xref: /llvm-project-15.0.7/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp (revision e1f613ce)

//===- AMDGPUInstructionSelector.cpp ----------------------------*- C++ -*-==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
/// \file
/// This file implements the targeting of the InstructionSelector class for
/// AMDGPU.
/// \todo This should be generated by TableGen.
//===----------------------------------------------------------------------===//

#include "AMDGPUInstructionSelector.h"
#include "AMDGPUInstrInfo.h"
#include "AMDGPUGlobalISelUtils.h"
#include "AMDGPURegisterBankInfo.h"
#include "AMDGPUSubtarget.h"
#include "AMDGPUTargetMachine.h"
#include "MCTargetDesc/AMDGPUMCTargetDesc.h"
#include "SIMachineFunctionInfo.h"
#include "llvm/CodeGen/GlobalISel/GISelKnownBits.h"
#include "llvm/CodeGen/GlobalISel/InstructionSelector.h"
#include "llvm/CodeGen/GlobalISel/InstructionSelectorImpl.h"
#include "llvm/CodeGen/GlobalISel/MachineIRBuilder.h"
#include "llvm/CodeGen/GlobalISel/MIPatternMatch.h"
#include "llvm/CodeGen/GlobalISel/Utils.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/IR/DiagnosticInfo.h"
#include "llvm/IR/Type.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"

#define DEBUG_TYPE "amdgpu-isel"

using namespace llvm;
using namespace MIPatternMatch;

static cl::opt<bool> AllowRiskySelect(
  "amdgpu-global-isel-risky-select",
  cl::desc("Allow GlobalISel to select cases that are likely to not work yet"),
  cl::init(false),
  cl::ReallyHidden);

#define GET_GLOBALISEL_IMPL
#define AMDGPUSubtarget GCNSubtarget
#include "AMDGPUGenGlobalISel.inc"
#undef GET_GLOBALISEL_IMPL
#undef AMDGPUSubtarget

AMDGPUInstructionSelector::AMDGPUInstructionSelector(
    const GCNSubtarget &STI, const AMDGPURegisterBankInfo &RBI,
    const AMDGPUTargetMachine &TM)
    : InstructionSelector(), TII(*STI.getInstrInfo()),
      TRI(*STI.getRegisterInfo()), RBI(RBI), TM(TM),
      STI(STI),
      EnableLateStructurizeCFG(AMDGPUTargetMachine::EnableLateStructurizeCFG),
#define GET_GLOBALISEL_PREDICATES_INIT
#include "AMDGPUGenGlobalISel.inc"
#undef GET_GLOBALISEL_PREDICATES_INIT
#define GET_GLOBALISEL_TEMPORARIES_INIT
#include "AMDGPUGenGlobalISel.inc"
#undef GET_GLOBALISEL_TEMPORARIES_INIT
{
}

const char *AMDGPUInstructionSelector::getName() { return DEBUG_TYPE; }

void AMDGPUInstructionSelector::setupMF(MachineFunction &MF, GISelKnownBits &KB,
                                        CodeGenCoverage &CoverageInfo) {
  MRI = &MF.getRegInfo();
  Subtarget = &MF.getSubtarget<GCNSubtarget>();
  InstructionSelector::setupMF(MF, KB, CoverageInfo);
}

bool AMDGPUInstructionSelector::isVCC(Register Reg,
                                      const MachineRegisterInfo &MRI) const {
  // The verifier is oblivious to s1 being a valid value for wavesize registers.
  if (Reg.isPhysical())
    return false;

  auto &RegClassOrBank = MRI.getRegClassOrRegBank(Reg);
  const TargetRegisterClass *RC =
      RegClassOrBank.dyn_cast<const TargetRegisterClass*>();
  if (RC) {
    const LLT Ty = MRI.getType(Reg);
    return RC->hasSuperClassEq(TRI.getBoolRC()) &&
           Ty.isValid() && Ty.getSizeInBits() == 1;
  }

  const RegisterBank *RB = RegClassOrBank.get<const RegisterBank *>();
  return RB->getID() == AMDGPU::VCCRegBankID;
}

bool AMDGPUInstructionSelector::constrainCopyLikeIntrin(MachineInstr &MI,
                                                        unsigned NewOpc) const {
  MI.setDesc(TII.get(NewOpc));
  MI.RemoveOperand(1); // Remove intrinsic ID.
  MI.addOperand(*MF, MachineOperand::CreateReg(AMDGPU::EXEC, false, true));

  MachineOperand &Dst = MI.getOperand(0);
  MachineOperand &Src = MI.getOperand(1);

  // TODO: This should be legalized to s32 if needed
  if (MRI->getType(Dst.getReg()) == LLT::scalar(1))
    return false;

  const TargetRegisterClass *DstRC
    = TRI.getConstrainedRegClassForOperand(Dst, *MRI);
  const TargetRegisterClass *SrcRC
    = TRI.getConstrainedRegClassForOperand(Src, *MRI);
  if (!DstRC || DstRC != SrcRC)
    return false;

  return RBI.constrainGenericRegister(Dst.getReg(), *DstRC, *MRI) &&
         RBI.constrainGenericRegister(Src.getReg(), *SrcRC, *MRI);
}

bool AMDGPUInstructionSelector::selectCOPY(MachineInstr &I) const {
  const DebugLoc &DL = I.getDebugLoc();
  MachineBasicBlock *BB = I.getParent();
  I.setDesc(TII.get(TargetOpcode::COPY));

  const MachineOperand &Src = I.getOperand(1);
  MachineOperand &Dst = I.getOperand(0);
  Register DstReg = Dst.getReg();
  Register SrcReg = Src.getReg();

  if (isVCC(DstReg, *MRI)) {
    if (SrcReg == AMDGPU::SCC) {
      const TargetRegisterClass *RC
        = TRI.getConstrainedRegClassForOperand(Dst, *MRI);
      if (!RC)
        return true;
      return RBI.constrainGenericRegister(DstReg, *RC, *MRI);
    }

    if (!isVCC(SrcReg, *MRI)) {
      // TODO: Should probably leave the copy and let copyPhysReg expand it.
      if (!RBI.constrainGenericRegister(DstReg, *TRI.getBoolRC(), *MRI))
        return false;

      const TargetRegisterClass *SrcRC
        = TRI.getConstrainedRegClassForOperand(Src, *MRI);

      Register MaskedReg = MRI->createVirtualRegister(SrcRC);

      // We can't trust the high bits at this point, so clear them.

      // TODO: Skip masking high bits if def is known boolean.

      unsigned AndOpc = TRI.isSGPRClass(SrcRC) ?
        AMDGPU::S_AND_B32 : AMDGPU::V_AND_B32_e32;
      BuildMI(*BB, &I, DL, TII.get(AndOpc), MaskedReg)
        .addImm(1)
        .addReg(SrcReg);
      BuildMI(*BB, &I, DL, TII.get(AMDGPU::V_CMP_NE_U32_e64), DstReg)
        .addImm(0)
        .addReg(MaskedReg);

      if (!MRI->getRegClassOrNull(SrcReg))
        MRI->setRegClass(SrcReg, SrcRC);
      I.eraseFromParent();
      return true;
    }

    const TargetRegisterClass *RC =
      TRI.getConstrainedRegClassForOperand(Dst, *MRI);
    if (RC && !RBI.constrainGenericRegister(DstReg, *RC, *MRI))
      return false;

    return true;
  }

  for (const MachineOperand &MO : I.operands()) {
    if (MO.getReg().isPhysical())
      continue;

    const TargetRegisterClass *RC =
            TRI.getConstrainedRegClassForOperand(MO, *MRI);
    if (!RC)
      continue;
    RBI.constrainGenericRegister(MO.getReg(), *RC, *MRI);
  }
  return true;
}

bool AMDGPUInstructionSelector::selectPHI(MachineInstr &I) const {
  const Register DefReg = I.getOperand(0).getReg();
  const LLT DefTy = MRI->getType(DefReg);
  if (DefTy == LLT::scalar(1)) {
    if (!AllowRiskySelect) {
      LLVM_DEBUG(dbgs() << "Skipping risky boolean phi\n");
      return false;
    }

    LLVM_DEBUG(dbgs() << "Selecting risky boolean phi\n");
  }

  // TODO: Verify this doesn't have insane operands (i.e. VGPR to SGPR copy)

  const RegClassOrRegBank &RegClassOrBank =
    MRI->getRegClassOrRegBank(DefReg);

  const TargetRegisterClass *DefRC
    = RegClassOrBank.dyn_cast<const TargetRegisterClass *>();
  if (!DefRC) {
    if (!DefTy.isValid()) {
      LLVM_DEBUG(dbgs() << "PHI operand has no type, not a gvreg?\n");
      return false;
    }

    const RegisterBank &RB = *RegClassOrBank.get<const RegisterBank *>();
    DefRC = TRI.getRegClassForTypeOnBank(DefTy, RB, *MRI);
    if (!DefRC) {
      LLVM_DEBUG(dbgs() << "PHI operand has unexpected size/bank\n");
      return false;
    }
  }

  // TODO: Verify that all registers have the same bank
  I.setDesc(TII.get(TargetOpcode::PHI));
  return RBI.constrainGenericRegister(DefReg, *DefRC, *MRI);
}

MachineOperand
AMDGPUInstructionSelector::getSubOperand64(MachineOperand &MO,
                                           const TargetRegisterClass &SubRC,
                                           unsigned SubIdx) const {

  MachineInstr *MI = MO.getParent();
  MachineBasicBlock *BB = MO.getParent()->getParent();
  Register DstReg = MRI->createVirtualRegister(&SubRC);

  if (MO.isReg()) {
    unsigned ComposedSubIdx = TRI.composeSubRegIndices(MO.getSubReg(), SubIdx);
    Register Reg = MO.getReg();
    BuildMI(*BB, MI, MI->getDebugLoc(), TII.get(AMDGPU::COPY), DstReg)
            .addReg(Reg, 0, ComposedSubIdx);

    return MachineOperand::CreateReg(DstReg, MO.isDef(), MO.isImplicit(),
                                     MO.isKill(), MO.isDead(), MO.isUndef(),
                                     MO.isEarlyClobber(), 0, MO.isDebug(),
                                     MO.isInternalRead());
  }

  assert(MO.isImm());

  APInt Imm(64, MO.getImm());

  switch (SubIdx) {
  default:
    llvm_unreachable("do not know to split immediate with this sub index.");
  case AMDGPU::sub0:
    return MachineOperand::CreateImm(Imm.getLoBits(32).getSExtValue());
  case AMDGPU::sub1:
    return MachineOperand::CreateImm(Imm.getHiBits(32).getSExtValue());
  }
}

static unsigned getLogicalBitOpcode(unsigned Opc, bool Is64) {
  switch (Opc) {
  case AMDGPU::G_AND:
    return Is64 ? AMDGPU::S_AND_B64 : AMDGPU::S_AND_B32;
  case AMDGPU::G_OR:
    return Is64 ? AMDGPU::S_OR_B64 : AMDGPU::S_OR_B32;
  case AMDGPU::G_XOR:
    return Is64 ? AMDGPU::S_XOR_B64 : AMDGPU::S_XOR_B32;
  default:
    llvm_unreachable("not a bit op");
  }
}

bool AMDGPUInstructionSelector::selectG_AND_OR_XOR(MachineInstr &I) const {
  Register DstReg = I.getOperand(0).getReg();
  unsigned Size = RBI.getSizeInBits(DstReg, *MRI, TRI);

  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  if (DstRB->getID() != AMDGPU::SGPRRegBankID &&
      DstRB->getID() != AMDGPU::VCCRegBankID)
    return false;

  bool Is64 = Size > 32 || (DstRB->getID() == AMDGPU::VCCRegBankID &&
                            STI.isWave64());
  I.setDesc(TII.get(getLogicalBitOpcode(I.getOpcode(), Is64)));

  // Dead implicit-def of scc
  I.addOperand(MachineOperand::CreateReg(AMDGPU::SCC, true, // isDef
                                         true, // isImp
                                         false, // isKill
                                         true)); // isDead
  return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
}

bool AMDGPUInstructionSelector::selectG_ADD_SUB(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  MachineFunction *MF = BB->getParent();
  Register DstReg = I.getOperand(0).getReg();
  const DebugLoc &DL = I.getDebugLoc();
  LLT Ty = MRI->getType(DstReg);
  if (Ty.isVector())
    return false;

  unsigned Size = Ty.getSizeInBits();
  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  const bool IsSALU = DstRB->getID() == AMDGPU::SGPRRegBankID;
  const bool Sub = I.getOpcode() == TargetOpcode::G_SUB;

  if (Size == 32) {
    if (IsSALU) {
      const unsigned Opc = Sub ? AMDGPU::S_SUB_U32 : AMDGPU::S_ADD_U32;
      MachineInstr *Add =
        BuildMI(*BB, &I, DL, TII.get(Opc), DstReg)
        .add(I.getOperand(1))
        .add(I.getOperand(2));
      I.eraseFromParent();
      return constrainSelectedInstRegOperands(*Add, TII, TRI, RBI);
    }

    if (STI.hasAddNoCarry()) {
      const unsigned Opc = Sub ? AMDGPU::V_SUB_U32_e64 : AMDGPU::V_ADD_U32_e64;
      I.setDesc(TII.get(Opc));
      I.addOperand(*MF, MachineOperand::CreateImm(0));
      I.addOperand(*MF, MachineOperand::CreateReg(AMDGPU::EXEC, false, true));
      return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
    }

    const unsigned Opc = Sub ? AMDGPU::V_SUB_CO_U32_e64 : AMDGPU::V_ADD_CO_U32_e64;

