AArch64InstrInfo.cpp

Go to the documentation of this file.

//===- AArch64InstrInfo.cpp - AArch64 Instruction Information -------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file contains the AArch64 implementation of the TargetInstrInfo class.
//
//===----------------------------------------------------------------------===//
 
#include "AArch64ExpandImm.h"
#include "AArch64InstrInfo.h"
#include "AArch64FrameLowering.h"
#include "AArch64MachineFunctionInfo.h"
#include "AArch64Subtarget.h"
#include "MCTargetDesc/AArch64AddressingModes.h"
#include "Utils/AArch64BaseInfo.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineCombinerPattern.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineMemOperand.h"
#include "llvm/CodeGen/MachineModuleInfo.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/RegisterScavenging.h"
#include "llvm/CodeGen/StackMaps.h"
#include "llvm/CodeGen/TargetRegisterInfo.h"
#include "llvm/CodeGen/TargetSubtargetInfo.h"
#include "llvm/IR/DebugInfoMetadata.h"
#include "llvm/IR/DebugLoc.h"
#include "llvm/IR/GlobalValue.h"
#include "llvm/MC/MCAsmInfo.h"
#include "llvm/MC/MCInst.h"
#include "llvm/MC/MCInstBuilder.h"
#include "llvm/MC/MCInstrDesc.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/CodeGen.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/LEB128.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetOptions.h"
#include <cassert>
#include <cstdint>
#include <iterator>
#include <utility>
 
using namespace llvm;
 
#define GET_INSTRINFO_CTOR_DTOR
#include "AArch64GenInstrInfo.inc"
 
static cl::opt<unsigned> TBZDisplacementBits(
    "aarch64-tbz-offset-bits", cl::Hidden, cl::init(14),
    cl::desc("Restrict range of TB[N]Z instructions (DEBUG)"));
 
static cl::opt<unsigned> CBZDisplacementBits(
    "aarch64-cbz-offset-bits", cl::Hidden, cl::init(19),
    cl::desc("Restrict range of CB[N]Z instructions (DEBUG)"));
 
static cl::opt<unsigned>
    BCCDisplacementBits("aarch64-bcc-offset-bits", cl::Hidden, cl::init(19),
                        cl::desc("Restrict range of Bcc instructions (DEBUG)"));
 
static cl::opt<unsigned>
    BDisplacementBits("aarch64-b-offset-bits", cl::Hidden, cl::init(26),
                      cl::desc("Restrict range of B instructions (DEBUG)"));
 
AArch64InstrInfo::AArch64InstrInfo(const AArch64Subtarget &STI)
    : AArch64GenInstrInfo(AArch64::ADJCALLSTACKDOWN, AArch64::ADJCALLSTACKUP,
                          AArch64::CATCHRET),
      RI(STI.getTargetTriple()), Subtarget(STI) {}
 
/// GetInstSize - Return the number of bytes of code the specified
/// instruction may be.  This returns the maximum number of bytes.
unsigned AArch64InstrInfo::getInstSizeInBytes(const MachineInstr &MI) const {
  const MachineBasicBlock &MBB = *MI.getParent();
  const MachineFunction *MF = MBB.getParent();
  const Function &F = MF->getFunction();
  const MCAsmInfo *MAI = MF->getTarget().getMCAsmInfo();
 
  {
    auto Op = MI.getOpcode();
    if (Op == AArch64::INLINEASM || Op == AArch64::INLINEASM_BR)
      return getInlineAsmLength(MI.getOperand(0).getSymbolName(), *MAI);
  }
 
  // Meta-instructions emit no code.
  if (MI.isMetaInstruction())
    return 0;
 
  // FIXME: We currently only handle pseudoinstructions that don't get expanded
  //        before the assembly printer.
  unsigned NumBytes = 0;
  const MCInstrDesc &Desc = MI.getDesc();
 
  // Size should be preferably set in
  // llvm/lib/Target/AArch64/AArch64InstrInfo.td (default case).
  // Specific cases handle instructions of variable sizes
  switch (Desc.getOpcode()) {
  default:
    if (Desc.getSize())
      return Desc.getSize();
 
    // Anything not explicitly designated otherwise (i.e. pseudo-instructions
    // with fixed constant size but not specified in .td file) is a normal
    // 4-byte insn.
    NumBytes = 4;
    break;
  case TargetOpcode::STACKMAP:
    // The upper bound for a stackmap intrinsic is the full length of its shadow
    NumBytes = StackMapOpers(&MI).getNumPatchBytes();
    assert(NumBytes % 4 == 0 && "Invalid number of NOP bytes requested!");
    break;
  case TargetOpcode::PATCHPOINT:
    // The size of the patchpoint intrinsic is the number of bytes requested
    NumBytes = PatchPointOpers(&MI).getNumPatchBytes();
    assert(NumBytes % 4 == 0 && "Invalid number of NOP bytes requested!");
    break;
  case TargetOpcode::STATEPOINT:
    NumBytes = StatepointOpers(&MI).getNumPatchBytes();
    assert(NumBytes % 4 == 0 && "Invalid number of NOP bytes requested!");
    // No patch bytes means a normal call inst is emitted
    if (NumBytes == 0)
      NumBytes = 4;
    break;
  case TargetOpcode::PATCHABLE_FUNCTION_ENTER:
    // If `patchable-function-entry` is set, PATCHABLE_FUNCTION_ENTER
    // instructions are expanded to the specified number of NOPs. Otherwise,
    // they are expanded to 36-byte XRay sleds.
    NumBytes =
        F.getFnAttributeAsParsedInteger("patchable-function-entry", 9) * 4;
    break;
  case TargetOpcode::PATCHABLE_FUNCTION_EXIT:
  case TargetOpcode::PATCHABLE_TYPED_EVENT_CALL:
    // An XRay sled can be 4 bytes of alignment plus a 32-byte block.
    NumBytes = 36;
    break;
  case TargetOpcode::PATCHABLE_EVENT_CALL:
    // EVENT_CALL XRay sleds are exactly 6 instructions long (no alignment).
    NumBytes = 24;
    break;
 
  case AArch64::SPACE:
    NumBytes = MI.getOperand(1).getImm();
    break;
  case TargetOpcode::BUNDLE:
    NumBytes = getInstBundleLength(MI);
    break;
  }
 
  return NumBytes;
}
 
unsigned AArch64InstrInfo::getInstBundleLength(const MachineInstr &MI) const {
  unsigned Size = 0;
  MachineBasicBlock::const_instr_iterator I = MI.getIterator();
  MachineBasicBlock::const_instr_iterator E = MI.getParent()->instr_end();
  while (++I != E && I->isInsideBundle()) {
    assert(!I->isBundle() && "No nested bundle!");
    Size += getInstSizeInBytes(*I);
  }
  return Size;
}
 
static void parseCondBranch(MachineInstr *LastInst, MachineBasicBlock *&Target,
                            SmallVectorImpl<MachineOperand> &Cond) {
  // Block ends with fall-through condbranch.
  switch (LastInst->getOpcode()) {
  default:
    llvm_unreachable("Unknown branch instruction?");
  case AArch64::Bcc:
    Target = LastInst->getOperand(1).getMBB();
    Cond.push_back(LastInst->getOperand(0));
    break;
  case AArch64::CBZW:
  case AArch64::CBZX:
  case AArch64::CBNZW:
  case AArch64::CBNZX:
    Target = LastInst->getOperand(1).getMBB();
    Cond.push_back(MachineOperand::CreateImm(-1));
    Cond.push_back(MachineOperand::CreateImm(LastInst->getOpcode()));
    Cond.push_back(LastInst->getOperand(0));
    break;
  case AArch64::TBZW:
  case AArch64::TBZX:
  case AArch64::TBNZW:
  case AArch64::TBNZX:
    Target = LastInst->getOperand(2).getMBB();
    Cond.push_back(MachineOperand::CreateImm(-1));
    Cond.push_back(MachineOperand::CreateImm(LastInst->getOpcode()));
    Cond.push_back(LastInst->getOperand(0));
    Cond.push_back(LastInst->getOperand(1));
  }
}
 
static unsigned getBranchDisplacementBits(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("unexpected opcode!");
  case AArch64::B:
    return BDisplacementBits;
  case AArch64::TBNZW:
  case AArch64::TBZW:
  case AArch64::TBNZX:
  case AArch64::TBZX:
    return TBZDisplacementBits;
  case AArch64::CBNZW:
  case AArch64::CBZW:
  case AArch64::CBNZX:
  case AArch64::CBZX:
    return CBZDisplacementBits;
  case AArch64::Bcc:
    return BCCDisplacementBits;
  }
}
 
bool AArch64InstrInfo::isBranchOffsetInRange(unsigned BranchOp,
                                             int64_t BrOffset) const {
  unsigned Bits = getBranchDisplacementBits(BranchOp);
  assert(Bits >= 3 && "max branch displacement must be enough to jump"
                      "over conditional branch expansion");
  return isIntN(Bits, BrOffset / 4);
}
 
MachineBasicBlock *
AArch64InstrInfo::getBranchDestBlock(const MachineInstr &MI) const {
  switch (MI.getOpcode()) {
  default:
    llvm_unreachable("unexpected opcode!");
  case AArch64::B:
    return MI.getOperand(0).getMBB();
  case AArch64::TBZW:
  case AArch64::TBNZW:
  case AArch64::TBZX:
  case AArch64::TBNZX:
    return MI.getOperand(2).getMBB();
  case AArch64::CBZW:
  case AArch64::CBNZW:
  case AArch64::CBZX:
  case AArch64::CBNZX:
  case AArch64::Bcc:
    return MI.getOperand(1).getMBB();
  }
}
 
void AArch64InstrInfo::insertIndirectBranch(MachineBasicBlock &MBB,
                                            MachineBasicBlock &NewDestBB,
                                            MachineBasicBlock &RestoreBB,
                                            const DebugLoc &DL,
                                            int64_t BrOffset,
                                            RegScavenger *RS) const {
  assert(RS && "RegScavenger required for long branching");
  assert(MBB.empty() &&
         "new block should be inserted for expanding unconditional branch");
  assert(MBB.pred_size() == 1);
  assert(RestoreBB.empty() &&
         "restore block should be inserted for restoring clobbered registers");
 
  auto buildIndirectBranch = [&](Register Reg, MachineBasicBlock &DestBB) {
    // Offsets outside of the signed 33-bit range are not supported for ADRP +
    // ADD.
    if (!isInt<33>(BrOffset))
      report_fatal_error(
          "Branch offsets outside of the signed 33-bit range not supported");
 
    BuildMI(MBB, MBB.end(), DL, get(AArch64::ADRP), Reg)
        .addSym(DestBB.getSymbol(), AArch64II::MO_PAGE);
    BuildMI(MBB, MBB.end(), DL, get(AArch64::ADDXri), Reg)
        .addReg(Reg)
        .addSym(DestBB.getSymbol(), AArch64II::MO_PAGEOFF | AArch64II::MO_NC)
        .addImm(0);
    BuildMI(MBB, MBB.end(), DL, get(AArch64::BR)).addReg(Reg);
  };
 
  RS->enterBasicBlockEnd(MBB);
  // If X16 is unused, we can rely on the linker to insert a range extension
  // thunk if NewDestBB is out of range of a single B instruction.
  constexpr Register Reg = AArch64::X16;
  if (!RS->isRegUsed(Reg)) {
    insertUnconditionalBranch(MBB, &NewDestBB, DL);
    RS->setRegUsed(Reg);
    return;
  }
 
  // If there's a free register and it's worth inflating the code size,
  // manually insert the indirect branch.
  Register Scavenged = RS->FindUnusedReg(&AArch64::GPR64RegClass);
  if (Scavenged != AArch64::NoRegister &&
      MBB.getSectionID() == MBBSectionID::ColdSectionID) {
    buildIndirectBranch(Scavenged, NewDestBB);
    RS->setRegUsed(Scavenged);
    return;
  }
 
  // Note: Spilling X16 briefly moves the stack pointer, making it incompatible
  // with red zones.
  AArch64FunctionInfo *AFI = MBB.getParent()->getInfo<AArch64FunctionInfo>();
  if (!AFI || AFI->hasRedZone().value_or(true))
    report_fatal_error(
        "Unable to insert indirect branch inside function that has red zone");
 
  // Otherwise, spill X16 and defer range extension to the linker.
  BuildMI(MBB, MBB.end(), DL, get(AArch64::STRXpre))
      .addReg(AArch64::SP, RegState::Define)
      .addReg(Reg)
      .addReg(AArch64::SP)
      .addImm(-16);
 
  BuildMI(MBB, MBB.end(), DL, get(AArch64::B)).addMBB(&RestoreBB);
 
  BuildMI(RestoreBB, RestoreBB.end(), DL, get(AArch64::LDRXpost))
      .addReg(AArch64::SP, RegState::Define)
      .addReg(Reg, RegState::Define)
      .addReg(AArch64::SP)
      .addImm(16);
}
 
// Branch analysis.
bool AArch64InstrInfo::analyzeBranch(MachineBasicBlock &MBB,
                                     MachineBasicBlock *&TBB,
                                     MachineBasicBlock *&FBB,
                                     SmallVectorImpl<MachineOperand> &Cond,
                                     bool AllowModify) const {
  // If the block has no terminators, it just falls into the block after it.
  MachineBasicBlock::iterator I = MBB.getLastNonDebugInstr();
  if (I == MBB.end())
    return false;
 
  // Skip over SpeculationBarrierEndBB terminators
  if (I->getOpcode() == AArch64::SpeculationBarrierISBDSBEndBB ||
      I->getOpcode() == AArch64::SpeculationBarrierSBEndBB) {
    --I;
  }
 
  if (!isUnpredicatedTerminator(*I))
    return false;
 
  // Get the last instruction in the block.
  MachineInstr *LastInst = &*I;
 
  // If there is only one terminator instruction, process it.
  unsigned LastOpc = LastInst->getOpcode();
  if (I == MBB.begin() || !isUnpredicatedTerminator(*--I)) {
    if (isUncondBranchOpcode(LastOpc)) {
      TBB = LastInst->getOperand(0).getMBB();
      return false;
    }
    if (isCondBranchOpcode(LastOpc)) {
      // Block ends with fall-through condbranch.
      parseCondBranch(LastInst, TBB, Cond);
      return false;
    }
    return true; // Can't handle indirect branch.
  }
 
  // Get the instruction before it if it is a terminator.
  MachineInstr *SecondLastInst = &*I;
  unsigned SecondLastOpc = SecondLastInst->getOpcode();
 
  // If AllowModify is true and the block ends with two or more unconditional
  // branches, delete all but the first unconditional branch.
  if (AllowModify && isUncondBranchOpcode(LastOpc)) {
    while (isUncondBranchOpcode(SecondLastOpc)) {
      LastInst->eraseFromParent();
      LastInst = SecondLastInst;
      LastOpc = LastInst->getOpcode();
      if (I == MBB.begin() || !isUnpredicatedTerminator(*--I)) {
        // Return now the only terminator is an unconditional branch.
        TBB = LastInst->getOperand(0).getMBB();
        return false;
      }
      SecondLastInst = &*I;
      SecondLastOpc = SecondLastInst->getOpcode();
    }
  }
 
  // If we're allowed to modify and the block ends in a unconditional branch
  // which could simply fallthrough, remove the branch.  (Note: This case only
  // matters when we can't understand the whole sequence, otherwise it's also
  // handled by BranchFolding.cpp.)
  if (AllowModify && isUncondBranchOpcode(LastOpc) &&
      MBB.isLayoutSuccessor(getBranchDestBlock(*LastInst))) {
    LastInst->eraseFromParent();
    LastInst = SecondLastInst;
    LastOpc = LastInst->getOpcode();
    if (I == MBB.begin() || !isUnpredicatedTerminator(*--I)) {
      assert(!isUncondBranchOpcode(LastOpc) &&
             "unreachable unconditional branches removed above");
 
      if (isCondBranchOpcode(LastOpc)) {
        // Block ends with fall-through condbranch.
        parseCondBranch(LastInst, TBB, Cond);
        return false;
      }
      return true; // Can't handle indirect branch.
    }
    SecondLastInst = &*I;
    SecondLastOpc = SecondLastInst->getOpcode();
  }
 
  // If there are three terminators, we don't know what sort of block this is.
  if (SecondLastInst && I != MBB.begin() && isUnpredicatedTerminator(*--I))
    return true;
 
  // If the block ends with a B and a Bcc, handle it.
  if (isCondBranchOpcode(SecondLastOpc) && isUncondBranchOpcode(LastOpc)) {
    parseCondBranch(SecondLastInst, TBB, Cond);
    FBB = LastInst->getOperand(0).getMBB();
    return false;
  }
 
  // If the block ends with two unconditional branches, handle it.  The second
  // one is not executed, so remove it.
  if (isUncondBranchOpcode(SecondLastOpc) && isUncondBranchOpcode(LastOpc)) {
    TBB = SecondLastInst->getOperand(0).getMBB();
    I = LastInst;
    if (AllowModify)
      I->eraseFromParent();
    return false;
  }
 
  // ...likewise if it ends with an indirect branch followed by an unconditional
  // branch.
  if (isIndirectBranchOpcode(SecondLastOpc) && isUncondBranchOpcode(LastOpc)) {
    I = LastInst;
    if (AllowModify)
      I->eraseFromParent();
    return true;
  }
 
  // Otherwise, can't handle this.
  return true;
}
 
bool AArch64InstrInfo::analyzeBranchPredicate(MachineBasicBlock &MBB,
                                              MachineBranchPredicate &MBP,
                                              bool AllowModify) const {
  // For the moment, handle only a block which ends with a cb(n)zx followed by
  // a fallthrough.  Why this?  Because it is a common form.
  // TODO: Should we handle b.cc?
 
  MachineBasicBlock::iterator I = MBB.getLastNonDebugInstr();
  if (I == MBB.end())
    return true;
 
  // Skip over SpeculationBarrierEndBB terminators
  if (I->getOpcode() == AArch64::SpeculationBarrierISBDSBEndBB ||
      I->getOpcode() == AArch64::SpeculationBarrierSBEndBB) {
    --I;
  }
 
  if (!isUnpredicatedTerminator(*I))
    return true;
 
  // Get the last instruction in the block.
  MachineInstr *LastInst = &*I;
  unsigned LastOpc = LastInst->getOpcode();
  if (!isCondBranchOpcode(LastOpc))
    return true;
 
  switch (LastOpc) {
  default:
    return true;
  case AArch64::CBZW:
  case AArch64::CBZX:
  case AArch64::CBNZW:
  case AArch64::CBNZX:
    break;
  };
 
  MBP.TrueDest = LastInst->getOperand(1).getMBB();
  assert(MBP.TrueDest && "expected!");
  MBP.FalseDest = MBB.getNextNode();
 
  MBP.ConditionDef = nullptr;
  MBP.SingleUseCondition = false;
 
  MBP.LHS = LastInst->getOperand(0);
  MBP.RHS = MachineOperand::CreateImm(0);
  MBP.Predicate = LastOpc == AArch64::CBNZX ? MachineBranchPredicate::PRED_NE
                                            : MachineBranchPredicate::PRED_EQ;
  return false;
}
 
bool AArch64InstrInfo::reverseBranchCondition(
    SmallVectorImpl<MachineOperand> &Cond) const {
  if (Cond[0].getImm() != -1) {
    // Regular Bcc
    AArch64CC::CondCode CC = (AArch64CC::CondCode)(int)Cond[0].getImm();
    Cond[0].setImm(AArch64CC::getInvertedCondCode(CC));
  } else {
    // Folded compare-and-branch
    switch (Cond[1].getImm()) {
    default:
      llvm_unreachable("Unknown conditional branch!");
    case AArch64::CBZW:
      Cond[1].setImm(AArch64::CBNZW);
      break;
    case AArch64::CBNZW:
      Cond[1].setImm(AArch64::CBZW);
      break;
    case AArch64::CBZX:
      Cond[1].setImm(AArch64::CBNZX);
      break;
    case AArch64::CBNZX:
      Cond[1].setImm(AArch64::CBZX);
      break;
    case AArch64::TBZW:
      Cond[1].setImm(AArch64::TBNZW);
      break;
    case AArch64::TBNZW:
      Cond[1].setImm(AArch64::TBZW);
      break;
    case AArch64::TBZX:
      Cond[1].setImm(AArch64::TBNZX);
      break;
    case AArch64::TBNZX:
      Cond[1].setImm(AArch64::TBZX);
      break;
    }
  }
 
  return false;
}
 
unsigned AArch64InstrInfo::removeBranch(MachineBasicBlock &MBB,
                                        int *BytesRemoved) const {
  MachineBasicBlock::iterator I = MBB.getLastNonDebugInstr();
  if (I == MBB.end())
    return 0;
 
  if (!isUncondBranchOpcode(I->getOpcode()) &&
      !isCondBranchOpcode(I->getOpcode()))
    return 0;
 
  // Remove the branch.
  I->eraseFromParent();
 
  I = MBB.end();
 
  if (I == MBB.begin()) {
    if (BytesRemoved)
      *BytesRemoved = 4;
    return 1;
  }
  --I;
  if (!isCondBranchOpcode(I->getOpcode())) {
    if (BytesRemoved)
      *BytesRemoved = 4;
    return 1;
  }
 
  // Remove the branch.
  I->eraseFromParent();
  if (BytesRemoved)
    *BytesRemoved = 8;
 
  return 2;
}
 
void AArch64InstrInfo::instantiateCondBranch(
    MachineBasicBlock &MBB, const DebugLoc &DL, MachineBasicBlock *TBB,
    ArrayRef<MachineOperand> Cond) const {
  if (Cond[0].getImm() != -1) {
    // Regular Bcc
    BuildMI(&MBB, DL, get(AArch64::Bcc)).addImm(Cond[0].getImm()).addMBB(TBB);
  } else {
    // Folded compare-and-branch
    // Note that we use addOperand instead of addReg to keep the flags.
    const MachineInstrBuilder MIB =
        BuildMI(&MBB, DL, get(Cond[1].getImm())).add(Cond[2]);
    if (Cond.size() > 3)
      MIB.addImm(Cond[3].getImm());
    MIB.addMBB(TBB);
  }
}
 
unsigned AArch64InstrInfo::insertBranch(
    MachineBasicBlock &MBB, MachineBasicBlock *TBB, MachineBasicBlock *FBB,
    ArrayRef<MachineOperand> Cond, const DebugLoc &DL, int *BytesAdded) const {
  // Shouldn't be a fall through.
  assert(TBB && "insertBranch must not be told to insert a fallthrough");
 
  if (!FBB) {
    if (Cond.empty()) // Unconditional branch?
      BuildMI(&MBB, DL, get(AArch64::B)).addMBB(TBB);
    else
      instantiateCondBranch(MBB, DL, TBB, Cond);
 
    if (BytesAdded)
      *BytesAdded = 4;
 
    return 1;
  }
 
  // Two-way conditional branch.
  instantiateCondBranch(MBB, DL, TBB, Cond);
  BuildMI(&MBB, DL, get(AArch64::B)).addMBB(FBB);
 
  if (BytesAdded)
    *BytesAdded = 8;
 
  return 2;
}
 
// Find the original register that VReg is copied from.
static unsigned removeCopies(const MachineRegisterInfo &MRI, unsigned VReg) {
  while (Register::isVirtualRegister(VReg)) {
    const MachineInstr *DefMI = MRI.getVRegDef(VReg);
    if (!DefMI->isFullCopy())
      return VReg;
    VReg = DefMI->getOperand(1).getReg();
  }
  return VReg;
}
 
