(DefReg, FPRRC, MRI)) {
        LLVM_DEBUG(dbgs() << "Failed to constrain G_FCONSTANT def operand\n");
        return false;
      }

      MachineOperand &ImmOp = I.getOperand(1);
      // FIXME: Is going through int64_t always correct?
      ImmOp.ChangeToImmediate(
          ImmOp.getFPImm()->getValueAPF().bitcastToAPInt().getZExtValue());
    } else if (I.getOperand(1).isCImm()) {
      uint64_t Val = I.getOperand(1).getCImm()->getZExtValue();
      I.getOperand(1).ChangeToImmediate(Val);
    } else if (I.getOperand(1).isImm()) {
      uint64_t Val = I.getOperand(1).getImm();
      I.getOperand(1).ChangeToImmediate(Val);
    }

    constrainSelectedInstRegOperands(I, TII, TRI, RBI);
    return true;
  }
  case TargetOpcode::G_EXTRACT: {
    LLT SrcTy = MRI.getType(I.getOperand(1).getReg());
    LLT DstTy = MRI.getType(I.getOperand(0).getReg());
    (void)DstTy;
    unsigned SrcSize = SrcTy.getSizeInBits();
    // Larger extracts are vectors, same-size extracts should be something else
    // by now (either split up or simplified to a COPY).
    if (SrcTy.getSizeInBits() > 64 || Ty.getSizeInBits() > 32)
      return false;

    I.setDesc(TII.get(SrcSize == 64 ? AArch64::UBFMXri : AArch64::UBFMWri));
    MachineInstrBuilder(MF, I).addImm(I.getOperand(2).getImm() +
                                      Ty.getSizeInBits() - 1);

    if (SrcSize < 64) {
      assert(SrcSize == 32 && DstTy.getSizeInBits() == 16 &&
             "unexpected G_EXTRACT types");
      return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
    }

    unsigned DstReg = MRI.createGenericVirtualRegister(LLT::scalar(64));
    MIB.setInsertPt(MIB.getMBB(), std::next(I.getIterator()));
    MIB.buildInstr(TargetOpcode::COPY, {I.getOperand(0).getReg()}, {})
        .addReg(DstReg, 0, AArch64::sub_32);
    RBI.constrainGenericRegister(I.getOperand(0).getReg(),
                                 AArch64::GPR32RegClass, MRI);
    I.getOperand(0).setReg(DstReg);

    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_INSERT: {
    LLT SrcTy = MRI.getType(I.getOperand(2).getReg());
    LLT DstTy = MRI.getType(I.getOperand(0).getReg());
    unsigned DstSize = DstTy.getSizeInBits();
    // Larger inserts are vectors, same-size ones should be something else by
    // now (split up or turned into COPYs).
    if (Ty.getSizeInBits() > 64 || SrcTy.getSizeInBits() > 32)
      return false;

    I.setDesc(TII.get(DstSize == 64 ? AArch64::BFMXri : AArch64::BFMWri));
    unsigned LSB = I.getOperand(3).getImm();
    unsigned Width = MRI.getType(I.getOperand(2).getReg()).getSizeInBits();
    I.getOperand(3).setImm((DstSize - LSB) % DstSize);
    MachineInstrBuilder(MF, I).addImm(Width - 1);

    if (DstSize < 64) {
      assert(DstSize == 32 && SrcTy.getSizeInBits() == 16 &&
             "unexpected G_INSERT types");
      return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
    }

    unsigned SrcReg = MRI.createGenericVirtualRegister(LLT::scalar(64));
    BuildMI(MBB, I.getIterator(), I.getDebugLoc(),
            TII.get(AArch64::SUBREG_TO_REG))
        .addDef(SrcReg)
        .addImm(0)
        .addUse(I.getOperand(2).getReg())
        .addImm(AArch64::sub_32);
    RBI.constrainGenericRegister(I.getOperand(2).getReg(),
                                 AArch64::GPR32RegClass, MRI);
    I.getOperand(2).setReg(SrcReg);

    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }
  case TargetOpcode::G_FRAME_INDEX: {
    // allocas and G_FRAME_INDEX are only supported in addrspace(0).
    if (Ty != LLT::pointer(0, 64)) {
      LLVM_DEBUG(dbgs() << "G_FRAME_INDEX pointer has type: " << Ty
                        << ", expected: " << LLT::pointer(0, 64) << '\n');
      return false;
    }
    I.setDesc(TII.get(AArch64::ADDXri));

    // MOs for a #0 shifted immediate.
    I.addOperand(MachineOperand::CreateImm(0));
    I.addOperand(MachineOperand::CreateImm(0));

    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_GLOBAL_VALUE: {
    auto GV = I.getOperand(1).getGlobal();
    if (GV->isThreadLocal()) {
      // FIXME: we don't support TLS yet.
      return false;
    }
    unsigned char OpFlags = STI.ClassifyGlobalReference(GV, TM);
    if (OpFlags & AArch64II::MO_GOT) {
      I.setDesc(TII.get(AArch64::LOADgot));
      I.getOperand(1).setTargetFlags(OpFlags);
    } else if (TM.getCodeModel() == CodeModel::Large) {
      // Materialize the global using movz/movk instructions.
      materializeLargeCMVal(I, GV, OpFlags);
      I.eraseFromParent();
      return true;
    } else if (TM.getCodeModel() == CodeModel::Tiny) {
      I.setDesc(TII.get(AArch64::ADR));
      I.getOperand(1).setTargetFlags(OpFlags);
    } else {
      I.setDesc(TII.get(AArch64::MOVaddr));
      I.getOperand(1).setTargetFlags(OpFlags | AArch64II::MO_PAGE);
      MachineInstrBuilder MIB(MF, I);
      MIB.addGlobalAddress(GV, I.getOperand(1).getOffset(),
                           OpFlags | AArch64II::MO_PAGEOFF | AArch64II::MO_NC);
    }
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_LOAD:
  case TargetOpcode::G_STORE: {
    LLT PtrTy = MRI.getType(I.getOperand(1).getReg());

    if (PtrTy != LLT::pointer(0, 64)) {
      LLVM_DEBUG(dbgs() << "Load/Store pointer has type: " << PtrTy
                        << ", expected: " << LLT::pointer(0, 64) << '\n');
      return false;
    }

    auto &MemOp = **I.memoperands_begin();
    if (MemOp.getOrdering() != AtomicOrdering::NotAtomic) {
      LLVM_DEBUG(dbgs() << "Atomic load/store not supported yet\n");
      return false;
    }
    unsigned MemSizeInBits = MemOp.getSize() * 8;

    const unsigned PtrReg = I.getOperand(1).getReg();
#ifndef NDEBUG
    const RegisterBank &PtrRB = *RBI.getRegBank(PtrReg, MRI, TRI);
    // Sanity-check the pointer register.
    assert(PtrRB.getID() == AArch64::GPRRegBankID &&
           "Load/Store pointer operand isn't a GPR");
    assert(MRI.getType(PtrReg).isPointer() &&
           "Load/Store pointer operand isn't a pointer");
#endif

    const unsigned ValReg = I.getOperand(0).getReg();
    const RegisterBank &RB = *RBI.getRegBank(ValReg, MRI, TRI);

    const unsigned NewOpc =
        selectLoadStoreUIOp(I.getOpcode(), RB.getID(), MemSizeInBits);
    if (NewOpc == I.getOpcode())
      return false;

    I.setDesc(TII.get(NewOpc));

    uint64_t Offset = 0;
    auto *PtrMI = MRI.getVRegDef(PtrReg);

    // Try to fold a GEP into our unsigned immediate addressing mode.
    if (PtrMI->getOpcode() == TargetOpcode::G_GEP) {
      if (auto COff = getConstantVRegVal(PtrMI->getOperand(2).getReg(), MRI)) {
        int64_t Imm = *COff;
        const unsigned Size = MemSizeInBits / 8;
        const unsigned Scale = Log2_32(Size);
        if ((Imm & (Size - 1)) == 0 && Imm >= 0 && Imm < (0x1000 << Scale)) {
          unsigned Ptr2Reg = PtrMI->getOperand(1).getReg();
          I.getOperand(1).setReg(Ptr2Reg);
          PtrMI = MRI.getVRegDef(Ptr2Reg);
          Offset = Imm / Size;
        }
      }
    }

    // If we haven't folded anything into our addressing mode yet, try to fold
    // a frame index into the base+offset.
    if (!Offset && PtrMI->getOpcode() == TargetOpcode::G_FRAME_INDEX)
      I.getOperand(1).ChangeToFrameIndex(PtrMI->getOperand(1).getIndex());

    I.addOperand(MachineOperand::CreateImm(Offset));

    // If we're storing a 0, use WZR/XZR.
    if (auto CVal = getConstantVRegVal(ValReg, MRI)) {
      if (*CVal == 0 && Opcode == TargetOpcode::G_STORE) {
        if (I.getOpcode() == AArch64::STRWui)
          I.getOperand(0).setReg(AArch64::WZR);
        else if (I.getOpcode() == AArch64::STRXui)
          I.getOperand(0).setReg(AArch64::XZR);
      }
    }