    Register UnusedCarry = MRI->createVirtualRegister(TRI.getWaveMaskRegClass());
    MachineInstr *Add
      = BuildMI(*BB, &I, DL, TII.get(Opc), DstReg)
      .addDef(UnusedCarry, RegState::Dead)
      .add(I.getOperand(1))
      .add(I.getOperand(2))
      .addImm(0);
    I.eraseFromParent();
    return constrainSelectedInstRegOperands(*Add, TII, TRI, RBI);
  }

  assert(!Sub && "illegal sub should not reach here");

  const TargetRegisterClass &RC
    = IsSALU ? AMDGPU::SReg_64_XEXECRegClass : AMDGPU::VReg_64RegClass;
  const TargetRegisterClass &HalfRC
    = IsSALU ? AMDGPU::SReg_32RegClass : AMDGPU::VGPR_32RegClass;

  MachineOperand Lo1(getSubOperand64(I.getOperand(1), HalfRC, AMDGPU::sub0));
  MachineOperand Lo2(getSubOperand64(I.getOperand(2), HalfRC, AMDGPU::sub0));
  MachineOperand Hi1(getSubOperand64(I.getOperand(1), HalfRC, AMDGPU::sub1));
  MachineOperand Hi2(getSubOperand64(I.getOperand(2), HalfRC, AMDGPU::sub1));

  Register DstLo = MRI->createVirtualRegister(&HalfRC);
  Register DstHi = MRI->createVirtualRegister(&HalfRC);

  if (IsSALU) {
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::S_ADD_U32), DstLo)
      .add(Lo1)
      .add(Lo2);
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::S_ADDC_U32), DstHi)
      .add(Hi1)
      .add(Hi2);
  } else {
    const TargetRegisterClass *CarryRC = TRI.getWaveMaskRegClass();
    Register CarryReg = MRI->createVirtualRegister(CarryRC);
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::V_ADD_CO_U32_e64), DstLo)
      .addDef(CarryReg)
      .add(Lo1)
      .add(Lo2)
      .addImm(0);
    MachineInstr *Addc = BuildMI(*BB, &I, DL, TII.get(AMDGPU::V_ADDC_U32_e64), DstHi)
      .addDef(MRI->createVirtualRegister(CarryRC), RegState::Dead)
      .add(Hi1)
      .add(Hi2)
      .addReg(CarryReg, RegState::Kill)
      .addImm(0);

    if (!constrainSelectedInstRegOperands(*Addc, TII, TRI, RBI))
      return false;
  }

  BuildMI(*BB, &I, DL, TII.get(AMDGPU::REG_SEQUENCE), DstReg)
    .addReg(DstLo)
    .addImm(AMDGPU::sub0)
    .addReg(DstHi)
    .addImm(AMDGPU::sub1);


  if (!RBI.constrainGenericRegister(DstReg, RC, *MRI))
    return false;

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectG_UADDO_USUBO_UADDE_USUBE(
  MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  MachineFunction *MF = BB->getParent();
  const DebugLoc &DL = I.getDebugLoc();
  Register Dst0Reg = I.getOperand(0).getReg();
  Register Dst1Reg = I.getOperand(1).getReg();
  const bool IsAdd = I.getOpcode() == AMDGPU::G_UADDO ||
                     I.getOpcode() == AMDGPU::G_UADDE;
  const bool HasCarryIn = I.getOpcode() == AMDGPU::G_UADDE ||
                          I.getOpcode() == AMDGPU::G_USUBE;

  if (isVCC(Dst1Reg, *MRI)) {
    unsigned NoCarryOpc =
        IsAdd ? AMDGPU::V_ADD_CO_U32_e64 : AMDGPU::V_SUB_CO_U32_e64;
    unsigned CarryOpc = IsAdd ? AMDGPU::V_ADDC_U32_e64 : AMDGPU::V_SUBB_U32_e64;
    I.setDesc(TII.get(HasCarryIn ? CarryOpc : NoCarryOpc));
    I.addOperand(*MF, MachineOperand::CreateReg(AMDGPU::EXEC, false, true));
    I.addOperand(*MF, MachineOperand::CreateImm(0));
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  Register Src0Reg = I.getOperand(2).getReg();
  Register Src1Reg = I.getOperand(3).getReg();

  if (HasCarryIn) {
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), AMDGPU::SCC)
      .addReg(I.getOperand(4).getReg());
  }

  unsigned NoCarryOpc = IsAdd ? AMDGPU::S_ADD_U32 : AMDGPU::S_SUB_U32;
  unsigned CarryOpc = IsAdd ? AMDGPU::S_ADDC_U32 : AMDGPU::S_SUBB_U32;

  BuildMI(*BB, &I, DL, TII.get(HasCarryIn ? CarryOpc : NoCarryOpc), Dst0Reg)
    .add(I.getOperand(2))
    .add(I.getOperand(3));
  BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), Dst1Reg)
    .addReg(AMDGPU::SCC);

  if (!MRI->getRegClassOrNull(Dst1Reg))
    MRI->setRegClass(Dst1Reg, &AMDGPU::SReg_32RegClass);

  if (!RBI.constrainGenericRegister(Dst0Reg, AMDGPU::SReg_32RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Src0Reg, AMDGPU::SReg_32RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Src1Reg, AMDGPU::SReg_32RegClass, *MRI))
    return false;

  if (HasCarryIn &&
      !RBI.constrainGenericRegister(I.getOperand(4).getReg(),
                                    AMDGPU::SReg_32RegClass, *MRI))
    return false;

  I.eraseFromParent();
  return true;
}

// TODO: We should probably legalize these to only using 32-bit results.
bool AMDGPUInstructionSelector::selectG_EXTRACT(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  Register DstReg = I.getOperand(0).getReg();
  Register SrcReg = I.getOperand(1).getReg();
  LLT DstTy = MRI->getType(DstReg);
  LLT SrcTy = MRI->getType(SrcReg);
  const unsigned SrcSize = SrcTy.getSizeInBits();
  unsigned DstSize = DstTy.getSizeInBits();

  // TODO: Should handle any multiple of 32 offset.
  unsigned Offset = I.getOperand(2).getImm();
  if (Offset % 32 != 0 || DstSize > 128)
    return false;

  // 16-bit operations really use 32-bit registers.
  // FIXME: Probably should not allow 16-bit G_EXTRACT results.
  if (DstSize == 16)
    DstSize = 32;

  const TargetRegisterClass *DstRC =
    TRI.getConstrainedRegClassForOperand(I.getOperand(0), *MRI);
  if (!DstRC || !RBI.constrainGenericRegister(DstReg, *DstRC, *MRI))
    return false;

  const RegisterBank *SrcBank = RBI.getRegBank(SrcReg, *MRI, TRI);
  const TargetRegisterClass *SrcRC =
    TRI.getRegClassForSizeOnBank(SrcSize, *SrcBank, *MRI);
  if (!SrcRC)
    return false;
  unsigned SubReg = SIRegisterInfo::getSubRegFromChannel(Offset / 32,
                                                         DstSize / 32);
  SrcRC = TRI.getSubClassWithSubReg(SrcRC, SubReg);
  if (!SrcRC)
    return false;

  SrcReg = constrainOperandRegClass(*MF, TRI, *MRI, TII, RBI, I,
                                    *SrcRC, I.getOperand(1));
  const DebugLoc &DL = I.getDebugLoc();
  BuildMI(*BB, &I, DL, TII.get(TargetOpcode::COPY), DstReg)
    .addReg(SrcReg, 0, SubReg);

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectG_MERGE_VALUES(MachineInstr &MI) const {
  MachineBasicBlock *BB = MI.getParent();
  Register DstReg = MI.getOperand(0).getReg();
  LLT DstTy = MRI->getType(DstReg);
  LLT SrcTy = MRI->getType(MI.getOperand(1).getReg());

  const unsigned SrcSize = SrcTy.getSizeInBits();
  if (SrcSize < 32)
    return selectImpl(MI, *CoverageInfo);

  const DebugLoc &DL = MI.getDebugLoc();
  const RegisterBank *DstBank = RBI.getRegBank(DstReg, *MRI, TRI);
  const unsigned DstSize = DstTy.getSizeInBits();
  const TargetRegisterClass *DstRC =
    TRI.getRegClassForSizeOnBank(DstSize, *DstBank, *MRI);
  if (!DstRC)
    return false;

  ArrayRef<int16_t> SubRegs = TRI.getRegSplitParts(DstRC, SrcSize / 8);
  MachineInstrBuilder MIB =
    BuildMI(*BB, &MI, DL, TII.get(TargetOpcode::REG_SEQUENCE), DstReg);
  for (int I = 0, E = MI.getNumOperands() - 1; I != E; ++I) {
    MachineOperand &Src = MI.getOperand(I + 1);
    MIB.addReg(Src.getReg(), getUndefRegState(Src.isUndef()));
    MIB.addImm(SubRegs[I]);

    const TargetRegisterClass *SrcRC
      = TRI.getConstrainedRegClassForOperand(Src, *MRI);
    if (SrcRC && !RBI.constrainGenericRegister(Src.getReg(), *SrcRC, *MRI))
      return false;
  }

  if (!RBI.constrainGenericRegister(DstReg, *DstRC, *MRI))
    return false;

  MI.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectG_UNMERGE_VALUES(MachineInstr &MI) const {
  MachineBasicBlock *BB = MI.getParent();
  const int NumDst = MI.getNumOperands() - 1;

  MachineOperand &Src = MI.getOperand(NumDst);

  Register SrcReg = Src.getReg();
  Register DstReg0 = MI.getOperand(0).getReg();
  LLT DstTy = MRI->getType(DstReg0);
  LLT SrcTy = MRI->getType(SrcReg);

  const unsigned DstSize = DstTy.getSizeInBits();
  const unsigned SrcSize = SrcTy.getSizeInBits();
  const DebugLoc &DL = MI.getDebugLoc();
  const RegisterBank *SrcBank = RBI.getRegBank(SrcReg, *MRI, TRI);

  const TargetRegisterClass *SrcRC =
    TRI.getRegClassForSizeOnBank(SrcSize, *SrcBank, *MRI);
  if (!SrcRC || !RBI.constrainGenericRegister(SrcReg, *SrcRC, *MRI))
    return false;

  // Note we could have mixed SGPR and VGPR destination banks for an SGPR
  // source, and this relies on the fact that the same subregister indices are
  // used for both.
  ArrayRef<int16_t> SubRegs = TRI.getRegSplitParts(SrcRC, DstSize / 8);
  for (int I = 0, E = NumDst; I != E; ++I) {
    MachineOperand &Dst = MI.getOperand(I);
    BuildMI(*BB, &MI, DL, TII.get(TargetOpcode::COPY), Dst.getReg())
      .addReg(SrcReg, 0, SubRegs[I]);

    // Make sure the subregister index is valid for the source register.
    SrcRC = TRI.getSubClassWithSubReg(SrcRC, SubRegs[I]);
    if (!SrcRC || !RBI.constrainGenericRegister(SrcReg, *SrcRC, *MRI))
      return false;

    const TargetRegisterClass *DstRC =
      TRI.getConstrainedRegClassForOperand(Dst, *MRI);
    if (DstRC && !RBI.constrainGenericRegister(Dst.getReg(), *DstRC, *MRI))
      return false;
  }

  MI.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectG_BUILD_VECTOR_TRUNC(
  MachineInstr &MI) const {
  if (selectImpl(MI, *CoverageInfo))
    return true;

  const LLT S32 = LLT::scalar(32);
  const LLT V2S16 = LLT::vector(2, 16);

  Register Dst = MI.getOperand(0).getReg();
  if (MRI->getType(Dst) != V2S16)
    return false;

  const RegisterBank *DstBank = RBI.getRegBank(Dst, *MRI, TRI);
  if (DstBank->getID() != AMDGPU::SGPRRegBankID)
    return false;

  Register Src0 = MI.getOperand(1).getReg();
  Register Src1 = MI.getOperand(2).getReg();
  if (MRI->getType(Src0) != S32)
    return false;

  const DebugLoc &DL = MI.getDebugLoc();
  MachineBasicBlock *BB = MI.getParent();

  auto ConstSrc1 = getConstantVRegValWithLookThrough(Src1, *MRI, true, true);
  if (ConstSrc1) {
    auto ConstSrc0 = getConstantVRegValWithLookThrough(Src0, *MRI, true, true);
    if (ConstSrc0) {
      uint32_t Lo16 = static_cast<uint32_t>(ConstSrc0->Value) & 0xffff;
      uint32_t Hi16 = static_cast<uint32_t>(ConstSrc1->Value) & 0xffff;

      BuildMI(*BB, &MI, DL, TII.get(AMDGPU::S_MOV_B32), Dst)
        .addImm(Lo16 | (Hi16 << 16));
      MI.eraseFromParent();
      return RBI.constrainGenericRegister(Dst, AMDGPU::SReg_32RegClass, *MRI);
    }
  }

  // TODO: This should probably be a combine somewhere
  // (build_vector_trunc $src0, undef -> copy $src0
  MachineInstr *Src1Def = getDefIgnoringCopies(Src1, *MRI);
  if (Src1Def && Src1Def->getOpcode() == AMDGPU::G_IMPLICIT_DEF) {
    MI.setDesc(TII.get(AMDGPU::COPY));
    MI.RemoveOperand(2);
    return RBI.constrainGenericRegister(Dst, AMDGPU::SReg_32RegClass, *MRI) &&
           RBI.constrainGenericRegister(Src0, AMDGPU::SReg_32RegClass, *MRI);
  }

  Register ShiftSrc0;
  Register ShiftSrc1;