// Determine if VReg is defined by an instruction that can be folded into a
// csel instruction. If so, return the folded opcode, and the replacement
// register.
static unsigned canFoldIntoCSel(const MachineRegisterInfo &MRI, unsigned VReg,
                                unsigned *NewVReg = nullptr) {
  VReg = removeCopies(MRI, VReg);
  if (!Register::isVirtualRegister(VReg))
    return 0;
 
  bool Is64Bit = AArch64::GPR64allRegClass.hasSubClassEq(MRI.getRegClass(VReg));
  const MachineInstr *DefMI = MRI.getVRegDef(VReg);
  unsigned Opc = 0;
  unsigned SrcOpNum = 0;
  switch (DefMI->getOpcode()) {
  case AArch64::ADDSXri:
  case AArch64::ADDSWri:
    // if NZCV is used, do not fold.
    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
      return 0;
    // fall-through to ADDXri and ADDWri.
    [[fallthrough]];
  case AArch64::ADDXri:
  case AArch64::ADDWri:
    // add x, 1 -> csinc.
    if (!DefMI->getOperand(2).isImm() || DefMI->getOperand(2).getImm() != 1 ||
        DefMI->getOperand(3).getImm() != 0)
      return 0;
    SrcOpNum = 1;
    Opc = Is64Bit ? AArch64::CSINCXr : AArch64::CSINCWr;
    break;
 
  case AArch64::ORNXrr:
  case AArch64::ORNWrr: {
    // not x -> csinv, represented as orn dst, xzr, src.
    unsigned ZReg = removeCopies(MRI, DefMI->getOperand(1).getReg());
    if (ZReg != AArch64::XZR && ZReg != AArch64::WZR)
      return 0;
    SrcOpNum = 2;
    Opc = Is64Bit ? AArch64::CSINVXr : AArch64::CSINVWr;
    break;
  }
 
  case AArch64::SUBSXrr:
  case AArch64::SUBSWrr:
    // if NZCV is used, do not fold.
    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
      return 0;
    // fall-through to SUBXrr and SUBWrr.
    [[fallthrough]];
  case AArch64::SUBXrr:
  case AArch64::SUBWrr: {
    // neg x -> csneg, represented as sub dst, xzr, src.
    unsigned ZReg = removeCopies(MRI, DefMI->getOperand(1).getReg());
    if (ZReg != AArch64::XZR && ZReg != AArch64::WZR)
      return 0;
    SrcOpNum = 2;
    Opc = Is64Bit ? AArch64::CSNEGXr : AArch64::CSNEGWr;
    break;
  }
  default:
    return 0;
  }
  assert(Opc && SrcOpNum && "Missing parameters");
 
  if (NewVReg)
    *NewVReg = DefMI->getOperand(SrcOpNum).getReg();
  return Opc;
}
 
bool AArch64InstrInfo::canInsertSelect(const MachineBasicBlock &MBB,
                                       ArrayRef<MachineOperand> Cond,
                                       Register DstReg, Register TrueReg,
                                       Register FalseReg, int &CondCycles,
                                       int &TrueCycles,
                                       int &FalseCycles) const {
  // Check register classes.
  const MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
  const TargetRegisterClass *RC =
      RI.getCommonSubClass(MRI.getRegClass(TrueReg), MRI.getRegClass(FalseReg));
  if (!RC)
    return false;
 
  // Also need to check the dest regclass, in case we're trying to optimize
  // something like:
  // %1(gpr) = PHI %2(fpr), bb1, %(fpr), bb2
  if (!RI.getCommonSubClass(RC, MRI.getRegClass(DstReg)))
    return false;
 
  // Expanding cbz/tbz requires an extra cycle of latency on the condition.
  unsigned ExtraCondLat = Cond.size() != 1;
 
  // GPRs are handled by csel.
  // FIXME: Fold in x+1, -x, and ~x when applicable.
  if (AArch64::GPR64allRegClass.hasSubClassEq(RC) ||
      AArch64::GPR32allRegClass.hasSubClassEq(RC)) {
    // Single-cycle csel, csinc, csinv, and csneg.
    CondCycles = 1 + ExtraCondLat;
    TrueCycles = FalseCycles = 1;
    if (canFoldIntoCSel(MRI, TrueReg))
      TrueCycles = 0;
    else if (canFoldIntoCSel(MRI, FalseReg))
      FalseCycles = 0;
    return true;
  }
 
  // Scalar floating point is handled by fcsel.
  // FIXME: Form fabs, fmin, and fmax when applicable.
  if (AArch64::FPR64RegClass.hasSubClassEq(RC) ||
      AArch64::FPR32RegClass.hasSubClassEq(RC)) {
    CondCycles = 5 + ExtraCondLat;
    TrueCycles = FalseCycles = 2;
    return true;
  }
 
  // Can't do vectors.
  return false;
}
 
void AArch64InstrInfo::insertSelect(MachineBasicBlock &MBB,
                                    MachineBasicBlock::iterator I,
                                    const DebugLoc &DL, Register DstReg,
                                    ArrayRef<MachineOperand> Cond,
                                    Register TrueReg, Register FalseReg) const {
  MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
 
  // Parse the condition code, see parseCondBranch() above.
  AArch64CC::CondCode CC;
  switch (Cond.size()) {
  default:
    llvm_unreachable("Unknown condition opcode in Cond");
  case 1: // b.cc
    CC = AArch64CC::CondCode(Cond[0].getImm());
    break;
  case 3: { // cbz/cbnz
    // We must insert a compare against 0.
    bool Is64Bit;
    switch (Cond[1].getImm()) {
    default:
      llvm_unreachable("Unknown branch opcode in Cond");
    case AArch64::CBZW:
      Is64Bit = false;
      CC = AArch64CC::EQ;
      break;
    case AArch64::CBZX:
      Is64Bit = true;
      CC = AArch64CC::EQ;
      break;
    case AArch64::CBNZW:
      Is64Bit = false;
      CC = AArch64CC::NE;
      break;
    case AArch64::CBNZX:
      Is64Bit = true;
      CC = AArch64CC::NE;
      break;
    }
    Register SrcReg = Cond[2].getReg();
    if (Is64Bit) {
      // cmp reg, #0 is actually subs xzr, reg, #0.
      MRI.constrainRegClass(SrcReg, &AArch64::GPR64spRegClass);
      BuildMI(MBB, I, DL, get(AArch64::SUBSXri), AArch64::XZR)
          .addReg(SrcReg)
          .addImm(0)
          .addImm(0);
    } else {
      MRI.constrainRegClass(SrcReg, &AArch64::GPR32spRegClass);
      BuildMI(MBB, I, DL, get(AArch64::SUBSWri), AArch64::WZR)
          .addReg(SrcReg)
          .addImm(0)
          .addImm(0);
    }
    break;
  }
  case 4: { // tbz/tbnz
    // We must insert a tst instruction.
    switch (Cond[1].getImm()) {
    default:
      llvm_unreachable("Unknown branch opcode in Cond");
    case AArch64::TBZW:
    case AArch64::TBZX:
      CC = AArch64CC::EQ;
      break;
    case AArch64::TBNZW:
    case AArch64::TBNZX:
      CC = AArch64CC::NE;
      break;
    }
    // cmp reg, #foo is actually ands xzr, reg, #1<<foo.
    if (Cond[1].getImm() == AArch64::TBZW || Cond[1].getImm() == AArch64::TBNZW)
      BuildMI(MBB, I, DL, get(AArch64::ANDSWri), AArch64::WZR)
          .addReg(Cond[2].getReg())
          .addImm(
              AArch64_AM::encodeLogicalImmediate(1ull << Cond[3].getImm(), 32));
    else
      BuildMI(MBB, I, DL, get(AArch64::ANDSXri), AArch64::XZR)
          .addReg(Cond[2].getReg())
          .addImm(
              AArch64_AM::encodeLogicalImmediate(1ull << Cond[3].getImm(), 64));
    break;
  }
  }
 
  unsigned Opc = 0;
  const TargetRegisterClass *RC = nullptr;
  bool TryFold = false;
  if (MRI.constrainRegClass(DstReg, &AArch64::GPR64RegClass)) {
    RC = &AArch64::GPR64RegClass;
    Opc = AArch64::CSELXr;
    TryFold = true;
  } else if (MRI.constrainRegClass(DstReg, &AArch64::GPR32RegClass)) {
    RC = &AArch64::GPR32RegClass;
    Opc = AArch64::CSELWr;
    TryFold = true;
  } else if (MRI.constrainRegClass(DstReg, &AArch64::FPR64RegClass)) {
    RC = &AArch64::FPR64RegClass;
    Opc = AArch64::FCSELDrrr;
  } else if (MRI.constrainRegClass(DstReg, &AArch64::FPR32RegClass)) {
    RC = &AArch64::FPR32RegClass;
    Opc = AArch64::FCSELSrrr;
  }
  assert(RC && "Unsupported regclass");
 
  // Try folding simple instructions into the csel.
  if (TryFold) {
    unsigned NewVReg = 0;
    unsigned FoldedOpc = canFoldIntoCSel(MRI, TrueReg, &NewVReg);
    if (FoldedOpc) {
      // The folded opcodes csinc, csinc and csneg apply the operation to
      // FalseReg, so we need to invert the condition.
      CC = AArch64CC::getInvertedCondCode(CC);
      TrueReg = FalseReg;
    } else
      FoldedOpc = canFoldIntoCSel(MRI, FalseReg, &NewVReg);
 
    // Fold the operation. Leave any dead instructions for DCE to clean up.
    if (FoldedOpc) {
      FalseReg = NewVReg;
      Opc = FoldedOpc;
      // The extends the live range of NewVReg.
      MRI.clearKillFlags(NewVReg);
    }
  }
 
  // Pull all virtual register into the appropriate class.
  MRI.constrainRegClass(TrueReg, RC);
  MRI.constrainRegClass(FalseReg, RC);
 
  // Insert the csel.
  BuildMI(MBB, I, DL, get(Opc), DstReg)
      .addReg(TrueReg)
      .addReg(FalseReg)
      .addImm(CC);
}
 
// Return true if Imm can be loaded into a register by a "cheap" sequence of
// instructions. For now, "cheap" means at most two instructions.
static bool isCheapImmediate(const MachineInstr &MI, unsigned BitSize) {
  if (BitSize == 32)
    return true;
 
  assert(BitSize == 64 && "Only bit sizes of 32 or 64 allowed");
  uint64_t Imm = static_cast<uint64_t>(MI.getOperand(1).getImm());
  SmallVector<AArch64_IMM::ImmInsnModel, 4> Is;
  AArch64_IMM::expandMOVImm(Imm, BitSize, Is);
 
  return Is.size() <= 2;
}
 
// FIXME: this implementation should be micro-architecture dependent, so a
// micro-architecture target hook should be introduced here in future.
bool AArch64InstrInfo::isAsCheapAsAMove(const MachineInstr &MI) const {
  if (Subtarget.hasExynosCheapAsMoveHandling()) {
    if (isExynosCheapAsMove(MI))
      return true;
    return MI.isAsCheapAsAMove();
  }
 
  switch (MI.getOpcode()) {
  default:
    return MI.isAsCheapAsAMove();
 
  case AArch64::ADDWrs:
  case AArch64::ADDXrs:
  case AArch64::SUBWrs:
  case AArch64::SUBXrs:
    return Subtarget.hasALULSLFast() && MI.getOperand(3).getImm() <= 4;
 
  // If MOVi32imm or MOVi64imm can be expanded into ORRWri or
  // ORRXri, it is as cheap as MOV.
  // Likewise if it can be expanded to MOVZ/MOVN/MOVK.
  case AArch64::MOVi32imm:
    return isCheapImmediate(MI, 32);
  case AArch64::MOVi64imm:
    return isCheapImmediate(MI, 64);
  }
}
 
bool AArch64InstrInfo::isFalkorShiftExtFast(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    return false;
 
  case AArch64::ADDWrs:
  case AArch64::ADDXrs:
  case AArch64::ADDSWrs:
  case AArch64::ADDSXrs: {
    unsigned Imm = MI.getOperand(3).getImm();
    unsigned ShiftVal = AArch64_AM::getShiftValue(Imm);
    if (ShiftVal == 0)
      return true;
    return AArch64_AM::getShiftType(Imm) == AArch64_AM::LSL && ShiftVal <= 5;
  }
 
  case AArch64::ADDWrx:
  case AArch64::ADDXrx:
  case AArch64::ADDXrx64:
  case AArch64::ADDSWrx:
  case AArch64::ADDSXrx:
  case AArch64::ADDSXrx64: {
    unsigned Imm = MI.getOperand(3).getImm();
    switch (AArch64_AM::getArithExtendType(Imm)) {
    default:
      return false;
    case AArch64_AM::UXTB:
    case AArch64_AM::UXTH:
    case AArch64_AM::UXTW:
    case AArch64_AM::UXTX:
      return AArch64_AM::getArithShiftValue(Imm) <= 4;
    }
  }
 
  case AArch64::SUBWrs:
  case AArch64::SUBSWrs: {
    unsigned Imm = MI.getOperand(3).getImm();
    unsigned ShiftVal = AArch64_AM::getShiftValue(Imm);
    return ShiftVal == 0 ||
           (AArch64_AM::getShiftType(Imm) == AArch64_AM::ASR && ShiftVal == 31);
  }
 
  case AArch64::SUBXrs:
  case AArch64::SUBSXrs: {
    unsigned Imm = MI.getOperand(3).getImm();
    unsigned ShiftVal = AArch64_AM::getShiftValue(Imm);
    return ShiftVal == 0 ||
           (AArch64_AM::getShiftType(Imm) == AArch64_AM::ASR && ShiftVal == 63);
  }
 
  case AArch64::SUBWrx:
  case AArch64::SUBXrx:
  case AArch64::SUBXrx64:
  case AArch64::SUBSWrx:
  case AArch64::SUBSXrx:
  case AArch64::SUBSXrx64: {
    unsigned Imm = MI.getOperand(3).getImm();
    switch (AArch64_AM::getArithExtendType(Imm)) {
    default:
      return false;
    case AArch64_AM::UXTB:
    case AArch64_AM::UXTH:
    case AArch64_AM::UXTW:
    case AArch64_AM::UXTX:
      return AArch64_AM::getArithShiftValue(Imm) == 0;
    }
  }
 
  case AArch64::LDRBBroW:
  case AArch64::LDRBBroX:
  case AArch64::LDRBroW:
  case AArch64::LDRBroX:
  case AArch64::LDRDroW:
  case AArch64::LDRDroX:
  case AArch64::LDRHHroW:
  case AArch64::LDRHHroX:
  case AArch64::LDRHroW:
  case AArch64::LDRHroX:
  case AArch64::LDRQroW:
  case AArch64::LDRQroX:
  case AArch64::LDRSBWroW:
  case AArch64::LDRSBWroX:
  case AArch64::LDRSBXroW:
  case AArch64::LDRSBXroX:
  case AArch64::LDRSHWroW:
  case AArch64::LDRSHWroX:
  case AArch64::LDRSHXroW:
  case AArch64::LDRSHXroX:
  case AArch64::LDRSWroW:
  case AArch64::LDRSWroX:
  case AArch64::LDRSroW:
  case AArch64::LDRSroX:
  case AArch64::LDRWroW:
  case AArch64::LDRWroX:
  case AArch64::LDRXroW:
  case AArch64::LDRXroX:
  case AArch64::PRFMroW:
  case AArch64::PRFMroX:
  case AArch64::STRBBroW:
  case AArch64::STRBBroX:
  case AArch64::STRBroW:
  case AArch64::STRBroX:
  case AArch64::STRDroW:
  case AArch64::STRDroX:
  case AArch64::STRHHroW:
  case AArch64::STRHHroX:
  case AArch64::STRHroW:
  case AArch64::STRHroX:
  case AArch64::STRQroW:
  case AArch64::STRQroX:
  case AArch64::STRSroW:
  case AArch64::STRSroX:
  case AArch64::STRWroW:
  case AArch64::STRWroX:
  case AArch64::STRXroW:
  case AArch64::STRXroX: {
    unsigned IsSigned = MI.getOperand(3).getImm();
    return !IsSigned;
  }
  }
}
 
bool AArch64InstrInfo::isSEHInstruction(const MachineInstr &MI) {
  unsigned Opc = MI.getOpcode();
  switch (Opc) {
    default:
      return false;
    case AArch64::SEH_StackAlloc:
    case AArch64::SEH_SaveFPLR:
    case AArch64::SEH_SaveFPLR_X:
    case AArch64::SEH_SaveReg:
    case AArch64::SEH_SaveReg_X:
    case AArch64::SEH_SaveRegP:
    case AArch64::SEH_SaveRegP_X:
    case AArch64::SEH_SaveFReg:
    case AArch64::SEH_SaveFReg_X:
    case AArch64::SEH_SaveFRegP:
    case AArch64::SEH_SaveFRegP_X:
    case AArch64::SEH_SetFP:
    case AArch64::SEH_AddFP:
    case AArch64::SEH_Nop:
    case AArch64::SEH_PrologEnd:
    case AArch64::SEH_EpilogStart:
    case AArch64::SEH_EpilogEnd:
    case AArch64::SEH_PACSignLR:
      return true;
  }
}
 
bool AArch64InstrInfo::isCoalescableExtInstr(const MachineInstr &MI,
                                             Register &SrcReg, Register &DstReg,
                                             unsigned &SubIdx) const {
  switch (MI.getOpcode()) {
  default:
    return false;
  case AArch64::SBFMXri: // aka sxtw
  case AArch64::UBFMXri: // aka uxtw
    // Check for the 32 -> 64 bit extension case, these instructions can do
    // much more.
    if (MI.getOperand(2).getImm() != 0 || MI.getOperand(3).getImm() != 31)
      return false;
    // This is a signed or unsigned 32 -> 64 bit extension.
    SrcReg = MI.getOperand(1).getReg();
    DstReg = MI.getOperand(0).getReg();
    SubIdx = AArch64::sub_32;
    return true;
  }
}
 
bool AArch64InstrInfo::areMemAccessesTriviallyDisjoint(
    const MachineInstr &MIa, const MachineInstr &MIb) const {
  const TargetRegisterInfo *TRI = &getRegisterInfo();
  const MachineOperand *BaseOpA = nullptr, *BaseOpB = nullptr;
  int64_t OffsetA = 0, OffsetB = 0;
  unsigned WidthA = 0, WidthB = 0;
  bool OffsetAIsScalable = false, OffsetBIsScalable = false;
 
  assert(MIa.mayLoadOrStore() && "MIa must be a load or store.");
  assert(MIb.mayLoadOrStore() && "MIb must be a load or store.");
 
  if (MIa.hasUnmodeledSideEffects() || MIb.hasUnmodeledSideEffects() ||
      MIa.hasOrderedMemoryRef() || MIb.hasOrderedMemoryRef())
    return false;
 
  // Retrieve the base, offset from the base and width. Width
  // is the size of memory that is being loaded/stored (e.g. 1, 2, 4, 8).  If
  // base are identical, and the offset of a lower memory access +
  // the width doesn't overlap the offset of a higher memory access,
  // then the memory accesses are different.
  // If OffsetAIsScalable and OffsetBIsScalable are both true, they
  // are assumed to have the same scale (vscale).
  if (getMemOperandWithOffsetWidth(MIa, BaseOpA, OffsetA, OffsetAIsScalable,
                                   WidthA, TRI) &&
      getMemOperandWithOffsetWidth(MIb, BaseOpB, OffsetB, OffsetBIsScalable,
                                   WidthB, TRI)) {
    if (BaseOpA->isIdenticalTo(*BaseOpB) &&
        OffsetAIsScalable == OffsetBIsScalable) {
      int LowOffset = OffsetA < OffsetB ? OffsetA : OffsetB;
      int HighOffset = OffsetA < OffsetB ? OffsetB : OffsetA;
      int LowWidth = (LowOffset == OffsetA) ? WidthA : WidthB;
      if (LowOffset + LowWidth <= HighOffset)
        return true;
    }
  }
  return false;
}
 
bool AArch64InstrInfo::isSchedulingBoundary(const MachineInstr &MI,
                                            const MachineBasicBlock *MBB,
                                            const MachineFunction &MF) const {
  if (TargetInstrInfo::isSchedulingBoundary(MI, MBB, MF))
    return true;
  switch (MI.getOpcode()) {
  case AArch64::HINT:
    // CSDB hints are scheduling barriers.
    if (MI.getOperand(0).getImm() == 0x14)
      return true;
    break;
  case AArch64::DSB:
  case AArch64::ISB:
    // DSB and ISB also are scheduling barriers.
    return true;
  case AArch64::MSRpstatesvcrImm1:
    // SMSTART and SMSTOP are also scheduling barriers.
    return true;
  default:;
  }
  if (isSEHInstruction(MI))
    return true;
  auto Next = std::next(MI.getIterator());
  return Next != MBB->end() && Next->isCFIInstruction();
}
 
/// analyzeCompare - For a comparison instruction, return the source registers
/// in SrcReg and SrcReg2, and the value it compares against in CmpValue.
/// Return true if the comparison instruction can be analyzed.
bool AArch64InstrInfo::analyzeCompare(const MachineInstr &MI, Register &SrcReg,
                                      Register &SrcReg2, int64_t &CmpMask,
                                      int64_t &CmpValue) const {
  // The first operand can be a frame index where we'd normally expect a
  // register.
  assert(MI.getNumOperands() >= 2 && "All AArch64 cmps should have 2 operands");
  if (!MI.getOperand(1).isReg())
    return false;
 
  switch (MI.getOpcode()) {
  default:
    break;
  case AArch64::PTEST_PP:
  case AArch64::PTEST_PP_ANY:
    SrcReg = MI.getOperand(0).getReg();
    SrcReg2 = MI.getOperand(1).getReg();
    // Not sure about the mask and value for now...
    CmpMask = ~0;
    CmpValue = 0;
    return true;
  case AArch64::SUBSWrr:
  case AArch64::SUBSWrs:
  case AArch64::SUBSWrx:
  case AArch64::SUBSXrr:
  case AArch64::SUBSXrs:
  case AArch64::SUBSXrx:
  case AArch64::ADDSWrr:
  case AArch64::ADDSWrs:
  case AArch64::ADDSWrx:
  case AArch64::ADDSXrr:
  case AArch64::ADDSXrs:
  case AArch64::ADDSXrx:
    // Replace SUBSWrr with SUBWrr if NZCV is not used.
    SrcReg = MI.getOperand(1).getReg();
    SrcReg2 = MI.getOperand(2).getReg();
    CmpMask = ~0;
    CmpValue = 0;
    return true;
  case AArch64::SUBSWri:
  case AArch64::ADDSWri:
  case AArch64::SUBSXri:
  case AArch64::ADDSXri:
    SrcReg = MI.getOperand(1).getReg();
    SrcReg2 = 0;
    CmpMask = ~0;
    CmpValue = MI.getOperand(2).getImm();
    return true;
  case AArch64::ANDSWri:
  case AArch64::ANDSXri:
    // ANDS does not use the same encoding scheme as the others xxxS
    // instructions.
    SrcReg = MI.getOperand(1).getReg();
    SrcReg2 = 0;
    CmpMask = ~0;
    CmpValue = AArch64_AM::decodeLogicalImmediate(
                   MI.getOperand(2).getImm(),
                   MI.getOpcode() == AArch64::ANDSWri ? 32 : 64);
    return true;
  }
 
  return false;
}
 
static bool UpdateOperandRegClass(MachineInstr &Instr) {
  MachineBasicBlock *MBB = Instr.getParent();
  assert(MBB && "Can't get MachineBasicBlock here");
  MachineFunction *MF = MBB->getParent();
  assert(MF && "Can't get MachineFunction here");
  const TargetInstrInfo *TII = MF->getSubtarget().getInstrInfo();
  const TargetRegisterInfo *TRI = MF->getSubtarget().getRegisterInfo();
  MachineRegisterInfo *MRI = &MF->getRegInfo();
 
  for (unsigned OpIdx = 0, EndIdx = Instr.getNumOperands(); OpIdx < EndIdx;
       ++OpIdx) {
    MachineOperand &MO = Instr.getOperand(OpIdx);
    const TargetRegisterClass *OpRegCstraints =
        Instr.getRegClassConstraint(OpIdx, TII, TRI);
 
    // If there's no constraint, there's nothing to do.
    if (!OpRegCstraints)
      continue;
    // If the operand is a frame index, there's nothing to do here.
    // A frame index operand will resolve correctly during PEI.
    if (MO.isFI())
      continue;
 
    assert(MO.isReg() &&
           "Operand has register constraints without being a register!");
 
    Register Reg = MO.getReg();
    if (Reg.isPhysical()) {
      if (!OpRegCstraints->contains(Reg))
        return false;
    } else if (!OpRegCstraints->hasSubClassEq(MRI->getRegClass(Reg)) &&
               !MRI->constrainRegClass(Reg, OpRegCstraints))
      return false;
  }
 
  return true;
}
 
/// Return the opcode that does not set flags when possible - otherwise
/// return the original opcode. The caller is responsible to do the actual
/// substitution and legality checking.
static unsigned convertToNonFlagSettingOpc(const MachineInstr &MI) {
  // Don't convert all compare instructions, because for some the zero register
  // encoding becomes the sp register.
  bool MIDefinesZeroReg = false;
  if (MI.definesRegister(AArch64::WZR) || MI.definesRegister(AArch64::XZR))
    MIDefinesZeroReg = true;
 
  switch (MI.getOpcode()) {
  default:
    return MI.getOpcode();
  case AArch64::ADDSWrr:
    return AArch64::ADDWrr;
  case AArch64::ADDSWri:
    return MIDefinesZeroReg ? AArch64::ADDSWri : AArch64::ADDWri;
  case AArch64::ADDSWrs:
    return MIDefinesZeroReg ? AArch64::ADDSWrs : AArch64::ADDWrs;
  case AArch64::ADDSWrx:
    return AArch64::ADDWrx;
  case AArch64::ADDSXrr:
    return AArch64::ADDXrr;
  case AArch64::ADDSXri:
    return MIDefinesZeroReg ? AArch64::ADDSXri : AArch64::ADDXri;
  case AArch64::ADDSXrs:
    return MIDefinesZeroReg ? AArch64::ADDSXrs : AArch64::ADDXrs;
  case AArch64::ADDSXrx:
    return AArch64::ADDXrx;
  case AArch64::SUBSWrr:
    return AArch64::SUBWrr;
  case AArch64::SUBSWri:
    return MIDefinesZeroReg ? AArch64::SUBSWri : AArch64::SUBWri;
  case AArch64::SUBSWrs:
    return MIDefinesZeroReg ? AArch64::SUBSWrs : AArch64::SUBWrs;
  case AArch64::SUBSWrx:
    return AArch64::SUBWrx;
  case AArch64::SUBSXrr:
    return AArch64::SUBXrr;
  case AArch64::SUBSXri:
    return MIDefinesZeroReg ? AArch64::SUBSXri : AArch64::SUBXri;
  case AArch64::SUBSXrs:
    return MIDefinesZeroReg ? AArch64::SUBSXrs : AArch64::SUBXrs;
  case AArch64::SUBSXrx:
    return AArch64::SUBXrx;
  }
}
 
enum AccessKind { AK_Write = 0x01, AK_Read = 0x10, AK_All = 0x11 };
 