    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_SMULH:
  case TargetOpcode::G_UMULH: {
    // Reject the various things we don't support yet.
    if (unsupportedBinOp(I, RBI, MRI, TRI))
      return false;

    const unsigned DefReg = I.getOperand(0).getReg();
    const RegisterBank &RB = *RBI.getRegBank(DefReg, MRI, TRI);

    if (RB.getID() != AArch64::GPRRegBankID) {
      LLVM_DEBUG(dbgs() << "G_[SU]MULH on bank: " << RB << ", expected: GPR\n");
      return false;
    }

    if (Ty != LLT::scalar(64)) {
      LLVM_DEBUG(dbgs() << "G_[SU]MULH has type: " << Ty
                        << ", expected: " << LLT::scalar(64) << '\n');
      return false;
    }

    unsigned NewOpc = I.getOpcode() == TargetOpcode::G_SMULH ? AArch64::SMULHrr
                                                             : AArch64::UMULHrr;
    I.setDesc(TII.get(NewOpc));

    // Now that we selected an opcode, we need to constrain the register
    // operands to use appropriate classes.
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }
  case TargetOpcode::G_FADD:
  case TargetOpcode::G_FSUB:
  case TargetOpcode::G_FMUL:
  case TargetOpcode::G_FDIV:

  case TargetOpcode::G_OR:
  case TargetOpcode::G_SHL:
  case TargetOpcode::G_LSHR:
  case TargetOpcode::G_ASHR:
  case TargetOpcode::G_GEP: {
    // Reject the various things we don't support yet.
    if (unsupportedBinOp(I, RBI, MRI, TRI))
      return false;

    const unsigned OpSize = Ty.getSizeInBits();

    const unsigned DefReg = I.getOperand(0).getReg();
    const RegisterBank &RB = *RBI.getRegBank(DefReg, MRI, TRI);

    const unsigned NewOpc = selectBinaryOp(I.getOpcode(), RB.getID(), OpSize);
    if (NewOpc == I.getOpcode())
      return false;

    I.setDesc(TII.get(NewOpc));
    // FIXME: Should the type be always reset in setDesc?

    // Now that we selected an opcode, we need to constrain the register
    // operands to use appropriate classes.
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  case TargetOpcode::G_UADDO: {
    // TODO: Support other types.
    unsigned OpSize = Ty.getSizeInBits();
    if (OpSize != 32 && OpSize != 64) {
      LLVM_DEBUG(
          dbgs()
          << "G_UADDO currently only supported for 32 and 64 b types.\n");
      return false;
    }

    // TODO: Support vectors.
    if (Ty.isVector()) {
      LLVM_DEBUG(dbgs() << "G_UADDO currently only supported for scalars.\n");
      return false;
    }

    // Add and set the set condition flag.
    unsigned AddsOpc = OpSize == 32 ? AArch64::ADDSWrr : AArch64::ADDSXrr;
    MachineIRBuilder MIRBuilder(I);
    auto AddsMI = MIRBuilder.buildInstr(
        AddsOpc, {I.getOperand(0).getReg()},
        {I.getOperand(2).getReg(), I.getOperand(3).getReg()});
    constrainSelectedInstRegOperands(*AddsMI, TII, TRI, RBI);

    // Now, put the overflow result in the register given by the first operand
    // to the G_UADDO. CSINC increments the result when the predicate is false,
    // so to get the increment when it's true, we need to use the inverse. In
    // this case, we want to increment when carry is set.
    auto CsetMI = MIRBuilder
                      .buildInstr(AArch64::CSINCWr, {I.getOperand(1).getReg()},
                                  {AArch64::WZR, AArch64::WZR})
                      .addImm(getInvertedCondCode(AArch64CC::HS));
    constrainSelectedInstRegOperands(*CsetMI, TII, TRI, RBI);
    I.eraseFromParent();
    return true;
  }

  case TargetOpcode::G_PTR_MASK: {
    uint64_t Align = I.getOperand(2).getImm();
    if (Align >= 64 || Align == 0)
      return false;

    uint64_t Mask = ~((1ULL << Align) - 1);
    I.setDesc(TII.get(AArch64::ANDXri));
    I.getOperand(2).setImm(AArch64_AM::encodeLogicalImmediate(Mask, 64));

    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }
  case TargetOpcode::G_PTRTOINT:
  case TargetOpcode::G_TRUNC: {
    const LLT DstTy = MRI.getType(I.getOperand(0).getReg());
    const LLT SrcTy = MRI.getType(I.getOperand(1).getReg());

    const unsigned DstReg = I.getOperand(0).getReg();
    const unsigned SrcReg = I.getOperand(1).getReg();

    const RegisterBank &DstRB = *RBI.getRegBank(DstReg, MRI, TRI);
    const RegisterBank &SrcRB = *RBI.getRegBank(SrcReg, MRI, TRI);

    if (DstRB.getID() != SrcRB.getID()) {
      LLVM_DEBUG(
          dbgs() << "G_TRUNC/G_PTRTOINT input/output on different banks\n");
      return false;
    }

    if (DstRB.getID() == AArch64::GPRRegBankID) {
      const TargetRegisterClass *DstRC =
          getRegClassForTypeOnBank(DstTy, DstRB, RBI);
      if (!DstRC)
        return false;

      const TargetRegisterClass *SrcRC =
          getRegClassForTypeOnBank(SrcTy, SrcRB, RBI);
      if (!SrcRC)
        return false;

      if (!RBI.constrainGenericRegister(SrcReg, *SrcRC, MRI) ||
          !RBI.constrainGenericRegister(DstReg, *DstRC, MRI)) {
        LLVM_DEBUG(dbgs() << "Failed to constrain G_TRUNC/G_PTRTOINT\n");
        return false;
      }

      if (DstRC == SrcRC) {
        // Nothing to be done
      } else if (Opcode == TargetOpcode::G_TRUNC && DstTy == LLT::scalar(32) &&
                 SrcTy == LLT::scalar(64)) {
        llvm_unreachable("TableGen can import this case");
        return false;
      } else if (DstRC == &AArch64::GPR32RegClass &&
                 SrcRC == &AArch64::GPR64RegClass) {
        I.getOperand(1).setSubReg(AArch64::sub_32);
      } else {
        LLVM_DEBUG(
            dbgs() << "Unhandled mismatched classes in G_TRUNC/G_PTRTOINT\n");
        return false;
      }

      I.setDesc(TII.get(TargetOpcode::COPY));
      return true;
    } else if (DstRB.getID() == AArch64::FPRRegBankID) {
      if (DstTy == LLT::vector(4, 16) && SrcTy == LLT::vector(4, 32)) {
        I.setDesc(TII.get(AArch64::XTNv4i16));
        constrainSelectedInstRegOperands(I, TII, TRI, RBI);
        return true;
      }
    }

    return false;
  }

  case TargetOpcode::G_ANYEXT: {
    const unsigned DstReg = I.getOperand(0).getReg();
    const unsigned SrcReg = I.getOperand(1).getReg();

    const RegisterBank &RBDst = *RBI.getRegBank(DstReg, MRI, TRI);
    if (RBDst.getID() != AArch64::GPRRegBankID) {
      LLVM_DEBUG(dbgs() << "G_ANYEXT on bank: " << RBDst
                        << ", expected: GPR\n");
      return false;
    }

    const RegisterBank &RBSrc = *RBI.getRegBank(SrcReg, MRI, TRI);
    if (RBSrc.getID() != AArch64::GPRRegBankID) {
      LLVM_DEBUG(dbgs() << "G_ANYEXT on bank: " << RBSrc
                        << ", expected: GPR\n");
      return false;
    }

    const unsigned DstSize = MRI.getType(DstReg).getSizeInBits();

    if (DstSize == 0) {
      LLVM_DEBUG(dbgs() << "G_ANYEXT operand has no size, not a gvreg?\n");
      return false;
    }

    if (DstSize != 64 && DstSize > 32) {
      LLVM_DEBUG(dbgs() << "G_ANYEXT to size: " << DstSize
                        << ", expected: 32 or 64\n");
      return false;
    }
    // At this point G_ANYEXT is just like a plain COPY, but we need
    // to explicitly form the 64-bit value if any.
    if (DstSize > 32) {
      unsigned ExtSrc = MRI.createVirtualRegister(&AArch64::GPR64allRegClass);
      BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::SUBREG_TO_REG))
          .addDef(ExtSrc)
          .addImm(0)
          .addUse(SrcReg)
          .addImm(AArch64::sub_32);
      I.getOperand(1).setReg(ExtSrc);
    }
    return selectCopy(I, TII, MRI, TRI, RBI);
  }

  case TargetOpcode::G_ZEXT:
  case TargetOpcode::G_SEXT: {
    unsigned Opcode = I.getOpcode();
    const LLT DstTy = MRI.getType(I.getOperand(0).getReg()),
              SrcTy = MRI.getType(I.getOperand(1).getReg());
    const bool isSigned = Opcode == TargetOpcode::G_SEXT;
    const unsigned DefReg = I.getOperand(0).getReg();
    const unsigned SrcReg = I.getOperand(1).getReg();
    const RegisterBank &RB = *RBI.getRegBank(DefReg, MRI, TRI);

    if (RB.getID() != AArch64::GPRRegBankID) {
      LLVM_DEBUG(dbgs() << TII.getName(I.getOpcode()) << " on bank: " << RB
                        << ", expected: GPR\n");
      return false;
    }