  // With multiple uses of the shift, this will duplicate the shift and
  // increase register pressure.
  //
  // (build_vector_trunc (lshr_oneuse $src0, 16), (lshr_oneuse $src1, 16)
  //  => (S_PACK_HH_B32_B16 $src0, $src1)
  // (build_vector_trunc $src0, (lshr_oneuse SReg_32:$src1, 16))
  //  => (S_PACK_LH_B32_B16 $src0, $src1)
  // (build_vector_trunc $src0, $src1)
  //  => (S_PACK_LL_B32_B16 $src0, $src1)

  bool Shift0 = mi_match(
      Src0, *MRI, m_OneUse(m_GLShr(m_Reg(ShiftSrc0), m_SpecificICst(16))));

  bool Shift1 = mi_match(
      Src1, *MRI, m_OneUse(m_GLShr(m_Reg(ShiftSrc1), m_SpecificICst(16))));

  unsigned Opc = AMDGPU::S_PACK_LL_B32_B16;
  if (Shift0 && Shift1) {
    Opc = AMDGPU::S_PACK_HH_B32_B16;
    MI.getOperand(1).setReg(ShiftSrc0);
    MI.getOperand(2).setReg(ShiftSrc1);
  } else if (Shift1) {
    Opc = AMDGPU::S_PACK_LH_B32_B16;
    MI.getOperand(2).setReg(ShiftSrc1);
  } else if (Shift0 && ConstSrc1 && ConstSrc1->Value == 0) {
    // build_vector_trunc (lshr $src0, 16), 0 -> s_lshr_b32 $src0, 16
    auto MIB = BuildMI(*BB, &MI, DL, TII.get(AMDGPU::S_LSHR_B32), Dst)
      .addReg(ShiftSrc0)
      .addImm(16);

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

  MI.setDesc(TII.get(Opc));
  return constrainSelectedInstRegOperands(MI, TII, TRI, RBI);
}

bool AMDGPUInstructionSelector::selectG_PTR_ADD(MachineInstr &I) const {
  return selectG_ADD_SUB(I);
}

bool AMDGPUInstructionSelector::selectG_IMPLICIT_DEF(MachineInstr &I) const {
  const MachineOperand &MO = I.getOperand(0);

  // FIXME: Interface for getConstrainedRegClassForOperand needs work. The
  // regbank check here is to know why getConstrainedRegClassForOperand failed.
  const TargetRegisterClass *RC = TRI.getConstrainedRegClassForOperand(MO, *MRI);
  if ((!RC && !MRI->getRegBankOrNull(MO.getReg())) ||
      (RC && RBI.constrainGenericRegister(MO.getReg(), *RC, *MRI))) {
    I.setDesc(TII.get(TargetOpcode::IMPLICIT_DEF));
    return true;
  }

  return false;
}

bool AMDGPUInstructionSelector::selectG_INSERT(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();

  Register DstReg = I.getOperand(0).getReg();
  Register Src0Reg = I.getOperand(1).getReg();
  Register Src1Reg = I.getOperand(2).getReg();
  LLT Src1Ty = MRI->getType(Src1Reg);

  unsigned DstSize = MRI->getType(DstReg).getSizeInBits();
  unsigned InsSize = Src1Ty.getSizeInBits();

  int64_t Offset = I.getOperand(3).getImm();

  // FIXME: These cases should have been illegal and unnecessary to check here.
  if (Offset % 32 != 0 || InsSize % 32 != 0)
    return false;

  // Currently not handled by getSubRegFromChannel.
  if (InsSize > 128)
    return false;

  unsigned SubReg = TRI.getSubRegFromChannel(Offset / 32, InsSize / 32);
  if (SubReg == AMDGPU::NoSubRegister)
    return false;

  const RegisterBank *DstBank = RBI.getRegBank(DstReg, *MRI, TRI);
  const TargetRegisterClass *DstRC =
    TRI.getRegClassForSizeOnBank(DstSize, *DstBank, *MRI);
  if (!DstRC)
    return false;

  const RegisterBank *Src0Bank = RBI.getRegBank(Src0Reg, *MRI, TRI);
  const RegisterBank *Src1Bank = RBI.getRegBank(Src1Reg, *MRI, TRI);
  const TargetRegisterClass *Src0RC =
    TRI.getRegClassForSizeOnBank(DstSize, *Src0Bank, *MRI);
  const TargetRegisterClass *Src1RC =
    TRI.getRegClassForSizeOnBank(InsSize, *Src1Bank, *MRI);

  // Deal with weird cases where the class only partially supports the subreg
  // index.
  Src0RC = TRI.getSubClassWithSubReg(Src0RC, SubReg);
  if (!Src0RC || !Src1RC)
    return false;

  if (!RBI.constrainGenericRegister(DstReg, *DstRC, *MRI) ||
      !RBI.constrainGenericRegister(Src0Reg, *Src0RC, *MRI) ||
      !RBI.constrainGenericRegister(Src1Reg, *Src1RC, *MRI))
    return false;

  const DebugLoc &DL = I.getDebugLoc();
  BuildMI(*BB, &I, DL, TII.get(TargetOpcode::INSERT_SUBREG), DstReg)
    .addReg(Src0Reg)
    .addReg(Src1Reg)
    .addImm(SubReg);

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectInterpP1F16(MachineInstr &MI) const {
  if (STI.getLDSBankCount() != 16)
    return selectImpl(MI, *CoverageInfo);

  Register Dst = MI.getOperand(0).getReg();
  Register Src0 = MI.getOperand(2).getReg();
  Register M0Val = MI.getOperand(6).getReg();
  if (!RBI.constrainGenericRegister(M0Val, AMDGPU::SReg_32RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Dst, AMDGPU::VGPR_32RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Src0, AMDGPU::VGPR_32RegClass, *MRI))
    return false;

  // This requires 2 instructions. It is possible to write a pattern to support
  // this, but the generated isel emitter doesn't correctly deal with multiple
  // output instructions using the same physical register input. The copy to m0
  // is incorrectly placed before the second instruction.
  //
  // TODO: Match source modifiers.

  Register InterpMov = MRI->createVirtualRegister(&AMDGPU::VGPR_32RegClass);
  const DebugLoc &DL = MI.getDebugLoc();
  MachineBasicBlock *MBB = MI.getParent();

  BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
    .addReg(M0Val);
  BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::V_INTERP_MOV_F32), InterpMov)
    .addImm(2)
    .addImm(MI.getOperand(4).getImm())  // $attr
    .addImm(MI.getOperand(3).getImm()); // $attrchan

  BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::V_INTERP_P1LV_F16), Dst)
    .addImm(0)                          // $src0_modifiers
    .addReg(Src0)                       // $src0
    .addImm(MI.getOperand(4).getImm())  // $attr
    .addImm(MI.getOperand(3).getImm())  // $attrchan
    .addImm(0)                          // $src2_modifiers
    .addReg(InterpMov)                  // $src2 - 2 f16 values selected by high
    .addImm(MI.getOperand(5).getImm())  // $high
    .addImm(0)                          // $clamp
    .addImm(0);                         // $omod

  MI.eraseFromParent();
  return true;
}

// Writelane is special in that it can use SGPR and M0 (which would normally
// count as using the constant bus twice - but in this case it is allowed since
// the lane selector doesn't count as a use of the constant bus). However, it is
// still required to abide by the 1 SGPR rule. Fix this up if we might have
// multiple SGPRs.
bool AMDGPUInstructionSelector::selectWritelane(MachineInstr &MI) const {
  // With a constant bus limit of at least 2, there's no issue.
  if (STI.getConstantBusLimit(AMDGPU::V_WRITELANE_B32) > 1)
    return selectImpl(MI, *CoverageInfo);

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();
  Register VDst = MI.getOperand(0).getReg();
  Register Val = MI.getOperand(2).getReg();
  Register LaneSelect = MI.getOperand(3).getReg();
  Register VDstIn = MI.getOperand(4).getReg();

  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::V_WRITELANE_B32), VDst);

  Optional<ValueAndVReg> ConstSelect =
    getConstantVRegValWithLookThrough(LaneSelect, *MRI, true, true);
  if (ConstSelect) {
    // The selector has to be an inline immediate, so we can use whatever for
    // the other operands.
    MIB.addReg(Val);
    MIB.addImm(ConstSelect->Value &
               maskTrailingOnes<uint64_t>(STI.getWavefrontSizeLog2()));
  } else {
    Optional<ValueAndVReg> ConstVal =
      getConstantVRegValWithLookThrough(Val, *MRI, true, true);

    // If the value written is an inline immediate, we can get away without a
    // copy to m0.
    if (ConstVal && AMDGPU::isInlinableLiteral32(ConstVal->Value,
                                                 STI.hasInv2PiInlineImm())) {
      MIB.addImm(ConstVal->Value);
      MIB.addReg(LaneSelect);
    } else {
      MIB.addReg(Val);

      // If the lane selector was originally in a VGPR and copied with
      // readfirstlane, there's a hazard to read the same SGPR from the
      // VALU. Constrain to a different SGPR to help avoid needing a nop later.
      RBI.constrainGenericRegister(LaneSelect, AMDGPU::SReg_32_XM0RegClass, *MRI);

      BuildMI(*MBB, *MIB, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
        .addReg(LaneSelect);
      MIB.addReg(AMDGPU::M0);
    }
  }

  MIB.addReg(VDstIn);

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

// We need to handle this here because tablegen doesn't support matching
// instructions with multiple outputs.
bool AMDGPUInstructionSelector::selectDivScale(MachineInstr &MI) const {
  Register Dst0 = MI.getOperand(0).getReg();
  Register Dst1 = MI.getOperand(1).getReg();

  LLT Ty = MRI->getType(Dst0);
  unsigned Opc;
  if (Ty == LLT::scalar(32))
    Opc = AMDGPU::V_DIV_SCALE_F32;
  else if (Ty == LLT::scalar(64))
    Opc = AMDGPU::V_DIV_SCALE_F64;
  else
    return false;

  // TODO: Match source modifiers.

  const DebugLoc &DL = MI.getDebugLoc();
  MachineBasicBlock *MBB = MI.getParent();

  Register Numer = MI.getOperand(3).getReg();
  Register Denom = MI.getOperand(4).getReg();
  unsigned ChooseDenom = MI.getOperand(5).getImm();

  Register Src0 = ChooseDenom != 0 ? Numer : Denom;

  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(Opc), Dst0)
    .addDef(Dst1)
    .addImm(0)     // $src0_modifiers
    .addUse(Src0)  // $src0
    .addImm(0)     // $src1_modifiers
    .addUse(Denom) // $src1
    .addImm(0)     // $src2_modifiers
    .addUse(Numer) // $src2
    .addImm(0)     // $clamp
    .addImm(0);    // $omod

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

bool AMDGPUInstructionSelector::selectG_INTRINSIC(MachineInstr &I) const {
  unsigned IntrinsicID = I.getIntrinsicID();
  switch (IntrinsicID) {
  case Intrinsic::amdgcn_if_break: {
    MachineBasicBlock *BB = I.getParent();

    // FIXME: Manually selecting to avoid dealiing with the SReg_1 trick
    // SelectionDAG uses for wave32 vs wave64.
    BuildMI(*BB, &I, I.getDebugLoc(), TII.get(AMDGPU::SI_IF_BREAK))
      .add(I.getOperand(0))
      .add(I.getOperand(2))
      .add(I.getOperand(3));

    Register DstReg = I.getOperand(0).getReg();
    Register Src0Reg = I.getOperand(2).getReg();
    Register Src1Reg = I.getOperand(3).getReg();

    I.eraseFromParent();

    for (Register Reg : { DstReg, Src0Reg, Src1Reg })
      MRI->setRegClass(Reg, TRI.getWaveMaskRegClass());

    return true;
  }
  case Intrinsic::amdgcn_interp_p1_f16:
    return selectInterpP1F16(I);
  case Intrinsic::amdgcn_wqm:
    return constrainCopyLikeIntrin(I, AMDGPU::WQM);
  case Intrinsic::amdgcn_softwqm:
    return constrainCopyLikeIntrin(I, AMDGPU::SOFT_WQM);
  case Intrinsic::amdgcn_wwm:
    return constrainCopyLikeIntrin(I, AMDGPU::WWM);
  case Intrinsic::amdgcn_writelane:
    return selectWritelane(I);
  case Intrinsic::amdgcn_div_scale:
    return selectDivScale(I);
  case Intrinsic::amdgcn_icmp:
    return selectIntrinsicIcmp(I);
  case Intrinsic::amdgcn_ballot:
    return selectBallot(I);
  case Intrinsic::amdgcn_reloc_constant:
    return selectRelocConstant(I);
  case Intrinsic::amdgcn_groupstaticsize:
    return selectGroupStaticSize(I);
  case Intrinsic::returnaddress:
    return selectReturnAddress(I);
  default:
    return selectImpl(I, *CoverageInfo);
  }
}