/// True when condition flags are accessed (either by writing or reading)
/// on the instruction trace starting at From and ending at To.
///
/// Note: If From and To are from different blocks it's assumed CC are accessed
///       on the path.
static bool areCFlagsAccessedBetweenInstrs(
    MachineBasicBlock::iterator From, MachineBasicBlock::iterator To,
    const TargetRegisterInfo *TRI, const AccessKind AccessToCheck = AK_All) {
  // Early exit if To is at the beginning of the BB.
  if (To == To->getParent()->begin())
    return true;
 
  // Check whether the instructions are in the same basic block
  // If not, assume the condition flags might get modified somewhere.
  if (To->getParent() != From->getParent())
    return true;
 
  // From must be above To.
  assert(std::any_of(
      ++To.getReverse(), To->getParent()->rend(),
      [From](MachineInstr &MI) { return MI.getIterator() == From; }));
 
  // We iterate backward starting at \p To until we hit \p From.
  for (const MachineInstr &Instr :
       instructionsWithoutDebug(++To.getReverse(), From.getReverse())) {
    if (((AccessToCheck & AK_Write) &&
         Instr.modifiesRegister(AArch64::NZCV, TRI)) ||
        ((AccessToCheck & AK_Read) && Instr.readsRegister(AArch64::NZCV, TRI)))
      return true;
  }
  return false;
}
 
/// optimizePTestInstr - Attempt to remove a ptest of a predicate-generating
/// operation which could set the flags in an identical manner
bool AArch64InstrInfo::optimizePTestInstr(
    MachineInstr *PTest, unsigned MaskReg, unsigned PredReg,
    const MachineRegisterInfo *MRI) const {
  auto *Mask = MRI->getUniqueVRegDef(MaskReg);
  auto *Pred = MRI->getUniqueVRegDef(PredReg);
  auto NewOp = Pred->getOpcode();
  bool OpChanged = false;
 
  unsigned MaskOpcode = Mask->getOpcode();
  unsigned PredOpcode = Pred->getOpcode();
  bool PredIsPTestLike = isPTestLikeOpcode(PredOpcode);
  bool PredIsWhileLike = isWhileOpcode(PredOpcode);
 
  if (isPTrueOpcode(MaskOpcode) && (PredIsPTestLike || PredIsWhileLike) &&
      getElementSizeForOpcode(MaskOpcode) ==
          getElementSizeForOpcode(PredOpcode) &&
      Mask->getOperand(1).getImm() == 31) {
    // For PTEST(PTRUE_ALL, WHILE), if the element size matches, the PTEST is
    // redundant since WHILE performs an implicit PTEST with an all active
    // mask. Must be an all active predicate of matching element size.
 
    // For PTEST(PTRUE_ALL, PTEST_LIKE), the PTEST is redundant if the
    // PTEST_LIKE instruction uses the same all active mask and the element
    // size matches. If the PTEST has a condition of any then it is always
    // redundant.
    if (PredIsPTestLike) {
      auto PTestLikeMask = MRI->getUniqueVRegDef(Pred->getOperand(1).getReg());
      if (Mask != PTestLikeMask && PTest->getOpcode() != AArch64::PTEST_PP_ANY)
        return false;
    }
 
    // Fallthough to simply remove the PTEST.
  } else if ((Mask == Pred) && (PredIsPTestLike || PredIsWhileLike) &&
             PTest->getOpcode() == AArch64::PTEST_PP_ANY) {
    // For PTEST(PG, PG), PTEST is redundant when PG is the result of an
    // instruction that sets the flags as PTEST would. This is only valid when
    // the condition is any.
 
    // Fallthough to simply remove the PTEST.
  } else if (PredIsPTestLike) {
    // For PTEST(PG, PTEST_LIKE(PG, ...)), the PTEST is redundant since the
    // flags are set based on the same mask 'PG', but PTEST_LIKE must operate
    // on 8-bit predicates like the PTEST.  Otherwise, for instructions like
    // compare that also support 16/32/64-bit predicates, the implicit PTEST
    // performed by the compare could consider fewer lanes for these element
    // sizes.
    //
    // For example, consider
    //
    //   ptrue p0.b                    ; P0=1111-1111-1111-1111
    //   index z0.s, #0, #1            ; Z0=<0,1,2,3>
    //   index z1.s, #1, #1            ; Z1=<1,2,3,4>
    //   cmphi p1.s, p0/z, z1.s, z0.s  ; P1=0001-0001-0001-0001
    //                                 ;       ^ last active
    //   ptest p0, p1.b                ; P1=0001-0001-0001-0001
    //                                 ;     ^ last active
    //
    // where the compare generates a canonical all active 32-bit predicate
    // (equivalent to 'ptrue p1.s, all'). The implicit PTEST sets the last
    // active flag, whereas the PTEST instruction with the same mask doesn't.
    // For PTEST_ANY this doesn't apply as the flags in this case would be
    // identical regardless of element size.
    auto PTestLikeMask = MRI->getUniqueVRegDef(Pred->getOperand(1).getReg());
    uint64_t PredElementSize = getElementSizeForOpcode(PredOpcode);
    if ((Mask != PTestLikeMask) ||
        (PredElementSize != AArch64::ElementSizeB &&
         PTest->getOpcode() != AArch64::PTEST_PP_ANY))
      return false;
 
    // Fallthough to simply remove the PTEST.
  } else {
    // If OP in PTEST(PG, OP(PG, ...)) has a flag-setting variant change the
    // opcode so the PTEST becomes redundant.
    switch (PredOpcode) {
    case AArch64::AND_PPzPP:
    case AArch64::BIC_PPzPP:
    case AArch64::EOR_PPzPP:
    case AArch64::NAND_PPzPP:
    case AArch64::NOR_PPzPP:
    case AArch64::ORN_PPzPP:
    case AArch64::ORR_PPzPP:
    case AArch64::BRKA_PPzP:
    case AArch64::BRKPA_PPzPP:
    case AArch64::BRKB_PPzP:
    case AArch64::BRKPB_PPzPP:
    case AArch64::RDFFR_PPz: {
      // Check to see if our mask is the same. If not the resulting flag bits
      // may be different and we can't remove the ptest.
      auto *PredMask = MRI->getUniqueVRegDef(Pred->getOperand(1).getReg());
      if (Mask != PredMask)
        return false;
      break;
    }
    case AArch64::BRKN_PPzP: {
      // BRKN uses an all active implicit mask to set flags unlike the other
      // flag-setting instructions.
      // PTEST(PTRUE_B(31), BRKN(PG, A, B)) -> BRKNS(PG, A, B).
      if ((MaskOpcode != AArch64::PTRUE_B) ||
          (Mask->getOperand(1).getImm() != 31))
        return false;
      break;
    }
    case AArch64::PTRUE_B:
      // PTEST(OP=PTRUE_B(A), OP) -> PTRUES_B(A)
      break;
    default:
      // Bail out if we don't recognize the input
      return false;
    }
 
    NewOp = convertToFlagSettingOpc(PredOpcode);
    OpChanged = true;
  }
 
  const TargetRegisterInfo *TRI = &getRegisterInfo();
 
  // If another instruction between Pred and PTest accesses flags, don't remove
  // the ptest or update the earlier instruction to modify them.
  if (areCFlagsAccessedBetweenInstrs(Pred, PTest, TRI))
    return false;
 
  // If we pass all the checks, it's safe to remove the PTEST and use the flags
  // as they are prior to PTEST. Sometimes this requires the tested PTEST
  // operand to be replaced with an equivalent instruction that also sets the
  // flags.
  Pred->setDesc(get(NewOp));
  PTest->eraseFromParent();
  if (OpChanged) {
    bool succeeded = UpdateOperandRegClass(*Pred);
    (void)succeeded;
    assert(succeeded && "Operands have incompatible register classes!");
    Pred->addRegisterDefined(AArch64::NZCV, TRI);
  }
 
  // Ensure that the flags def is live.
  if (Pred->registerDefIsDead(AArch64::NZCV, TRI)) {
    unsigned i = 0, e = Pred->getNumOperands();
    for (; i != e; ++i) {
      MachineOperand &MO = Pred->getOperand(i);
      if (MO.isReg() && MO.isDef() && MO.getReg() == AArch64::NZCV) {
        MO.setIsDead(false);
        break;
      }
    }
  }
  return true;
}
 
/// Try to optimize a compare instruction. A compare instruction is an
/// instruction which produces AArch64::NZCV. It can be truly compare
/// instruction
/// when there are no uses of its destination register.
///
/// The following steps are tried in order:
/// 1. Convert CmpInstr into an unconditional version.
/// 2. Remove CmpInstr if above there is an instruction producing a needed
///    condition code or an instruction which can be converted into such an
///    instruction.
///    Only comparison with zero is supported.
bool AArch64InstrInfo::optimizeCompareInstr(
    MachineInstr &CmpInstr, Register SrcReg, Register SrcReg2, int64_t CmpMask,
    int64_t CmpValue, const MachineRegisterInfo *MRI) const {
  assert(CmpInstr.getParent());
  assert(MRI);
 
  // Replace SUBSWrr with SUBWrr if NZCV is not used.
  int DeadNZCVIdx = CmpInstr.findRegisterDefOperandIdx(AArch64::NZCV, true);
  if (DeadNZCVIdx != -1) {
    if (CmpInstr.definesRegister(AArch64::WZR) ||
        CmpInstr.definesRegister(AArch64::XZR)) {
      CmpInstr.eraseFromParent();
      return true;
    }
    unsigned Opc = CmpInstr.getOpcode();
    unsigned NewOpc = convertToNonFlagSettingOpc(CmpInstr);
    if (NewOpc == Opc)
      return false;
    const MCInstrDesc &MCID = get(NewOpc);
    CmpInstr.setDesc(MCID);
    CmpInstr.removeOperand(DeadNZCVIdx);
    bool succeeded = UpdateOperandRegClass(CmpInstr);
    (void)succeeded;
    assert(succeeded && "Some operands reg class are incompatible!");
    return true;
  }
 
  if (CmpInstr.getOpcode() == AArch64::PTEST_PP ||
      CmpInstr.getOpcode() == AArch64::PTEST_PP_ANY)
    return optimizePTestInstr(&CmpInstr, SrcReg, SrcReg2, MRI);
 
  if (SrcReg2 != 0)
    return false;
 
  // CmpInstr is a Compare instruction if destination register is not used.
  if (!MRI->use_nodbg_empty(CmpInstr.getOperand(0).getReg()))
    return false;
 
  if (CmpValue == 0 && substituteCmpToZero(CmpInstr, SrcReg, *MRI))
    return true;
  return (CmpValue == 0 || CmpValue == 1) &&
         removeCmpToZeroOrOne(CmpInstr, SrcReg, CmpValue, *MRI);
}
 
/// Get opcode of S version of Instr.
/// If Instr is S version its opcode is returned.
/// AArch64::INSTRUCTION_LIST_END is returned if Instr does not have S version
/// or we are not interested in it.
static unsigned sForm(MachineInstr &Instr) {
  switch (Instr.getOpcode()) {
  default:
    return AArch64::INSTRUCTION_LIST_END;
 
  case AArch64::ADDSWrr:
  case AArch64::ADDSWri:
  case AArch64::ADDSXrr:
  case AArch64::ADDSXri:
  case AArch64::SUBSWrr:
  case AArch64::SUBSWri:
  case AArch64::SUBSXrr:
  case AArch64::SUBSXri:
    return Instr.getOpcode();
 
  case AArch64::ADDWrr:
    return AArch64::ADDSWrr;
  case AArch64::ADDWri:
    return AArch64::ADDSWri;
  case AArch64::ADDXrr:
    return AArch64::ADDSXrr;
  case AArch64::ADDXri:
    return AArch64::ADDSXri;
  case AArch64::ADCWr:
    return AArch64::ADCSWr;
  case AArch64::ADCXr:
    return AArch64::ADCSXr;
  case AArch64::SUBWrr:
    return AArch64::SUBSWrr;
  case AArch64::SUBWri:
    return AArch64::SUBSWri;
  case AArch64::SUBXrr:
    return AArch64::SUBSXrr;
  case AArch64::SUBXri:
    return AArch64::SUBSXri;
  case AArch64::SBCWr:
    return AArch64::SBCSWr;
  case AArch64::SBCXr:
    return AArch64::SBCSXr;
  case AArch64::ANDWri:
    return AArch64::ANDSWri;
  case AArch64::ANDXri:
    return AArch64::ANDSXri;
  }
}
 
/// Check if AArch64::NZCV should be alive in successors of MBB.
static bool areCFlagsAliveInSuccessors(const MachineBasicBlock *MBB) {
  for (auto *BB : MBB->successors())
    if (BB->isLiveIn(AArch64::NZCV))
      return true;
  return false;
}
 
/// \returns The condition code operand index for \p Instr if it is a branch
/// or select and -1 otherwise.
static int
findCondCodeUseOperandIdxForBranchOrSelect(const MachineInstr &Instr) {
  switch (Instr.getOpcode()) {
  default:
    return -1;
 
  case AArch64::Bcc: {
    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV);
    assert(Idx >= 2);
    return Idx - 2;
  }
 
  case AArch64::CSINVWr:
  case AArch64::CSINVXr:
  case AArch64::CSINCWr:
  case AArch64::CSINCXr:
  case AArch64::CSELWr:
  case AArch64::CSELXr:
  case AArch64::CSNEGWr:
  case AArch64::CSNEGXr:
  case AArch64::FCSELSrrr:
  case AArch64::FCSELDrrr: {
    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV);
    assert(Idx >= 1);
    return Idx - 1;
  }
  }
}
 
/// Find a condition code used by the instruction.
/// Returns AArch64CC::Invalid if either the instruction does not use condition
/// codes or we don't optimize CmpInstr in the presence of such instructions.
static AArch64CC::CondCode findCondCodeUsedByInstr(const MachineInstr &Instr) {
  int CCIdx = findCondCodeUseOperandIdxForBranchOrSelect(Instr);
  return CCIdx >= 0 ? static_cast<AArch64CC::CondCode>(
                          Instr.getOperand(CCIdx).getImm())
                    : AArch64CC::Invalid;
}
 
static UsedNZCV getUsedNZCV(AArch64CC::CondCode CC) {
  assert(CC != AArch64CC::Invalid);
  UsedNZCV UsedFlags;
  switch (CC) {
  default:
    break;
 
  case AArch64CC::EQ: // Z set
  case AArch64CC::NE: // Z clear
    UsedFlags.Z = true;
    break;
 
  case AArch64CC::HI: // Z clear and C set
  case AArch64CC::LS: // Z set   or  C clear
    UsedFlags.Z = true;
    [[fallthrough]];
  case AArch64CC::HS: // C set
  case AArch64CC::LO: // C clear
    UsedFlags.C = true;
    break;
 
  case AArch64CC::MI: // N set
  case AArch64CC::PL: // N clear
    UsedFlags.N = true;
    break;
 
  case AArch64CC::VS: // V set
  case AArch64CC::VC: // V clear
    UsedFlags.V = true;
    break;
 
  case AArch64CC::GT: // Z clear, N and V the same
  case AArch64CC::LE: // Z set,   N and V differ
    UsedFlags.Z = true;
    [[fallthrough]];
  case AArch64CC::GE: // N and V the same
  case AArch64CC::LT: // N and V differ
    UsedFlags.N = true;
    UsedFlags.V = true;
    break;
  }
  return UsedFlags;
}
 
/// \returns Conditions flags used after \p CmpInstr in its MachineBB if NZCV
/// flags are not alive in successors of the same \p CmpInstr and \p MI parent.
/// \returns std::nullopt otherwise.
///
/// Collect instructions using that flags in \p CCUseInstrs if provided.
std::optional<UsedNZCV>
llvm::examineCFlagsUse(MachineInstr &MI, MachineInstr &CmpInstr,
                       const TargetRegisterInfo &TRI,
                       SmallVectorImpl<MachineInstr *> *CCUseInstrs) {
  MachineBasicBlock *CmpParent = CmpInstr.getParent();
  if (MI.getParent() != CmpParent)
    return std::nullopt;
 
  if (areCFlagsAliveInSuccessors(CmpParent))
    return std::nullopt;
 
  UsedNZCV NZCVUsedAfterCmp;
  for (MachineInstr &Instr : instructionsWithoutDebug(
           std::next(CmpInstr.getIterator()), CmpParent->instr_end())) {
    if (Instr.readsRegister(AArch64::NZCV, &TRI)) {
      AArch64CC::CondCode CC = findCondCodeUsedByInstr(Instr);
      if (CC == AArch64CC::Invalid) // Unsupported conditional instruction
        return std::nullopt;
      NZCVUsedAfterCmp |= getUsedNZCV(CC);
      if (CCUseInstrs)
        CCUseInstrs->push_back(&Instr);
    }
    if (Instr.modifiesRegister(AArch64::NZCV, &TRI))
      break;
  }
  return NZCVUsedAfterCmp;
}
 
static bool isADDSRegImm(unsigned Opcode) {
  return Opcode == AArch64::ADDSWri || Opcode == AArch64::ADDSXri;
}
 
static bool isSUBSRegImm(unsigned Opcode) {
  return Opcode == AArch64::SUBSWri || Opcode == AArch64::SUBSXri;
}
 
/// Check if CmpInstr can be substituted by MI.
///
/// CmpInstr can be substituted:
/// - CmpInstr is either 'ADDS %vreg, 0' or 'SUBS %vreg, 0'
/// - and, MI and CmpInstr are from the same MachineBB
/// - and, condition flags are not alive in successors of the CmpInstr parent
/// - and, if MI opcode is the S form there must be no defs of flags between
///        MI and CmpInstr
///        or if MI opcode is not the S form there must be neither defs of flags
///        nor uses of flags between MI and CmpInstr.
/// - and, if C/V flags are not used after CmpInstr
///        or if N flag is used but MI produces poison value if signed overflow
///        occurs.
static bool canInstrSubstituteCmpInstr(MachineInstr &MI, MachineInstr &CmpInstr,
                                       const TargetRegisterInfo &TRI) {
  // NOTE this assertion guarantees that MI.getOpcode() is add or subtraction
  // that may or may not set flags.
  assert(sForm(MI) != AArch64::INSTRUCTION_LIST_END);
 
  const unsigned CmpOpcode = CmpInstr.getOpcode();
  if (!isADDSRegImm(CmpOpcode) && !isSUBSRegImm(CmpOpcode))
    return false;
 
  assert((CmpInstr.getOperand(2).isImm() &&
          CmpInstr.getOperand(2).getImm() == 0) &&
         "Caller guarantees that CmpInstr compares with constant 0");
 
  std::optional<UsedNZCV> NZVCUsed = examineCFlagsUse(MI, CmpInstr, TRI);
  if (!NZVCUsed || NZVCUsed->C)
    return false;
 
  // CmpInstr is either 'ADDS %vreg, 0' or 'SUBS %vreg, 0', and MI is either
  // '%vreg = add ...' or '%vreg = sub ...'.
  // Condition flag V is used to indicate signed overflow.
  // 1) MI and CmpInstr set N and V to the same value.
  // 2) If MI is add/sub with no-signed-wrap, it produces a poison value when
  //    signed overflow occurs, so CmpInstr could still be simplified away.
  if (NZVCUsed->V && !MI.getFlag(MachineInstr::NoSWrap))
    return false;
 
  AccessKind AccessToCheck = AK_Write;
  if (sForm(MI) != MI.getOpcode())
    AccessToCheck = AK_All;
  return !areCFlagsAccessedBetweenInstrs(&MI, &CmpInstr, &TRI, AccessToCheck);
}
 
/// Substitute an instruction comparing to zero with another instruction
/// which produces needed condition flags.
///
/// Return true on success.
bool AArch64InstrInfo::substituteCmpToZero(
    MachineInstr &CmpInstr, unsigned SrcReg,
    const MachineRegisterInfo &MRI) const {
  // Get the unique definition of SrcReg.
  MachineInstr *MI = MRI.getUniqueVRegDef(SrcReg);
  if (!MI)
    return false;
 
  const TargetRegisterInfo &TRI = getRegisterInfo();
 
  unsigned NewOpc = sForm(*MI);
  if (NewOpc == AArch64::INSTRUCTION_LIST_END)
    return false;
 
  if (!canInstrSubstituteCmpInstr(*MI, CmpInstr, TRI))
    return false;
 
  // Update the instruction to set NZCV.
  MI->setDesc(get(NewOpc));
  CmpInstr.eraseFromParent();
  bool succeeded = UpdateOperandRegClass(*MI);
  (void)succeeded;
  assert(succeeded && "Some operands reg class are incompatible!");
  MI->addRegisterDefined(AArch64::NZCV, &TRI);
  return true;
}
 
/// \returns True if \p CmpInstr can be removed.
///
/// \p IsInvertCC is true if, after removing \p CmpInstr, condition
/// codes used in \p CCUseInstrs must be inverted.
static bool canCmpInstrBeRemoved(MachineInstr &MI, MachineInstr &CmpInstr,
                                 int CmpValue, const TargetRegisterInfo &TRI,
                                 SmallVectorImpl<MachineInstr *> &CCUseInstrs,
                                 bool &IsInvertCC) {
  assert((CmpValue == 0 || CmpValue == 1) &&
         "Only comparisons to 0 or 1 considered for removal!");
 
  // MI is 'CSINCWr %vreg, wzr, wzr, <cc>' or 'CSINCXr %vreg, xzr, xzr, <cc>'
  unsigned MIOpc = MI.getOpcode();
  if (MIOpc == AArch64::CSINCWr) {
    if (MI.getOperand(1).getReg() != AArch64::WZR ||
        MI.getOperand(2).getReg() != AArch64::WZR)
      return false;
  } else if (MIOpc == AArch64::CSINCXr) {
    if (MI.getOperand(1).getReg() != AArch64::XZR ||
        MI.getOperand(2).getReg() != AArch64::XZR)
      return false;
  } else {
    return false;
  }
  AArch64CC::CondCode MICC = findCondCodeUsedByInstr(MI);
  if (MICC == AArch64CC::Invalid)
    return false;
 
  // NZCV needs to be defined
  if (MI.findRegisterDefOperandIdx(AArch64::NZCV, true) != -1)
    return false;
 
  // CmpInstr is 'ADDS %vreg, 0' or 'SUBS %vreg, 0' or 'SUBS %vreg, 1'
  const unsigned CmpOpcode = CmpInstr.getOpcode();
  bool IsSubsRegImm = isSUBSRegImm(CmpOpcode);
  if (CmpValue && !IsSubsRegImm)
    return false;
  if (!CmpValue && !IsSubsRegImm && !isADDSRegImm(CmpOpcode))
    return false;
 
  // MI conditions allowed: eq, ne, mi, pl
  UsedNZCV MIUsedNZCV = getUsedNZCV(MICC);
  if (MIUsedNZCV.C || MIUsedNZCV.V)
    return false;
 
  std::optional<UsedNZCV> NZCVUsedAfterCmp =
      examineCFlagsUse(MI, CmpInstr, TRI, &CCUseInstrs);
  // Condition flags are not used in CmpInstr basic block successors and only
  // Z or N flags allowed to be used after CmpInstr within its basic block
  if (!NZCVUsedAfterCmp || NZCVUsedAfterCmp->C || NZCVUsedAfterCmp->V)
    return false;
  // Z or N flag used after CmpInstr must correspond to the flag used in MI
  if ((MIUsedNZCV.Z && NZCVUsedAfterCmp->N) ||
      (MIUsedNZCV.N && NZCVUsedAfterCmp->Z))
    return false;
  // If CmpInstr is comparison to zero MI conditions are limited to eq, ne
  if (MIUsedNZCV.N && !CmpValue)
    return false;
 