    MachineInstr *ExtI;
    if (DstTy == LLT::scalar(64)) {
      // FIXME: Can we avoid manually doing this?
      if (!RBI.constrainGenericRegister(SrcReg, AArch64::GPR32RegClass, MRI)) {
        LLVM_DEBUG(dbgs() << "Failed to constrain " << TII.getName(Opcode)
                          << " operand\n");
        return false;
      }

      const unsigned SrcXReg =
          MRI.createVirtualRegister(&AArch64::GPR64RegClass);
      BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::SUBREG_TO_REG))
          .addDef(SrcXReg)
          .addImm(0)
          .addUse(SrcReg)
          .addImm(AArch64::sub_32);

      const unsigned NewOpc = isSigned ? AArch64::SBFMXri : AArch64::UBFMXri;
      ExtI = BuildMI(MBB, I, I.getDebugLoc(), TII.get(NewOpc))
                 .addDef(DefReg)
                 .addUse(SrcXReg)
                 .addImm(0)
                 .addImm(SrcTy.getSizeInBits() - 1);
    } else if (DstTy.isScalar() && DstTy.getSizeInBits() <= 32) {
      const unsigned NewOpc = isSigned ? AArch64::SBFMWri : AArch64::UBFMWri;
      ExtI = BuildMI(MBB, I, I.getDebugLoc(), TII.get(NewOpc))
                 .addDef(DefReg)
                 .addUse(SrcReg)
                 .addImm(0)
                 .addImm(SrcTy.getSizeInBits() - 1);
    } else {
      return false;
    }

    constrainSelectedInstRegOperands(*ExtI, TII, TRI, RBI);

    I.eraseFromParent();
    return true;
  }

  case TargetOpcode::G_SITOFP:
  case TargetOpcode::G_UITOFP:
  case TargetOpcode::G_FPTOSI:
  case TargetOpcode::G_FPTOUI: {
    const LLT DstTy = MRI.getType(I.getOperand(0).getReg()),
              SrcTy = MRI.getType(I.getOperand(1).getReg());
    const unsigned NewOpc = selectFPConvOpc(Opcode, DstTy, SrcTy);
    if (NewOpc == Opcode)
      return false;

    I.setDesc(TII.get(NewOpc));
    constrainSelectedInstRegOperands(I, TII, TRI, RBI);

    return true;
  }


  case TargetOpcode::G_INTTOPTR:
    // The importer is currently unable to import pointer types since they
    // didn't exist in SelectionDAG.
    return selectCopy(I, TII, MRI, TRI, RBI);

  case TargetOpcode::G_BITCAST:
    // Imported SelectionDAG rules can handle every bitcast except those that
    // bitcast from a type to the same type. Ideally, these shouldn't occur
    // but we might not run an optimizer that deletes them.
    if (MRI.getType(I.getOperand(0).getReg()) ==
        MRI.getType(I.getOperand(1).getReg()))
      return selectCopy(I, TII, MRI, TRI, RBI);
    return false;

  case TargetOpcode::G_SELECT: {
    if (MRI.getType(I.getOperand(1).getReg()) != LLT::scalar(1)) {
      LLVM_DEBUG(dbgs() << "G_SELECT cond has type: " << Ty
                        << ", expected: " << LLT::scalar(1) << '\n');
      return false;
    }

    const unsigned CondReg = I.getOperand(1).getReg();
    const unsigned TReg = I.getOperand(2).getReg();
    const unsigned FReg = I.getOperand(3).getReg();

    unsigned CSelOpc = 0;

    if (Ty == LLT::scalar(32)) {
      CSelOpc = AArch64::CSELWr;
    } else if (Ty == LLT::scalar(64) || Ty == LLT::pointer(0, 64)) {
      CSelOpc = AArch64::CSELXr;
    } else {
      return false;
    }

    MachineInstr &TstMI =
        *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::ANDSWri))
             .addDef(AArch64::WZR)
             .addUse(CondReg)
             .addImm(AArch64_AM::encodeLogicalImmediate(1, 32));

    MachineInstr &CSelMI = *BuildMI(MBB, I, I.getDebugLoc(), TII.get(CSelOpc))
                                .addDef(I.getOperand(0).getReg())
                                .addUse(TReg)
                                .addUse(FReg)
                                .addImm(AArch64CC::NE);

    constrainSelectedInstRegOperands(TstMI, TII, TRI, RBI);
    constrainSelectedInstRegOperands(CSelMI, TII, TRI, RBI);

    I.eraseFromParent();
    return true;
  }
  case TargetOpcode::G_ICMP: {
    if (Ty != LLT::scalar(32)) {
      LLVM_DEBUG(dbgs() << "G_ICMP result has type: " << Ty
                        << ", expected: " << LLT::scalar(32) << '\n');
      return false;
    }

    unsigned CmpOpc = 0;
    unsigned ZReg = 0;

    LLT CmpTy = MRI.getType(I.getOperand(2).getReg());
    if (CmpTy == LLT::scalar(32)) {
      CmpOpc = AArch64::SUBSWrr;
      ZReg = AArch64::WZR;
    } else if (CmpTy == LLT::scalar(64) || CmpTy.isPointer()) {
      CmpOpc = AArch64::SUBSXrr;
      ZReg = AArch64::XZR;
    } else {
      return false;
    }

    // CSINC increments the result by one when the condition code is false.
    // Therefore, we have to invert the predicate to get an increment by 1 when
    // the predicate is true.
    const AArch64CC::CondCode invCC =
        changeICMPPredToAArch64CC(CmpInst::getInversePredicate(
            (CmpInst::Predicate)I.getOperand(1).getPredicate()));

    MachineInstr &CmpMI = *BuildMI(MBB, I, I.getDebugLoc(), TII.get(CmpOpc))
                               .addDef(ZReg)
                               .addUse(I.getOperand(2).getReg())
                               .addUse(I.getOperand(3).getReg());

    MachineInstr &CSetMI =
        *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::CSINCWr))
             .addDef(I.getOperand(0).getReg())
             .addUse(AArch64::WZR)
             .addUse(AArch64::WZR)
             .addImm(invCC);

    constrainSelectedInstRegOperands(CmpMI, TII, TRI, RBI);
    constrainSelectedInstRegOperands(CSetMI, TII, TRI, RBI);

    I.eraseFromParent();
    return true;
  }

  case TargetOpcode::G_FCMP: {
    if (Ty != LLT::scalar(32)) {
      LLVM_DEBUG(dbgs() << "G_FCMP result has type: " << Ty
                        << ", expected: " << LLT::scalar(32) << '\n');
      return false;
    }

    unsigned CmpOpc = 0;
    LLT CmpTy = MRI.getType(I.getOperand(2).getReg());
    if (CmpTy == LLT::scalar(32)) {
      CmpOpc = AArch64::FCMPSrr;
    } else if (CmpTy == LLT::scalar(64)) {
      CmpOpc = AArch64::FCMPDrr;
    } else {
      return false;
    }

    // FIXME: regbank

    AArch64CC::CondCode CC1, CC2;
    changeFCMPPredToAArch64CC(
        (CmpInst::Predicate)I.getOperand(1).getPredicate(), CC1, CC2);

    MachineInstr &CmpMI = *BuildMI(MBB, I, I.getDebugLoc(), TII.get(CmpOpc))
                               .addUse(I.getOperand(2).getReg())
                               .addUse(I.getOperand(3).getReg());

    const unsigned DefReg = I.getOperand(0).getReg();
    unsigned Def1Reg = DefReg;
    if (CC2 != AArch64CC::AL)
      Def1Reg = MRI.createVirtualRegister(&AArch64::GPR32RegClass);

    MachineInstr &CSetMI =
        *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::CSINCWr))
             .addDef(Def1Reg)
             .addUse(AArch64::WZR)
             .addUse(AArch64::WZR)
             .addImm(getInvertedCondCode(CC1));

    if (CC2 != AArch64CC::AL) {
      unsigned Def2Reg = MRI.createVirtualRegister(&AArch64::GPR32RegClass);
      MachineInstr &CSet2MI =
          *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::CSINCWr))
               .addDef(Def2Reg)
               .addUse(AArch64::WZR)
               .addUse(AArch64::WZR)
               .addImm(getInvertedCondCode(CC2));
      MachineInstr &OrMI =
          *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::ORRWrr))
               .addDef(DefReg)
               .addUse(Def1Reg)
               .addUse(Def2Reg);
      constrainSelectedInstRegOperands(OrMI, TII, TRI, RBI);
      constrainSelectedInstRegOperands(CSet2MI, TII, TRI, RBI);
    }

    constrainSelectedInstRegOperands(CmpMI, TII, TRI, RBI);
    constrainSelectedInstRegOperands(CSetMI, TII, TRI, RBI);