static int getV_CMPOpcode(CmpInst::Predicate P, unsigned Size) {
  if (Size != 32 && Size != 64)
    return -1;
  switch (P) {
  default:
    llvm_unreachable("Unknown condition code!");
  case CmpInst::ICMP_NE:
    return Size == 32 ? AMDGPU::V_CMP_NE_U32_e64 : AMDGPU::V_CMP_NE_U64_e64;
  case CmpInst::ICMP_EQ:
    return Size == 32 ? AMDGPU::V_CMP_EQ_U32_e64 : AMDGPU::V_CMP_EQ_U64_e64;
  case CmpInst::ICMP_SGT:
    return Size == 32 ? AMDGPU::V_CMP_GT_I32_e64 : AMDGPU::V_CMP_GT_I64_e64;
  case CmpInst::ICMP_SGE:
    return Size == 32 ? AMDGPU::V_CMP_GE_I32_e64 : AMDGPU::V_CMP_GE_I64_e64;
  case CmpInst::ICMP_SLT:
    return Size == 32 ? AMDGPU::V_CMP_LT_I32_e64 : AMDGPU::V_CMP_LT_I64_e64;
  case CmpInst::ICMP_SLE:
    return Size == 32 ? AMDGPU::V_CMP_LE_I32_e64 : AMDGPU::V_CMP_LE_I64_e64;
  case CmpInst::ICMP_UGT:
    return Size == 32 ? AMDGPU::V_CMP_GT_U32_e64 : AMDGPU::V_CMP_GT_U64_e64;
  case CmpInst::ICMP_UGE:
    return Size == 32 ? AMDGPU::V_CMP_GE_U32_e64 : AMDGPU::V_CMP_GE_U64_e64;
  case CmpInst::ICMP_ULT:
    return Size == 32 ? AMDGPU::V_CMP_LT_U32_e64 : AMDGPU::V_CMP_LT_U64_e64;
  case CmpInst::ICMP_ULE:
    return Size == 32 ? AMDGPU::V_CMP_LE_U32_e64 : AMDGPU::V_CMP_LE_U64_e64;
  }
}

int AMDGPUInstructionSelector::getS_CMPOpcode(CmpInst::Predicate P,
                                              unsigned Size) const {
  if (Size == 64) {
    if (!STI.hasScalarCompareEq64())
      return -1;

    switch (P) {
    case CmpInst::ICMP_NE:
      return AMDGPU::S_CMP_LG_U64;
    case CmpInst::ICMP_EQ:
      return AMDGPU::S_CMP_EQ_U64;
    default:
      return -1;
    }
  }

  if (Size != 32)
    return -1;

  switch (P) {
  case CmpInst::ICMP_NE:
    return AMDGPU::S_CMP_LG_U32;
  case CmpInst::ICMP_EQ:
    return AMDGPU::S_CMP_EQ_U32;
  case CmpInst::ICMP_SGT:
    return AMDGPU::S_CMP_GT_I32;
  case CmpInst::ICMP_SGE:
    return AMDGPU::S_CMP_GE_I32;
  case CmpInst::ICMP_SLT:
    return AMDGPU::S_CMP_LT_I32;
  case CmpInst::ICMP_SLE:
    return AMDGPU::S_CMP_LE_I32;
  case CmpInst::ICMP_UGT:
    return AMDGPU::S_CMP_GT_U32;
  case CmpInst::ICMP_UGE:
    return AMDGPU::S_CMP_GE_U32;
  case CmpInst::ICMP_ULT:
    return AMDGPU::S_CMP_LT_U32;
  case CmpInst::ICMP_ULE:
    return AMDGPU::S_CMP_LE_U32;
  default:
    llvm_unreachable("Unknown condition code!");
  }
}

bool AMDGPUInstructionSelector::selectG_ICMP(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();

  Register SrcReg = I.getOperand(2).getReg();
  unsigned Size = RBI.getSizeInBits(SrcReg, *MRI, TRI);

  auto Pred = (CmpInst::Predicate)I.getOperand(1).getPredicate();

  Register CCReg = I.getOperand(0).getReg();
  if (!isVCC(CCReg, *MRI)) {
    int Opcode = getS_CMPOpcode(Pred, Size);
    if (Opcode == -1)
      return false;
    MachineInstr *ICmp = BuildMI(*BB, &I, DL, TII.get(Opcode))
            .add(I.getOperand(2))
            .add(I.getOperand(3));
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), CCReg)
      .addReg(AMDGPU::SCC);
    bool Ret =
        constrainSelectedInstRegOperands(*ICmp, TII, TRI, RBI) &&
        RBI.constrainGenericRegister(CCReg, AMDGPU::SReg_32RegClass, *MRI);
    I.eraseFromParent();
    return Ret;
  }

  int Opcode = getV_CMPOpcode(Pred, Size);
  if (Opcode == -1)
    return false;

  MachineInstr *ICmp = BuildMI(*BB, &I, DL, TII.get(Opcode),
            I.getOperand(0).getReg())
            .add(I.getOperand(2))
            .add(I.getOperand(3));
  RBI.constrainGenericRegister(ICmp->getOperand(0).getReg(),
                               *TRI.getBoolRC(), *MRI);
  bool Ret = constrainSelectedInstRegOperands(*ICmp, TII, TRI, RBI);
  I.eraseFromParent();
  return Ret;
}

bool AMDGPUInstructionSelector::selectIntrinsicIcmp(MachineInstr &I) const {
  Register Dst = I.getOperand(0).getReg();
  if (isVCC(Dst, *MRI))
    return false;

  if (MRI->getType(Dst).getSizeInBits() != STI.getWavefrontSize())
    return false;

  MachineBasicBlock *BB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();
  Register SrcReg = I.getOperand(2).getReg();
  unsigned Size = RBI.getSizeInBits(SrcReg, *MRI, TRI);
  auto Pred = static_cast<CmpInst::Predicate>(I.getOperand(4).getImm());

  int Opcode = getV_CMPOpcode(Pred, Size);
  if (Opcode == -1)
    return false;

  MachineInstr *ICmp = BuildMI(*BB, &I, DL, TII.get(Opcode), Dst)
                           .add(I.getOperand(2))
                           .add(I.getOperand(3));
  RBI.constrainGenericRegister(ICmp->getOperand(0).getReg(), *TRI.getBoolRC(),
                               *MRI);
  bool Ret = constrainSelectedInstRegOperands(*ICmp, TII, TRI, RBI);
  I.eraseFromParent();
  return Ret;
}

bool AMDGPUInstructionSelector::selectBallot(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();
  Register DstReg = I.getOperand(0).getReg();
  const unsigned Size = MRI->getType(DstReg).getSizeInBits();
  const bool Is64 = Size == 64;

  if (Size != STI.getWavefrontSize())
    return false;

  Optional<ValueAndVReg> Arg =
      getConstantVRegValWithLookThrough(I.getOperand(2).getReg(), *MRI, true);

  if (Arg.hasValue()) {
    const int64_t Value = Arg.getValue().Value;
    if (Value == 0) {
      unsigned Opcode = Is64 ? AMDGPU::S_MOV_B64 : AMDGPU::S_MOV_B32;
      BuildMI(*BB, &I, DL, TII.get(Opcode), DstReg).addImm(0);
    } else if (Value == -1) { // all ones
      Register SrcReg = Is64 ? AMDGPU::EXEC : AMDGPU::EXEC_LO;
      BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), DstReg).addReg(SrcReg);
    } else
      return false;
  } else {
    Register SrcReg = I.getOperand(2).getReg();
    BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), DstReg).addReg(SrcReg);
  }

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectRelocConstant(MachineInstr &I) const {
  Register DstReg = I.getOperand(0).getReg();
  const RegisterBank *DstBank = RBI.getRegBank(DstReg, *MRI, TRI);
  const TargetRegisterClass *DstRC =
    TRI.getRegClassForSizeOnBank(32, *DstBank, *MRI);
  if (!DstRC || !RBI.constrainGenericRegister(DstReg, *DstRC, *MRI))
    return false;

  const bool IsVALU = DstBank->getID() == AMDGPU::VGPRRegBankID;

  Module *M = MF->getFunction().getParent();
  const MDNode *Metadata = I.getOperand(2).getMetadata();
  auto SymbolName = cast<MDString>(Metadata->getOperand(0))->getString();
  auto RelocSymbol = cast<GlobalVariable>(
    M->getOrInsertGlobal(SymbolName, Type::getInt32Ty(M->getContext())));

  MachineBasicBlock *BB = I.getParent();
  BuildMI(*BB, &I, I.getDebugLoc(),
          TII.get(IsVALU ? AMDGPU::V_MOV_B32_e32 : AMDGPU::S_MOV_B32), DstReg)
    .addGlobalAddress(RelocSymbol, 0, SIInstrInfo::MO_ABS32_LO);

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectGroupStaticSize(MachineInstr &I) const {
  Triple::OSType OS = MF->getTarget().getTargetTriple().getOS();

  Register DstReg = I.getOperand(0).getReg();
  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  unsigned Mov = DstRB->getID() == AMDGPU::SGPRRegBankID ?
    AMDGPU::S_MOV_B32 : AMDGPU::V_MOV_B32_e32;

  MachineBasicBlock *MBB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();

  auto MIB = BuildMI(*MBB, &I, DL, TII.get(Mov), DstReg);

  if (OS == Triple::AMDHSA || OS == Triple::AMDPAL) {
    const SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
    MIB.addImm(MFI->getLDSSize());
  } else {
    Module *M = MF->getFunction().getParent();
    const GlobalValue *GV
      = Intrinsic::getDeclaration(M, Intrinsic::amdgcn_groupstaticsize);
    MIB.addGlobalAddress(GV, 0, SIInstrInfo::MO_ABS32_LO);
  }

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

bool AMDGPUInstructionSelector::selectReturnAddress(MachineInstr &I) const {
  MachineBasicBlock *MBB = I.getParent();
  MachineFunction &MF = *MBB->getParent();
  const DebugLoc &DL = I.getDebugLoc();

  MachineOperand &Dst = I.getOperand(0);
  Register DstReg = Dst.getReg();
  unsigned Depth = I.getOperand(2).getImm();

  const TargetRegisterClass *RC
    = TRI.getConstrainedRegClassForOperand(Dst, *MRI);
  if (!RC->hasSubClassEq(&AMDGPU::SGPR_64RegClass) ||
      !RBI.constrainGenericRegister(DstReg, *RC, *MRI))
    return false;

  // Check for kernel and shader functions
  if (Depth != 0 ||
      MF.getInfo<SIMachineFunctionInfo>()->isEntryFunction()) {
    BuildMI(*MBB, &I, DL, TII.get(AMDGPU::S_MOV_B64), DstReg)
      .addImm(0);
    I.eraseFromParent();
    return true;
  }

  MachineFrameInfo &MFI = MF.getFrameInfo();
  // There is a call to @llvm.returnaddress in this function
  MFI.setReturnAddressIsTaken(true);

  // Get the return address reg and mark it as an implicit live-in
  Register ReturnAddrReg = TRI.getReturnAddressReg(MF);
  Register LiveIn = getFunctionLiveInPhysReg(MF, TII, ReturnAddrReg,
                                             AMDGPU::SReg_64RegClass);
  BuildMI(*MBB, &I, DL, TII.get(AMDGPU::COPY), DstReg)
    .addReg(LiveIn);
  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectEndCfIntrinsic(MachineInstr &MI) const {
  // FIXME: Manually selecting to avoid dealiing with the SReg_1 trick
  // SelectionDAG uses for wave32 vs wave64.
  MachineBasicBlock *BB = MI.getParent();
  BuildMI(*BB, &MI, MI.getDebugLoc(), TII.get(AMDGPU::SI_END_CF))
      .add(MI.getOperand(1));

  Register Reg = MI.getOperand(1).getReg();
  MI.eraseFromParent();

  if (!MRI->getRegClassOrNull(Reg))
    MRI->setRegClass(Reg, TRI.getWaveMaskRegClass());
  return true;
}

bool AMDGPUInstructionSelector::selectDSOrderedIntrinsic(
  MachineInstr &MI, Intrinsic::ID IntrID) const {
  MachineBasicBlock *MBB = MI.getParent();
  MachineFunction *MF = MBB->getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  unsigned IndexOperand = MI.getOperand(7).getImm();
  bool WaveRelease = MI.getOperand(8).getImm() != 0;
  bool WaveDone = MI.getOperand(9).getImm() != 0;

  if (WaveDone && !WaveRelease)
    report_fatal_error("ds_ordered_count: wave_done requires wave_release");

  unsigned OrderedCountIndex = IndexOperand & 0x3f;
  IndexOperand &= ~0x3f;
  unsigned CountDw = 0;

  if (STI.getGeneration() >= AMDGPUSubtarget::GFX10) {
    CountDw = (IndexOperand >> 24) & 0xf;
    IndexOperand &= ~(0xf << 24);

    if (CountDw < 1 || CountDw > 4) {
      report_fatal_error(
        "ds_ordered_count: dword count must be between 1 and 4");
    }
  }

  if (IndexOperand)
    report_fatal_error("ds_ordered_count: bad index operand");

  unsigned Instruction = IntrID == Intrinsic::amdgcn_ds_ordered_add ? 0 : 1;
  unsigned ShaderType = SIInstrInfo::getDSShaderTypeValue(*MF);

  unsigned Offset0 = OrderedCountIndex << 2;
  unsigned Offset1 = WaveRelease | (WaveDone << 1) | (ShaderType << 2) |
                     (Instruction << 4);

  if (STI.getGeneration() >= AMDGPUSubtarget::GFX10)
    Offset1 |= (CountDw - 1) << 6;

  unsigned Offset = Offset0 | (Offset1 << 8);

  Register M0Val = MI.getOperand(2).getReg();
  BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
    .addReg(M0Val);