  // There must be no defs of flags between MI and CmpInstr
  if (areCFlagsAccessedBetweenInstrs(&MI, &CmpInstr, &TRI, AK_Write))
    return false;
 
  // Condition code is inverted in the following cases:
  // 1. MI condition is ne; CmpInstr is 'ADDS %vreg, 0' or 'SUBS %vreg, 0'
  // 2. MI condition is eq, pl; CmpInstr is 'SUBS %vreg, 1'
  IsInvertCC = (CmpValue && (MICC == AArch64CC::EQ || MICC == AArch64CC::PL)) ||
               (!CmpValue && MICC == AArch64CC::NE);
  return true;
}
 
/// Remove comparison in csinc-cmp sequence
///
/// Examples:
/// 1. \code
///   csinc w9, wzr, wzr, ne
///   cmp   w9, #0
///   b.eq
///    \endcode
/// to
///    \code
///   csinc w9, wzr, wzr, ne
///   b.ne
///    \endcode
///
/// 2. \code
///   csinc x2, xzr, xzr, mi
///   cmp   x2, #1
///   b.pl
///    \endcode
/// to
///    \code
///   csinc x2, xzr, xzr, mi
///   b.pl
///    \endcode
///
/// \param  CmpInstr comparison instruction
/// \return True when comparison removed
bool AArch64InstrInfo::removeCmpToZeroOrOne(
    MachineInstr &CmpInstr, unsigned SrcReg, int CmpValue,
    const MachineRegisterInfo &MRI) const {
  MachineInstr *MI = MRI.getUniqueVRegDef(SrcReg);
  if (!MI)
    return false;
  const TargetRegisterInfo &TRI = getRegisterInfo();
  SmallVector<MachineInstr *, 4> CCUseInstrs;
  bool IsInvertCC = false;
  if (!canCmpInstrBeRemoved(*MI, CmpInstr, CmpValue, TRI, CCUseInstrs,
                            IsInvertCC))
    return false;
  // Make transformation
  CmpInstr.eraseFromParent();
  if (IsInvertCC) {
    // Invert condition codes in CmpInstr CC users
    for (MachineInstr *CCUseInstr : CCUseInstrs) {
      int Idx = findCondCodeUseOperandIdxForBranchOrSelect(*CCUseInstr);
      assert(Idx >= 0 && "Unexpected instruction using CC.");
      MachineOperand &CCOperand = CCUseInstr->getOperand(Idx);
      AArch64CC::CondCode CCUse = AArch64CC::getInvertedCondCode(
          static_cast<AArch64CC::CondCode>(CCOperand.getImm()));
      CCOperand.setImm(CCUse);
    }
  }
  return true;
}
 
bool AArch64InstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
  if (MI.getOpcode() != TargetOpcode::LOAD_STACK_GUARD &&
      MI.getOpcode() != AArch64::CATCHRET)
    return false;
 
  MachineBasicBlock &MBB = *MI.getParent();
  auto &Subtarget = MBB.getParent()->getSubtarget<AArch64Subtarget>();
  auto TRI = Subtarget.getRegisterInfo();
  DebugLoc DL = MI.getDebugLoc();
 
  if (MI.getOpcode() == AArch64::CATCHRET) {
    // Skip to the first instruction before the epilog.
    const TargetInstrInfo *TII =
      MBB.getParent()->getSubtarget().getInstrInfo();
    MachineBasicBlock *TargetMBB = MI.getOperand(0).getMBB();
    auto MBBI = MachineBasicBlock::iterator(MI);
    MachineBasicBlock::iterator FirstEpilogSEH = std::prev(MBBI);
    while (FirstEpilogSEH->getFlag(MachineInstr::FrameDestroy) &&
           FirstEpilogSEH != MBB.begin())
      FirstEpilogSEH = std::prev(FirstEpilogSEH);
    if (FirstEpilogSEH != MBB.begin())
      FirstEpilogSEH = std::next(FirstEpilogSEH);
    BuildMI(MBB, FirstEpilogSEH, DL, TII->get(AArch64::ADRP))
        .addReg(AArch64::X0, RegState::Define)
        .addMBB(TargetMBB);
    BuildMI(MBB, FirstEpilogSEH, DL, TII->get(AArch64::ADDXri))
        .addReg(AArch64::X0, RegState::Define)
        .addReg(AArch64::X0)
        .addMBB(TargetMBB)
        .addImm(0);
    return true;
  }
 
  Register Reg = MI.getOperand(0).getReg();
  Module &M = *MBB.getParent()->getFunction().getParent();
  if (M.getStackProtectorGuard() == "sysreg") {
    const AArch64SysReg::SysReg *SrcReg =
        AArch64SysReg::lookupSysRegByName(M.getStackProtectorGuardReg());
    if (!SrcReg)
      report_fatal_error("Unknown SysReg for Stack Protector Guard Register");
 
    // mrs xN, sysreg
    BuildMI(MBB, MI, DL, get(AArch64::MRS))
        .addDef(Reg, RegState::Renamable)
        .addImm(SrcReg->Encoding);
    int Offset = M.getStackProtectorGuardOffset();
    if (Offset >= 0 && Offset <= 32760 && Offset % 8 == 0) {
      // ldr xN, [xN, #offset]
      BuildMI(MBB, MI, DL, get(AArch64::LDRXui))
          .addDef(Reg)
          .addUse(Reg, RegState::Kill)
          .addImm(Offset / 8);
    } else if (Offset >= -256 && Offset <= 255) {
      // ldur xN, [xN, #offset]
      BuildMI(MBB, MI, DL, get(AArch64::LDURXi))
          .addDef(Reg)
          .addUse(Reg, RegState::Kill)
          .addImm(Offset);
    } else if (Offset >= -4095 && Offset <= 4095) {
      if (Offset > 0) {
        // add xN, xN, #offset
        BuildMI(MBB, MI, DL, get(AArch64::ADDXri))
            .addDef(Reg)
            .addUse(Reg, RegState::Kill)
            .addImm(Offset)
            .addImm(0);
      } else {
        // sub xN, xN, #offset
        BuildMI(MBB, MI, DL, get(AArch64::SUBXri))
            .addDef(Reg)
            .addUse(Reg, RegState::Kill)
            .addImm(-Offset)
            .addImm(0);
      }
      // ldr xN, [xN]
      BuildMI(MBB, MI, DL, get(AArch64::LDRXui))
          .addDef(Reg)
          .addUse(Reg, RegState::Kill)
          .addImm(0);
    } else {
      // Cases that are larger than +/- 4095 and not a multiple of 8, or larger
      // than 23760.
      // It might be nice to use AArch64::MOVi32imm here, which would get
      // expanded in PreSched2 after PostRA, but our lone scratch Reg already
      // contains the MRS result. findScratchNonCalleeSaveRegister() in
      // AArch64FrameLowering might help us find such a scratch register
      // though. If we failed to find a scratch register, we could emit a
      // stream of add instructions to build up the immediate. Or, we could try
      // to insert a AArch64::MOVi32imm before register allocation so that we
      // didn't need to scavenge for a scratch register.
      report_fatal_error("Unable to encode Stack Protector Guard Offset");
    }
    MBB.erase(MI);
    return true;
  }
 
  const GlobalValue *GV =
      cast<GlobalValue>((*MI.memoperands_begin())->getValue());
  const TargetMachine &TM = MBB.getParent()->getTarget();
  unsigned OpFlags = Subtarget.ClassifyGlobalReference(GV, TM);
  const unsigned char MO_NC = AArch64II::MO_NC;
 
  if ((OpFlags & AArch64II::MO_GOT) != 0) {
    BuildMI(MBB, MI, DL, get(AArch64::LOADgot), Reg)
        .addGlobalAddress(GV, 0, OpFlags);
    if (Subtarget.isTargetILP32()) {
      unsigned Reg32 = TRI->getSubReg(Reg, AArch64::sub_32);
      BuildMI(MBB, MI, DL, get(AArch64::LDRWui))
          .addDef(Reg32, RegState::Dead)
          .addUse(Reg, RegState::Kill)
          .addImm(0)
          .addMemOperand(*MI.memoperands_begin())
          .addDef(Reg, RegState::Implicit);
    } else {
      BuildMI(MBB, MI, DL, get(AArch64::LDRXui), Reg)
          .addReg(Reg, RegState::Kill)
          .addImm(0)
          .addMemOperand(*MI.memoperands_begin());
    }
  } else if (TM.getCodeModel() == CodeModel::Large) {
    assert(!Subtarget.isTargetILP32() && "how can large exist in ILP32?");
    BuildMI(MBB, MI, DL, get(AArch64::MOVZXi), Reg)
        .addGlobalAddress(GV, 0, AArch64II::MO_G0 | MO_NC)
        .addImm(0);
    BuildMI(MBB, MI, DL, get(AArch64::MOVKXi), Reg)
        .addReg(Reg, RegState::Kill)
        .addGlobalAddress(GV, 0, AArch64II::MO_G1 | MO_NC)
        .addImm(16);
    BuildMI(MBB, MI, DL, get(AArch64::MOVKXi), Reg)
        .addReg(Reg, RegState::Kill)
        .addGlobalAddress(GV, 0, AArch64II::MO_G2 | MO_NC)
        .addImm(32);
    BuildMI(MBB, MI, DL, get(AArch64::MOVKXi), Reg)
        .addReg(Reg, RegState::Kill)
        .addGlobalAddress(GV, 0, AArch64II::MO_G3)
        .addImm(48);
    BuildMI(MBB, MI, DL, get(AArch64::LDRXui), Reg)
        .addReg(Reg, RegState::Kill)
        .addImm(0)
        .addMemOperand(*MI.memoperands_begin());
  } else if (TM.getCodeModel() == CodeModel::Tiny) {
    BuildMI(MBB, MI, DL, get(AArch64::ADR), Reg)
        .addGlobalAddress(GV, 0, OpFlags);
  } else {
    BuildMI(MBB, MI, DL, get(AArch64::ADRP), Reg)
        .addGlobalAddress(GV, 0, OpFlags | AArch64II::MO_PAGE);
    unsigned char LoFlags = OpFlags | AArch64II::MO_PAGEOFF | MO_NC;
    if (Subtarget.isTargetILP32()) {
      unsigned Reg32 = TRI->getSubReg(Reg, AArch64::sub_32);
      BuildMI(MBB, MI, DL, get(AArch64::LDRWui))
          .addDef(Reg32, RegState::Dead)
          .addUse(Reg, RegState::Kill)
          .addGlobalAddress(GV, 0, LoFlags)
          .addMemOperand(*MI.memoperands_begin())
          .addDef(Reg, RegState::Implicit);
    } else {
      BuildMI(MBB, MI, DL, get(AArch64::LDRXui), Reg)
          .addReg(Reg, RegState::Kill)
          .addGlobalAddress(GV, 0, LoFlags)
          .addMemOperand(*MI.memoperands_begin());
    }
  }
 
  MBB.erase(MI);
 
  return true;
}
 
// Return true if this instruction simply sets its single destination register
// to zero. This is equivalent to a register rename of the zero-register.
bool AArch64InstrInfo::isGPRZero(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    break;
  case AArch64::MOVZWi:
  case AArch64::MOVZXi: // movz Rd, #0 (LSL #0)
    if (MI.getOperand(1).isImm() && MI.getOperand(1).getImm() == 0) {
      assert(MI.getDesc().getNumOperands() == 3 &&
             MI.getOperand(2).getImm() == 0 && "invalid MOVZi operands");
      return true;
    }
    break;
  case AArch64::ANDWri: // and Rd, Rzr, #imm
    return MI.getOperand(1).getReg() == AArch64::WZR;
  case AArch64::ANDXri:
    return MI.getOperand(1).getReg() == AArch64::XZR;
  case TargetOpcode::COPY:
    return MI.getOperand(1).getReg() == AArch64::WZR;
  }
  return false;
}
 
// Return true if this instruction simply renames a general register without
// modifying bits.
bool AArch64InstrInfo::isGPRCopy(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    break;
  case TargetOpcode::COPY: {
    // GPR32 copies will by lowered to ORRXrs
    Register DstReg = MI.getOperand(0).getReg();
    return (AArch64::GPR32RegClass.contains(DstReg) ||
            AArch64::GPR64RegClass.contains(DstReg));
  }
  case AArch64::ORRXrs: // orr Xd, Xzr, Xm (LSL #0)
    if (MI.getOperand(1).getReg() == AArch64::XZR) {
      assert(MI.getDesc().getNumOperands() == 4 &&
             MI.getOperand(3).getImm() == 0 && "invalid ORRrs operands");
      return true;
    }
    break;
  case AArch64::ADDXri: // add Xd, Xn, #0 (LSL #0)
    if (MI.getOperand(2).getImm() == 0) {
      assert(MI.getDesc().getNumOperands() == 4 &&
             MI.getOperand(3).getImm() == 0 && "invalid ADDXri operands");
      return true;
    }
    break;
  }
  return false;
}
 
// Return true if this instruction simply renames a general register without
// modifying bits.
bool AArch64InstrInfo::isFPRCopy(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    break;
  case TargetOpcode::COPY: {
    Register DstReg = MI.getOperand(0).getReg();
    return AArch64::FPR128RegClass.contains(DstReg);
  }
  case AArch64::ORRv16i8:
    if (MI.getOperand(1).getReg() == MI.getOperand(2).getReg()) {
      assert(MI.getDesc().getNumOperands() == 3 && MI.getOperand(0).isReg() &&
             "invalid ORRv16i8 operands");
      return true;
    }
    break;
  }
  return false;
}
 
unsigned AArch64InstrInfo::isLoadFromStackSlot(const MachineInstr &MI,
                                               int &FrameIndex) const {
  switch (MI.getOpcode()) {
  default:
    break;
  case AArch64::LDRWui:
  case AArch64::LDRXui:
  case AArch64::LDRBui:
  case AArch64::LDRHui:
  case AArch64::LDRSui:
  case AArch64::LDRDui:
  case AArch64::LDRQui:
  case AArch64::LDR_PXI:
    if (MI.getOperand(0).getSubReg() == 0 && MI.getOperand(1).isFI() &&
        MI.getOperand(2).isImm() && MI.getOperand(2).getImm() == 0) {
      FrameIndex = MI.getOperand(1).getIndex();
      return MI.getOperand(0).getReg();
    }
    break;
  }
 
  return 0;
}
 
unsigned AArch64InstrInfo::isStoreToStackSlot(const MachineInstr &MI,
                                              int &FrameIndex) const {
  switch (MI.getOpcode()) {
  default:
    break;
  case AArch64::STRWui:
  case AArch64::STRXui:
  case AArch64::STRBui:
  case AArch64::STRHui:
  case AArch64::STRSui:
  case AArch64::STRDui:
  case AArch64::STRQui:
  case AArch64::STR_PXI:
    if (MI.getOperand(0).getSubReg() == 0 && MI.getOperand(1).isFI() &&
        MI.getOperand(2).isImm() && MI.getOperand(2).getImm() == 0) {
      FrameIndex = MI.getOperand(1).getIndex();
      return MI.getOperand(0).getReg();
    }
    break;
  }
  return 0;
}
 
/// Check all MachineMemOperands for a hint to suppress pairing.
bool AArch64InstrInfo::isLdStPairSuppressed(const MachineInstr &MI) {
  return llvm::any_of(MI.memoperands(), [](MachineMemOperand *MMO) {
    return MMO->getFlags() & MOSuppressPair;
  });
}
 
/// Set a flag on the first MachineMemOperand to suppress pairing.
void AArch64InstrInfo::suppressLdStPair(MachineInstr &MI) {
  if (MI.memoperands_empty())
    return;
  (*MI.memoperands_begin())->setFlags(MOSuppressPair);
}
 
/// Check all MachineMemOperands for a hint that the load/store is strided.
bool AArch64InstrInfo::isStridedAccess(const MachineInstr &MI) {
  return llvm::any_of(MI.memoperands(), [](MachineMemOperand *MMO) {
    return MMO->getFlags() & MOStridedAccess;
  });
}
 
bool AArch64InstrInfo::hasUnscaledLdStOffset(unsigned Opc) {
  switch (Opc) {
  default:
    return false;
  case AArch64::STURSi:
  case AArch64::STRSpre:
  case AArch64::STURDi:
  case AArch64::STRDpre:
  case AArch64::STURQi:
  case AArch64::STRQpre:
  case AArch64::STURBBi:
  case AArch64::STURHHi:
  case AArch64::STURWi:
  case AArch64::STRWpre:
  case AArch64::STURXi:
  case AArch64::STRXpre:
  case AArch64::LDURSi:
  case AArch64::LDRSpre:
  case AArch64::LDURDi:
  case AArch64::LDRDpre:
  case AArch64::LDURQi:
  case AArch64::LDRQpre:
  case AArch64::LDURWi:
  case AArch64::LDRWpre:
  case AArch64::LDURXi:
  case AArch64::LDRXpre:
  case AArch64::LDRSWpre:
  case AArch64::LDURSWi:
  case AArch64::LDURHHi:
  case AArch64::LDURBBi:
  case AArch64::LDURSBWi:
  case AArch64::LDURSHWi:
    return true;
  }
}
 
std::optional<unsigned> AArch64InstrInfo::getUnscaledLdSt(unsigned Opc) {
  switch (Opc) {
  default: return {};
  case AArch64::PRFMui: return AArch64::PRFUMi;
  case AArch64::LDRXui: return AArch64::LDURXi;
  case AArch64::LDRWui: return AArch64::LDURWi;
  case AArch64::LDRBui: return AArch64::LDURBi;
  case AArch64::LDRHui: return AArch64::LDURHi;
  case AArch64::LDRSui: return AArch64::LDURSi;
  case AArch64::LDRDui: return AArch64::LDURDi;
  case AArch64::LDRQui: return AArch64::LDURQi;
  case AArch64::LDRBBui: return AArch64::LDURBBi;
  case AArch64::LDRHHui: return AArch64::LDURHHi;
  case AArch64::LDRSBXui: return AArch64::LDURSBXi;
  case AArch64::LDRSBWui: return AArch64::LDURSBWi;
  case AArch64::LDRSHXui: return AArch64::LDURSHXi;
  case AArch64::LDRSHWui: return AArch64::LDURSHWi;
  case AArch64::LDRSWui: return AArch64::LDURSWi;
  case AArch64::STRXui: return AArch64::STURXi;
  case AArch64::STRWui: return AArch64::STURWi;
  case AArch64::STRBui: return AArch64::STURBi;
  case AArch64::STRHui: return AArch64::STURHi;
  case AArch64::STRSui: return AArch64::STURSi;
  case AArch64::STRDui: return AArch64::STURDi;
  case AArch64::STRQui: return AArch64::STURQi;
  case AArch64::STRBBui: return AArch64::STURBBi;
  case AArch64::STRHHui: return AArch64::STURHHi;
  }
}
 
unsigned AArch64InstrInfo::getLoadStoreImmIdx(unsigned Opc) {
  switch (Opc) {
  default:
    return 2;
  case AArch64::LDPXi:
  case AArch64::LDPDi:
  case AArch64::STPXi:
  case AArch64::STPDi:
  case AArch64::LDNPXi:
  case AArch64::LDNPDi:
  case AArch64::STNPXi:
  case AArch64::STNPDi:
  case AArch64::LDPQi:
  case AArch64::STPQi:
  case AArch64::LDNPQi:
  case AArch64::STNPQi:
  case AArch64::LDPWi:
  case AArch64::LDPSi:
  case AArch64::STPWi:
  case AArch64::STPSi:
  case AArch64::LDNPWi:
  case AArch64::LDNPSi:
  case AArch64::STNPWi:
  case AArch64::STNPSi:
  case AArch64::LDG:
  case AArch64::STGPi:
 
  case AArch64::LD1B_IMM:
  case AArch64::LD1B_H_IMM:
  case AArch64::LD1B_S_IMM:
  case AArch64::LD1B_D_IMM:
  case AArch64::LD1SB_H_IMM:
  case AArch64::LD1SB_S_IMM:
  case AArch64::LD1SB_D_IMM:
  case AArch64::LD1H_IMM:
  case AArch64::LD1H_S_IMM:
  case AArch64::LD1H_D_IMM:
  case AArch64::LD1SH_S_IMM:
  case AArch64::LD1SH_D_IMM:
  case AArch64::LD1W_IMM:
  case AArch64::LD1W_D_IMM:
  case AArch64::LD1SW_D_IMM:
  case AArch64::LD1D_IMM:
 
  case AArch64::LD2B_IMM:
  case AArch64::LD2H_IMM:
  case AArch64::LD2W_IMM:
  case AArch64::LD2D_IMM:
  case AArch64::LD3B_IMM:
  case AArch64::LD3H_IMM:
  case AArch64::LD3W_IMM:
  case AArch64::LD3D_IMM:
  case AArch64::LD4B_IMM:
  case AArch64::LD4H_IMM:
  case AArch64::LD4W_IMM:
  case AArch64::LD4D_IMM:
 
  case AArch64::ST1B_IMM:
  case AArch64::ST1B_H_IMM:
  case AArch64::ST1B_S_IMM:
  case AArch64::ST1B_D_IMM:
  case AArch64::ST1H_IMM:
  case AArch64::ST1H_S_IMM:
  case AArch64::ST1H_D_IMM:
  case AArch64::ST1W_IMM:
  case AArch64::ST1W_D_IMM:
  case AArch64::ST1D_IMM:
 
  case AArch64::ST2B_IMM:
  case AArch64::ST2H_IMM:
  case AArch64::ST2W_IMM:
  case AArch64::ST2D_IMM:
  case AArch64::ST3B_IMM:
  case AArch64::ST3H_IMM:
  case AArch64::ST3W_IMM:
  case AArch64::ST3D_IMM:
  case AArch64::ST4B_IMM:
  case AArch64::ST4H_IMM:
  case AArch64::ST4W_IMM:
  case AArch64::ST4D_IMM:
 
  case AArch64::LD1RB_IMM:
  case AArch64::LD1RB_H_IMM:
  case AArch64::LD1RB_S_IMM:
  case AArch64::LD1RB_D_IMM:
  case AArch64::LD1RSB_H_IMM:
  case AArch64::LD1RSB_S_IMM:
  case AArch64::LD1RSB_D_IMM:
  case AArch64::LD1RH_IMM:
  case AArch64::LD1RH_S_IMM:
  case AArch64::LD1RH_D_IMM:
  case AArch64::LD1RSH_S_IMM:
  case AArch64::LD1RSH_D_IMM:
  case AArch64::LD1RW_IMM:
  case AArch64::LD1RW_D_IMM:
  case AArch64::LD1RSW_IMM:
  case AArch64::LD1RD_IMM:
 
  case AArch64::LDNT1B_ZRI:
  case AArch64::LDNT1H_ZRI:
  case AArch64::LDNT1W_ZRI:
  case AArch64::LDNT1D_ZRI:
  case AArch64::STNT1B_ZRI:
  case AArch64::STNT1H_ZRI:
  case AArch64::STNT1W_ZRI:
  case AArch64::STNT1D_ZRI:
 
  case AArch64::LDNF1B_IMM:
  case AArch64::LDNF1B_H_IMM:
  case AArch64::LDNF1B_S_IMM:
  case AArch64::LDNF1B_D_IMM:
  case AArch64::LDNF1SB_H_IMM:
  case AArch64::LDNF1SB_S_IMM:
  case AArch64::LDNF1SB_D_IMM:
  case AArch64::LDNF1H_IMM:
  case AArch64::LDNF1H_S_IMM:
  case AArch64::LDNF1H_D_IMM:
  case AArch64::LDNF1SH_S_IMM:
  case AArch64::LDNF1SH_D_IMM:
  case AArch64::LDNF1W_IMM:
  case AArch64::LDNF1W_D_IMM:
  case AArch64::LDNF1SW_D_IMM:
  case AArch64::LDNF1D_IMM:
    return 3;
  case AArch64::ADDG:
  case AArch64::STGi:
  case AArch64::LDR_PXI:
  case AArch64::STR_PXI:
    return 2;
  }
}
 
bool AArch64InstrInfo::isPairableLdStInst(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    return false;
  // Scaled instructions.
  case AArch64::STRSui:
  case AArch64::STRDui:
  case AArch64::STRQui:
  case AArch64::STRXui:
  case AArch64::STRWui:
  case AArch64::LDRSui:
  case AArch64::LDRDui:
  case AArch64::LDRQui:
  case AArch64::LDRXui:
  case AArch64::LDRWui:
  case AArch64::LDRSWui:
  // Unscaled instructions.
  case AArch64::STURSi:
  case AArch64::STRSpre:
  case AArch64::STURDi:
  case AArch64::STRDpre:
  case AArch64::STURQi:
  case AArch64::STRQpre:
  case AArch64::STURWi:
  case AArch64::STRWpre:
  case AArch64::STURXi:
  case AArch64::STRXpre:
  case AArch64::LDURSi:
  case AArch64::LDRSpre:
  case AArch64::LDURDi:
  case AArch64::LDRDpre:
  case AArch64::LDURQi:
  case AArch64::LDRQpre:
  case AArch64::LDURWi:
  case AArch64::LDRWpre:
  case AArch64::LDURXi:
  case AArch64::LDRXpre:
  case AArch64::LDURSWi:
  case AArch64::LDRSWpre:
    return true;
  }
}
 