    I.eraseFromParent();
    return true;
  }
  case TargetOpcode::G_VASTART:
    return STI.isTargetDarwin() ? selectVaStartDarwin(I, MF, MRI)
                                : selectVaStartAAPCS(I, MF, MRI);
  case TargetOpcode::G_INTRINSIC_W_SIDE_EFFECTS:
    if (!I.getOperand(0).isIntrinsicID())
      return false;
    if (I.getOperand(0).getIntrinsicID() != Intrinsic::trap)
      return false;
    BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::BRK))
      .addImm(1);
    I.eraseFromParent();
    return true;
  case TargetOpcode::G_IMPLICIT_DEF: {
    I.setDesc(TII.get(TargetOpcode::IMPLICIT_DEF));
    const LLT DstTy = MRI.getType(I.getOperand(0).getReg());
    const unsigned DstReg = I.getOperand(0).getReg();
    const RegisterBank &DstRB = *RBI.getRegBank(DstReg, MRI, TRI);
    const TargetRegisterClass *DstRC =
        getRegClassForTypeOnBank(DstTy, DstRB, RBI);
    RBI.constrainGenericRegister(DstReg, *DstRC, MRI);
    return true;
  }
  case TargetOpcode::G_BLOCK_ADDR: {
    if (TM.getCodeModel() == CodeModel::Large) {
      materializeLargeCMVal(I, I.getOperand(1).getBlockAddress(), 0);
      I.eraseFromParent();
      return true;
    } else {
      I.setDesc(TII.get(AArch64::MOVaddrBA));
      auto MovMI = BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::MOVaddrBA),
                           I.getOperand(0).getReg())
                       .addBlockAddress(I.getOperand(1).getBlockAddress(),
                                        /* Offset */ 0, AArch64II::MO_PAGE)
                       .addBlockAddress(
                           I.getOperand(1).getBlockAddress(), /* Offset */ 0,
                           AArch64II::MO_NC | AArch64II::MO_PAGEOFF);
      I.eraseFromParent();
      return constrainSelectedInstRegOperands(*MovMI, TII, TRI, RBI);
    }
  }
  case TargetOpcode::G_BUILD_VECTOR:
    return selectBuildVector(I, MRI);
  case TargetOpcode::G_MERGE_VALUES:
    return selectMergeValues(I, MRI);
  case TargetOpcode::G_UNMERGE_VALUES:
    return selectUnmergeValues(I, MRI);
  case TargetOpcode::G_SHUFFLE_VECTOR:
    return selectShuffleVector(I, MRI);
  case TargetOpcode::G_EXTRACT_VECTOR_ELT:
    return selectExtractElt(I, MRI);
  case TargetOpcode::G_INSERT_VECTOR_ELT:
    return selectInsertElt(I, MRI);
  case TargetOpcode::G_CONCAT_VECTORS:
    return selectConcatVectors(I, MRI);
  }

  return false;
}

MachineInstr *AArch64InstructionSelector::emitScalarToVector(
    unsigned EltSize, const TargetRegisterClass *DstRC, unsigned Scalar,
    MachineIRBuilder &MIRBuilder) const {
  auto Undef = MIRBuilder.buildInstr(TargetOpcode::IMPLICIT_DEF, {DstRC}, {});

  auto BuildFn = [&](unsigned SubregIndex) {
    auto Ins =
        MIRBuilder
            .buildInstr(TargetOpcode::INSERT_SUBREG, {DstRC}, {Undef, Scalar})
            .addImm(SubregIndex);
    constrainSelectedInstRegOperands(*Undef, TII, TRI, RBI);
    constrainSelectedInstRegOperands(*Ins, TII, TRI, RBI);
    return &*Ins;
  };

  switch (EltSize) {
  case 16:
    return BuildFn(AArch64::hsub);
  case 32:
    return BuildFn(AArch64::ssub);
  case 64:
    return BuildFn(AArch64::dsub);
  default:
    return nullptr;
  }
}

bool AArch64InstructionSelector::selectMergeValues(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_MERGE_VALUES && "unexpected opcode");
  const LLT DstTy = MRI.getType(I.getOperand(0).getReg());
  const LLT SrcTy = MRI.getType(I.getOperand(1).getReg());
  assert(!DstTy.isVector() && !SrcTy.isVector() && "invalid merge operation");

  // At the moment we only support merging two s32s into an s64.
  if (I.getNumOperands() != 3)
    return false;
  if (DstTy.getSizeInBits() != 64 || SrcTy.getSizeInBits() != 32)
    return false;
  const RegisterBank &RB = *RBI.getRegBank(I.getOperand(1).getReg(), MRI, TRI);
  if (RB.getID() != AArch64::GPRRegBankID)
    return false;

  auto *DstRC = &AArch64::GPR64RegClass;
  unsigned SubToRegDef = MRI.createVirtualRegister(DstRC);
  MachineInstr &SubRegMI = *BuildMI(*I.getParent(), I, I.getDebugLoc(),
                                    TII.get(TargetOpcode::SUBREG_TO_REG))
                                .addDef(SubToRegDef)
                                .addImm(0)
                                .addUse(I.getOperand(1).getReg())
                                .addImm(AArch64::sub_32);
  unsigned SubToRegDef2 = MRI.createVirtualRegister(DstRC);
  // Need to anyext the second scalar before we can use bfm
  MachineInstr &SubRegMI2 = *BuildMI(*I.getParent(), I, I.getDebugLoc(),
                                    TII.get(TargetOpcode::SUBREG_TO_REG))
                                .addDef(SubToRegDef2)
                                .addImm(0)
                                .addUse(I.getOperand(2).getReg())
                                .addImm(AArch64::sub_32);
  MachineInstr &BFM =
      *BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(AArch64::BFMXri))
           .addDef(I.getOperand(0).getReg())
           .addUse(SubToRegDef)
           .addUse(SubToRegDef2)
           .addImm(32)
           .addImm(31);
  constrainSelectedInstRegOperands(SubRegMI, TII, TRI, RBI);
  constrainSelectedInstRegOperands(SubRegMI2, TII, TRI, RBI);
  constrainSelectedInstRegOperands(BFM, TII, TRI, RBI);
  I.eraseFromParent();
  return true;
}

static bool getLaneCopyOpcode(unsigned &CopyOpc, unsigned &ExtractSubReg,
                              const unsigned EltSize) {
  // Choose a lane copy opcode and subregister based off of the size of the
  // vector's elements.
  switch (EltSize) {
  case 16:
    CopyOpc = AArch64::CPYi16;
    ExtractSubReg = AArch64::hsub;
    break;
  case 32:
    CopyOpc = AArch64::CPYi32;
    ExtractSubReg = AArch64::ssub;
    break;
  case 64:
    CopyOpc = AArch64::CPYi64;
    ExtractSubReg = AArch64::dsub;
    break;
  default:
    // Unknown size, bail out.
    LLVM_DEBUG(dbgs() << "Elt size '" << EltSize << "' unsupported.\n");
    return false;
  }
  return true;
}

/// Given a register \p Reg, find the value of a constant defining \p Reg.
/// Return true if one could be found, and store it in \p Val. Return false
/// otherwise.
static bool getConstantValueForReg(unsigned Reg, MachineRegisterInfo &MRI,
                                   unsigned &Val) {
  // Look at the def of the register.
  MachineInstr *Def = MRI.getVRegDef(Reg);
  if (!Def)
    return false;

  // Find the first definition which isn't a copy.
  if (Def->isCopy()) {
    Reg = Def->getOperand(1).getReg();
    auto It = find_if_not(MRI.reg_nodbg_instructions(Reg),
                          [](const MachineInstr &MI) { return MI.isCopy(); });
    if (It == MRI.reg_instr_nodbg_end()) {
      LLVM_DEBUG(dbgs() << "Couldn't find non-copy def for register\n");
      return false;
    }
    Def = &*It;
  }

  // TODO: Handle opcodes other than G_CONSTANT.
  if (Def->getOpcode() != TargetOpcode::G_CONSTANT) {
    LLVM_DEBUG(dbgs() << "VRegs defined by anything other than G_CONSTANT "
                         "currently unsupported.\n");
    return false;
  }

  // Return the constant value associated with the operand.
  Val = Def->getOperand(1).getCImm()->getLimitedValue();
  return true;
}