  Register DstReg = MI.getOperand(0).getReg();
  Register ValReg = MI.getOperand(3).getReg();
  MachineInstrBuilder DS =
    BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::DS_ORDERED_COUNT), DstReg)
      .addReg(ValReg)
      .addImm(Offset)
      .cloneMemRefs(MI);

  if (!RBI.constrainGenericRegister(M0Val, AMDGPU::SReg_32RegClass, *MRI))
    return false;

  bool Ret = constrainSelectedInstRegOperands(*DS, TII, TRI, RBI);
  MI.eraseFromParent();
  return Ret;
}

static unsigned gwsIntrinToOpcode(unsigned IntrID) {
  switch (IntrID) {
  case Intrinsic::amdgcn_ds_gws_init:
    return AMDGPU::DS_GWS_INIT;
  case Intrinsic::amdgcn_ds_gws_barrier:
    return AMDGPU::DS_GWS_BARRIER;
  case Intrinsic::amdgcn_ds_gws_sema_v:
    return AMDGPU::DS_GWS_SEMA_V;
  case Intrinsic::amdgcn_ds_gws_sema_br:
    return AMDGPU::DS_GWS_SEMA_BR;
  case Intrinsic::amdgcn_ds_gws_sema_p:
    return AMDGPU::DS_GWS_SEMA_P;
  case Intrinsic::amdgcn_ds_gws_sema_release_all:
    return AMDGPU::DS_GWS_SEMA_RELEASE_ALL;
  default:
    llvm_unreachable("not a gws intrinsic");
  }
}

bool AMDGPUInstructionSelector::selectDSGWSIntrinsic(MachineInstr &MI,
                                                     Intrinsic::ID IID) const {
  if (IID == Intrinsic::amdgcn_ds_gws_sema_release_all &&
      !STI.hasGWSSemaReleaseAll())
    return false;

  // intrinsic ID, vsrc, offset
  const bool HasVSrc = MI.getNumOperands() == 3;
  assert(HasVSrc || MI.getNumOperands() == 2);

  Register BaseOffset = MI.getOperand(HasVSrc ? 2 : 1).getReg();
  const RegisterBank *OffsetRB = RBI.getRegBank(BaseOffset, *MRI, TRI);
  if (OffsetRB->getID() != AMDGPU::SGPRRegBankID)
    return false;

  MachineInstr *OffsetDef = getDefIgnoringCopies(BaseOffset, *MRI);
  assert(OffsetDef);

  unsigned ImmOffset;

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  MachineInstr *Readfirstlane = nullptr;

  // If we legalized the VGPR input, strip out the readfirstlane to analyze the
  // incoming offset, in case there's an add of a constant. We'll have to put it
  // back later.
  if (OffsetDef->getOpcode() == AMDGPU::V_READFIRSTLANE_B32) {
    Readfirstlane = OffsetDef;
    BaseOffset = OffsetDef->getOperand(1).getReg();
    OffsetDef = getDefIgnoringCopies(BaseOffset, *MRI);
  }

  if (OffsetDef->getOpcode() == AMDGPU::G_CONSTANT) {
    // If we have a constant offset, try to use the 0 in m0 as the base.
    // TODO: Look into changing the default m0 initialization value. If the
    // default -1 only set the low 16-bits, we could leave it as-is and add 1 to
    // the immediate offset.

    ImmOffset = OffsetDef->getOperand(1).getCImm()->getZExtValue();
    BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::S_MOV_B32), AMDGPU::M0)
      .addImm(0);
  } else {
    std::tie(BaseOffset, ImmOffset) =
        AMDGPU::getBaseWithConstantOffset(*MRI, BaseOffset);

    if (Readfirstlane) {
      // We have the constant offset now, so put the readfirstlane back on the
      // variable component.
      if (!RBI.constrainGenericRegister(BaseOffset, AMDGPU::VGPR_32RegClass, *MRI))
        return false;

      Readfirstlane->getOperand(1).setReg(BaseOffset);
      BaseOffset = Readfirstlane->getOperand(0).getReg();
    } else {
      if (!RBI.constrainGenericRegister(BaseOffset,
                                        AMDGPU::SReg_32RegClass, *MRI))
        return false;
    }

    Register M0Base = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
    BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::S_LSHL_B32), M0Base)
      .addReg(BaseOffset)
      .addImm(16);

    BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
      .addReg(M0Base);
  }

  // The resource id offset is computed as (<isa opaque base> + M0[21:16] +
  // offset field) % 64. Some versions of the programming guide omit the m0
  // part, or claim it's from offset 0.
  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(gwsIntrinToOpcode(IID)));

  if (HasVSrc) {
    Register VSrc = MI.getOperand(1).getReg();
    MIB.addReg(VSrc);
    if (!RBI.constrainGenericRegister(VSrc, AMDGPU::VGPR_32RegClass, *MRI))
      return false;
  }

  MIB.addImm(ImmOffset)
     .cloneMemRefs(MI);

  MI.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectDSAppendConsume(MachineInstr &MI,
                                                      bool IsAppend) const {
  Register PtrBase = MI.getOperand(2).getReg();
  LLT PtrTy = MRI->getType(PtrBase);
  bool IsGDS = PtrTy.getAddressSpace() == AMDGPUAS::REGION_ADDRESS;

  unsigned Offset;
  std::tie(PtrBase, Offset) = selectDS1Addr1OffsetImpl(MI.getOperand(2));

  // TODO: Should this try to look through readfirstlane like GWS?
  if (!isDSOffsetLegal(PtrBase, Offset)) {
    PtrBase = MI.getOperand(2).getReg();
    Offset = 0;
  }

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();
  const unsigned Opc = IsAppend ? AMDGPU::DS_APPEND : AMDGPU::DS_CONSUME;

  BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
    .addReg(PtrBase);
  if (!RBI.constrainGenericRegister(PtrBase, AMDGPU::SReg_32RegClass, *MRI))
    return false;

  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(Opc), MI.getOperand(0).getReg())
    .addImm(Offset)
    .addImm(IsGDS ? -1 : 0)
    .cloneMemRefs(MI);
  MI.eraseFromParent();
  return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}

bool AMDGPUInstructionSelector::selectSBarrier(MachineInstr &MI) const {
  if (TM.getOptLevel() > CodeGenOpt::None) {
    unsigned WGSize = STI.getFlatWorkGroupSizes(MF->getFunction()).second;
    if (WGSize <= STI.getWavefrontSize()) {
      MachineBasicBlock *MBB = MI.getParent();
      const DebugLoc &DL = MI.getDebugLoc();
      BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::WAVE_BARRIER));
      MI.eraseFromParent();
      return true;
    }
  }
  return selectImpl(MI, *CoverageInfo);
}

static bool parseTexFail(uint64_t TexFailCtrl, bool &TFE, bool &LWE,
                         bool &IsTexFail) {
  if (TexFailCtrl)
    IsTexFail = true;

  TFE = (TexFailCtrl & 0x1) ? 1 : 0;
  TexFailCtrl &= ~(uint64_t)0x1;
  LWE = (TexFailCtrl & 0x2) ? 1 : 0;
  TexFailCtrl &= ~(uint64_t)0x2;

  return TexFailCtrl == 0;
}

static bool parseCachePolicy(uint64_t Value,
                             bool *GLC, bool *SLC, bool *DLC) {
  if (GLC) {
    *GLC = (Value & 0x1) ? 1 : 0;
    Value &= ~(uint64_t)0x1;
  }
  if (SLC) {
    *SLC = (Value & 0x2) ? 1 : 0;
    Value &= ~(uint64_t)0x2;
  }
  if (DLC) {
    *DLC = (Value & 0x4) ? 1 : 0;
    Value &= ~(uint64_t)0x4;
  }

  return Value == 0;
}

bool AMDGPUInstructionSelector::selectImageIntrinsic(
  MachineInstr &MI, const AMDGPU::ImageDimIntrinsicInfo *Intr) const {
  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  const AMDGPU::MIMGBaseOpcodeInfo *BaseOpcode =
    AMDGPU::getMIMGBaseOpcodeInfo(Intr->BaseOpcode);

  const AMDGPU::MIMGDimInfo *DimInfo = AMDGPU::getMIMGDimInfo(Intr->Dim);
  const AMDGPU::MIMGLZMappingInfo *LZMappingInfo =
      AMDGPU::getMIMGLZMappingInfo(Intr->BaseOpcode);
  const AMDGPU::MIMGMIPMappingInfo *MIPMappingInfo =
      AMDGPU::getMIMGMIPMappingInfo(Intr->BaseOpcode);
  unsigned IntrOpcode = Intr->BaseOpcode;
  const bool IsGFX10Plus = AMDGPU::isGFX10Plus(STI);

  const unsigned ArgOffset = MI.getNumExplicitDefs() + 1;

  Register VDataIn, VDataOut;
  LLT VDataTy;
  int NumVDataDwords = -1;
  bool IsD16 = false;

  bool Unorm;
  if (!BaseOpcode->Sampler)
    Unorm = true;
  else
    Unorm = MI.getOperand(ArgOffset + Intr->UnormIndex).getImm() != 0;

  bool TFE;
  bool LWE;
  bool IsTexFail = false;
  if (!parseTexFail(MI.getOperand(ArgOffset + Intr->TexFailCtrlIndex).getImm(),
                    TFE, LWE, IsTexFail))
    return false;

  const int Flags = MI.getOperand(ArgOffset + Intr->NumArgs).getImm();
  const bool IsA16 = (Flags & 1) != 0;
  const bool IsG16 = (Flags & 2) != 0;

  // A16 implies 16 bit gradients
  if (IsA16 && !IsG16)
    return false;

  unsigned DMask = 0;
  unsigned DMaskLanes = 0;

  if (BaseOpcode->Atomic) {
    VDataOut = MI.getOperand(0).getReg();
    VDataIn = MI.getOperand(2).getReg();
    LLT Ty = MRI->getType(VDataIn);

    // Be careful to allow atomic swap on 16-bit element vectors.
    const bool Is64Bit = BaseOpcode->AtomicX2 ?
      Ty.getSizeInBits() == 128 :
      Ty.getSizeInBits() == 64;

    if (BaseOpcode->AtomicX2) {
      assert(MI.getOperand(3).getReg() == AMDGPU::NoRegister);

      DMask = Is64Bit ? 0xf : 0x3;
      NumVDataDwords = Is64Bit ? 4 : 2;
    } else {
      DMask = Is64Bit ? 0x3 : 0x1;
      NumVDataDwords = Is64Bit ? 2 : 1;
    }
  } else {
    DMask = MI.getOperand(ArgOffset + Intr->DMaskIndex).getImm();
    DMaskLanes = BaseOpcode->Gather4 ? 4 : countPopulation(DMask);

    // One memoperand is mandatory, except for getresinfo.
    // FIXME: Check this in verifier.
    if (!MI.memoperands_empty()) {
      const MachineMemOperand *MMO = *MI.memoperands_begin();

      // Infer d16 from the memory size, as the register type will be mangled by
      // unpacked subtargets, or by TFE.
      IsD16 = ((8 * MMO->getSize()) / DMaskLanes) < 32;
    }

    if (BaseOpcode->Store) {
      VDataIn = MI.getOperand(1).getReg();
      VDataTy = MRI->getType(VDataIn);
      NumVDataDwords = (VDataTy.getSizeInBits() + 31) / 32;
    } else {
      VDataOut = MI.getOperand(0).getReg();
      VDataTy = MRI->getType(VDataOut);
      NumVDataDwords = DMaskLanes;

      if (IsD16 && !STI.hasUnpackedD16VMem())
        NumVDataDwords = (DMaskLanes + 1) / 2;
    }
  }

  // Optimize _L to _LZ when _L is zero
  if (LZMappingInfo) {
    // The legalizer replaced the register with an immediate 0 if we need to
    // change the opcode.
    const MachineOperand &Lod = MI.getOperand(ArgOffset + Intr->LodIndex);
    if (Lod.isImm()) {
      assert(Lod.getImm() == 0);
      IntrOpcode = LZMappingInfo->LZ;  // set new opcode to _lz variant of _l
    }
  }

  // Optimize _mip away, when 'lod' is zero
  if (MIPMappingInfo) {
    const MachineOperand &Lod = MI.getOperand(ArgOffset + Intr->MipIndex);
    if (Lod.isImm()) {
      assert(Lod.getImm() == 0);
      IntrOpcode = MIPMappingInfo->NONMIP;  // set new opcode to variant without _mip
    }
  }

  // Set G16 opcode
  if (IsG16 && !IsA16) {
    const AMDGPU::MIMGG16MappingInfo *G16MappingInfo =
        AMDGPU::getMIMGG16MappingInfo(Intr->BaseOpcode);
    assert(G16MappingInfo);
    IntrOpcode = G16MappingInfo->G16; // set opcode to variant with _g16
  }

  // TODO: Check this in verifier.
  assert((!IsTexFail || DMaskLanes >= 1) && "should have legalized this");

  bool GLC = false;
  bool SLC = false;
  bool DLC = false;
  if (BaseOpcode->Atomic) {
    GLC = true; // TODO no-return optimization
    if (!parseCachePolicy(
            MI.getOperand(ArgOffset + Intr->CachePolicyIndex).getImm(), nullptr,
            &SLC, IsGFX10Plus ? &DLC : nullptr))
      return false;
  } else {
    if (!parseCachePolicy(
            MI.getOperand(ArgOffset + Intr->CachePolicyIndex).getImm(), &GLC,
            &SLC, IsGFX10Plus ? &DLC : nullptr))
      return false;
  }

  int NumVAddrRegs = 0;
  int NumVAddrDwords = 0;
  for (unsigned I = Intr->VAddrStart; I < Intr->VAddrEnd; I++) {
    // Skip the $noregs and 0s inserted during legalization.
    MachineOperand &AddrOp = MI.getOperand(ArgOffset + I);
    if (!AddrOp.isReg())
      continue; // XXX - Break?