unsigned AArch64InstrInfo::convertToFlagSettingOpc(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has no flag setting equivalent!");
  // 32-bit cases:
  case AArch64::ADDWri:
    return AArch64::ADDSWri;
  case AArch64::ADDWrr:
    return AArch64::ADDSWrr;
  case AArch64::ADDWrs:
    return AArch64::ADDSWrs;
  case AArch64::ADDWrx:
    return AArch64::ADDSWrx;
  case AArch64::ANDWri:
    return AArch64::ANDSWri;
  case AArch64::ANDWrr:
    return AArch64::ANDSWrr;
  case AArch64::ANDWrs:
    return AArch64::ANDSWrs;
  case AArch64::BICWrr:
    return AArch64::BICSWrr;
  case AArch64::BICWrs:
    return AArch64::BICSWrs;
  case AArch64::SUBWri:
    return AArch64::SUBSWri;
  case AArch64::SUBWrr:
    return AArch64::SUBSWrr;
  case AArch64::SUBWrs:
    return AArch64::SUBSWrs;
  case AArch64::SUBWrx:
    return AArch64::SUBSWrx;
  // 64-bit cases:
  case AArch64::ADDXri:
    return AArch64::ADDSXri;
  case AArch64::ADDXrr:
    return AArch64::ADDSXrr;
  case AArch64::ADDXrs:
    return AArch64::ADDSXrs;
  case AArch64::ADDXrx:
    return AArch64::ADDSXrx;
  case AArch64::ANDXri:
    return AArch64::ANDSXri;
  case AArch64::ANDXrr:
    return AArch64::ANDSXrr;
  case AArch64::ANDXrs:
    return AArch64::ANDSXrs;
  case AArch64::BICXrr:
    return AArch64::BICSXrr;
  case AArch64::BICXrs:
    return AArch64::BICSXrs;
  case AArch64::SUBXri:
    return AArch64::SUBSXri;
  case AArch64::SUBXrr:
    return AArch64::SUBSXrr;
  case AArch64::SUBXrs:
    return AArch64::SUBSXrs;
  case AArch64::SUBXrx:
    return AArch64::SUBSXrx;
  // SVE instructions:
  case AArch64::AND_PPzPP:
    return AArch64::ANDS_PPzPP;
  case AArch64::BIC_PPzPP:
    return AArch64::BICS_PPzPP;
  case AArch64::EOR_PPzPP:
    return AArch64::EORS_PPzPP;
  case AArch64::NAND_PPzPP:
    return AArch64::NANDS_PPzPP;
  case AArch64::NOR_PPzPP:
    return AArch64::NORS_PPzPP;
  case AArch64::ORN_PPzPP:
    return AArch64::ORNS_PPzPP;
  case AArch64::ORR_PPzPP:
    return AArch64::ORRS_PPzPP;
  case AArch64::BRKA_PPzP:
    return AArch64::BRKAS_PPzP;
  case AArch64::BRKPA_PPzPP:
    return AArch64::BRKPAS_PPzPP;
  case AArch64::BRKB_PPzP:
    return AArch64::BRKBS_PPzP;
  case AArch64::BRKPB_PPzPP:
    return AArch64::BRKPBS_PPzPP;
  case AArch64::BRKN_PPzP:
    return AArch64::BRKNS_PPzP;
  case AArch64::RDFFR_PPz:
    return AArch64::RDFFRS_PPz;
  case AArch64::PTRUE_B:
    return AArch64::PTRUES_B;
  }
}
 
// Is this a candidate for ld/st merging or pairing?  For example, we don't
// touch volatiles or load/stores that have a hint to avoid pair formation.
bool AArch64InstrInfo::isCandidateToMergeOrPair(const MachineInstr &MI) const {
 
  bool IsPreLdSt = isPreLdSt(MI);
 
  // If this is a volatile load/store, don't mess with it.
  if (MI.hasOrderedMemoryRef())
    return false;
 
  // Make sure this is a reg/fi+imm (as opposed to an address reloc).
  // For Pre-inc LD/ST, the operand is shifted by one.
  assert((MI.getOperand(IsPreLdSt ? 2 : 1).isReg() ||
          MI.getOperand(IsPreLdSt ? 2 : 1).isFI()) &&
         "Expected a reg or frame index operand.");
 
  // For Pre-indexed addressing quadword instructions, the third operand is the
  // immediate value.
  bool IsImmPreLdSt = IsPreLdSt && MI.getOperand(3).isImm();
 
  if (!MI.getOperand(2).isImm() && !IsImmPreLdSt)
    return false;
 
  // Can't merge/pair if the instruction modifies the base register.
  // e.g., ldr x0, [x0]
  // This case will never occur with an FI base.
  // However, if the instruction is an LDR<S,D,Q,W,X,SW>pre or
  // STR<S,D,Q,W,X>pre, it can be merged.
  // For example:
  //   ldr q0, [x11, #32]!
  //   ldr q1, [x11, #16]
  //   to
  //   ldp q0, q1, [x11, #32]!
  if (MI.getOperand(1).isReg() && !IsPreLdSt) {
    Register BaseReg = MI.getOperand(1).getReg();
    const TargetRegisterInfo *TRI = &getRegisterInfo();
    if (MI.modifiesRegister(BaseReg, TRI))
      return false;
  }
 
  // Check if this load/store has a hint to avoid pair formation.
  // MachineMemOperands hints are set by the AArch64StorePairSuppress pass.
  if (isLdStPairSuppressed(MI))
    return false;
 
  // Do not pair any callee-save store/reload instructions in the
  // prologue/epilogue if the CFI information encoded the operations as separate
  // instructions, as that will cause the size of the actual prologue to mismatch
  // with the prologue size recorded in the Windows CFI.
  const MCAsmInfo *MAI = MI.getMF()->getTarget().getMCAsmInfo();
  bool NeedsWinCFI = MAI->usesWindowsCFI() &&
                     MI.getMF()->getFunction().needsUnwindTableEntry();
  if (NeedsWinCFI && (MI.getFlag(MachineInstr::FrameSetup) ||
                      MI.getFlag(MachineInstr::FrameDestroy)))
    return false;
 
  // On some CPUs quad load/store pairs are slower than two single load/stores.
  if (Subtarget.isPaired128Slow()) {
    switch (MI.getOpcode()) {
    default:
      break;
    case AArch64::LDURQi:
    case AArch64::STURQi:
    case AArch64::LDRQui:
    case AArch64::STRQui:
      return false;
    }
  }
 
  return true;
}
 
bool AArch64InstrInfo::getMemOperandsWithOffsetWidth(
    const MachineInstr &LdSt, SmallVectorImpl<const MachineOperand *> &BaseOps,
    int64_t &Offset, bool &OffsetIsScalable, unsigned &Width,
    const TargetRegisterInfo *TRI) const {
  if (!LdSt.mayLoadOrStore())
    return false;
 
  const MachineOperand *BaseOp;
  if (!getMemOperandWithOffsetWidth(LdSt, BaseOp, Offset, OffsetIsScalable,
                                    Width, TRI))
    return false;
  BaseOps.push_back(BaseOp);
  return true;
}
 
std::optional<ExtAddrMode>
AArch64InstrInfo::getAddrModeFromMemoryOp(const MachineInstr &MemI,
                                          const TargetRegisterInfo *TRI) const {
  const MachineOperand *Base; // Filled with the base operand of MI.
  int64_t Offset;             // Filled with the offset of MI.
  bool OffsetIsScalable;
  if (!getMemOperandWithOffset(MemI, Base, Offset, OffsetIsScalable, TRI))
    return std::nullopt;
 
  if (!Base->isReg())
    return std::nullopt;
  ExtAddrMode AM;
  AM.BaseReg = Base->getReg();
  AM.Displacement = Offset;
  AM.ScaledReg = 0;
  AM.Scale = 0;
  return AM;
}
 
bool AArch64InstrInfo::canFoldIntoAddrMode(const MachineInstr &MemI,
                                           Register Reg,
                                           const MachineInstr &AddrI,
                                           ExtAddrMode &AM) const {
  // Filter out instructions into which we cannot fold.
  unsigned NumBytes;
  int64_t OffsetScale = 1;
  switch (MemI.getOpcode()) {
  default:
    return false;
 
  case AArch64::LDURQi:
  case AArch64::STURQi:
    NumBytes = 16;
    break;
 
  case AArch64::LDURDi:
  case AArch64::STURDi:
  case AArch64::LDURXi:
  case AArch64::STURXi:
    NumBytes = 8;
    break;
 
  case AArch64::LDURWi:
  case AArch64::LDURSWi:
  case AArch64::STURWi:
    NumBytes = 4;
    break;
 
  case AArch64::LDURHi:
  case AArch64::STURHi:
  case AArch64::LDURHHi:
  case AArch64::STURHHi:
  case AArch64::LDURSHXi:
  case AArch64::LDURSHWi:
    NumBytes = 2;
    break;
 
  case AArch64::LDRBroX:
  case AArch64::LDRBBroX:
  case AArch64::LDRSBXroX:
  case AArch64::LDRSBWroX:
  case AArch64::STRBroX:
  case AArch64::STRBBroX:
  case AArch64::LDURBi:
  case AArch64::LDURBBi:
  case AArch64::LDURSBXi:
  case AArch64::LDURSBWi:
  case AArch64::STURBi:
  case AArch64::STURBBi:
  case AArch64::LDRBui:
  case AArch64::LDRBBui:
  case AArch64::LDRSBXui:
  case AArch64::LDRSBWui:
  case AArch64::STRBui:
  case AArch64::STRBBui:
    NumBytes = 1;
    break;
 
  case AArch64::LDRQroX:
  case AArch64::STRQroX:
  case AArch64::LDRQui:
  case AArch64::STRQui:
    NumBytes = 16;
    OffsetScale = 16;
    break;
 
  case AArch64::LDRDroX:
  case AArch64::STRDroX:
  case AArch64::LDRXroX:
  case AArch64::STRXroX:
  case AArch64::LDRDui:
  case AArch64::STRDui:
  case AArch64::LDRXui:
  case AArch64::STRXui:
    NumBytes = 8;
    OffsetScale = 8;
    break;
 
  case AArch64::LDRWroX:
  case AArch64::LDRSWroX:
  case AArch64::STRWroX:
  case AArch64::LDRWui:
  case AArch64::LDRSWui:
  case AArch64::STRWui:
    NumBytes = 4;
    OffsetScale = 4;
    break;
 
  case AArch64::LDRHroX:
  case AArch64::STRHroX:
  case AArch64::LDRHHroX:
  case AArch64::STRHHroX:
  case AArch64::LDRSHXroX:
  case AArch64::LDRSHWroX:
  case AArch64::LDRHui:
  case AArch64::STRHui:
  case AArch64::LDRHHui:
  case AArch64::STRHHui:
  case AArch64::LDRSHXui:
  case AArch64::LDRSHWui:
    NumBytes = 2;
    OffsetScale = 2;
    break;
  }
 
  // Check the fold operand is not the loaded/stored value.
  const MachineOperand &BaseRegOp = MemI.getOperand(0);
  if (BaseRegOp.isReg() && BaseRegOp.getReg() == Reg)
    return false;
 
  // Handle memory instructions with a [Reg, Reg] addressing mode.
  if (MemI.getOperand(2).isReg()) {
    // Bail if the addressing mode already includes extension of the offset
    // register.
    if (MemI.getOperand(3).getImm())
      return false;
 
    // Check if we actually have a scaled offset.
    if (MemI.getOperand(4).getImm() == 0)
      OffsetScale = 1;
 
    // If the address instructions is folded into the base register, then the
    // addressing mode must not have a scale. Then we can swap the base and the
    // scaled registers.
    if (MemI.getOperand(1).getReg() == Reg && OffsetScale != 1)
      return false;
 
    switch (AddrI.getOpcode()) {
    default:
      return false;
 
    case AArch64::SBFMXri:
      // sxtw Xa, Wm
      // ldr Xd, [Xn, Xa, lsl #N]
      // ->
      // ldr Xd, [Xn, Wm, sxtw #N]
      if (AddrI.getOperand(2).getImm() != 0 ||
          AddrI.getOperand(3).getImm() != 31)
        return false;
 
      AM.BaseReg = MemI.getOperand(1).getReg();
      if (AM.BaseReg == Reg)
        AM.BaseReg = MemI.getOperand(2).getReg();
      AM.ScaledReg = AddrI.getOperand(1).getReg();
      AM.Scale = OffsetScale;
      AM.Displacement = 0;
      AM.Form = ExtAddrMode::Formula::SExtScaledReg;
      return true;
 
    case TargetOpcode::SUBREG_TO_REG: {
      // mov Wa, Wm
      // ldr Xd, [Xn, Xa, lsl #N]
      // ->
      // ldr Xd, [Xn, Wm, uxtw #N]
 
      // Zero-extension looks like an ORRWrs followed by a SUBREG_TO_REG.
      if (AddrI.getOperand(1).getImm() != 0 ||
          AddrI.getOperand(3).getImm() != AArch64::sub_32)
        return false;
 
      const MachineRegisterInfo &MRI = AddrI.getMF()->getRegInfo();
      Register OffsetReg = AddrI.getOperand(2).getReg();
      if (!OffsetReg.isVirtual() || !MRI.hasOneNonDBGUse(OffsetReg))
        return false;
 
      const MachineInstr &DefMI = *MRI.getVRegDef(OffsetReg);
      if (DefMI.getOpcode() != AArch64::ORRWrs ||
          DefMI.getOperand(1).getReg() != AArch64::WZR ||
          DefMI.getOperand(3).getImm() != 0)
        return false;
 
      AM.BaseReg = MemI.getOperand(1).getReg();
      if (AM.BaseReg == Reg)
        AM.BaseReg = MemI.getOperand(2).getReg();
      AM.ScaledReg = DefMI.getOperand(2).getReg();
      AM.Scale = OffsetScale;
      AM.Displacement = 0;
      AM.Form = ExtAddrMode::Formula::ZExtScaledReg;
      return true;
    }
    }
  }
 
  // Handle memory instructions with a [Reg, #Imm] addressing mode.
 
  // Check we are not breaking a potential conversion to an LDP.
  auto validateOffsetForLDP = [](unsigned NumBytes, int64_t OldOffset,
                                 int64_t NewOffset) -> bool {
    int64_t MinOffset, MaxOffset;
    switch (NumBytes) {
    default:
      return true;
    case 4:
      MinOffset = -256;
      MaxOffset = 252;
      break;
    case 8:
      MinOffset = -512;
      MaxOffset = 504;
      break;
    case 16:
      MinOffset = -1024;
      MaxOffset = 1008;
      break;
    }
    return OldOffset < MinOffset || OldOffset > MaxOffset ||
           (NewOffset >= MinOffset && NewOffset <= MaxOffset);
  };
  auto canFoldAddSubImmIntoAddrMode = [&](int64_t Disp) -> bool {
    int64_t OldOffset = MemI.getOperand(2).getImm() * OffsetScale;
    int64_t NewOffset = OldOffset + Disp;
    if (!isLegalAddressingMode(NumBytes, NewOffset, /* Scale */ 0))
      return false;
    // If the old offset would fit into an LDP, but the new offset wouldn't,
    // bail out.
    if (!validateOffsetForLDP(NumBytes, OldOffset, NewOffset))
      return false;
    AM.BaseReg = AddrI.getOperand(1).getReg();
    AM.ScaledReg = 0;
    AM.Scale = 0;
    AM.Displacement = NewOffset;
    AM.Form = ExtAddrMode::Formula::Basic;
    return true;
  };
 
  auto canFoldAddRegIntoAddrMode =
      [&](int64_t Scale,
          ExtAddrMode::Formula Form = ExtAddrMode::Formula::Basic) -> bool {
    if (MemI.getOperand(2).getImm() != 0)
      return false;
    if (!isLegalAddressingMode(NumBytes, /* Offset */ 0, Scale))
      return false;
    AM.BaseReg = AddrI.getOperand(1).getReg();
    AM.ScaledReg = AddrI.getOperand(2).getReg();
    AM.Scale = Scale;
    AM.Displacement = 0;
    AM.Form = Form;
    return true;
  };
 
  auto avoidSlowSTRQ = [&](const MachineInstr &MemI) {
    unsigned Opcode = MemI.getOpcode();
    return (Opcode == AArch64::STURQi || Opcode == AArch64::STRQui) &&
           Subtarget.isSTRQroSlow();
  };
 
  int64_t Disp = 0;
  const bool OptSize = MemI.getMF()->getFunction().hasOptSize();
  switch (AddrI.getOpcode()) {
  default:
    return false;
 
  case AArch64::ADDXri:
    // add Xa, Xn, #N
    // ldr Xd, [Xa, #M]
    // ->
    // ldr Xd, [Xn, #N'+M]
    Disp = AddrI.getOperand(2).getImm() << AddrI.getOperand(3).getImm();
    return canFoldAddSubImmIntoAddrMode(Disp);
 
  case AArch64::SUBXri:
    // sub Xa, Xn, #N
    // ldr Xd, [Xa, #M]
    // ->
    // ldr Xd, [Xn, #N'+M]
    Disp = AddrI.getOperand(2).getImm() << AddrI.getOperand(3).getImm();
    return canFoldAddSubImmIntoAddrMode(-Disp);
 
  case AArch64::ADDXrs: {
    // add Xa, Xn, Xm, lsl #N
    // ldr Xd, [Xa]
    // ->
    // ldr Xd, [Xn, Xm, lsl #N]
 
    // Don't fold the add if the result would be slower, unless optimising for
    // size.
    int64_t Shift = AddrI.getOperand(3).getImm();
    if (!OptSize) {
      if ((Shift != 2 && Shift != 3) || !Subtarget.hasAddrLSLFast())
        return false;
      if (avoidSlowSTRQ(MemI))
        return false;
    }
    return canFoldAddRegIntoAddrMode(1ULL << Shift);
  }
 
  case AArch64::ADDXrr:
    // add Xa, Xn, Xm
    // ldr Xd, [Xa]
    // ->
    // ldr Xd, [Xn, Xm, lsl #0]
 
    // Don't fold the add if the result would be slower, unless optimising for
    // size.
    if (!OptSize && avoidSlowSTRQ(MemI))
      return false;
    return canFoldAddRegIntoAddrMode(1);
 
  case AArch64::ADDXrx:
    // add Xa, Xn, Wm, {s,u}xtw #N
    // ldr Xd, [Xa]
    // ->
    // ldr Xd, [Xn, Wm, {s,u}xtw #N]
 
    // Don't fold the add if the result would be slower, unless optimising for
    // size.
    if (!OptSize && avoidSlowSTRQ(MemI))
      return false;
 
    // Can fold only sign-/zero-extend of a word.
    unsigned Imm = static_cast<unsigned>(AddrI.getOperand(3).getImm());
    AArch64_AM::ShiftExtendType Extend = AArch64_AM::getArithExtendType(Imm);
    if (Extend != AArch64_AM::UXTW && Extend != AArch64_AM::SXTW)
      return false;
 
    return canFoldAddRegIntoAddrMode(
        1ULL << AArch64_AM::getArithShiftValue(Imm),
        (Extend == AArch64_AM::SXTW) ? ExtAddrMode::Formula::SExtScaledReg
                                     : ExtAddrMode::Formula::ZExtScaledReg);
  }
}
 
// Given an opcode for an instruction with a [Reg, #Imm] addressing mode,
// return the opcode of an instruction performing the same operation, but using
// the [Reg, Reg] addressing mode.
static unsigned regOffsetOpcode(unsigned Opcode) {
  switch (Opcode) {
  default:
    llvm_unreachable("Address folding not implemented for instruction");
 
  case AArch64::LDURQi:
  case AArch64::LDRQui:
    return AArch64::LDRQroX;
  case AArch64::STURQi:
  case AArch64::STRQui:
    return AArch64::STRQroX;
  case AArch64::LDURDi:
  case AArch64::LDRDui:
    return AArch64::LDRDroX;
  case AArch64::STURDi:
  case AArch64::STRDui:
    return AArch64::STRDroX;
  case AArch64::LDURXi:
  case AArch64::LDRXui:
    return AArch64::LDRXroX;
  case AArch64::STURXi:
  case AArch64::STRXui:
    return AArch64::STRXroX;
  case AArch64::LDURWi:
  case AArch64::LDRWui:
    return AArch64::LDRWroX;
  case AArch64::LDURSWi:
  case AArch64::LDRSWui:
    return AArch64::LDRSWroX;
  case AArch64::STURWi:
  case AArch64::STRWui:
    return AArch64::STRWroX;
  case AArch64::LDURHi:
  case AArch64::LDRHui:
    return AArch64::LDRHroX;
  case AArch64::STURHi:
  case AArch64::STRHui:
    return AArch64::STRHroX;
  case AArch64::LDURHHi:
  case AArch64::LDRHHui:
    return AArch64::LDRHHroX;
  case AArch64::STURHHi:
  case AArch64::STRHHui:
    return AArch64::STRHHroX;
  case AArch64::LDURSHXi:
  case AArch64::LDRSHXui:
    return AArch64::LDRSHXroX;
  case AArch64::LDURSHWi:
  case AArch64::LDRSHWui:
    return AArch64::LDRSHWroX;
  case AArch64::LDURBi:
  case AArch64::LDRBui:
    return AArch64::LDRBroX;
  case AArch64::LDURBBi:
  case AArch64::LDRBBui:
    return AArch64::LDRBBroX;
  case AArch64::LDURSBXi:
  case AArch64::LDRSBXui:
    return AArch64::LDRSBXroX;
  case AArch64::LDURSBWi:
  case AArch64::LDRSBWui:
    return AArch64::LDRSBWroX;
  case AArch64::STURBi:
  case AArch64::STRBui:
    return AArch64::STRBroX;
  case AArch64::STURBBi:
  case AArch64::STRBBui:
    return AArch64::STRBBroX;
  }
}
 
// Given an opcode for an instruction with a [Reg, #Imm] addressing mode, return
// the opcode of an instruction performing the same operation, but using the
// [Reg, #Imm] addressing mode with scaled offset.
unsigned scaledOffsetOpcode(unsigned Opcode, unsigned &Scale) {
  switch (Opcode) {
  default:
    llvm_unreachable("Address folding not implemented for instruction");
 
  case AArch64::LDURQi:
    Scale = 16;
    return AArch64::LDRQui;
  case AArch64::STURQi:
    Scale = 16;
    return AArch64::STRQui;
  case AArch64::LDURDi:
    Scale = 8;
    return AArch64::LDRDui;
  case AArch64::STURDi:
    Scale = 8;
    return AArch64::STRDui;
  case AArch64::LDURXi:
    Scale = 8;
    return AArch64::LDRXui;
  case AArch64::STURXi:
    Scale = 8;
    return AArch64::STRXui;
  case AArch64::LDURWi:
    Scale = 4;
    return AArch64::LDRWui;
  case AArch64::LDURSWi:
    Scale = 4;
    return AArch64::LDRSWui;
  case AArch64::STURWi:
    Scale = 4;
    return AArch64::STRWui;
  case AArch64::LDURHi:
    Scale = 2;
    return AArch64::LDRHui;
  case AArch64::STURHi:
    Scale = 2;
    return AArch64::STRHui;
  case AArch64::LDURHHi:
    Scale = 2;
    return AArch64::LDRHHui;
  case AArch64::STURHHi:
    Scale = 2;
    return AArch64::STRHHui;
  case AArch64::LDURSHXi:
    Scale = 2;
    return AArch64::LDRSHXui;
  case AArch64::LDURSHWi:
    Scale = 2;
    return AArch64::LDRSHWui;
  case AArch64::LDURBi:
    Scale = 1;
    return AArch64::LDRBui;
  case AArch64::LDURBBi:
    Scale = 1;
    return AArch64::LDRBBui;
  case AArch64::LDURSBXi:
    Scale = 1;
    return AArch64::LDRSBXui;
  case AArch64::LDURSBWi:
    Scale = 1;
    return AArch64::LDRSBWui;
  case AArch64::STURBi:
    Scale = 1;
    return AArch64::STRBui;
  case AArch64::STURBBi:
    Scale = 1;
    return AArch64::STRBBui;
  case AArch64::LDRQui:
  case AArch64::STRQui:
    Scale = 16;
    return Opcode;
  case AArch64::LDRDui:
  case AArch64::STRDui:
  case AArch64::LDRXui:
  case AArch64::STRXui:
    Scale = 8;
    return Opcode;
  case AArch64::LDRWui:
  case AArch64::LDRSWui:
  case AArch64::STRWui:
    Scale = 4;
    return Opcode;
  case AArch64::LDRHui:
  case AArch64::STRHui:
  case AArch64::LDRHHui:
  case AArch64::STRHHui:
  case AArch64::LDRSHXui:
  case AArch64::LDRSHWui:
    Scale = 2;
    return Opcode;
  case AArch64::LDRBui:
  case AArch64::LDRBBui:
  case AArch64::LDRSBXui:
  case AArch64::LDRSBWui:
  case AArch64::STRBui:
  case AArch64::STRBBui:
    Scale = 1;
    return Opcode;
  }
}
 