MachineInstr *AArch64InstructionSelector::emitExtractVectorElt(
    Optional<unsigned> DstReg, const RegisterBank &DstRB, LLT ScalarTy,
    unsigned VecReg, unsigned LaneIdx, MachineIRBuilder &MIRBuilder) const {
  MachineRegisterInfo &MRI = *MIRBuilder.getMRI();
  unsigned CopyOpc = 0;
  unsigned ExtractSubReg = 0;
  if (!getLaneCopyOpcode(CopyOpc, ExtractSubReg, ScalarTy.getSizeInBits())) {
    LLVM_DEBUG(
        dbgs() << "Couldn't determine lane copy opcode for instruction.\n");
    return nullptr;
  }

  const TargetRegisterClass *DstRC =
      getRegClassForTypeOnBank(ScalarTy, DstRB, RBI, true);
  if (!DstRC) {
    LLVM_DEBUG(dbgs() << "Could not determine destination register class.\n");
    return nullptr;
  }

  const RegisterBank &VecRB = *RBI.getRegBank(VecReg, MRI, TRI);
  const LLT &VecTy = MRI.getType(VecReg);
  const TargetRegisterClass *VecRC =
      getRegClassForTypeOnBank(VecTy, VecRB, RBI, true);
  if (!VecRC) {
    LLVM_DEBUG(dbgs() << "Could not determine source register class.\n");
    return nullptr;
  }

  // The register that we're going to copy into.
  unsigned InsertReg = VecReg;
  if (!DstReg)
    DstReg = MRI.createVirtualRegister(DstRC);
  // If the lane index is 0, we just use a subregister COPY.
  if (LaneIdx == 0) {
    auto Copy = MIRBuilder.buildInstr(TargetOpcode::COPY, {*DstReg}, {})
                    .addReg(VecReg, 0, ExtractSubReg);
    RBI.constrainGenericRegister(*DstReg, *DstRC, MRI);
    return &*Copy;
  }

  // Lane copies require 128-bit wide registers. If we're dealing with an
  // unpacked vector, then we need to move up to that width. Insert an implicit
  // def and a subregister insert to get us there.
  if (VecTy.getSizeInBits() != 128) {
    MachineInstr *ScalarToVector = emitScalarToVector(
        VecTy.getSizeInBits(), &AArch64::FPR128RegClass, VecReg, MIRBuilder);
    if (!ScalarToVector)
      return nullptr;
    InsertReg = ScalarToVector->getOperand(0).getReg();
  }

  MachineInstr *LaneCopyMI =
      MIRBuilder.buildInstr(CopyOpc, {*DstReg}, {InsertReg}).addImm(LaneIdx);
  constrainSelectedInstRegOperands(*LaneCopyMI, TII, TRI, RBI);

  // Make sure that we actually constrain the initial copy.
  RBI.constrainGenericRegister(*DstReg, *DstRC, MRI);
  return LaneCopyMI;
}

bool AArch64InstructionSelector::selectExtractElt(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_EXTRACT_VECTOR_ELT &&
         "unexpected opcode!");
  unsigned DstReg = I.getOperand(0).getReg();
  const LLT NarrowTy = MRI.getType(DstReg);
  const unsigned SrcReg = I.getOperand(1).getReg();
  const LLT WideTy = MRI.getType(SrcReg);
  (void)WideTy;
  assert(WideTy.getSizeInBits() >= NarrowTy.getSizeInBits() &&
         "source register size too small!");
  assert(NarrowTy.isScalar() && "cannot extract vector into vector!");

  // Need the lane index to determine the correct copy opcode.
  MachineOperand &LaneIdxOp = I.getOperand(2);
  assert(LaneIdxOp.isReg() && "Lane index operand was not a register?");

  if (RBI.getRegBank(DstReg, MRI, TRI)->getID() != AArch64::FPRRegBankID) {
    LLVM_DEBUG(dbgs() << "Cannot extract into GPR.\n");
    return false;
  }

  // Find the index to extract from.
  unsigned LaneIdx = 0;
  if (!getConstantValueForReg(LaneIdxOp.getReg(), MRI, LaneIdx))
    return false;

  MachineIRBuilder MIRBuilder(I);

  const RegisterBank &DstRB = *RBI.getRegBank(DstReg, MRI, TRI);
  MachineInstr *Extract = emitExtractVectorElt(DstReg, DstRB, NarrowTy, SrcReg,
                                               LaneIdx, MIRBuilder);
  if (!Extract)
    return false;

  I.eraseFromParent();
  return true;
}

bool AArch64InstructionSelector::selectSplitVectorUnmerge(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  unsigned NumElts = I.getNumOperands() - 1;
  unsigned SrcReg = I.getOperand(NumElts).getReg();
  const LLT NarrowTy = MRI.getType(I.getOperand(0).getReg());
  const LLT SrcTy = MRI.getType(SrcReg);

  assert(NarrowTy.isVector() && "Expected an unmerge into vectors");
  if (SrcTy.getSizeInBits() > 128) {
    LLVM_DEBUG(dbgs() << "Unexpected vector type for vec split unmerge");
    return false;
  }

  MachineIRBuilder MIB(I);

  // We implement a split vector operation by treating the sub-vectors as
  // scalars and extracting them.
  const RegisterBank &DstRB =
      *RBI.getRegBank(I.getOperand(0).getReg(), MRI, TRI);
  for (unsigned OpIdx = 0; OpIdx < NumElts; ++OpIdx) {
    unsigned Dst = I.getOperand(OpIdx).getReg();
    MachineInstr *Extract =
        emitExtractVectorElt(Dst, DstRB, NarrowTy, SrcReg, OpIdx, MIB);
    if (!Extract)
      return false;
  }
  I.eraseFromParent();
  return true;
}

bool AArch64InstructionSelector::selectUnmergeValues(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_UNMERGE_VALUES &&
         "unexpected opcode");

  // TODO: Handle unmerging into GPRs and from scalars to scalars.
  if (RBI.getRegBank(I.getOperand(0).getReg(), MRI, TRI)->getID() !=
          AArch64::FPRRegBankID ||
      RBI.getRegBank(I.getOperand(1).getReg(), MRI, TRI)->getID() !=
          AArch64::FPRRegBankID) {
    LLVM_DEBUG(dbgs() << "Unmerging vector-to-gpr and scalar-to-scalar "
                         "currently unsupported.\n");
    return false;
  }

  // The last operand is the vector source register, and every other operand is
  // a register to unpack into.
  unsigned NumElts = I.getNumOperands() - 1;
  unsigned SrcReg = I.getOperand(NumElts).getReg();
  const LLT NarrowTy = MRI.getType(I.getOperand(0).getReg());
  const LLT WideTy = MRI.getType(SrcReg);
  (void)WideTy;
  assert(WideTy.isVector() && "can only unmerge from vector types!");
  assert(WideTy.getSizeInBits() > NarrowTy.getSizeInBits() &&
         "source register size too small!");

  if (!NarrowTy.isScalar())
    return selectSplitVectorUnmerge(I, MRI);

  MachineIRBuilder MIB(I);

  // Choose a lane copy opcode and subregister based off of the size of the
  // vector's elements.
  unsigned CopyOpc = 0;
  unsigned ExtractSubReg = 0;
  if (!getLaneCopyOpcode(CopyOpc, ExtractSubReg, NarrowTy.getSizeInBits()))
    return false;

  // Set up for the lane copies.
  MachineBasicBlock &MBB = *I.getParent();

  // Stores the registers we'll be copying from.
  SmallVector<unsigned, 4> InsertRegs;

  // We'll use the first register twice, so we only need NumElts-1 registers.
  unsigned NumInsertRegs = NumElts - 1;

  // If our elements fit into exactly 128 bits, then we can copy from the source
  // directly. Otherwise, we need to do a bit of setup with some subregister
  // inserts.
  if (NarrowTy.getSizeInBits() * NumElts == 128) {
    InsertRegs = SmallVector<unsigned, 4>(NumInsertRegs, SrcReg);
  } else {
    // No. We have to perform subregister inserts. For each insert, create an
    // implicit def and a subregister insert, and save the register we create.
    for (unsigned Idx = 0; Idx < NumInsertRegs; ++Idx) {
      unsigned ImpDefReg = MRI.createVirtualRegister(&AArch64::FPR128RegClass);
      MachineInstr &ImpDefMI =
          *BuildMI(MBB, I, I.getDebugLoc(), TII.get(TargetOpcode::IMPLICIT_DEF),
                   ImpDefReg);

      // Now, create the subregister insert from SrcReg.
      unsigned InsertReg = MRI.createVirtualRegister(&AArch64::FPR128RegClass);
      MachineInstr &InsMI =
          *BuildMI(MBB, I, I.getDebugLoc(),
                   TII.get(TargetOpcode::INSERT_SUBREG), InsertReg)
               .addUse(ImpDefReg)
               .addUse(SrcReg)
               .addImm(AArch64::dsub);

      constrainSelectedInstRegOperands(ImpDefMI, TII, TRI, RBI);
      constrainSelectedInstRegOperands(InsMI, TII, TRI, RBI);

      // Save the register so that we can copy from it after.
      InsertRegs.push_back(InsertReg);
    }
  }

  // Now that we've created any necessary subregister inserts, we can
  // create the copies.
  //
  // Perform the first copy separately as a subregister copy.
  unsigned CopyTo = I.getOperand(0).getReg();
  auto FirstCopy = MIB.buildInstr(TargetOpcode::COPY, {CopyTo}, {})
                       .addReg(InsertRegs[0], 0, ExtractSubReg);
  constrainSelectedInstRegOperands(*FirstCopy, TII, TRI, RBI);