    Register Addr = AddrOp.getReg();
    if (!Addr)
      break;

    ++NumVAddrRegs;
    NumVAddrDwords += (MRI->getType(Addr).getSizeInBits() + 31) / 32;
  }

  // The legalizer preprocessed the intrinsic arguments. If we aren't using
  // NSA, these should have beeen packed into a single value in the first
  // address register
  const bool UseNSA = NumVAddrRegs != 1 && NumVAddrDwords == NumVAddrRegs;
  if (UseNSA && !STI.hasFeature(AMDGPU::FeatureNSAEncoding)) {
    LLVM_DEBUG(dbgs() << "Trying to use NSA on non-NSA target\n");
    return false;
  }

  if (IsTexFail)
    ++NumVDataDwords;

  int Opcode = -1;
  if (IsGFX10Plus) {
    Opcode = AMDGPU::getMIMGOpcode(IntrOpcode,
                                   UseNSA ? AMDGPU::MIMGEncGfx10NSA
                                          : AMDGPU::MIMGEncGfx10Default,
                                   NumVDataDwords, NumVAddrDwords);
  } else {
    if (STI.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS)
      Opcode = AMDGPU::getMIMGOpcode(IntrOpcode, AMDGPU::MIMGEncGfx8,
                                     NumVDataDwords, NumVAddrDwords);
    if (Opcode == -1)
      Opcode = AMDGPU::getMIMGOpcode(IntrOpcode, AMDGPU::MIMGEncGfx6,
                                     NumVDataDwords, NumVAddrDwords);
  }
  assert(Opcode != -1);

  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(Opcode))
    .cloneMemRefs(MI);

  if (VDataOut) {
    if (BaseOpcode->AtomicX2) {
      const bool Is64 = MRI->getType(VDataOut).getSizeInBits() == 64;

      Register TmpReg = MRI->createVirtualRegister(
        Is64 ? &AMDGPU::VReg_128RegClass : &AMDGPU::VReg_64RegClass);
      unsigned SubReg = Is64 ? AMDGPU::sub0_sub1 : AMDGPU::sub0;

      MIB.addDef(TmpReg);
      BuildMI(*MBB, &MI, DL, TII.get(AMDGPU::COPY), VDataOut)
        .addReg(TmpReg, RegState::Kill, SubReg);

    } else {
      MIB.addDef(VDataOut); // vdata output
    }
  }

  if (VDataIn)
    MIB.addReg(VDataIn); // vdata input

  for (int I = 0; I != NumVAddrRegs; ++I) {
    MachineOperand &SrcOp = MI.getOperand(ArgOffset + Intr->VAddrStart + I);
    if (SrcOp.isReg()) {
      assert(SrcOp.getReg() != 0);
      MIB.addReg(SrcOp.getReg());
    }
  }

  MIB.addReg(MI.getOperand(ArgOffset + Intr->RsrcIndex).getReg());
  if (BaseOpcode->Sampler)
    MIB.addReg(MI.getOperand(ArgOffset + Intr->SampIndex).getReg());

  MIB.addImm(DMask); // dmask

  if (IsGFX10Plus)
    MIB.addImm(DimInfo->Encoding);
  MIB.addImm(Unorm);
  if (IsGFX10Plus)
    MIB.addImm(DLC);

  MIB.addImm(GLC);
  MIB.addImm(SLC);
  MIB.addImm(IsA16 &&  // a16 or r128
             STI.hasFeature(AMDGPU::FeatureR128A16) ? -1 : 0);
  if (IsGFX10Plus)
    MIB.addImm(IsA16 ? -1 : 0);

  MIB.addImm(TFE); // tfe
  MIB.addImm(LWE); // lwe
  if (!IsGFX10Plus)
    MIB.addImm(DimInfo->DA ? -1 : 0);
  if (BaseOpcode->HasD16)
    MIB.addImm(IsD16 ? -1 : 0);

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

bool AMDGPUInstructionSelector::selectG_INTRINSIC_W_SIDE_EFFECTS(
    MachineInstr &I) const {
  unsigned IntrinsicID = I.getIntrinsicID();
  switch (IntrinsicID) {
  case Intrinsic::amdgcn_end_cf:
    return selectEndCfIntrinsic(I);
  case Intrinsic::amdgcn_ds_ordered_add:
  case Intrinsic::amdgcn_ds_ordered_swap:
    return selectDSOrderedIntrinsic(I, IntrinsicID);
  case Intrinsic::amdgcn_ds_gws_init:
  case Intrinsic::amdgcn_ds_gws_barrier:
  case Intrinsic::amdgcn_ds_gws_sema_v:
  case Intrinsic::amdgcn_ds_gws_sema_br:
  case Intrinsic::amdgcn_ds_gws_sema_p:
  case Intrinsic::amdgcn_ds_gws_sema_release_all:
    return selectDSGWSIntrinsic(I, IntrinsicID);
  case Intrinsic::amdgcn_ds_append:
    return selectDSAppendConsume(I, true);
  case Intrinsic::amdgcn_ds_consume:
    return selectDSAppendConsume(I, false);
  case Intrinsic::amdgcn_s_barrier:
    return selectSBarrier(I);
  case Intrinsic::amdgcn_global_atomic_fadd:
    return selectGlobalAtomicFaddIntrinsic(I);
  default: {
    return selectImpl(I, *CoverageInfo);
  }
  }
}

bool AMDGPUInstructionSelector::selectG_SELECT(MachineInstr &I) const {
  if (selectImpl(I, *CoverageInfo))
    return true;

  MachineBasicBlock *BB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();

  Register DstReg = I.getOperand(0).getReg();
  unsigned Size = RBI.getSizeInBits(DstReg, *MRI, TRI);
  assert(Size <= 32 || Size == 64);
  const MachineOperand &CCOp = I.getOperand(1);
  Register CCReg = CCOp.getReg();
  if (!isVCC(CCReg, *MRI)) {
    unsigned SelectOpcode = Size == 64 ? AMDGPU::S_CSELECT_B64 :
                                         AMDGPU::S_CSELECT_B32;
    MachineInstr *CopySCC = BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), AMDGPU::SCC)
            .addReg(CCReg);

    // The generic constrainSelectedInstRegOperands doesn't work for the scc register
    // bank, because it does not cover the register class that we used to represent
    // for it.  So we need to manually set the register class here.
    if (!MRI->getRegClassOrNull(CCReg))
        MRI->setRegClass(CCReg, TRI.getConstrainedRegClassForOperand(CCOp, *MRI));
    MachineInstr *Select = BuildMI(*BB, &I, DL, TII.get(SelectOpcode), DstReg)
            .add(I.getOperand(2))
            .add(I.getOperand(3));

    bool Ret = constrainSelectedInstRegOperands(*Select, TII, TRI, RBI) |
               constrainSelectedInstRegOperands(*CopySCC, TII, TRI, RBI);
    I.eraseFromParent();
    return Ret;
  }

  // Wide VGPR select should have been split in RegBankSelect.
  if (Size > 32)
    return false;

  MachineInstr *Select =
      BuildMI(*BB, &I, DL, TII.get(AMDGPU::V_CNDMASK_B32_e64), DstReg)
              .addImm(0)
              .add(I.getOperand(3))
              .addImm(0)
              .add(I.getOperand(2))
              .add(I.getOperand(1));

  bool Ret = constrainSelectedInstRegOperands(*Select, TII, TRI, RBI);
  I.eraseFromParent();
  return Ret;
}

static int sizeToSubRegIndex(unsigned Size) {
  switch (Size) {
  case 32:
    return AMDGPU::sub0;
  case 64:
    return AMDGPU::sub0_sub1;
  case 96:
    return AMDGPU::sub0_sub1_sub2;
  case 128:
    return AMDGPU::sub0_sub1_sub2_sub3;
  case 256:
    return AMDGPU::sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7;
  default:
    if (Size < 32)
      return AMDGPU::sub0;
    if (Size > 256)
      return -1;
    return sizeToSubRegIndex(PowerOf2Ceil(Size));
  }
}

bool AMDGPUInstructionSelector::selectG_TRUNC(MachineInstr &I) const {
  Register DstReg = I.getOperand(0).getReg();
  Register SrcReg = I.getOperand(1).getReg();
  const LLT DstTy = MRI->getType(DstReg);
  const LLT SrcTy = MRI->getType(SrcReg);
  const LLT S1 = LLT::scalar(1);

  const RegisterBank *SrcRB = RBI.getRegBank(SrcReg, *MRI, TRI);
  const RegisterBank *DstRB;
  if (DstTy == S1) {
    // This is a special case. We don't treat s1 for legalization artifacts as
    // vcc booleans.
    DstRB = SrcRB;
  } else {
    DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
    if (SrcRB != DstRB)
      return false;
  }

  const bool IsVALU = DstRB->getID() == AMDGPU::VGPRRegBankID;

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

  const TargetRegisterClass *SrcRC
    = TRI.getRegClassForSizeOnBank(SrcSize, *SrcRB, *MRI);
  const TargetRegisterClass *DstRC
    = TRI.getRegClassForSizeOnBank(DstSize, *DstRB, *MRI);
  if (!SrcRC || !DstRC)
    return false;

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

  if (DstTy == LLT::vector(2, 16) && SrcTy == LLT::vector(2, 32)) {
    MachineBasicBlock *MBB = I.getParent();
    const DebugLoc &DL = I.getDebugLoc();

    Register LoReg = MRI->createVirtualRegister(DstRC);
    Register HiReg = MRI->createVirtualRegister(DstRC);
    BuildMI(*MBB, I, DL, TII.get(AMDGPU::COPY), LoReg)
      .addReg(SrcReg, 0, AMDGPU::sub0);
    BuildMI(*MBB, I, DL, TII.get(AMDGPU::COPY), HiReg)
      .addReg(SrcReg, 0, AMDGPU::sub1);

    if (IsVALU && STI.hasSDWA()) {
      // Write the low 16-bits of the high element into the high 16-bits of the
      // low element.
      MachineInstr *MovSDWA =
        BuildMI(*MBB, I, DL, TII.get(AMDGPU::V_MOV_B32_sdwa), DstReg)
        .addImm(0)                             // $src0_modifiers
        .addReg(HiReg)                         // $src0
        .addImm(0)                             // $clamp
        .addImm(AMDGPU::SDWA::WORD_1)          // $dst_sel
        .addImm(AMDGPU::SDWA::UNUSED_PRESERVE) // $dst_unused
        .addImm(AMDGPU::SDWA::WORD_0)          // $src0_sel
        .addReg(LoReg, RegState::Implicit);
      MovSDWA->tieOperands(0, MovSDWA->getNumOperands() - 1);
    } else {
      Register TmpReg0 = MRI->createVirtualRegister(DstRC);
      Register TmpReg1 = MRI->createVirtualRegister(DstRC);
      Register ImmReg = MRI->createVirtualRegister(DstRC);
      if (IsVALU) {
        BuildMI(*MBB, I, DL, TII.get(AMDGPU::V_LSHLREV_B32_e64), TmpReg0)
          .addImm(16)
          .addReg(HiReg);
      } else {
        BuildMI(*MBB, I, DL, TII.get(AMDGPU::S_LSHL_B32), TmpReg0)
          .addReg(HiReg)
          .addImm(16);
      }

      unsigned MovOpc = IsVALU ? AMDGPU::V_MOV_B32_e32 : AMDGPU::S_MOV_B32;
      unsigned AndOpc = IsVALU ? AMDGPU::V_AND_B32_e64 : AMDGPU::S_AND_B32;
      unsigned OrOpc = IsVALU ? AMDGPU::V_OR_B32_e64 : AMDGPU::S_OR_B32;

      BuildMI(*MBB, I, DL, TII.get(MovOpc), ImmReg)
        .addImm(0xffff);
      BuildMI(*MBB, I, DL, TII.get(AndOpc), TmpReg1)
        .addReg(LoReg)
        .addReg(ImmReg);
      BuildMI(*MBB, I, DL, TII.get(OrOpc), DstReg)
        .addReg(TmpReg0)
        .addReg(TmpReg1);
    }

    I.eraseFromParent();
    return true;
  }

  if (!DstTy.isScalar())
    return false;

  if (SrcSize > 32) {
    int SubRegIdx = sizeToSubRegIndex(DstSize);
    if (SubRegIdx == -1)
      return false;

    // Deal with weird cases where the class only partially supports the subreg
    // index.
    const TargetRegisterClass *SrcWithSubRC
      = TRI.getSubClassWithSubReg(SrcRC, SubRegIdx);
    if (!SrcWithSubRC)
      return false;

    if (SrcWithSubRC != SrcRC) {
      if (!RBI.constrainGenericRegister(SrcReg, *SrcWithSubRC, *MRI))
        return false;
    }

    I.getOperand(1).setSubReg(SubRegIdx);
  }

  I.setDesc(TII.get(TargetOpcode::COPY));
  return true;
}

/// \returns true if a bitmask for \p Size bits will be an inline immediate.
static bool shouldUseAndMask(unsigned Size, unsigned &Mask) {
  Mask = maskTrailingOnes<unsigned>(Size);
  int SignedMask = static_cast<int>(Mask);
  return SignedMask >= -16 && SignedMask <= 64;
}

// Like RegisterBankInfo::getRegBank, but don't assume vcc for s1.
const RegisterBank *AMDGPUInstructionSelector::getArtifactRegBank(
  Register Reg, const MachineRegisterInfo &MRI,
  const TargetRegisterInfo &TRI) const {
  const RegClassOrRegBank &RegClassOrBank = MRI.getRegClassOrRegBank(Reg);
  if (auto *RB = RegClassOrBank.dyn_cast<const RegisterBank *>())
    return RB;