// Given an opcode for an instruction with a [Reg, #Imm] addressing mode, return
// the opcode of an instruction performing the same operation, but using the
// [Reg, #Imm] addressing mode with unscaled offset.
unsigned unscaledOffsetOpcode(unsigned Opcode) {
  switch (Opcode) {
  default:
    llvm_unreachable("Address folding not implemented for instruction");
 
  case AArch64::LDURQi:
  case AArch64::STURQi:
  case AArch64::LDURDi:
  case AArch64::STURDi:
  case AArch64::LDURXi:
  case AArch64::STURXi:
  case AArch64::LDURWi:
  case AArch64::LDURSWi:
  case AArch64::STURWi:
  case AArch64::LDURHi:
  case AArch64::STURHi:
  case AArch64::LDURHHi:
  case AArch64::STURHHi:
  case AArch64::LDURSHXi:
  case AArch64::LDURSHWi:
  case AArch64::LDURBi:
  case AArch64::STURBi:
  case AArch64::LDURBBi:
  case AArch64::STURBBi:
  case AArch64::LDURSBWi:
  case AArch64::LDURSBXi:
    return Opcode;
  case AArch64::LDRQui:
    return AArch64::LDURQi;
  case AArch64::STRQui:
    return AArch64::STURQi;
  case AArch64::LDRDui:
    return AArch64::LDURDi;
  case AArch64::STRDui:
    return AArch64::STURDi;
  case AArch64::LDRXui:
    return AArch64::LDURXi;
  case AArch64::STRXui:
    return AArch64::STURXi;
  case AArch64::LDRWui:
    return AArch64::LDURWi;
  case AArch64::LDRSWui:
    return AArch64::LDURSWi;
  case AArch64::STRWui:
    return AArch64::STURWi;
  case AArch64::LDRHui:
    return AArch64::LDURHi;
  case AArch64::STRHui:
    return AArch64::STURHi;
  case AArch64::LDRHHui:
    return AArch64::LDURHHi;
  case AArch64::STRHHui:
    return AArch64::STURHHi;
  case AArch64::LDRSHXui:
    return AArch64::LDURSHXi;
  case AArch64::LDRSHWui:
    return AArch64::LDURSHWi;
  case AArch64::LDRBBui:
    return AArch64::LDURBBi;
  case AArch64::LDRBui:
    return AArch64::LDURBi;
  case AArch64::STRBBui:
    return AArch64::STURBBi;
  case AArch64::STRBui:
    return AArch64::STURBi;
  case AArch64::LDRSBWui:
    return AArch64::LDURSBWi;
  case AArch64::LDRSBXui:
    return AArch64::LDURSBXi;
  }
}
 
// Given the opcode of a memory load/store instruction, return the opcode of an
// instruction performing the same operation, but using
// the [Reg, Reg, {s,u}xtw #N] addressing mode with sign-/zero-extend of the
// offset register.
static unsigned offsetExtendOpcode(unsigned Opcode) {
  switch (Opcode) {
  default:
    llvm_unreachable("Address folding not implemented for instruction");
 
  case AArch64::LDRQroX:
  case AArch64::LDURQi:
  case AArch64::LDRQui:
    return AArch64::LDRQroW;
  case AArch64::STRQroX:
  case AArch64::STURQi:
  case AArch64::STRQui:
    return AArch64::STRQroW;
  case AArch64::LDRDroX:
  case AArch64::LDURDi:
  case AArch64::LDRDui:
    return AArch64::LDRDroW;
  case AArch64::STRDroX:
  case AArch64::STURDi:
  case AArch64::STRDui:
    return AArch64::STRDroW;
  case AArch64::LDRXroX:
  case AArch64::LDURXi:
  case AArch64::LDRXui:
    return AArch64::LDRXroW;
  case AArch64::STRXroX:
  case AArch64::STURXi:
  case AArch64::STRXui:
    return AArch64::STRXroW;
  case AArch64::LDRWroX:
  case AArch64::LDURWi:
  case AArch64::LDRWui:
    return AArch64::LDRWroW;
  case AArch64::LDRSWroX:
  case AArch64::LDURSWi:
  case AArch64::LDRSWui:
    return AArch64::LDRSWroW;
  case AArch64::STRWroX:
  case AArch64::STURWi:
  case AArch64::STRWui:
    return AArch64::STRWroW;
  case AArch64::LDRHroX:
  case AArch64::LDURHi:
  case AArch64::LDRHui:
    return AArch64::LDRHroW;
  case AArch64::STRHroX:
  case AArch64::STURHi:
  case AArch64::STRHui:
    return AArch64::STRHroW;
  case AArch64::LDRHHroX:
  case AArch64::LDURHHi:
  case AArch64::LDRHHui:
    return AArch64::LDRHHroW;
  case AArch64::STRHHroX:
  case AArch64::STURHHi:
  case AArch64::STRHHui:
    return AArch64::STRHHroW;
  case AArch64::LDRSHXroX:
  case AArch64::LDURSHXi:
  case AArch64::LDRSHXui:
    return AArch64::LDRSHXroW;
  case AArch64::LDRSHWroX:
  case AArch64::LDURSHWi:
  case AArch64::LDRSHWui:
    return AArch64::LDRSHWroW;
  case AArch64::LDRBroX:
  case AArch64::LDURBi:
  case AArch64::LDRBui:
    return AArch64::LDRBroW;
  case AArch64::LDRBBroX:
  case AArch64::LDURBBi:
  case AArch64::LDRBBui:
    return AArch64::LDRBBroW;
  case AArch64::LDRSBXroX:
  case AArch64::LDURSBXi:
  case AArch64::LDRSBXui:
    return AArch64::LDRSBXroW;
  case AArch64::LDRSBWroX:
  case AArch64::LDURSBWi:
  case AArch64::LDRSBWui:
    return AArch64::LDRSBWroW;
  case AArch64::STRBroX:
  case AArch64::STURBi:
  case AArch64::STRBui:
    return AArch64::STRBroW;
  case AArch64::STRBBroX:
  case AArch64::STURBBi:
  case AArch64::STRBBui:
    return AArch64::STRBBroW;
  }
}
 
MachineInstr *AArch64InstrInfo::emitLdStWithAddr(MachineInstr &MemI,
                                                 const ExtAddrMode &AM) const {
 
  const DebugLoc &DL = MemI.getDebugLoc();
  MachineBasicBlock &MBB = *MemI.getParent();
  MachineRegisterInfo &MRI = MemI.getMF()->getRegInfo();
 
  if (AM.Form == ExtAddrMode::Formula::Basic) {
    if (AM.ScaledReg) {
      // The new instruction will be in the form `ldr Rt, [Xn, Xm, lsl #imm]`.
      unsigned Opcode = regOffsetOpcode(MemI.getOpcode());
      MRI.constrainRegClass(AM.BaseReg, &AArch64::GPR64spRegClass);
      auto B = BuildMI(MBB, MemI, DL, get(Opcode))
                   .addReg(MemI.getOperand(0).getReg(),
                           MemI.mayLoad() ? RegState::Define : 0)
                   .addReg(AM.BaseReg)
                   .addReg(AM.ScaledReg)
                   .addImm(0)
                   .addImm(AM.Scale > 1)
                   .setMemRefs(MemI.memoperands())
                   .setMIFlags(MemI.getFlags());
      return B.getInstr();
    }
 
    assert(AM.ScaledReg == 0 && AM.Scale == 0 &&
           "Addressing mode not supported for folding");
 
    // The new instruction will be in the form `ld[u]r Rt, [Xn, #imm]`.
    unsigned Scale = 1;
    unsigned Opcode = MemI.getOpcode();
    if (isInt<9>(AM.Displacement))
      Opcode = unscaledOffsetOpcode(Opcode);
    else
      Opcode = scaledOffsetOpcode(Opcode, Scale);
 
    auto B = BuildMI(MBB, MemI, DL, get(Opcode))
                 .addReg(MemI.getOperand(0).getReg(),
                         MemI.mayLoad() ? RegState::Define : 0)
                 .addReg(AM.BaseReg)
                 .addImm(AM.Displacement / Scale)
                 .setMemRefs(MemI.memoperands())
                 .setMIFlags(MemI.getFlags());
    return B.getInstr();
  }
 
  if (AM.Form == ExtAddrMode::Formula::SExtScaledReg ||
      AM.Form == ExtAddrMode::Formula::ZExtScaledReg) {
    // The new instruction will be in the form `ldr Rt, [Xn, Wm, {s,u}xtw #N]`.
    assert(AM.ScaledReg && !AM.Displacement &&
           "Address offset can be a register or an immediate, but not both");
    unsigned Opcode = offsetExtendOpcode(MemI.getOpcode());
    MRI.constrainRegClass(AM.BaseReg, &AArch64::GPR64spRegClass);
    // Make sure the offset register is in the correct register class.
    Register OffsetReg = AM.ScaledReg;
    const TargetRegisterClass *RC = MRI.getRegClass(OffsetReg);
    if (RC->hasSuperClassEq(&AArch64::GPR64RegClass)) {
      OffsetReg = MRI.createVirtualRegister(&AArch64::GPR32RegClass);
      BuildMI(MBB, MemI, DL, get(TargetOpcode::COPY), OffsetReg)
          .addReg(AM.ScaledReg, 0, AArch64::sub_32);
    }
    auto B = BuildMI(MBB, MemI, DL, get(Opcode))
                 .addReg(MemI.getOperand(0).getReg(),
                         MemI.mayLoad() ? RegState::Define : 0)
                 .addReg(AM.BaseReg)
                 .addReg(OffsetReg)
                 .addImm(AM.Form == ExtAddrMode::Formula::SExtScaledReg)
                 .addImm(AM.Scale != 1)
                 .setMemRefs(MemI.memoperands())
                 .setMIFlags(MemI.getFlags());
 
    return B.getInstr();
  }
 
  llvm_unreachable(
      "Function must not be called with an addressing mode it can't handle");
}
 
bool AArch64InstrInfo::getMemOperandWithOffsetWidth(
    const MachineInstr &LdSt, const MachineOperand *&BaseOp, int64_t &Offset,
    bool &OffsetIsScalable, unsigned &Width,
    const TargetRegisterInfo *TRI) const {
  assert(LdSt.mayLoadOrStore() && "Expected a memory operation.");
  // Handle only loads/stores with base register followed by immediate offset.
  if (LdSt.getNumExplicitOperands() == 3) {
    // Non-paired instruction (e.g., ldr x1, [x0, #8]).
    if ((!LdSt.getOperand(1).isReg() && !LdSt.getOperand(1).isFI()) ||
        !LdSt.getOperand(2).isImm())
      return false;
  } else if (LdSt.getNumExplicitOperands() == 4) {
    // Paired instruction (e.g., ldp x1, x2, [x0, #8]).
    if (!LdSt.getOperand(1).isReg() ||
        (!LdSt.getOperand(2).isReg() && !LdSt.getOperand(2).isFI()) ||
        !LdSt.getOperand(3).isImm())
      return false;
  } else
    return false;
 
  // Get the scaling factor for the instruction and set the width for the
  // instruction.
  TypeSize Scale(0U, false);
  int64_t Dummy1, Dummy2;
 
  // If this returns false, then it's an instruction we don't want to handle.
  if (!getMemOpInfo(LdSt.getOpcode(), Scale, Width, Dummy1, Dummy2))
    return false;
 
  // Compute the offset. Offset is calculated as the immediate operand
  // multiplied by the scaling factor. Unscaled instructions have scaling factor
  // set to 1.
  if (LdSt.getNumExplicitOperands() == 3) {
    BaseOp = &LdSt.getOperand(1);
    Offset = LdSt.getOperand(2).getImm() * Scale.getKnownMinValue();
  } else {
    assert(LdSt.getNumExplicitOperands() == 4 && "invalid number of operands");
    BaseOp = &LdSt.getOperand(2);
    Offset = LdSt.getOperand(3).getImm() * Scale.getKnownMinValue();
  }
  OffsetIsScalable = Scale.isScalable();
 
  if (!BaseOp->isReg() && !BaseOp->isFI())
    return false;
 
  return true;
}
 
MachineOperand &
AArch64InstrInfo::getMemOpBaseRegImmOfsOffsetOperand(MachineInstr &LdSt) const {
  assert(LdSt.mayLoadOrStore() && "Expected a memory operation.");
  MachineOperand &OfsOp = LdSt.getOperand(LdSt.getNumExplicitOperands() - 1);
  assert(OfsOp.isImm() && "Offset operand wasn't immediate.");
  return OfsOp;
}
 
bool AArch64InstrInfo::getMemOpInfo(unsigned Opcode, TypeSize &Scale,
                                    unsigned &Width, int64_t &MinOffset,
                                    int64_t &MaxOffset) {
  const unsigned SVEMaxBytesPerVector = AArch64::SVEMaxBitsPerVector / 8;
  switch (Opcode) {
  // Not a memory operation or something we want to handle.
  default:
    Scale = TypeSize::Fixed(0);
    Width = 0;
    MinOffset = MaxOffset = 0;
    return false;
  case AArch64::STRWpost:
  case AArch64::LDRWpost:
    Width = 32;
    Scale = TypeSize::Fixed(4);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDURQi:
  case AArch64::STURQi:
    Width = 16;
    Scale = TypeSize::Fixed(1);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::PRFUMi:
  case AArch64::LDURXi:
  case AArch64::LDURDi:
  case AArch64::STURXi:
  case AArch64::STURDi:
    Width = 8;
    Scale = TypeSize::Fixed(1);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDURWi:
  case AArch64::LDURSi:
  case AArch64::LDURSWi:
  case AArch64::STURWi:
  case AArch64::STURSi:
    Width = 4;
    Scale = TypeSize::Fixed(1);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDURHi:
  case AArch64::LDURHHi:
  case AArch64::LDURSHXi:
  case AArch64::LDURSHWi:
  case AArch64::STURHi:
  case AArch64::STURHHi:
    Width = 2;
    Scale = TypeSize::Fixed(1);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDURBi:
  case AArch64::LDURBBi:
  case AArch64::LDURSBXi:
  case AArch64::LDURSBWi:
  case AArch64::STURBi:
  case AArch64::STURBBi:
    Width = 1;
    Scale = TypeSize::Fixed(1);
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDPQi:
  case AArch64::LDNPQi:
  case AArch64::STPQi:
  case AArch64::STNPQi:
    Scale = TypeSize::Fixed(16);
    Width = 32;
    MinOffset = -64;
    MaxOffset = 63;
    break;
  case AArch64::LDRQui:
  case AArch64::STRQui:
    Scale = TypeSize::Fixed(16);
    Width = 16;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::LDPXi:
  case AArch64::LDPDi:
  case AArch64::LDNPXi:
  case AArch64::LDNPDi:
  case AArch64::STPXi:
  case AArch64::STPDi:
  case AArch64::STNPXi:
  case AArch64::STNPDi:
    Scale = TypeSize::Fixed(8);
    Width = 16;
    MinOffset = -64;
    MaxOffset = 63;
    break;
  case AArch64::PRFMui:
  case AArch64::LDRXui:
  case AArch64::LDRDui:
  case AArch64::STRXui:
  case AArch64::STRDui:
    Scale = TypeSize::Fixed(8);
    Width = 8;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::StoreSwiftAsyncContext:
    // Store is an STRXui, but there might be an ADDXri in the expansion too.
    Scale = TypeSize::Fixed(1);
    Width = 8;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::LDPWi:
  case AArch64::LDPSi:
  case AArch64::LDNPWi:
  case AArch64::LDNPSi:
  case AArch64::STPWi:
  case AArch64::STPSi:
  case AArch64::STNPWi:
  case AArch64::STNPSi:
    Scale = TypeSize::Fixed(4);
    Width = 8;
    MinOffset = -64;
    MaxOffset = 63;
    break;
  case AArch64::LDRWui:
  case AArch64::LDRSui:
  case AArch64::LDRSWui:
  case AArch64::STRWui:
  case AArch64::STRSui:
    Scale = TypeSize::Fixed(4);
    Width = 4;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::LDRHui:
  case AArch64::LDRHHui:
  case AArch64::LDRSHWui:
  case AArch64::LDRSHXui:
  case AArch64::STRHui:
  case AArch64::STRHHui:
    Scale = TypeSize::Fixed(2);
    Width = 2;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::LDRBui:
  case AArch64::LDRBBui:
  case AArch64::LDRSBWui:
  case AArch64::LDRSBXui:
  case AArch64::STRBui:
  case AArch64::STRBBui:
    Scale = TypeSize::Fixed(1);
    Width = 1;
    MinOffset = 0;
    MaxOffset = 4095;
    break;
  case AArch64::STPXpre:
  case AArch64::LDPXpost:
  case AArch64::STPDpre:
  case AArch64::LDPDpost:
    Scale = TypeSize::Fixed(8);
    Width = 8;
    MinOffset = -512;
    MaxOffset = 504;
    break;
  case AArch64::STPQpre:
  case AArch64::LDPQpost:
    Scale = TypeSize::Fixed(16);
    Width = 16;
    MinOffset = -1024;
    MaxOffset = 1008;
    break;
  case AArch64::STRXpre:
  case AArch64::STRDpre:
  case AArch64::LDRXpost:
  case AArch64::LDRDpost:
    Scale = TypeSize::Fixed(1);
    Width = 8;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::STRQpre:
  case AArch64::LDRQpost:
    Scale = TypeSize::Fixed(1);
    Width = 16;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::ADDG:
    Scale = TypeSize::Fixed(16);
    Width = 0;
    MinOffset = 0;
    MaxOffset = 63;
    break;
  case AArch64::TAGPstack:
    Scale = TypeSize::Fixed(16);
    Width = 0;
    // TAGP with a negative offset turns into SUBP, which has a maximum offset
    // of 63 (not 64!).
    MinOffset = -63;
    MaxOffset = 63;
    break;
  case AArch64::LDG:
  case AArch64::STGi:
  case AArch64::STZGi:
    Scale = TypeSize::Fixed(16);
    Width = 16;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::STR_ZZZZXI:
  case AArch64::LDR_ZZZZXI:
    Scale = TypeSize::Scalable(16);
    Width = SVEMaxBytesPerVector * 4;
    MinOffset = -256;
    MaxOffset = 252;
    break;
  case AArch64::STR_ZZZXI:
  case AArch64::LDR_ZZZXI:
    Scale = TypeSize::Scalable(16);
    Width = SVEMaxBytesPerVector * 3;
    MinOffset = -256;
    MaxOffset = 253;
    break;
  case AArch64::STR_ZZXI:
  case AArch64::LDR_ZZXI:
    Scale = TypeSize::Scalable(16);
    Width = SVEMaxBytesPerVector * 2;
    MinOffset = -256;
    MaxOffset = 254;
    break;
  case AArch64::LDR_PXI:
  case AArch64::STR_PXI:
    Scale = TypeSize::Scalable(2);
    Width = SVEMaxBytesPerVector / 8;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LDR_ZXI:
  case AArch64::STR_ZXI:
    Scale = TypeSize::Scalable(16);
    Width = SVEMaxBytesPerVector;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::LD1B_IMM:
  case AArch64::LD1H_IMM:
  case AArch64::LD1W_IMM:
  case AArch64::LD1D_IMM:
  case AArch64::LDNT1B_ZRI:
  case AArch64::LDNT1H_ZRI:
  case AArch64::LDNT1W_ZRI:
  case AArch64::LDNT1D_ZRI:
  case AArch64::ST1B_IMM:
  case AArch64::ST1H_IMM:
  case AArch64::ST1W_IMM:
  case AArch64::ST1D_IMM:
  case AArch64::STNT1B_ZRI:
  case AArch64::STNT1H_ZRI:
  case AArch64::STNT1W_ZRI:
  case AArch64::STNT1D_ZRI:
  case AArch64::LDNF1B_IMM:
  case AArch64::LDNF1H_IMM:
  case AArch64::LDNF1W_IMM:
  case AArch64::LDNF1D_IMM:
    // A full vectors worth of data
    // Width = mbytes * elements
    Scale = TypeSize::Scalable(16);
    Width = SVEMaxBytesPerVector;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD2B_IMM:
  case AArch64::LD2H_IMM:
  case AArch64::LD2W_IMM:
  case AArch64::LD2D_IMM:
  case AArch64::ST2B_IMM:
  case AArch64::ST2H_IMM:
  case AArch64::ST2W_IMM:
  case AArch64::ST2D_IMM:
    Scale = TypeSize::Scalable(32);
    Width = SVEMaxBytesPerVector * 2;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD3B_IMM:
  case AArch64::LD3H_IMM:
  case AArch64::LD3W_IMM:
  case AArch64::LD3D_IMM:
  case AArch64::ST3B_IMM:
  case AArch64::ST3H_IMM:
  case AArch64::ST3W_IMM:
  case AArch64::ST3D_IMM:
    Scale = TypeSize::Scalable(48);
    Width = SVEMaxBytesPerVector * 3;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD4B_IMM:
  case AArch64::LD4H_IMM:
  case AArch64::LD4W_IMM:
  case AArch64::LD4D_IMM:
  case AArch64::ST4B_IMM:
  case AArch64::ST4H_IMM:
  case AArch64::ST4W_IMM:
  case AArch64::ST4D_IMM:
    Scale = TypeSize::Scalable(64);
    Width = SVEMaxBytesPerVector * 4;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD1B_H_IMM:
  case AArch64::LD1SB_H_IMM:
  case AArch64::LD1H_S_IMM:
  case AArch64::LD1SH_S_IMM:
  case AArch64::LD1W_D_IMM:
  case AArch64::LD1SW_D_IMM:
  case AArch64::ST1B_H_IMM:
  case AArch64::ST1H_S_IMM:
  case AArch64::ST1W_D_IMM:
  case AArch64::LDNF1B_H_IMM:
  case AArch64::LDNF1SB_H_IMM:
  case AArch64::LDNF1H_S_IMM:
  case AArch64::LDNF1SH_S_IMM:
  case AArch64::LDNF1W_D_IMM:
  case AArch64::LDNF1SW_D_IMM:
    // A half vector worth of data
    // Width = mbytes * elements
    Scale = TypeSize::Scalable(8);
    Width = SVEMaxBytesPerVector / 2;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD1B_S_IMM:
  case AArch64::LD1SB_S_IMM:
  case AArch64::LD1H_D_IMM:
  case AArch64::LD1SH_D_IMM:
  case AArch64::ST1B_S_IMM:
  case AArch64::ST1H_D_IMM:
  case AArch64::LDNF1B_S_IMM:
  case AArch64::LDNF1SB_S_IMM:
  case AArch64::LDNF1H_D_IMM:
  case AArch64::LDNF1SH_D_IMM:
    // A quarter vector worth of data
    // Width = mbytes * elements
    Scale = TypeSize::Scalable(4);
    Width = SVEMaxBytesPerVector / 4;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::LD1B_D_IMM:
  case AArch64::LD1SB_D_IMM:
  case AArch64::ST1B_D_IMM:
  case AArch64::LDNF1B_D_IMM:
  case AArch64::LDNF1SB_D_IMM:
    // A eighth vector worth of data
    // Width = mbytes * elements
    Scale = TypeSize::Scalable(2);
    Width = SVEMaxBytesPerVector / 8;
    MinOffset = -8;
    MaxOffset = 7;
    break;
  case AArch64::ST2Gi:
  case AArch64::STZ2Gi:
    Scale = TypeSize::Fixed(16);
    Width = 32;
    MinOffset = -256;
    MaxOffset = 255;
    break;
  case AArch64::STGPi:
    Scale = TypeSize::Fixed(16);
    Width = 16;
    MinOffset = -64;
    MaxOffset = 63;
    break;
  case AArch64::LD1RB_IMM:
  case AArch64::LD1RB_H_IMM:
  case AArch64::LD1RB_S_IMM:
  case AArch64::LD1RB_D_IMM:
  case AArch64::LD1RSB_H_IMM:
  case AArch64::LD1RSB_S_IMM:
  case AArch64::LD1RSB_D_IMM:
    Scale = TypeSize::Fixed(1);
    Width = 1;
    MinOffset = 0;
    MaxOffset = 63;
    break;
  case AArch64::LD1RH_IMM:
  case AArch64::LD1RH_S_IMM:
  case AArch64::LD1RH_D_IMM:
  case AArch64::LD1RSH_S_IMM:
  case AArch64::LD1RSH_D_IMM:
    Scale = TypeSize::Fixed(2);
    Width = 2;
    MinOffset = 0;
    MaxOffset = 63;
    break;
  case AArch64::LD1RW_IMM:
  case AArch64::LD1RW_D_IMM:
  case AArch64::LD1RSW_IMM:
    Scale = TypeSize::Fixed(4);
    Width = 4;
    MinOffset = 0;
    MaxOffset = 63;
    break;
  case AArch64::LD1RD_IMM:
    Scale = TypeSize::Fixed(8);
    Width = 8;
    MinOffset = 0;
    MaxOffset = 63;
    break;
  }
 