  // Now, perform the remaining copies as vector lane copies.
  unsigned LaneIdx = 1;
  for (unsigned InsReg : InsertRegs) {
    unsigned CopyTo = I.getOperand(LaneIdx).getReg();
    MachineInstr &CopyInst =
        *BuildMI(MBB, I, I.getDebugLoc(), TII.get(CopyOpc), CopyTo)
             .addUse(InsReg)
             .addImm(LaneIdx);
    constrainSelectedInstRegOperands(CopyInst, TII, TRI, RBI);
    ++LaneIdx;
  }

  // Separately constrain the first copy's destination. Because of the
  // limitation in constrainOperandRegClass, we can't guarantee that this will
  // actually be constrained. So, do it ourselves using the second operand.
  const TargetRegisterClass *RC =
      MRI.getRegClassOrNull(I.getOperand(1).getReg());
  if (!RC) {
    LLVM_DEBUG(dbgs() << "Couldn't constrain copy destination.\n");
    return false;
  }

  RBI.constrainGenericRegister(CopyTo, *RC, MRI);
  I.eraseFromParent();
  return true;
}

bool AArch64InstructionSelector::selectConcatVectors(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_CONCAT_VECTORS &&
         "Unexpected opcode");
  unsigned Dst = I.getOperand(0).getReg();
  unsigned Op1 = I.getOperand(1).getReg();
  unsigned Op2 = I.getOperand(2).getReg();
  MachineIRBuilder MIRBuilder(I);
  MachineInstr *ConcatMI = emitVectorConcat(Dst, Op1, Op2, MIRBuilder);
  if (!ConcatMI)
    return false;
  I.eraseFromParent();
  return true;
}

void AArch64InstructionSelector::collectShuffleMaskIndices(
    MachineInstr &I, MachineRegisterInfo &MRI,
    SmallVectorImpl<int> &Idxs) const {
  MachineInstr *MaskDef = MRI.getVRegDef(I.getOperand(3).getReg());
  assert(
      MaskDef->getOpcode() == TargetOpcode::G_BUILD_VECTOR &&
      "G_SHUFFLE_VECTOR should have a constant mask operand as G_BUILD_VECTOR");
  // Find the constant indices.
  for (unsigned i = 1, e = MaskDef->getNumOperands(); i < e; ++i) {
    MachineInstr *ScalarDef = MRI.getVRegDef(MaskDef->getOperand(i).getReg());
    assert(ScalarDef && "Could not find vreg def of shufflevec index op");
    // Look through copies.
    while (ScalarDef->getOpcode() == TargetOpcode::COPY) {
      ScalarDef = MRI.getVRegDef(ScalarDef->getOperand(1).getReg());
      assert(ScalarDef && "Could not find def of copy operand");
    }
    assert(ScalarDef->getOpcode() == TargetOpcode::G_CONSTANT);
    Idxs.push_back(ScalarDef->getOperand(1).getCImm()->getSExtValue());
  }
}

unsigned
AArch64InstructionSelector::emitConstantPoolEntry(Constant *CPVal,
                                                  MachineFunction &MF) const {
  Type *CPTy = CPVal->getType()->getPointerTo();
  unsigned Align = MF.getDataLayout().getPrefTypeAlignment(CPTy);
  if (Align == 0)
    Align = MF.getDataLayout().getTypeAllocSize(CPTy);

  MachineConstantPool *MCP = MF.getConstantPool();
  return MCP->getConstantPoolIndex(CPVal, Align);
}

MachineInstr *AArch64InstructionSelector::emitLoadFromConstantPool(
    Constant *CPVal, MachineIRBuilder &MIRBuilder) const {
  unsigned CPIdx = emitConstantPoolEntry(CPVal, MIRBuilder.getMF());

  auto Adrp =
      MIRBuilder.buildInstr(AArch64::ADRP, {&AArch64::GPR64RegClass}, {})
          .addConstantPoolIndex(CPIdx, 0, AArch64II::MO_PAGE);

  MachineInstr *LoadMI = nullptr;
  switch (MIRBuilder.getDataLayout().getTypeStoreSize(CPVal->getType())) {
  case 16:
    LoadMI =
        &*MIRBuilder
              .buildInstr(AArch64::LDRQui, {&AArch64::FPR128RegClass}, {Adrp})
              .addConstantPoolIndex(CPIdx, 0,
                                    AArch64II::MO_PAGEOFF | AArch64II::MO_NC);
    break;
  case 8:
    LoadMI = &*MIRBuilder
                 .buildInstr(AArch64::LDRDui, {&AArch64::FPR64RegClass}, {Adrp})
                 .addConstantPoolIndex(
                     CPIdx, 0, AArch64II::MO_PAGEOFF | AArch64II::MO_NC);
    break;
  default:
    LLVM_DEBUG(dbgs() << "Could not load from constant pool of type "
                      << *CPVal->getType());
    return nullptr;
  }
  constrainSelectedInstRegOperands(*Adrp, TII, TRI, RBI);
  constrainSelectedInstRegOperands(*LoadMI, TII, TRI, RBI);
  return LoadMI;
}

/// Return an <Opcode, SubregIndex> pair to do an vector elt insert of a given
/// size and RB.
static std::pair<unsigned, unsigned>
getInsertVecEltOpInfo(const RegisterBank &RB, unsigned EltSize) {
  unsigned Opc, SubregIdx;
  if (RB.getID() == AArch64::GPRRegBankID) {
    if (EltSize == 32) {
      Opc = AArch64::INSvi32gpr;
      SubregIdx = AArch64::ssub;
    } else if (EltSize == 64) {
      Opc = AArch64::INSvi64gpr;
      SubregIdx = AArch64::dsub;
    } else {
      llvm_unreachable("invalid elt size!");
    }
  } else {
    if (EltSize == 8) {
      Opc = AArch64::INSvi8lane;
      SubregIdx = AArch64::bsub;
    } else if (EltSize == 16) {
      Opc = AArch64::INSvi16lane;
      SubregIdx = AArch64::hsub;
    } else if (EltSize == 32) {
      Opc = AArch64::INSvi32lane;
      SubregIdx = AArch64::ssub;
    } else if (EltSize == 64) {
      Opc = AArch64::INSvi64lane;
      SubregIdx = AArch64::dsub;
    } else {
      llvm_unreachable("invalid elt size!");
    }
  }
  return std::make_pair(Opc, SubregIdx);
}

MachineInstr *AArch64InstructionSelector::emitVectorConcat(
    Optional<unsigned> Dst, unsigned Op1, unsigned Op2,
    MachineIRBuilder &MIRBuilder) const {
  // We implement a vector concat by:
  // 1. Use scalar_to_vector to insert the lower vector into the larger dest
  // 2. Insert the upper vector into the destination's upper element
  // TODO: some of this code is common with G_BUILD_VECTOR handling.
  MachineRegisterInfo &MRI = MIRBuilder.getMF().getRegInfo();

  const LLT Op1Ty = MRI.getType(Op1);
  const LLT Op2Ty = MRI.getType(Op2);

  if (Op1Ty != Op2Ty) {
    LLVM_DEBUG(dbgs() << "Could not do vector concat of differing vector tys");
    return nullptr;
  }
  assert(Op1Ty.isVector() && "Expected a vector for vector concat");

  if (Op1Ty.getSizeInBits() >= 128) {
    LLVM_DEBUG(dbgs() << "Vector concat not supported for full size vectors");
    return nullptr;
  }

  // At the moment we just support 64 bit vector concats.
  if (Op1Ty.getSizeInBits() != 64) {
    LLVM_DEBUG(dbgs() << "Vector concat supported for 64b vectors");
    return nullptr;
  }

  const LLT ScalarTy = LLT::scalar(Op1Ty.getSizeInBits());
  const RegisterBank &FPRBank = *RBI.getRegBank(Op1, MRI, TRI);
  const TargetRegisterClass *DstRC =
      getMinClassForRegBank(FPRBank, Op1Ty.getSizeInBits() * 2);

  MachineInstr *WidenedOp1 =
      emitScalarToVector(ScalarTy.getSizeInBits(), DstRC, Op1, MIRBuilder);
  MachineInstr *WidenedOp2 =
      emitScalarToVector(ScalarTy.getSizeInBits(), DstRC, Op2, MIRBuilder);
  if (!WidenedOp1 || !WidenedOp2) {
    LLVM_DEBUG(dbgs() << "Could not emit a vector from scalar value");
    return nullptr;
  }