  // Ignore the type, since we don't use vcc in artifacts.
  if (auto *RC = RegClassOrBank.dyn_cast<const TargetRegisterClass *>())
    return &RBI.getRegBankFromRegClass(*RC, LLT());
  return nullptr;
}

bool AMDGPUInstructionSelector::selectG_SZA_EXT(MachineInstr &I) const {
  bool InReg = I.getOpcode() == AMDGPU::G_SEXT_INREG;
  bool Signed = I.getOpcode() == AMDGPU::G_SEXT || InReg;
  const DebugLoc &DL = I.getDebugLoc();
  MachineBasicBlock &MBB = *I.getParent();
  const Register DstReg = I.getOperand(0).getReg();
  const Register SrcReg = I.getOperand(1).getReg();

  const LLT DstTy = MRI->getType(DstReg);
  const LLT SrcTy = MRI->getType(SrcReg);
  const unsigned SrcSize = I.getOpcode() == AMDGPU::G_SEXT_INREG ?
    I.getOperand(2).getImm() : SrcTy.getSizeInBits();
  const unsigned DstSize = DstTy.getSizeInBits();
  if (!DstTy.isScalar())
    return false;

  // Artifact casts should never use vcc.
  const RegisterBank *SrcBank = getArtifactRegBank(SrcReg, *MRI, TRI);

  // FIXME: This should probably be illegal and split earlier.
  if (I.getOpcode() == AMDGPU::G_ANYEXT) {
    if (DstSize <= 32)
      return selectCOPY(I);

    const TargetRegisterClass *SrcRC =
        TRI.getRegClassForTypeOnBank(SrcTy, *SrcBank, *MRI);
    const RegisterBank *DstBank = RBI.getRegBank(DstReg, *MRI, TRI);
    const TargetRegisterClass *DstRC =
        TRI.getRegClassForSizeOnBank(DstSize, *DstBank, *MRI);

    Register UndefReg = MRI->createVirtualRegister(SrcRC);
    BuildMI(MBB, I, DL, TII.get(AMDGPU::IMPLICIT_DEF), UndefReg);
    BuildMI(MBB, I, DL, TII.get(AMDGPU::REG_SEQUENCE), DstReg)
      .addReg(SrcReg)
      .addImm(AMDGPU::sub0)
      .addReg(UndefReg)
      .addImm(AMDGPU::sub1);
    I.eraseFromParent();

    return RBI.constrainGenericRegister(DstReg, *DstRC, *MRI) &&
           RBI.constrainGenericRegister(SrcReg, *SrcRC, *MRI);
  }

  if (SrcBank->getID() == AMDGPU::VGPRRegBankID && DstSize <= 32) {
    // 64-bit should have been split up in RegBankSelect

    // Try to use an and with a mask if it will save code size.
    unsigned Mask;
    if (!Signed && shouldUseAndMask(SrcSize, Mask)) {
      MachineInstr *ExtI =
      BuildMI(MBB, I, DL, TII.get(AMDGPU::V_AND_B32_e32), DstReg)
        .addImm(Mask)
        .addReg(SrcReg);
      I.eraseFromParent();
      return constrainSelectedInstRegOperands(*ExtI, TII, TRI, RBI);
    }

    const unsigned BFE = Signed ? AMDGPU::V_BFE_I32 : AMDGPU::V_BFE_U32;
    MachineInstr *ExtI =
      BuildMI(MBB, I, DL, TII.get(BFE), DstReg)
      .addReg(SrcReg)
      .addImm(0) // Offset
      .addImm(SrcSize); // Width
    I.eraseFromParent();
    return constrainSelectedInstRegOperands(*ExtI, TII, TRI, RBI);
  }

  if (SrcBank->getID() == AMDGPU::SGPRRegBankID && DstSize <= 64) {
    const TargetRegisterClass &SrcRC = InReg && DstSize > 32 ?
      AMDGPU::SReg_64RegClass : AMDGPU::SReg_32RegClass;
    if (!RBI.constrainGenericRegister(SrcReg, SrcRC, *MRI))
      return false;

    if (Signed && DstSize == 32 && (SrcSize == 8 || SrcSize == 16)) {
      const unsigned SextOpc = SrcSize == 8 ?
        AMDGPU::S_SEXT_I32_I8 : AMDGPU::S_SEXT_I32_I16;
      BuildMI(MBB, I, DL, TII.get(SextOpc), DstReg)
        .addReg(SrcReg);
      I.eraseFromParent();
      return RBI.constrainGenericRegister(DstReg, AMDGPU::SReg_32RegClass, *MRI);
    }

    const unsigned BFE64 = Signed ? AMDGPU::S_BFE_I64 : AMDGPU::S_BFE_U64;
    const unsigned BFE32 = Signed ? AMDGPU::S_BFE_I32 : AMDGPU::S_BFE_U32;

    // Scalar BFE is encoded as S1[5:0] = offset, S1[22:16]= width.
    if (DstSize > 32 && (SrcSize <= 32 || InReg)) {
      // We need a 64-bit register source, but the high bits don't matter.
      Register ExtReg = MRI->createVirtualRegister(&AMDGPU::SReg_64RegClass);
      Register UndefReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
      unsigned SubReg = InReg ? AMDGPU::sub0 : 0;

      BuildMI(MBB, I, DL, TII.get(AMDGPU::IMPLICIT_DEF), UndefReg);
      BuildMI(MBB, I, DL, TII.get(AMDGPU::REG_SEQUENCE), ExtReg)
        .addReg(SrcReg, 0, SubReg)
        .addImm(AMDGPU::sub0)
        .addReg(UndefReg)
        .addImm(AMDGPU::sub1);

      BuildMI(MBB, I, DL, TII.get(BFE64), DstReg)
        .addReg(ExtReg)
        .addImm(SrcSize << 16);

      I.eraseFromParent();
      return RBI.constrainGenericRegister(DstReg, AMDGPU::SReg_64RegClass, *MRI);
    }

    unsigned Mask;
    if (!Signed && shouldUseAndMask(SrcSize, Mask)) {
      BuildMI(MBB, I, DL, TII.get(AMDGPU::S_AND_B32), DstReg)
        .addReg(SrcReg)
        .addImm(Mask);
    } else {
      BuildMI(MBB, I, DL, TII.get(BFE32), DstReg)
        .addReg(SrcReg)
        .addImm(SrcSize << 16);
    }

    I.eraseFromParent();
    return RBI.constrainGenericRegister(DstReg, AMDGPU::SReg_32RegClass, *MRI);
  }

  return false;
}

bool AMDGPUInstructionSelector::selectG_CONSTANT(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  MachineOperand &ImmOp = I.getOperand(1);
  Register DstReg = I.getOperand(0).getReg();
  unsigned Size = MRI->getType(DstReg).getSizeInBits();

  // The AMDGPU backend only supports Imm operands and not CImm or FPImm.
  if (ImmOp.isFPImm()) {
    const APInt &Imm = ImmOp.getFPImm()->getValueAPF().bitcastToAPInt();
    ImmOp.ChangeToImmediate(Imm.getZExtValue());
  } else if (ImmOp.isCImm()) {
    ImmOp.ChangeToImmediate(ImmOp.getCImm()->getSExtValue());
  } else {
    llvm_unreachable("Not supported by g_constants");
  }

  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  const bool IsSgpr = DstRB->getID() == AMDGPU::SGPRRegBankID;

  unsigned Opcode;
  if (DstRB->getID() == AMDGPU::VCCRegBankID) {
    Opcode = STI.isWave32() ? AMDGPU::S_MOV_B32 : AMDGPU::S_MOV_B64;
  } else {
    Opcode = IsSgpr ? AMDGPU::S_MOV_B32 : AMDGPU::V_MOV_B32_e32;

    // We should never produce s1 values on banks other than VCC. If the user of
    // this already constrained the register, we may incorrectly think it's VCC
    // if it wasn't originally.
    if (Size == 1)
      return false;
  }

  if (Size != 64) {
    I.setDesc(TII.get(Opcode));
    I.addImplicitDefUseOperands(*MF);
    return constrainSelectedInstRegOperands(I, TII, TRI, RBI);
  }

  const DebugLoc &DL = I.getDebugLoc();

  APInt Imm(Size, I.getOperand(1).getImm());

  MachineInstr *ResInst;
  if (IsSgpr && TII.isInlineConstant(Imm)) {
    ResInst = BuildMI(*BB, &I, DL, TII.get(AMDGPU::S_MOV_B64), DstReg)
      .addImm(I.getOperand(1).getImm());
  } else {
    const TargetRegisterClass *RC = IsSgpr ?
      &AMDGPU::SReg_32RegClass : &AMDGPU::VGPR_32RegClass;
    Register LoReg = MRI->createVirtualRegister(RC);
    Register HiReg = MRI->createVirtualRegister(RC);

    BuildMI(*BB, &I, DL, TII.get(Opcode), LoReg)
      .addImm(Imm.trunc(32).getZExtValue());

    BuildMI(*BB, &I, DL, TII.get(Opcode), HiReg)
      .addImm(Imm.ashr(32).getZExtValue());

    ResInst = BuildMI(*BB, &I, DL, TII.get(AMDGPU::REG_SEQUENCE), DstReg)
      .addReg(LoReg)
      .addImm(AMDGPU::sub0)
      .addReg(HiReg)
      .addImm(AMDGPU::sub1);
  }

  // We can't call constrainSelectedInstRegOperands here, because it doesn't
  // work for target independent opcodes
  I.eraseFromParent();
  const TargetRegisterClass *DstRC =
    TRI.getConstrainedRegClassForOperand(ResInst->getOperand(0), *MRI);
  if (!DstRC)
    return true;
  return RBI.constrainGenericRegister(DstReg, *DstRC, *MRI);
}

bool AMDGPUInstructionSelector::selectG_FNEG(MachineInstr &MI) const {
  // Only manually handle the f64 SGPR case.
  //
  // FIXME: This is a workaround for 2.5 different tablegen problems. Because
  // the bit ops theoretically have a second result due to the implicit def of
  // SCC, the GlobalISelEmitter is overly conservative and rejects it. Fixing
  // that is easy by disabling the check. The result works, but uses a
  // nonsensical sreg32orlds_and_sreg_1 regclass.
  //
  // The DAG emitter is more problematic, and incorrectly adds both S_XOR_B32 to
  // the variadic REG_SEQUENCE operands.

  Register Dst = MI.getOperand(0).getReg();
  const RegisterBank *DstRB = RBI.getRegBank(Dst, *MRI, TRI);
  if (DstRB->getID() != AMDGPU::SGPRRegBankID ||
      MRI->getType(Dst) != LLT::scalar(64))
    return false;

  Register Src = MI.getOperand(1).getReg();
  MachineInstr *Fabs = getOpcodeDef(TargetOpcode::G_FABS, Src, *MRI);
  if (Fabs)
    Src = Fabs->getOperand(1).getReg();

  if (!RBI.constrainGenericRegister(Src, AMDGPU::SReg_64RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Dst, AMDGPU::SReg_64RegClass, *MRI))
    return false;

  MachineBasicBlock *BB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();
  Register LoReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register HiReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register ConstReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register OpReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);

  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), LoReg)
    .addReg(Src, 0, AMDGPU::sub0);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), HiReg)
    .addReg(Src, 0, AMDGPU::sub1);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::S_MOV_B32), ConstReg)
    .addImm(0x80000000);

  // Set or toggle sign bit.
  unsigned Opc = Fabs ? AMDGPU::S_OR_B32 : AMDGPU::S_XOR_B32;
  BuildMI(*BB, &MI, DL, TII.get(Opc), OpReg)
    .addReg(HiReg)
    .addReg(ConstReg);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::REG_SEQUENCE), Dst)
    .addReg(LoReg)
    .addImm(AMDGPU::sub0)
    .addReg(OpReg)
    .addImm(AMDGPU::sub1);
  MI.eraseFromParent();
  return true;
}

// FIXME: This is a workaround for the same tablegen problems as G_FNEG
bool AMDGPUInstructionSelector::selectG_FABS(MachineInstr &MI) const {
  Register Dst = MI.getOperand(0).getReg();
  const RegisterBank *DstRB = RBI.getRegBank(Dst, *MRI, TRI);
  if (DstRB->getID() != AMDGPU::SGPRRegBankID ||
      MRI->getType(Dst) != LLT::scalar(64))
    return false;

  Register Src = MI.getOperand(1).getReg();
  MachineBasicBlock *BB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();
  Register LoReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register HiReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register ConstReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register OpReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);

  if (!RBI.constrainGenericRegister(Src, AMDGPU::SReg_64RegClass, *MRI) ||
      !RBI.constrainGenericRegister(Dst, AMDGPU::SReg_64RegClass, *MRI))
    return false;

  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), LoReg)
    .addReg(Src, 0, AMDGPU::sub0);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), HiReg)
    .addReg(Src, 0, AMDGPU::sub1);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::S_MOV_B32), ConstReg)
    .addImm(0x7fffffff);

  // Clear sign bit.
  // TODO: Should this used S_BITSET0_*?
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::S_AND_B32), OpReg)
    .addReg(HiReg)
    .addReg(ConstReg);
  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::REG_SEQUENCE), Dst)
    .addReg(LoReg)
    .addImm(AMDGPU::sub0)
    .addReg(OpReg)
    .addImm(AMDGPU::sub1);

  MI.eraseFromParent();
  return true;
}

static bool isConstant(const MachineInstr &MI) {
  return MI.getOpcode() == TargetOpcode::G_CONSTANT;
}

void AMDGPUInstructionSelector::getAddrModeInfo(const MachineInstr &Load,
    const MachineRegisterInfo &MRI, SmallVectorImpl<GEPInfo> &AddrInfo) const {

  const MachineInstr *PtrMI = MRI.getUniqueVRegDef(Load.getOperand(1).getReg());

  assert(PtrMI);

  if (PtrMI->getOpcode() != TargetOpcode::G_PTR_ADD)
    return;