  return true;
}
 
// Scaling factor for unscaled load or store.
int AArch64InstrInfo::getMemScale(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has unknown scale!");
  case AArch64::LDRBBui:
  case AArch64::LDURBBi:
  case AArch64::LDRSBWui:
  case AArch64::LDURSBWi:
  case AArch64::STRBBui:
  case AArch64::STURBBi:
    return 1;
  case AArch64::LDRHHui:
  case AArch64::LDURHHi:
  case AArch64::LDRSHWui:
  case AArch64::LDURSHWi:
  case AArch64::STRHHui:
  case AArch64::STURHHi:
    return 2;
  case AArch64::LDRSui:
  case AArch64::LDURSi:
  case AArch64::LDRSpre:
  case AArch64::LDRSWui:
  case AArch64::LDURSWi:
  case AArch64::LDRSWpre:
  case AArch64::LDRWpre:
  case AArch64::LDRWui:
  case AArch64::LDURWi:
  case AArch64::STRSui:
  case AArch64::STURSi:
  case AArch64::STRSpre:
  case AArch64::STRWui:
  case AArch64::STURWi:
  case AArch64::STRWpre:
  case AArch64::LDPSi:
  case AArch64::LDPSWi:
  case AArch64::LDPWi:
  case AArch64::STPSi:
  case AArch64::STPWi:
    return 4;
  case AArch64::LDRDui:
  case AArch64::LDURDi:
  case AArch64::LDRDpre:
  case AArch64::LDRXui:
  case AArch64::LDURXi:
  case AArch64::LDRXpre:
  case AArch64::STRDui:
  case AArch64::STURDi:
  case AArch64::STRDpre:
  case AArch64::STRXui:
  case AArch64::STURXi:
  case AArch64::STRXpre:
  case AArch64::LDPDi:
  case AArch64::LDPXi:
  case AArch64::STPDi:
  case AArch64::STPXi:
    return 8;
  case AArch64::LDRQui:
  case AArch64::LDURQi:
  case AArch64::STRQui:
  case AArch64::STURQi:
  case AArch64::STRQpre:
  case AArch64::LDPQi:
  case AArch64::LDRQpre:
  case AArch64::STPQi:
  case AArch64::STGi:
  case AArch64::STZGi:
  case AArch64::ST2Gi:
  case AArch64::STZ2Gi:
  case AArch64::STGPi:
    return 16;
  }
}
 
bool AArch64InstrInfo::isPreLd(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    return false;
  case AArch64::LDRWpre:
  case AArch64::LDRXpre:
  case AArch64::LDRSWpre:
  case AArch64::LDRSpre:
  case AArch64::LDRDpre:
  case AArch64::LDRQpre:
    return true;
  }
}
 
bool AArch64InstrInfo::isPreSt(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    return false;
  case AArch64::STRWpre:
  case AArch64::STRXpre:
  case AArch64::STRSpre:
  case AArch64::STRDpre:
  case AArch64::STRQpre:
    return true;
  }
}
 
bool AArch64InstrInfo::isPreLdSt(const MachineInstr &MI) {
  return isPreLd(MI) || isPreSt(MI);
}
 
bool AArch64InstrInfo::isPairedLdSt(const MachineInstr &MI) {
  switch (MI.getOpcode()) {
  default:
    return false;
  case AArch64::LDPSi:
  case AArch64::LDPSWi:
  case AArch64::LDPDi:
  case AArch64::LDPQi:
  case AArch64::LDPWi:
  case AArch64::LDPXi:
  case AArch64::STPSi:
  case AArch64::STPDi:
  case AArch64::STPQi:
  case AArch64::STPWi:
  case AArch64::STPXi:
  case AArch64::STGPi:
    return true;
  }
}
 
const MachineOperand &AArch64InstrInfo::getLdStBaseOp(const MachineInstr &MI) {
  unsigned Idx =
      AArch64InstrInfo::isPairedLdSt(MI) || AArch64InstrInfo::isPreLdSt(MI) ? 2
                                                                            : 1;
  return MI.getOperand(Idx);
}
 
const MachineOperand &
AArch64InstrInfo::getLdStOffsetOp(const MachineInstr &MI) {
  unsigned Idx =
      AArch64InstrInfo::isPairedLdSt(MI) || AArch64InstrInfo::isPreLdSt(MI) ? 3
                                                                            : 2;
  return MI.getOperand(Idx);
}
 
static const TargetRegisterClass *getRegClass(const MachineInstr &MI,
                                              Register Reg) {
  if (MI.getParent() == nullptr)
    return nullptr;
  const MachineFunction *MF = MI.getParent()->getParent();
  return MF ? MF->getRegInfo().getRegClassOrNull(Reg) : nullptr;
}
 
bool AArch64InstrInfo::isHForm(const MachineInstr &MI) {
  auto IsHFPR = [&](const MachineOperand &Op) {
    if (!Op.isReg())
      return false;
    auto Reg = Op.getReg();
    if (Reg.isPhysical())
      return AArch64::FPR16RegClass.contains(Reg);
    const TargetRegisterClass *TRC = ::getRegClass(MI, Reg);
    return TRC == &AArch64::FPR16RegClass ||
           TRC == &AArch64::FPR16_loRegClass;
  };
  return llvm::any_of(MI.operands(), IsHFPR);
}
 
bool AArch64InstrInfo::isQForm(const MachineInstr &MI) {
  auto IsQFPR = [&](const MachineOperand &Op) {
    if (!Op.isReg())
      return false;
    auto Reg = Op.getReg();
    if (Reg.isPhysical())
      return AArch64::FPR128RegClass.contains(Reg);
    const TargetRegisterClass *TRC = ::getRegClass(MI, Reg);
    return TRC == &AArch64::FPR128RegClass ||
           TRC == &AArch64::FPR128_loRegClass;
  };
  return llvm::any_of(MI.operands(), IsQFPR);
}
 
bool AArch64InstrInfo::isFpOrNEON(const MachineInstr &MI) {
  auto IsFPR = [&](const MachineOperand &Op) {
    if (!Op.isReg())
      return false;
    auto Reg = Op.getReg();
    if (Reg.isPhysical())
      return AArch64::FPR128RegClass.contains(Reg) ||
             AArch64::FPR64RegClass.contains(Reg) ||
             AArch64::FPR32RegClass.contains(Reg) ||
             AArch64::FPR16RegClass.contains(Reg) ||
             AArch64::FPR8RegClass.contains(Reg);
 
    const TargetRegisterClass *TRC = ::getRegClass(MI, Reg);
    return TRC == &AArch64::FPR128RegClass ||
           TRC == &AArch64::FPR128_loRegClass ||
           TRC == &AArch64::FPR64RegClass ||
           TRC == &AArch64::FPR64_loRegClass ||
           TRC == &AArch64::FPR32RegClass || TRC == &AArch64::FPR16RegClass ||
           TRC == &AArch64::FPR8RegClass;
  };
  return llvm::any_of(MI.operands(), IsFPR);
}
 
// Scale the unscaled offsets.  Returns false if the unscaled offset can't be
// scaled.
static bool scaleOffset(unsigned Opc, int64_t &Offset) {
  int Scale = AArch64InstrInfo::getMemScale(Opc);
 
  // If the byte-offset isn't a multiple of the stride, we can't scale this
  // offset.
  if (Offset % Scale != 0)
    return false;
 
  // Convert the byte-offset used by unscaled into an "element" offset used
  // by the scaled pair load/store instructions.
  Offset /= Scale;
  return true;
}
 
static bool canPairLdStOpc(unsigned FirstOpc, unsigned SecondOpc) {
  if (FirstOpc == SecondOpc)
    return true;
  // We can also pair sign-ext and zero-ext instructions.
  switch (FirstOpc) {
  default:
    return false;
  case AArch64::LDRWui:
  case AArch64::LDURWi:
    return SecondOpc == AArch64::LDRSWui || SecondOpc == AArch64::LDURSWi;
  case AArch64::LDRSWui:
  case AArch64::LDURSWi:
    return SecondOpc == AArch64::LDRWui || SecondOpc == AArch64::LDURWi;
  }
  // These instructions can't be paired based on their opcodes.
  return false;
}
 
static bool shouldClusterFI(const MachineFrameInfo &MFI, int FI1,
                            int64_t Offset1, unsigned Opcode1, int FI2,
                            int64_t Offset2, unsigned Opcode2) {
  // Accesses through fixed stack object frame indices may access a different
  // fixed stack slot. Check that the object offsets + offsets match.
  if (MFI.isFixedObjectIndex(FI1) && MFI.isFixedObjectIndex(FI2)) {
    int64_t ObjectOffset1 = MFI.getObjectOffset(FI1);
    int64_t ObjectOffset2 = MFI.getObjectOffset(FI2);
    assert(ObjectOffset1 <= ObjectOffset2 && "Object offsets are not ordered.");
    // Convert to scaled object offsets.
    int Scale1 = AArch64InstrInfo::getMemScale(Opcode1);
    if (ObjectOffset1 % Scale1 != 0)
      return false;
    ObjectOffset1 /= Scale1;
    int Scale2 = AArch64InstrInfo::getMemScale(Opcode2);
    if (ObjectOffset2 % Scale2 != 0)
      return false;
    ObjectOffset2 /= Scale2;
    ObjectOffset1 += Offset1;
    ObjectOffset2 += Offset2;
    return ObjectOffset1 + 1 == ObjectOffset2;
  }
 
  return FI1 == FI2;
}
 
/// Detect opportunities for ldp/stp formation.
///
/// Only called for LdSt for which getMemOperandWithOffset returns true.
bool AArch64InstrInfo::shouldClusterMemOps(
    ArrayRef<const MachineOperand *> BaseOps1,
    ArrayRef<const MachineOperand *> BaseOps2, unsigned NumLoads,
    unsigned NumBytes) const {
  assert(BaseOps1.size() == 1 && BaseOps2.size() == 1);
  const MachineOperand &BaseOp1 = *BaseOps1.front();
  const MachineOperand &BaseOp2 = *BaseOps2.front();
  const MachineInstr &FirstLdSt = *BaseOp1.getParent();
  const MachineInstr &SecondLdSt = *BaseOp2.getParent();
  if (BaseOp1.getType() != BaseOp2.getType())
    return false;
 
  assert((BaseOp1.isReg() || BaseOp1.isFI()) &&
         "Only base registers and frame indices are supported.");
 
  // Check for both base regs and base FI.
  if (BaseOp1.isReg() && BaseOp1.getReg() != BaseOp2.getReg())
    return false;
 
  // Only cluster up to a single pair.
  if (NumLoads > 2)
    return false;
 
  if (!isPairableLdStInst(FirstLdSt) || !isPairableLdStInst(SecondLdSt))
    return false;
 
  // Can we pair these instructions based on their opcodes?
  unsigned FirstOpc = FirstLdSt.getOpcode();
  unsigned SecondOpc = SecondLdSt.getOpcode();
  if (!canPairLdStOpc(FirstOpc, SecondOpc))
    return false;
 
  // Can't merge volatiles or load/stores that have a hint to avoid pair
  // formation, for example.
  if (!isCandidateToMergeOrPair(FirstLdSt) ||
      !isCandidateToMergeOrPair(SecondLdSt))
    return false;
 
  // isCandidateToMergeOrPair guarantees that operand 2 is an immediate.
  int64_t Offset1 = FirstLdSt.getOperand(2).getImm();
  if (hasUnscaledLdStOffset(FirstOpc) && !scaleOffset(FirstOpc, Offset1))
    return false;
 
  int64_t Offset2 = SecondLdSt.getOperand(2).getImm();
  if (hasUnscaledLdStOffset(SecondOpc) && !scaleOffset(SecondOpc, Offset2))
    return false;
 
  // Pairwise instructions have a 7-bit signed offset field.
  if (Offset1 > 63 || Offset1 < -64)
    return false;
 
  // The caller should already have ordered First/SecondLdSt by offset.
  // Note: except for non-equal frame index bases
  if (BaseOp1.isFI()) {
    assert((!BaseOp1.isIdenticalTo(BaseOp2) || Offset1 <= Offset2) &&
           "Caller should have ordered offsets.");
 
    const MachineFrameInfo &MFI =
        FirstLdSt.getParent()->getParent()->getFrameInfo();
    return shouldClusterFI(MFI, BaseOp1.getIndex(), Offset1, FirstOpc,
                           BaseOp2.getIndex(), Offset2, SecondOpc);
  }
 
  assert(Offset1 <= Offset2 && "Caller should have ordered offsets.");
 
  return Offset1 + 1 == Offset2;
}
 
static const MachineInstrBuilder &AddSubReg(const MachineInstrBuilder &MIB,
                                            unsigned Reg, unsigned SubIdx,
                                            unsigned State,
                                            const TargetRegisterInfo *TRI) {
  if (!SubIdx)
    return MIB.addReg(Reg, State);
 
  if (Register::isPhysicalRegister(Reg))
    return MIB.addReg(TRI->getSubReg(Reg, SubIdx), State);
  return MIB.addReg(Reg, State, SubIdx);
}
 
static bool forwardCopyWillClobberTuple(unsigned DestReg, unsigned SrcReg,
                                        unsigned NumRegs) {
  // We really want the positive remainder mod 32 here, that happens to be
  // easily obtainable with a mask.
  return ((DestReg - SrcReg) & 0x1f) < NumRegs;
}
 
void AArch64InstrInfo::copyPhysRegTuple(MachineBasicBlock &MBB,
                                        MachineBasicBlock::iterator I,
                                        const DebugLoc &DL, MCRegister DestReg,
                                        MCRegister SrcReg, bool KillSrc,
                                        unsigned Opcode,
                                        ArrayRef<unsigned> Indices) const {
  assert(Subtarget.hasNEON() && "Unexpected register copy without NEON");
  const TargetRegisterInfo *TRI = &getRegisterInfo();
  uint16_t DestEncoding = TRI->getEncodingValue(DestReg);
  uint16_t SrcEncoding = TRI->getEncodingValue(SrcReg);
  unsigned NumRegs = Indices.size();
 
  int SubReg = 0, End = NumRegs, Incr = 1;
  if (forwardCopyWillClobberTuple(DestEncoding, SrcEncoding, NumRegs)) {
    SubReg = NumRegs - 1;
    End = -1;
    Incr = -1;
  }
 
  for (; SubReg != End; SubReg += Incr) {
    const MachineInstrBuilder MIB = BuildMI(MBB, I, DL, get(Opcode));
    AddSubReg(MIB, DestReg, Indices[SubReg], RegState::Define, TRI);
    AddSubReg(MIB, SrcReg, Indices[SubReg], 0, TRI);
    AddSubReg(MIB, SrcReg, Indices[SubReg], getKillRegState(KillSrc), TRI);
  }
}
 
void AArch64InstrInfo::copyGPRRegTuple(MachineBasicBlock &MBB,
                                       MachineBasicBlock::iterator I,
                                       DebugLoc DL, unsigned DestReg,
                                       unsigned SrcReg, bool KillSrc,
                                       unsigned Opcode, unsigned ZeroReg,
                                       llvm::ArrayRef<unsigned> Indices) const {
  const TargetRegisterInfo *TRI = &getRegisterInfo();
  unsigned NumRegs = Indices.size();
 
#ifndef NDEBUG
  uint16_t DestEncoding = TRI->getEncodingValue(DestReg);
  uint16_t SrcEncoding = TRI->getEncodingValue(SrcReg);
  assert(DestEncoding % NumRegs == 0 && SrcEncoding % NumRegs == 0 &&
         "GPR reg sequences should not be able to overlap");
#endif
 
  for (unsigned SubReg = 0; SubReg != NumRegs; ++SubReg) {
    const MachineInstrBuilder MIB = BuildMI(MBB, I, DL, get(Opcode));
    AddSubReg(MIB, DestReg, Indices[SubReg], RegState::Define, TRI);
    MIB.addReg(ZeroReg);
    AddSubReg(MIB, SrcReg, Indices[SubReg], getKillRegState(KillSrc), TRI);
    MIB.addImm(0);
  }
}
 
void AArch64InstrInfo::copyPhysReg(MachineBasicBlock &MBB,
                                   MachineBasicBlock::iterator I,
                                   const DebugLoc &DL, MCRegister DestReg,
                                   MCRegister SrcReg, bool KillSrc) const {
  if (AArch64::GPR32spRegClass.contains(DestReg) &&
      (AArch64::GPR32spRegClass.contains(SrcReg) || SrcReg == AArch64::WZR)) {
    const TargetRegisterInfo *TRI = &getRegisterInfo();
 
    if (DestReg == AArch64::WSP || SrcReg == AArch64::WSP) {
      // If either operand is WSP, expand to ADD #0.
      if (Subtarget.hasZeroCycleRegMove()) {
        // Cyclone recognizes "ADD Xd, Xn, #0" as a zero-cycle register move.
        MCRegister DestRegX = TRI->getMatchingSuperReg(
            DestReg, AArch64::sub_32, &AArch64::GPR64spRegClass);
        MCRegister SrcRegX = TRI->getMatchingSuperReg(
            SrcReg, AArch64::sub_32, &AArch64::GPR64spRegClass);
        // This instruction is reading and writing X registers.  This may upset
        // the register scavenger and machine verifier, so we need to indicate
        // that we are reading an undefined value from SrcRegX, but a proper
        // value from SrcReg.
        BuildMI(MBB, I, DL, get(AArch64::ADDXri), DestRegX)
            .addReg(SrcRegX, RegState::Undef)
            .addImm(0)
            .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0))
            .addReg(SrcReg, RegState::Implicit | getKillRegState(KillSrc));
      } else {
        BuildMI(MBB, I, DL, get(AArch64::ADDWri), DestReg)
            .addReg(SrcReg, getKillRegState(KillSrc))
            .addImm(0)
            .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0));
      }
    } else if (SrcReg == AArch64::WZR && Subtarget.hasZeroCycleZeroingGP()) {
      BuildMI(MBB, I, DL, get(AArch64::MOVZWi), DestReg)
          .addImm(0)
          .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0));
    } else {
      if (Subtarget.hasZeroCycleRegMove()) {
        // Cyclone recognizes "ORR Xd, XZR, Xm" as a zero-cycle register move.
        MCRegister DestRegX = TRI->getMatchingSuperReg(
            DestReg, AArch64::sub_32, &AArch64::GPR64spRegClass);
        MCRegister SrcRegX = TRI->getMatchingSuperReg(
            SrcReg, AArch64::sub_32, &AArch64::GPR64spRegClass);
        // This instruction is reading and writing X registers.  This may upset
        // the register scavenger and machine verifier, so we need to indicate
        // that we are reading an undefined value from SrcRegX, but a proper
        // value from SrcReg.
        BuildMI(MBB, I, DL, get(AArch64::ORRXrr), DestRegX)
            .addReg(AArch64::XZR)
            .addReg(SrcRegX, RegState::Undef)
            .addReg(SrcReg, RegState::Implicit | getKillRegState(KillSrc));
      } else {
        // Otherwise, expand to ORR WZR.
        BuildMI(MBB, I, DL, get(AArch64::ORRWrr), DestReg)
            .addReg(AArch64::WZR)
            .addReg(SrcReg, getKillRegState(KillSrc));
      }
    }
    return;
  }
 
  // Copy a Predicate register by ORRing with itself.
  if (AArch64::PPRRegClass.contains(DestReg) &&
      AArch64::PPRRegClass.contains(SrcReg)) {
    assert(Subtarget.hasSVEorSME() && "Unexpected SVE register.");
    BuildMI(MBB, I, DL, get(AArch64::ORR_PPzPP), DestReg)
      .addReg(SrcReg) // Pg
      .addReg(SrcReg)
      .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  if (AArch64::PNRRegClass.contains(DestReg) &&
      AArch64::PPRRegClass.contains(SrcReg)) {
    assert((Subtarget.hasSVE2p1() || Subtarget.hasSME2()) &&
           "Unexpected predicate-as-counter register.");
    // Copy from pX to pnX is a no-op
    if ((DestReg.id() - AArch64::PN0) == (SrcReg.id() - AArch64::P0))
      return;
    MCRegister PPRDestReg = (DestReg - AArch64::PN0) + AArch64::P0;
    BuildMI(MBB, I, DL, get(AArch64::ORR_PPzPP), PPRDestReg)
        .addReg(SrcReg)
        .addReg(SrcReg)
        .addReg(SrcReg, getKillRegState(KillSrc))
        .addDef(DestReg, RegState::Implicit);
    return;
  }
 
  if (AArch64::PPRRegClass.contains(DestReg) &&
      AArch64::PNRRegClass.contains(SrcReg)) {
    assert((Subtarget.hasSVE2p1() || Subtarget.hasSME2()) &&
           "Unexpected predicate-as-counter register.");
    // Copy from pnX to pX is a no-op
    if ((DestReg.id() - AArch64::P0) == (SrcReg.id() - AArch64::PN0))
      return;
    MCRegister PNRDestReg = (DestReg - AArch64::P0) + AArch64::PN0;
    MCRegister PPRSrcReg = (SrcReg - AArch64::PN0) + AArch64::P0;
    BuildMI(MBB, I, DL, get(AArch64::ORR_PPzPP), DestReg)
        .addReg(PPRSrcReg)
        .addReg(PPRSrcReg)
        .addReg(PPRSrcReg, getKillRegState(KillSrc))
        .addDef(PNRDestReg, RegState::Implicit);
    return;
  }
 
  // Copy a Z register by ORRing with itself.
  if (AArch64::ZPRRegClass.contains(DestReg) &&
      AArch64::ZPRRegClass.contains(SrcReg)) {
    assert(Subtarget.hasSVEorSME() && "Unexpected SVE register.");
    BuildMI(MBB, I, DL, get(AArch64::ORR_ZZZ), DestReg)
      .addReg(SrcReg)
      .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  // Copy a Z register pair by copying the individual sub-registers.
  if ((AArch64::ZPR2RegClass.contains(DestReg) ||
       AArch64::ZPR2StridedOrContiguousRegClass.contains(DestReg)) &&
      (AArch64::ZPR2RegClass.contains(SrcReg) ||
       AArch64::ZPR2StridedOrContiguousRegClass.contains(SrcReg))) {
    assert(Subtarget.hasSVEorSME() && "Unexpected SVE register.");
    static const unsigned Indices[] = {AArch64::zsub0, AArch64::zsub1};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORR_ZZZ,
                     Indices);
    return;
  }
 
  // Copy a Z register triple by copying the individual sub-registers.
  if (AArch64::ZPR3RegClass.contains(DestReg) &&
      AArch64::ZPR3RegClass.contains(SrcReg)) {
    assert(Subtarget.hasSVEorSME() && "Unexpected SVE register.");
    static const unsigned Indices[] = {AArch64::zsub0, AArch64::zsub1,
                                       AArch64::zsub2};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORR_ZZZ,
                     Indices);
    return;
  }
 
  // Copy a Z register quad by copying the individual sub-registers.
  if ((AArch64::ZPR4RegClass.contains(DestReg) ||
       AArch64::ZPR4StridedOrContiguousRegClass.contains(DestReg)) &&
      (AArch64::ZPR4RegClass.contains(SrcReg) ||
       AArch64::ZPR4StridedOrContiguousRegClass.contains(SrcReg))) {
    assert(Subtarget.hasSVEorSME() && "Unexpected SVE register.");
    static const unsigned Indices[] = {AArch64::zsub0, AArch64::zsub1,
                                       AArch64::zsub2, AArch64::zsub3};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORR_ZZZ,
                     Indices);
    return;
  }
 
  if (AArch64::GPR64spRegClass.contains(DestReg) &&
      (AArch64::GPR64spRegClass.contains(SrcReg) || SrcReg == AArch64::XZR)) {
    if (DestReg == AArch64::SP || SrcReg == AArch64::SP) {
      // If either operand is SP, expand to ADD #0.
      BuildMI(MBB, I, DL, get(AArch64::ADDXri), DestReg)
          .addReg(SrcReg, getKillRegState(KillSrc))
          .addImm(0)
          .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0));
    } else if (SrcReg == AArch64::XZR && Subtarget.hasZeroCycleZeroingGP()) {
      BuildMI(MBB, I, DL, get(AArch64::MOVZXi), DestReg)
          .addImm(0)
          .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0));
    } else {
      // Otherwise, expand to ORR XZR.
      BuildMI(MBB, I, DL, get(AArch64::ORRXrr), DestReg)
          .addReg(AArch64::XZR)
          .addReg(SrcReg, getKillRegState(KillSrc));
    }
    return;
  }
 