  // Now do the insert of the upper element.
  unsigned InsertOpc, InsSubRegIdx;
  std::tie(InsertOpc, InsSubRegIdx) =
      getInsertVecEltOpInfo(FPRBank, ScalarTy.getSizeInBits());

  if (!Dst)
    Dst = MRI.createVirtualRegister(DstRC);
  auto InsElt =
      MIRBuilder
          .buildInstr(InsertOpc, {*Dst}, {WidenedOp1->getOperand(0).getReg()})
          .addImm(1) /* Lane index */
          .addUse(WidenedOp2->getOperand(0).getReg())
          .addImm(0);
  constrainSelectedInstRegOperands(*InsElt, TII, TRI, RBI);
  return &*InsElt;
}

bool AArch64InstructionSelector::selectShuffleVector(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  const LLT DstTy = MRI.getType(I.getOperand(0).getReg());
  unsigned Src1Reg = I.getOperand(1).getReg();
  const LLT Src1Ty = MRI.getType(Src1Reg);
  unsigned Src2Reg = I.getOperand(2).getReg();
  const LLT Src2Ty = MRI.getType(Src2Reg);

  MachineBasicBlock &MBB = *I.getParent();
  MachineFunction &MF = *MBB.getParent();
  LLVMContext &Ctx = MF.getFunction().getContext();

  // G_SHUFFLE_VECTOR doesn't really have a strictly enforced constant mask
  // operand, it comes in as a normal vector value which we have to analyze to
  // find the mask indices.
  SmallVector<int, 8> Mask;
  collectShuffleMaskIndices(I, MRI, Mask);
  assert(!Mask.empty() && "Expected to find mask indices");

  // G_SHUFFLE_VECTOR is weird in that the source operands can be scalars, if
  // it's originated from a <1 x T> type. Those should have been lowered into
  // G_BUILD_VECTOR earlier.
  if (!Src1Ty.isVector() || !Src2Ty.isVector()) {
    LLVM_DEBUG(dbgs() << "Could not select a \"scalar\" G_SHUFFLE_VECTOR\n");
    return false;
  }

  unsigned BytesPerElt = DstTy.getElementType().getSizeInBits() / 8;

  SmallVector<Constant *, 64> CstIdxs;
  for (int Val : Mask) {
    for (unsigned Byte = 0; Byte < BytesPerElt; ++Byte) {
      unsigned Offset = Byte + Val * BytesPerElt;
      CstIdxs.emplace_back(ConstantInt::get(Type::getInt8Ty(Ctx), Offset));
    }
  }

  MachineIRBuilder MIRBuilder(I);

  // Use a constant pool to load the index vector for TBL.
  Constant *CPVal = ConstantVector::get(CstIdxs);
  MachineInstr *IndexLoad = emitLoadFromConstantPool(CPVal, MIRBuilder);
  if (!IndexLoad) {
    LLVM_DEBUG(dbgs() << "Could not load from a constant pool");
    return false;
  }

  if (DstTy.getSizeInBits() != 128) {
    assert(DstTy.getSizeInBits() == 64 && "Unexpected shuffle result ty");
    // This case can be done with TBL1.
    MachineInstr *Concat = emitVectorConcat(None, Src1Reg, Src2Reg, MIRBuilder);
    if (!Concat) {
      LLVM_DEBUG(dbgs() << "Could not do vector concat for tbl1");
      return false;
    }

    // The constant pool load will be 64 bits, so need to convert to FPR128 reg.
    IndexLoad =
        emitScalarToVector(64, &AArch64::FPR128RegClass,
                           IndexLoad->getOperand(0).getReg(), MIRBuilder);

    auto TBL1 = MIRBuilder.buildInstr(
        AArch64::TBLv16i8One, {&AArch64::FPR128RegClass},
        {Concat->getOperand(0).getReg(), IndexLoad->getOperand(0).getReg()});
    constrainSelectedInstRegOperands(*TBL1, TII, TRI, RBI);

    auto Copy =
        MIRBuilder
            .buildInstr(TargetOpcode::COPY, {I.getOperand(0).getReg()}, {})
            .addReg(TBL1.getReg(0), 0, AArch64::dsub);
    RBI.constrainGenericRegister(Copy.getReg(0), AArch64::FPR64RegClass, MRI);
    I.eraseFromParent();
    return true;
  }

  // For TBL2 we need to emit a REG_SEQUENCE to tie together two consecutive
  // Q registers for regalloc.
  auto RegSeq = MIRBuilder
                    .buildInstr(TargetOpcode::REG_SEQUENCE,
                                {&AArch64::QQRegClass}, {Src1Reg})
                    .addImm(AArch64::qsub0)
                    .addUse(Src2Reg)
                    .addImm(AArch64::qsub1);

  auto TBL2 =
      MIRBuilder.buildInstr(AArch64::TBLv16i8Two, {I.getOperand(0).getReg()},
                            {RegSeq, IndexLoad->getOperand(0).getReg()});
  constrainSelectedInstRegOperands(*RegSeq, TII, TRI, RBI);
  constrainSelectedInstRegOperands(*TBL2, TII, TRI, RBI);
  I.eraseFromParent();
  return true;
}

MachineInstr *AArch64InstructionSelector::emitLaneInsert(
    Optional<unsigned> DstReg, unsigned SrcReg, unsigned EltReg,
    unsigned LaneIdx, const RegisterBank &RB,
    MachineIRBuilder &MIRBuilder) const {
  MachineInstr *InsElt = nullptr;
  const TargetRegisterClass *DstRC = &AArch64::FPR128RegClass;
  MachineRegisterInfo &MRI = *MIRBuilder.getMRI();

  // Create a register to define with the insert if one wasn't passed in.
  if (!DstReg)
    DstReg = MRI.createVirtualRegister(DstRC);

  unsigned EltSize = MRI.getType(EltReg).getSizeInBits();
  unsigned Opc = getInsertVecEltOpInfo(RB, EltSize).first;

  if (RB.getID() == AArch64::FPRRegBankID) {
    auto InsSub = emitScalarToVector(EltSize, DstRC, EltReg, MIRBuilder);
    InsElt = MIRBuilder.buildInstr(Opc, {*DstReg}, {SrcReg})
                 .addImm(LaneIdx)
                 .addUse(InsSub->getOperand(0).getReg())
                 .addImm(0);
  } else {
    InsElt = MIRBuilder.buildInstr(Opc, {*DstReg}, {SrcReg})
                 .addImm(LaneIdx)
                 .addUse(EltReg);
  }

  constrainSelectedInstRegOperands(*InsElt, TII, TRI, RBI);
  return InsElt;
}

bool AArch64InstructionSelector::selectInsertElt(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_INSERT_VECTOR_ELT);

  // Get information on the destination.
  unsigned DstReg = I.getOperand(0).getReg();
  const LLT DstTy = MRI.getType(DstReg);
  if (DstTy.getSizeInBits() < 128) {
    // TODO: Handle unpacked vectors.
    LLVM_DEBUG(dbgs() << "Unpacked vectors not supported yet!");
    return false;
  }

  // Get information on the element we want to insert into the destination.
  unsigned EltReg = I.getOperand(2).getReg();
  const LLT EltTy = MRI.getType(EltReg);
  unsigned EltSize = EltTy.getSizeInBits();
  if (EltSize < 16 || EltSize > 64)
    return false; // Don't support all element types yet.

  // Find the definition of the index. Bail out if it's not defined by a
  // G_CONSTANT.
  unsigned IdxReg = I.getOperand(3).getReg();
  unsigned LaneIdx = 0;
  if (!getConstantValueForReg(IdxReg, MRI, LaneIdx))
    return false;

  // Perform the lane insert.
  unsigned SrcReg = I.getOperand(1).getReg();
  const RegisterBank &EltRB = *RBI.getRegBank(EltReg, MRI, TRI);
  MachineIRBuilder MIRBuilder(I);
  emitLaneInsert(DstReg, SrcReg, EltReg, LaneIdx, EltRB, MIRBuilder);
  I.eraseFromParent();
  return true;
}

bool AArch64InstructionSelector::selectBuildVector(
    MachineInstr &I, MachineRegisterInfo &MRI) const {
  assert(I.getOpcode() == TargetOpcode::G_BUILD_VECTOR);
  // Until we port more of the optimized selections, for now just use a vector
  // insert sequence.
  const LLT DstTy = MRI.getType(I.getOperand(0).getReg());
  const LLT EltTy = MRI.getType(I.getOperand(1).getReg());
  unsigned EltSize = EltTy.getSizeInBits();
  if (EltSize < 16 || EltSize > 64)
    return false; // Don't support all element types yet.
  const RegisterBank &RB = *RBI.getRegBank(I.getOperand(1).getReg(), MRI, TRI);
  MachineIRBuilder MIRBuilder(I);

  const TargetRegisterClass *DstRC = &AArch64::FPR128RegClass;
  MachineInstr *ScalarToVec =
      emitScalarToVector(DstTy.getElementType().getSizeInBits(), DstRC,
                         I.getOperand(1).getReg(), MIRBuilder);
  if (!ScalarToVec)
    return false;

  unsigned DstVec = ScalarToVec->getOperand(0).getReg();
  unsigned DstSize = DstTy.getSizeInBits();