  // Copy a DDDD register quad by copying the individual sub-registers.
  if (AArch64::DDDDRegClass.contains(DestReg) &&
      AArch64::DDDDRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::dsub0, AArch64::dsub1,
                                       AArch64::dsub2, AArch64::dsub3};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv8i8,
                     Indices);
    return;
  }
 
  // Copy a DDD register triple by copying the individual sub-registers.
  if (AArch64::DDDRegClass.contains(DestReg) &&
      AArch64::DDDRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::dsub0, AArch64::dsub1,
                                       AArch64::dsub2};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv8i8,
                     Indices);
    return;
  }
 
  // Copy a DD register pair by copying the individual sub-registers.
  if (AArch64::DDRegClass.contains(DestReg) &&
      AArch64::DDRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::dsub0, AArch64::dsub1};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv8i8,
                     Indices);
    return;
  }
 
  // Copy a QQQQ register quad by copying the individual sub-registers.
  if (AArch64::QQQQRegClass.contains(DestReg) &&
      AArch64::QQQQRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::qsub0, AArch64::qsub1,
                                       AArch64::qsub2, AArch64::qsub3};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv16i8,
                     Indices);
    return;
  }
 
  // Copy a QQQ register triple by copying the individual sub-registers.
  if (AArch64::QQQRegClass.contains(DestReg) &&
      AArch64::QQQRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::qsub0, AArch64::qsub1,
                                       AArch64::qsub2};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv16i8,
                     Indices);
    return;
  }
 
  // Copy a QQ register pair by copying the individual sub-registers.
  if (AArch64::QQRegClass.contains(DestReg) &&
      AArch64::QQRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::qsub0, AArch64::qsub1};
    copyPhysRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRv16i8,
                     Indices);
    return;
  }
 
  if (AArch64::XSeqPairsClassRegClass.contains(DestReg) &&
      AArch64::XSeqPairsClassRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::sube64, AArch64::subo64};
    copyGPRRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRXrs,
                    AArch64::XZR, Indices);
    return;
  }
 
  if (AArch64::WSeqPairsClassRegClass.contains(DestReg) &&
      AArch64::WSeqPairsClassRegClass.contains(SrcReg)) {
    static const unsigned Indices[] = {AArch64::sube32, AArch64::subo32};
    copyGPRRegTuple(MBB, I, DL, DestReg, SrcReg, KillSrc, AArch64::ORRWrs,
                    AArch64::WZR, Indices);
    return;
  }
 
  if (AArch64::FPR128RegClass.contains(DestReg) &&
      AArch64::FPR128RegClass.contains(SrcReg)) {
    if (Subtarget.hasSVEorSME() && !Subtarget.isNeonAvailable())
      BuildMI(MBB, I, DL, get(AArch64::ORR_ZZZ))
          .addReg(AArch64::Z0 + (DestReg - AArch64::Q0), RegState::Define)
          .addReg(AArch64::Z0 + (SrcReg - AArch64::Q0))
          .addReg(AArch64::Z0 + (SrcReg - AArch64::Q0));
    else if (Subtarget.hasNEON())
      BuildMI(MBB, I, DL, get(AArch64::ORRv16i8), DestReg)
          .addReg(SrcReg)
          .addReg(SrcReg, getKillRegState(KillSrc));
    else {
      BuildMI(MBB, I, DL, get(AArch64::STRQpre))
          .addReg(AArch64::SP, RegState::Define)
          .addReg(SrcReg, getKillRegState(KillSrc))
          .addReg(AArch64::SP)
          .addImm(-16);
      BuildMI(MBB, I, DL, get(AArch64::LDRQpre))
          .addReg(AArch64::SP, RegState::Define)
          .addReg(DestReg, RegState::Define)
          .addReg(AArch64::SP)
          .addImm(16);
    }
    return;
  }
 
  if (AArch64::FPR64RegClass.contains(DestReg) &&
      AArch64::FPR64RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVDr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  if (AArch64::FPR32RegClass.contains(DestReg) &&
      AArch64::FPR32RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVSr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  if (AArch64::FPR16RegClass.contains(DestReg) &&
      AArch64::FPR16RegClass.contains(SrcReg)) {
    DestReg =
        RI.getMatchingSuperReg(DestReg, AArch64::hsub, &AArch64::FPR32RegClass);
    SrcReg =
        RI.getMatchingSuperReg(SrcReg, AArch64::hsub, &AArch64::FPR32RegClass);
    BuildMI(MBB, I, DL, get(AArch64::FMOVSr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  if (AArch64::FPR8RegClass.contains(DestReg) &&
      AArch64::FPR8RegClass.contains(SrcReg)) {
    DestReg =
        RI.getMatchingSuperReg(DestReg, AArch64::bsub, &AArch64::FPR32RegClass);
    SrcReg =
        RI.getMatchingSuperReg(SrcReg, AArch64::bsub, &AArch64::FPR32RegClass);
    BuildMI(MBB, I, DL, get(AArch64::FMOVSr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  // Copies between GPR64 and FPR64.
  if (AArch64::FPR64RegClass.contains(DestReg) &&
      AArch64::GPR64RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVXDr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
  if (AArch64::GPR64RegClass.contains(DestReg) &&
      AArch64::FPR64RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVDXr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
  // Copies between GPR32 and FPR32.
  if (AArch64::FPR32RegClass.contains(DestReg) &&
      AArch64::GPR32RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVWSr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
  if (AArch64::GPR32RegClass.contains(DestReg) &&
      AArch64::FPR32RegClass.contains(SrcReg)) {
    BuildMI(MBB, I, DL, get(AArch64::FMOVSWr), DestReg)
        .addReg(SrcReg, getKillRegState(KillSrc));
    return;
  }
 
  if (DestReg == AArch64::NZCV) {
    assert(AArch64::GPR64RegClass.contains(SrcReg) && "Invalid NZCV copy");
    BuildMI(MBB, I, DL, get(AArch64::MSR))
        .addImm(AArch64SysReg::NZCV)
        .addReg(SrcReg, getKillRegState(KillSrc))
        .addReg(AArch64::NZCV, RegState::Implicit | RegState::Define);
    return;
  }
 
  if (SrcReg == AArch64::NZCV) {
    assert(AArch64::GPR64RegClass.contains(DestReg) && "Invalid NZCV copy");
    BuildMI(MBB, I, DL, get(AArch64::MRS), DestReg)
        .addImm(AArch64SysReg::NZCV)
        .addReg(AArch64::NZCV, RegState::Implicit | getKillRegState(KillSrc));
    return;
  }
 
#ifndef NDEBUG
  const TargetRegisterInfo &TRI = getRegisterInfo();
  errs() << TRI.getRegAsmName(DestReg) << " = COPY "
         << TRI.getRegAsmName(SrcReg) << "\n";
#endif
  llvm_unreachable("unimplemented reg-to-reg copy");
}
 
static void storeRegPairToStackSlot(const TargetRegisterInfo &TRI,
                                    MachineBasicBlock &MBB,
                                    MachineBasicBlock::iterator InsertBefore,
                                    const MCInstrDesc &MCID,
                                    Register SrcReg, bool IsKill,
                                    unsigned SubIdx0, unsigned SubIdx1, int FI,
                                    MachineMemOperand *MMO) {
  Register SrcReg0 = SrcReg;
  Register SrcReg1 = SrcReg;
  if (SrcReg.isPhysical()) {
    SrcReg0 = TRI.getSubReg(SrcReg, SubIdx0);
    SubIdx0 = 0;
    SrcReg1 = TRI.getSubReg(SrcReg, SubIdx1);
    SubIdx1 = 0;
  }
  BuildMI(MBB, InsertBefore, DebugLoc(), MCID)
      .addReg(SrcReg0, getKillRegState(IsKill), SubIdx0)
      .addReg(SrcReg1, getKillRegState(IsKill), SubIdx1)
      .addFrameIndex(FI)
      .addImm(0)
      .addMemOperand(MMO);
}
 
void AArch64InstrInfo::storeRegToStackSlot(MachineBasicBlock &MBB,
                                           MachineBasicBlock::iterator MBBI,
                                           Register SrcReg, bool isKill, int FI,
                                           const TargetRegisterClass *RC,
                                           const TargetRegisterInfo *TRI,
                                           Register VReg) const {
  MachineFunction &MF = *MBB.getParent();
  MachineFrameInfo &MFI = MF.getFrameInfo();
 
  MachinePointerInfo PtrInfo = MachinePointerInfo::getFixedStack(MF, FI);
  MachineMemOperand *MMO =
      MF.getMachineMemOperand(PtrInfo, MachineMemOperand::MOStore,
                              MFI.getObjectSize(FI), MFI.getObjectAlign(FI));
  unsigned Opc = 0;
  bool Offset = true;
  MCRegister PNRReg = MCRegister::NoRegister;
  unsigned StackID = TargetStackID::Default;
  switch (TRI->getSpillSize(*RC)) {
  case 1:
    if (AArch64::FPR8RegClass.hasSubClassEq(RC))
      Opc = AArch64::STRBui;
    break;
  case 2:
    if (AArch64::FPR16RegClass.hasSubClassEq(RC))
      Opc = AArch64::STRHui;
    else if (AArch64::PPRRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register store without SVE store instructions");
      Opc = AArch64::STR_PXI;
      StackID = TargetStackID::ScalableVector;
    } else if (AArch64::PNRRegClass.hasSubClassEq(RC)) {
      assert((Subtarget.hasSVE2p1() || Subtarget.hasSME2()) &&
             "Unexpected register store without SVE2p1 or SME2");
      SrcReg = (SrcReg - AArch64::PN0) + AArch64::P0;
      Opc = AArch64::STR_PXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 4:
    if (AArch64::GPR32allRegClass.hasSubClassEq(RC)) {
      Opc = AArch64::STRWui;
      if (SrcReg.isVirtual())
        MF.getRegInfo().constrainRegClass(SrcReg, &AArch64::GPR32RegClass);
      else
        assert(SrcReg != AArch64::WSP);
    } else if (AArch64::FPR32RegClass.hasSubClassEq(RC))
      Opc = AArch64::STRSui;
    break;
  case 8:
    if (AArch64::GPR64allRegClass.hasSubClassEq(RC)) {
      Opc = AArch64::STRXui;
      if (SrcReg.isVirtual())
        MF.getRegInfo().constrainRegClass(SrcReg, &AArch64::GPR64RegClass);
      else
        assert(SrcReg != AArch64::SP);
    } else if (AArch64::FPR64RegClass.hasSubClassEq(RC)) {
      Opc = AArch64::STRDui;
    } else if (AArch64::WSeqPairsClassRegClass.hasSubClassEq(RC)) {
      storeRegPairToStackSlot(getRegisterInfo(), MBB, MBBI,
                              get(AArch64::STPWi), SrcReg, isKill,
                              AArch64::sube32, AArch64::subo32, FI, MMO);
      return;
    }
    break;
  case 16:
    if (AArch64::FPR128RegClass.hasSubClassEq(RC))
      Opc = AArch64::STRQui;
    else if (AArch64::DDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Twov1d;
      Offset = false;
    } else if (AArch64::XSeqPairsClassRegClass.hasSubClassEq(RC)) {
      storeRegPairToStackSlot(getRegisterInfo(), MBB, MBBI,
                              get(AArch64::STPXi), SrcReg, isKill,
                              AArch64::sube64, AArch64::subo64, FI, MMO);
      return;
    } else if (AArch64::ZPRRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register store without SVE store instructions");
      Opc = AArch64::STR_ZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 24:
    if (AArch64::DDDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Threev1d;
      Offset = false;
    }
    break;
  case 32:
    if (AArch64::DDDDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Fourv1d;
      Offset = false;
    } else if (AArch64::QQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Twov2d;
      Offset = false;
    } else if (AArch64::ZPR2RegClass.hasSubClassEq(RC) ||
               AArch64::ZPR2StridedOrContiguousRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register store without SVE store instructions");
      Opc = AArch64::STR_ZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 48:
    if (AArch64::QQQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Threev2d;
      Offset = false;
    } else if (AArch64::ZPR3RegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register store without SVE store instructions");
      Opc = AArch64::STR_ZZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 64:
    if (AArch64::QQQQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register store without NEON");
      Opc = AArch64::ST1Fourv2d;
      Offset = false;
    } else if (AArch64::ZPR4RegClass.hasSubClassEq(RC) ||
               AArch64::ZPR4StridedOrContiguousRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register store without SVE store instructions");
      Opc = AArch64::STR_ZZZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  }
  assert(Opc && "Unknown register class");
  MFI.setStackID(FI, StackID);
 
  const MachineInstrBuilder MI = BuildMI(MBB, MBBI, DebugLoc(), get(Opc))
                                     .addReg(SrcReg, getKillRegState(isKill))
                                     .addFrameIndex(FI);
 
  if (Offset)
    MI.addImm(0);
  if (PNRReg.isValid())
    MI.addDef(PNRReg, RegState::Implicit);
  MI.addMemOperand(MMO);
}
 
static void loadRegPairFromStackSlot(const TargetRegisterInfo &TRI,
                                     MachineBasicBlock &MBB,
                                     MachineBasicBlock::iterator InsertBefore,
                                     const MCInstrDesc &MCID,
                                     Register DestReg, unsigned SubIdx0,
                                     unsigned SubIdx1, int FI,
                                     MachineMemOperand *MMO) {
  Register DestReg0 = DestReg;
  Register DestReg1 = DestReg;
  bool IsUndef = true;
  if (DestReg.isPhysical()) {
    DestReg0 = TRI.getSubReg(DestReg, SubIdx0);
    SubIdx0 = 0;
    DestReg1 = TRI.getSubReg(DestReg, SubIdx1);
    SubIdx1 = 0;
    IsUndef = false;
  }
  BuildMI(MBB, InsertBefore, DebugLoc(), MCID)
      .addReg(DestReg0, RegState::Define | getUndefRegState(IsUndef), SubIdx0)
      .addReg(DestReg1, RegState::Define | getUndefRegState(IsUndef), SubIdx1)
      .addFrameIndex(FI)
      .addImm(0)
      .addMemOperand(MMO);
}
 
void AArch64InstrInfo::loadRegFromStackSlot(MachineBasicBlock &MBB,
                                            MachineBasicBlock::iterator MBBI,
                                            Register DestReg, int FI,
                                            const TargetRegisterClass *RC,
                                            const TargetRegisterInfo *TRI,
                                            Register VReg) const {
  MachineFunction &MF = *MBB.getParent();
  MachineFrameInfo &MFI = MF.getFrameInfo();
  MachinePointerInfo PtrInfo = MachinePointerInfo::getFixedStack(MF, FI);
  MachineMemOperand *MMO =
      MF.getMachineMemOperand(PtrInfo, MachineMemOperand::MOLoad,
                              MFI.getObjectSize(FI), MFI.getObjectAlign(FI));
 
  unsigned Opc = 0;
  bool Offset = true;
  unsigned StackID = TargetStackID::Default;
  MCRegister PNRReg = MCRegister::NoRegister;
  switch (TRI->getSpillSize(*RC)) {
  case 1:
    if (AArch64::FPR8RegClass.hasSubClassEq(RC))
      Opc = AArch64::LDRBui;
    break;
  case 2:
    if (AArch64::FPR16RegClass.hasSubClassEq(RC))
      Opc = AArch64::LDRHui;
    else if (AArch64::PPRRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register load without SVE load instructions");
      Opc = AArch64::LDR_PXI;
      StackID = TargetStackID::ScalableVector;
    } else if (AArch64::PNRRegClass.hasSubClassEq(RC)) {
      assert((Subtarget.hasSVE2p1() || Subtarget.hasSME2()) &&
             "Unexpected register load without SVE2p1 or SME2");
      PNRReg = DestReg;
      DestReg = (DestReg - AArch64::PN0) + AArch64::P0;
      Opc = AArch64::LDR_PXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 4:
    if (AArch64::GPR32allRegClass.hasSubClassEq(RC)) {
      Opc = AArch64::LDRWui;
      if (DestReg.isVirtual())
        MF.getRegInfo().constrainRegClass(DestReg, &AArch64::GPR32RegClass);
      else
        assert(DestReg != AArch64::WSP);
    } else if (AArch64::FPR32RegClass.hasSubClassEq(RC))
      Opc = AArch64::LDRSui;
    break;
  case 8:
    if (AArch64::GPR64allRegClass.hasSubClassEq(RC)) {
      Opc = AArch64::LDRXui;
      if (DestReg.isVirtual())
        MF.getRegInfo().constrainRegClass(DestReg, &AArch64::GPR64RegClass);
      else
        assert(DestReg != AArch64::SP);
    } else if (AArch64::FPR64RegClass.hasSubClassEq(RC)) {
      Opc = AArch64::LDRDui;
    } else if (AArch64::WSeqPairsClassRegClass.hasSubClassEq(RC)) {
      loadRegPairFromStackSlot(getRegisterInfo(), MBB, MBBI,
                               get(AArch64::LDPWi), DestReg, AArch64::sube32,
                               AArch64::subo32, FI, MMO);
      return;
    }
    break;
  case 16:
    if (AArch64::FPR128RegClass.hasSubClassEq(RC))
      Opc = AArch64::LDRQui;
    else if (AArch64::DDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Twov1d;
      Offset = false;
    } else if (AArch64::XSeqPairsClassRegClass.hasSubClassEq(RC)) {
      loadRegPairFromStackSlot(getRegisterInfo(), MBB, MBBI,
                               get(AArch64::LDPXi), DestReg, AArch64::sube64,
                               AArch64::subo64, FI, MMO);
      return;
    } else if (AArch64::ZPRRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register load without SVE load instructions");
      Opc = AArch64::LDR_ZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 24:
    if (AArch64::DDDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Threev1d;
      Offset = false;
    }
    break;
  case 32:
    if (AArch64::DDDDRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Fourv1d;
      Offset = false;
    } else if (AArch64::QQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Twov2d;
      Offset = false;
    } else if (AArch64::ZPR2RegClass.hasSubClassEq(RC) ||
               AArch64::ZPR2StridedOrContiguousRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register load without SVE load instructions");
      Opc = AArch64::LDR_ZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 48:
    if (AArch64::QQQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Threev2d;
      Offset = false;
    } else if (AArch64::ZPR3RegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register load without SVE load instructions");
      Opc = AArch64::LDR_ZZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  case 64:
    if (AArch64::QQQQRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasNEON() && "Unexpected register load without NEON");
      Opc = AArch64::LD1Fourv2d;
      Offset = false;
    } else if (AArch64::ZPR4RegClass.hasSubClassEq(RC) ||
               AArch64::ZPR4StridedOrContiguousRegClass.hasSubClassEq(RC)) {
      assert(Subtarget.hasSVEorSME() &&
             "Unexpected register load without SVE load instructions");
      Opc = AArch64::LDR_ZZZZXI;
      StackID = TargetStackID::ScalableVector;
    }
    break;
  }
 
  assert(Opc && "Unknown register class");
  MFI.setStackID(FI, StackID);
 
  const MachineInstrBuilder MI = BuildMI(MBB, MBBI, DebugLoc(), get(Opc))
                                     .addReg(DestReg, getDefRegState(true))
                                     .addFrameIndex(FI);
  if (Offset)
    MI.addImm(0);
  if (PNRReg.isValid())
    MI.addDef(PNRReg, RegState::Implicit);
  MI.addMemOperand(MMO);
}
 
bool llvm::isNZCVTouchedInInstructionRange(const MachineInstr &DefMI,
                                           const MachineInstr &UseMI,
                                           const TargetRegisterInfo *TRI) {
  return any_of(instructionsWithoutDebug(std::next(DefMI.getIterator()),
                                         UseMI.getIterator()),
                [TRI](const MachineInstr &I) {
                  return I.modifiesRegister(AArch64::NZCV, TRI) ||
                         I.readsRegister(AArch64::NZCV, TRI);
                });
}
 
void AArch64InstrInfo::decomposeStackOffsetForDwarfOffsets(
    const StackOffset &Offset, int64_t &ByteSized, int64_t &VGSized) {
  // The smallest scalable element supported by scaled SVE addressing
  // modes are predicates, which are 2 scalable bytes in size. So the scalable
  // byte offset must always be a multiple of 2.
  assert(Offset.getScalable() % 2 == 0 && "Invalid frame offset");
 
  // VGSized offsets are divided by '2', because the VG register is the
  // the number of 64bit granules as opposed to 128bit vector chunks,
  // which is how the 'n' in e.g. MVT::nxv1i8 is modelled.
  // So, for a stack offset of 16 MVT::nxv1i8's, the size is n x 16 bytes.
  // VG = n * 2 and the dwarf offset must be VG * 8 bytes.
  ByteSized = Offset.getFixed();
  VGSized = Offset.getScalable() / 2;
}
 
/// Returns the offset in parts to which this frame offset can be
/// decomposed for the purpose of describing a frame offset.
/// For non-scalable offsets this is simply its byte size.
void AArch64InstrInfo::decomposeStackOffsetForFrameOffsets(
    const StackOffset &Offset, int64_t &NumBytes, int64_t &NumPredicateVectors,
    int64_t &NumDataVectors) {
  // The smallest scalable element supported by scaled SVE addressing
  // modes are predicates, which are 2 scalable bytes in size. So the scalable
  // byte offset must always be a multiple of 2.
  assert(Offset.getScalable() % 2 == 0 && "Invalid frame offset");
 
  NumBytes = Offset.getFixed();
  NumDataVectors = 0;
  NumPredicateVectors = Offset.getScalable() / 2;
  // This method is used to get the offsets to adjust the frame offset.
  // If the function requires ADDPL to be used and needs more than two ADDPL
  // instructions, part of the offset is folded into NumDataVectors so that it
  // uses ADDVL for part of it, reducing the number of ADDPL instructions.
  if (NumPredicateVectors % 8 == 0 || NumPredicateVectors < -64 ||
      NumPredicateVectors > 62) {
    NumDataVectors = NumPredicateVectors / 8;
    NumPredicateVectors -= NumDataVectors * 8;
  }
}
 
// Convenience function to create a DWARF expression for
//   Expr + NumBytes + NumVGScaledBytes * AArch64::VG
static void appendVGScaledOffsetExpr(SmallVectorImpl<char> &Expr, int NumBytes,
                                     int NumVGScaledBytes, unsigned VG,
                                     llvm::raw_string_ostream &Comment) {
  uint8_t buffer[16];
 
  if (NumBytes) {
    Expr.push_back(dwarf::DW_OP_consts);
    Expr.append(buffer, buffer + encodeSLEB128(NumBytes, buffer));
    Expr.push_back((uint8_t)dwarf::DW_OP_plus);
    Comment << (NumBytes < 0 ? " - " : " + ") << std::abs(NumBytes);
  }
 
  if (NumVGScaledBytes) {
    Expr.push_back((uint8_t)dwarf::DW_OP_consts);
    Expr.append(buffer, buffer + encodeSLEB128(NumVGScaledBytes, buffer));
 
    Expr.push_back((uint8_t)dwarf::DW_OP_bregx);
    Expr.append(buffer, buffer + encodeULEB128(VG, buffer));
    Expr.push_back(0);
 
    Expr.push_back((uint8_t)dwarf::DW_OP_mul);
    Expr.push_back((uint8_t)dwarf::DW_OP_plus);
 
    Comment << (NumVGScaledBytes < 0 ? " - " : " + ")
            << std::abs(NumVGScaledBytes) << " * VG";
  }
}
 
// Creates an MCCFIInstruction:
//    { DW_CFA_def_cfa_expression, ULEB128 (sizeof expr), expr }
static MCCFIInstruction createDefCFAExpression(const TargetRegisterInfo &TRI,
                                               unsigned Reg,
                                               const StackOffset &Offset) {
  int64_t NumBytes, NumVGScaledBytes;
  AArch64InstrInfo::decomposeStackOffsetForDwarfOffsets(Offset, NumBytes,
                                                        NumVGScaledBytes);
  std::string CommentBuffer;
  llvm::raw_string_ostream Comment(CommentBuffer);
 
  if (Reg == AArch64::SP)
    Comment << "sp";
  else if (Reg == AArch64::FP)
    Comment << "fp";