  // Keep track of the last MI we inserted. Later on, we might be able to save
  // a copy using it.
  MachineInstr *PrevMI = nullptr;
  for (unsigned i = 2, e = DstSize / EltSize + 1; i < e; ++i) {
    // Note that if we don't do a subregister copy, we can end up making an
    // extra register.
    PrevMI = &*emitLaneInsert(None, DstVec, I.getOperand(i).getReg(), i - 1, RB,
                              MIRBuilder);
    DstVec = PrevMI->getOperand(0).getReg();
  }

  // If DstTy's size in bits is less than 128, then emit a subregister copy
  // from DstVec to the last register we've defined.
  if (DstSize < 128) {
    // Force this to be FPR using the destination vector.
    const TargetRegisterClass *RC =
        getMinClassForRegBank(*RBI.getRegBank(DstVec, MRI, TRI), DstSize);
    if (!RC)
      return false;
    if (RC != &AArch64::FPR32RegClass && RC != &AArch64::FPR64RegClass) {
      LLVM_DEBUG(dbgs() << "Unsupported register class!\n");
      return false;
    }

    unsigned SubReg = 0;
    if (!getSubRegForClass(RC, TRI, SubReg))
      return false;
    if (SubReg != AArch64::ssub && SubReg != AArch64::dsub) {
      LLVM_DEBUG(dbgs() << "Unsupported destination size! (" << DstSize
                        << "\n");
      return false;
    }

    unsigned Reg = MRI.createVirtualRegister(RC);
    unsigned DstReg = I.getOperand(0).getReg();

    MIRBuilder.buildInstr(TargetOpcode::COPY, {DstReg}, {})
        .addReg(DstVec, 0, SubReg);
    MachineOperand &RegOp = I.getOperand(1);
    RegOp.setReg(Reg);
    RBI.constrainGenericRegister(DstReg, *RC, MRI);
  } else {
    // We don't need a subregister copy. Save a copy by re-using the
    // destination register on the final insert.
    assert(PrevMI && "PrevMI was null?");
    PrevMI->getOperand(0).setReg(I.getOperand(0).getReg());
    constrainSelectedInstRegOperands(*PrevMI, TII, TRI, RBI);
  }

  I.eraseFromParent();
  return true;
}

/// SelectArithImmed - Select an immediate value that can be represented as
/// a 12-bit value shifted left by either 0 or 12.  If so, return true with
/// Val set to the 12-bit value and Shift set to the shifter operand.
InstructionSelector::ComplexRendererFns
AArch64InstructionSelector::selectArithImmed(MachineOperand &Root) const {
  MachineInstr &MI = *Root.getParent();
  MachineBasicBlock &MBB = *MI.getParent();
  MachineFunction &MF = *MBB.getParent();
  MachineRegisterInfo &MRI = MF.getRegInfo();

  // This function is called from the addsub_shifted_imm ComplexPattern,
  // which lists [imm] as the list of opcode it's interested in, however
  // we still need to check whether the operand is actually an immediate
  // here because the ComplexPattern opcode list is only used in
  // root-level opcode matching.
  uint64_t Immed;
  if (Root.isImm())
    Immed = Root.getImm();
  else if (Root.isCImm())
    Immed = Root.getCImm()->getZExtValue();
  else if (Root.isReg()) {
    MachineInstr *Def = MRI.getVRegDef(Root.getReg());
    if (Def->getOpcode() != TargetOpcode::G_CONSTANT)
      return None;
    MachineOperand &Op1 = Def->getOperand(1);
    if (!Op1.isCImm() || Op1.getCImm()->getBitWidth() > 64)
      return None;
    Immed = Op1.getCImm()->getZExtValue();
  } else
    return None;

  unsigned ShiftAmt;

  if (Immed >> 12 == 0) {
    ShiftAmt = 0;
  } else if ((Immed & 0xfff) == 0 && Immed >> 24 == 0) {
    ShiftAmt = 12;
    Immed = Immed >> 12;
  } else
    return None;

  unsigned ShVal = AArch64_AM::getShifterImm(AArch64_AM::LSL, ShiftAmt);
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Immed); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(ShVal); },
  }};
}

/// Select a "register plus unscaled signed 9-bit immediate" address.  This
/// should only match when there is an offset that is not valid for a scaled
/// immediate addressing mode.  The "Size" argument is the size in bytes of the
/// memory reference, which is needed here to know what is valid for a scaled
/// immediate.
InstructionSelector::ComplexRendererFns
AArch64InstructionSelector::selectAddrModeUnscaled(MachineOperand &Root,
                                                   unsigned Size) const {
  MachineRegisterInfo &MRI =
      Root.getParent()->getParent()->getParent()->getRegInfo();

  if (!Root.isReg())
    return None;

  if (!isBaseWithConstantOffset(Root, MRI))
    return None;

  MachineInstr *RootDef = MRI.getVRegDef(Root.getReg());
  if (!RootDef)
    return None;

  MachineOperand &OffImm = RootDef->getOperand(2);
  if (!OffImm.isReg())
    return None;
  MachineInstr *RHS = MRI.getVRegDef(OffImm.getReg());
  if (!RHS || RHS->getOpcode() != TargetOpcode::G_CONSTANT)
    return None;
  int64_t RHSC;
  MachineOperand &RHSOp1 = RHS->getOperand(1);
  if (!RHSOp1.isCImm() || RHSOp1.getCImm()->getBitWidth() > 64)
    return None;
  RHSC = RHSOp1.getCImm()->getSExtValue();

  // If the offset is valid as a scaled immediate, don't match here.
  if ((RHSC & (Size - 1)) == 0 && RHSC >= 0 && RHSC < (0x1000 << Log2_32(Size)))
    return None;
  if (RHSC >= -256 && RHSC < 256) {
    MachineOperand &Base = RootDef->getOperand(1);
    return {{
        [=](MachineInstrBuilder &MIB) { MIB.add(Base); },
        [=](MachineInstrBuilder &MIB) { MIB.addImm(RHSC); },
    }};
  }
  return None;
}

/// Select a "register plus scaled unsigned 12-bit immediate" address.  The
/// "Size" argument is the size in bytes of the memory reference, which
/// determines the scale.
InstructionSelector::ComplexRendererFns
AArch64InstructionSelector::selectAddrModeIndexed(MachineOperand &Root,
                                                  unsigned Size) const {
  MachineRegisterInfo &MRI =
      Root.getParent()->getParent()->getParent()->getRegInfo();

  if (!Root.isReg())
    return None;

  MachineInstr *RootDef = MRI.getVRegDef(Root.getReg());
  if (!RootDef)
    return None;

  if (RootDef->getOpcode() == TargetOpcode::G_FRAME_INDEX) {
    return {{
        [=](MachineInstrBuilder &MIB) { MIB.add(RootDef->getOperand(1)); },
        [=](MachineInstrBuilder &MIB) { MIB.addImm(0); },
    }};
  }

  if (isBaseWithConstantOffset(Root, MRI)) {
    MachineOperand &LHS = RootDef->getOperand(1);
    MachineOperand &RHS = RootDef->getOperand(2);
    MachineInstr *LHSDef = MRI.getVRegDef(LHS.getReg());
    MachineInstr *RHSDef = MRI.getVRegDef(RHS.getReg());
    if (LHSDef && RHSDef) {
      int64_t RHSC = (int64_t)RHSDef->getOperand(1).getCImm()->getZExtValue();
      unsigned Scale = Log2_32(Size);
      if ((RHSC & (Size - 1)) == 0 && RHSC >= 0 && RHSC < (0x1000 << Scale)) {
        if (LHSDef->getOpcode() == TargetOpcode::G_FRAME_INDEX)
          return {{
              [=](MachineInstrBuilder &MIB) { MIB.add(LHSDef->getOperand(1)); },
              [=](MachineInstrBuilder &MIB) { MIB.addImm(RHSC >> Scale); },
          }};

        return {{
            [=](MachineInstrBuilder &MIB) { MIB.add(LHS); },
            [=](MachineInstrBuilder &MIB) { MIB.addImm(RHSC >> Scale); },
        }};
      }
    }
  }

  // Before falling back to our general case, check if the unscaled
  // instructions can handle this. If so, that's preferable.
  if (selectAddrModeUnscaled(Root, Size).hasValue())
    return None;

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.add(Root); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); },
  }};
}

void AArch64InstructionSelector::renderTruncImm(MachineInstrBuilder &MIB,
                                                const MachineInstr &MI) const {
  const MachineRegisterInfo &MRI = MI.getParent()->getParent()->getRegInfo();
  assert(MI.getOpcode() == TargetOpcode::G_CONSTANT && "Expected G_CONSTANT");
  Optional<int64_t> CstVal = getConstantVRegVal(MI.getOperand(0).getReg(), MRI);
  assert(CstVal && "Expected constant value");
  MIB.addImm(CstVal.getValue());
}

namespace llvm {
InstructionSelector *
createAArch64InstructionSelector(const AArch64TargetMachine &TM,
                                 AArch64Subtarget &Subtarget,
                                 AArch64RegisterBankInfo &RBI) {
  return new AArch64InstructionSelector(TM, Subtarget, RBI);
}
}