  GEPInfo GEPInfo(*PtrMI);

  for (unsigned i = 1; i != 3; ++i) {
    const MachineOperand &GEPOp = PtrMI->getOperand(i);
    const MachineInstr *OpDef = MRI.getUniqueVRegDef(GEPOp.getReg());
    assert(OpDef);
    if (i == 2 && isConstant(*OpDef)) {
      // TODO: Could handle constant base + variable offset, but a combine
      // probably should have commuted it.
      assert(GEPInfo.Imm == 0);
      GEPInfo.Imm = OpDef->getOperand(1).getCImm()->getSExtValue();
      continue;
    }
    const RegisterBank *OpBank = RBI.getRegBank(GEPOp.getReg(), MRI, TRI);
    if (OpBank->getID() == AMDGPU::SGPRRegBankID)
      GEPInfo.SgprParts.push_back(GEPOp.getReg());
    else
      GEPInfo.VgprParts.push_back(GEPOp.getReg());
  }

  AddrInfo.push_back(GEPInfo);
  getAddrModeInfo(*PtrMI, MRI, AddrInfo);
}

bool AMDGPUInstructionSelector::isSGPR(Register Reg) const {
  return RBI.getRegBank(Reg, *MRI, TRI)->getID() == AMDGPU::SGPRRegBankID;
}

bool AMDGPUInstructionSelector::isInstrUniform(const MachineInstr &MI) const {
  if (!MI.hasOneMemOperand())
    return false;

  const MachineMemOperand *MMO = *MI.memoperands_begin();
  const Value *Ptr = MMO->getValue();

  // UndefValue means this is a load of a kernel input.  These are uniform.
  // Sometimes LDS instructions have constant pointers.
  // If Ptr is null, then that means this mem operand contains a
  // PseudoSourceValue like GOT.
  if (!Ptr || isa<UndefValue>(Ptr) || isa<Argument>(Ptr) ||
      isa<Constant>(Ptr) || isa<GlobalValue>(Ptr))
    return true;

  if (MMO->getAddrSpace() == AMDGPUAS::CONSTANT_ADDRESS_32BIT)
    return true;

  const Instruction *I = dyn_cast<Instruction>(Ptr);
  return I && I->getMetadata("amdgpu.uniform");
}

bool AMDGPUInstructionSelector::hasVgprParts(ArrayRef<GEPInfo> AddrInfo) const {
  for (const GEPInfo &GEPInfo : AddrInfo) {
    if (!GEPInfo.VgprParts.empty())
      return true;
  }
  return false;
}

void AMDGPUInstructionSelector::initM0(MachineInstr &I) const {
  const LLT PtrTy = MRI->getType(I.getOperand(1).getReg());
  unsigned AS = PtrTy.getAddressSpace();
  if ((AS == AMDGPUAS::LOCAL_ADDRESS || AS == AMDGPUAS::REGION_ADDRESS) &&
      STI.ldsRequiresM0Init()) {
    MachineBasicBlock *BB = I.getParent();

    // If DS instructions require M0 initializtion, insert it before selecting.
    BuildMI(*BB, &I, I.getDebugLoc(), TII.get(AMDGPU::S_MOV_B32), AMDGPU::M0)
      .addImm(-1);
  }
}

bool AMDGPUInstructionSelector::selectG_LOAD_STORE_ATOMICRMW(
  MachineInstr &I) const {
  initM0(I);
  return selectImpl(I, *CoverageInfo);
}

// TODO: No rtn optimization.
bool AMDGPUInstructionSelector::selectG_AMDGPU_ATOMIC_CMPXCHG(
  MachineInstr &MI) const {
  Register PtrReg = MI.getOperand(1).getReg();
  const LLT PtrTy = MRI->getType(PtrReg);
  if (PtrTy.getAddressSpace() == AMDGPUAS::FLAT_ADDRESS ||
      STI.useFlatForGlobal())
    return selectImpl(MI, *CoverageInfo);

  Register DstReg = MI.getOperand(0).getReg();
  const LLT Ty = MRI->getType(DstReg);
  const bool Is64 = Ty.getSizeInBits() == 64;
  const unsigned SubReg = Is64 ? AMDGPU::sub0_sub1 : AMDGPU::sub0;
  Register TmpReg = MRI->createVirtualRegister(
    Is64 ? &AMDGPU::VReg_128RegClass : &AMDGPU::VReg_64RegClass);

  const DebugLoc &DL = MI.getDebugLoc();
  MachineBasicBlock *BB = MI.getParent();

  Register VAddr, RSrcReg, SOffset;
  int64_t Offset = 0;

  unsigned Opcode;
  if (selectMUBUFOffsetImpl(MI.getOperand(1), RSrcReg, SOffset, Offset)) {
    Opcode = Is64 ? AMDGPU::BUFFER_ATOMIC_CMPSWAP_X2_OFFSET_RTN :
                             AMDGPU::BUFFER_ATOMIC_CMPSWAP_OFFSET_RTN;
  } else if (selectMUBUFAddr64Impl(MI.getOperand(1), VAddr,
                                   RSrcReg, SOffset, Offset)) {
    Opcode = Is64 ? AMDGPU::BUFFER_ATOMIC_CMPSWAP_X2_ADDR64_RTN :
                    AMDGPU::BUFFER_ATOMIC_CMPSWAP_ADDR64_RTN;
  } else
    return selectImpl(MI, *CoverageInfo);

  auto MIB = BuildMI(*BB, &MI, DL, TII.get(Opcode), TmpReg)
    .addReg(MI.getOperand(2).getReg());

  if (VAddr)
    MIB.addReg(VAddr);

  MIB.addReg(RSrcReg);
  if (SOffset)
    MIB.addReg(SOffset);
  else
    MIB.addImm(0);

  MIB.addImm(Offset);
  MIB.addImm(1); // glc
  MIB.addImm(0); // slc
  MIB.cloneMemRefs(MI);

  BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), DstReg)
    .addReg(TmpReg, RegState::Kill, SubReg);

  MI.eraseFromParent();

  MRI->setRegClass(
    DstReg, Is64 ? &AMDGPU::VReg_64RegClass : &AMDGPU::VGPR_32RegClass);
  return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}

bool AMDGPUInstructionSelector::selectG_BRCOND(MachineInstr &I) const {
  MachineBasicBlock *BB = I.getParent();
  MachineOperand &CondOp = I.getOperand(0);
  Register CondReg = CondOp.getReg();
  const DebugLoc &DL = I.getDebugLoc();

  unsigned BrOpcode;
  Register CondPhysReg;
  const TargetRegisterClass *ConstrainRC;

  // In SelectionDAG, we inspect the IR block for uniformity metadata to decide
  // whether the branch is uniform when selecting the instruction. In
  // GlobalISel, we should push that decision into RegBankSelect. Assume for now
  // RegBankSelect knows what it's doing if the branch condition is scc, even
  // though it currently does not.
  if (!isVCC(CondReg, *MRI)) {
    if (MRI->getType(CondReg) != LLT::scalar(32))
      return false;

    CondPhysReg = AMDGPU::SCC;
    BrOpcode = AMDGPU::S_CBRANCH_SCC1;
    ConstrainRC = &AMDGPU::SReg_32RegClass;
  } else {
    // FIXME: Do we have to insert an and with exec here, like in SelectionDAG?
    // We sort of know that a VCC producer based on the register bank, that ands
    // inactive lanes with 0. What if there was a logical operation with vcc
    // producers in different blocks/with different exec masks?
    // FIXME: Should scc->vcc copies and with exec?
    CondPhysReg = TRI.getVCC();
    BrOpcode = AMDGPU::S_CBRANCH_VCCNZ;
    ConstrainRC = TRI.getBoolRC();
  }

  if (!MRI->getRegClassOrNull(CondReg))
    MRI->setRegClass(CondReg, ConstrainRC);

  BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), CondPhysReg)
    .addReg(CondReg);
  BuildMI(*BB, &I, DL, TII.get(BrOpcode))
    .addMBB(I.getOperand(1).getMBB());

  I.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectG_GLOBAL_VALUE(
  MachineInstr &I) const {
  Register DstReg = I.getOperand(0).getReg();
  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  const bool IsVGPR = DstRB->getID() == AMDGPU::VGPRRegBankID;
  I.setDesc(TII.get(IsVGPR ? AMDGPU::V_MOV_B32_e32 : AMDGPU::S_MOV_B32));
  if (IsVGPR)
    I.addOperand(*MF, MachineOperand::CreateReg(AMDGPU::EXEC, false, true));

  return RBI.constrainGenericRegister(
    DstReg, IsVGPR ? AMDGPU::VGPR_32RegClass : AMDGPU::SReg_32RegClass, *MRI);
}

bool AMDGPUInstructionSelector::selectG_PTRMASK(MachineInstr &I) const {
  Register DstReg = I.getOperand(0).getReg();
  Register SrcReg = I.getOperand(1).getReg();
  Register MaskReg = I.getOperand(2).getReg();
  LLT Ty = MRI->getType(DstReg);
  LLT MaskTy = MRI->getType(MaskReg);

  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  const RegisterBank *SrcRB = RBI.getRegBank(SrcReg, *MRI, TRI);
  const RegisterBank *MaskRB = RBI.getRegBank(MaskReg, *MRI, TRI);
  const bool IsVGPR = DstRB->getID() == AMDGPU::VGPRRegBankID;
  if (DstRB != SrcRB) // Should only happen for hand written MIR.
    return false;

  unsigned NewOpc = IsVGPR ? AMDGPU::V_AND_B32_e64 : AMDGPU::S_AND_B32;
  const TargetRegisterClass &RegRC
    = IsVGPR ? AMDGPU::VGPR_32RegClass : AMDGPU::SReg_32RegClass;

  const TargetRegisterClass *DstRC = TRI.getRegClassForTypeOnBank(Ty, *DstRB,
                                                                  *MRI);
  const TargetRegisterClass *SrcRC = TRI.getRegClassForTypeOnBank(Ty, *SrcRB,
                                                                  *MRI);
  const TargetRegisterClass *MaskRC =
      TRI.getRegClassForTypeOnBank(MaskTy, *MaskRB, *MRI);

  if (!RBI.constrainGenericRegister(DstReg, *DstRC, *MRI) ||
      !RBI.constrainGenericRegister(SrcReg, *SrcRC, *MRI) ||
      !RBI.constrainGenericRegister(MaskReg, *MaskRC, *MRI))
    return false;

  MachineBasicBlock *BB = I.getParent();
  const DebugLoc &DL = I.getDebugLoc();
  if (Ty.getSizeInBits() == 32) {
    assert(MaskTy.getSizeInBits() == 32 &&
           "ptrmask should have been narrowed during legalize");

    BuildMI(*BB, &I, DL, TII.get(NewOpc), DstReg)
      .addReg(SrcReg)
      .addReg(MaskReg);
    I.eraseFromParent();
    return true;
  }

  Register HiReg = MRI->createVirtualRegister(&RegRC);
  Register LoReg = MRI->createVirtualRegister(&RegRC);

  // Extract the subregisters from the source pointer.
  BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), LoReg)
    .addReg(SrcReg, 0, AMDGPU::sub0);
  BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), HiReg)
    .addReg(SrcReg, 0, AMDGPU::sub1);

  Register MaskedLo, MaskedHi;

  // Try to avoid emitting a bit operation when we only need to touch half of
  // the 64-bit pointer.
  APInt MaskOnes = KnownBits->getKnownOnes(MaskReg).zextOrSelf(64);

  const APInt MaskHi32 = APInt::getHighBitsSet(64, 32);
  const APInt MaskLo32 = APInt::getLowBitsSet(64, 32);
  if ((MaskOnes & MaskLo32) == MaskLo32) {
    // If all the bits in the low half are 1, we only need a copy for it.
    MaskedLo = LoReg;
  } else {
    // Extract the mask subregister and apply the and.
    Register MaskLo = MRI->createVirtualRegister(&RegRC);
    MaskedLo = MRI->createVirtualRegister(&RegRC);

    BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), MaskLo)
      .addReg(MaskReg, 0, AMDGPU::sub0);
    BuildMI(*BB, &I, DL, TII.get(NewOpc), MaskedLo)
      .addReg(LoReg)
      .addReg(MaskLo);
  }

  if ((MaskOnes & MaskHi32) == MaskHi32) {
    // If all the bits in the high half are 1, we only need a copy for it.
    MaskedHi = HiReg;
  } else {
    Register MaskHi = MRI->createVirtualRegister(&RegRC);
    MaskedHi = MRI->createVirtualRegister(&RegRC);

    BuildMI(*BB, &I, DL, TII.get(AMDGPU::COPY), MaskHi)
      .addReg(MaskReg, 0, AMDGPU::sub1);
    BuildMI(*BB, &I, DL, TII.get(NewOpc), MaskedHi)
      .addReg(HiReg)
      .addReg(MaskHi);
  }

  BuildMI(*BB, &I, DL, TII.get(AMDGPU::REG_SEQUENCE), DstReg)
    .addReg(MaskedLo)
    .addImm(AMDGPU::sub0)
    .addReg(MaskedHi)
    .addImm(AMDGPU::sub1);
  I.eraseFromParent();
  return true;
}

/// Return the register to use for the index value, and the subregister to use
/// for the indirectly accessed register.
static std::pair<Register, unsigned>
computeIndirectRegIndex(MachineRegisterInfo &MRI,
                        const SIRegisterInfo &TRI,
                        const TargetRegisterClass *SuperRC,
                        Register IdxReg,
                        unsigned EltSize) {
  Register IdxBaseReg;
  int Offset;

  std::tie(IdxBaseReg, Offset) = AMDGPU::getBaseWithConstantOffset(MRI, IdxReg);
  if (IdxBaseReg == AMDGPU::NoRegister) {
    // This will happen if the index is a known constant. This should ordinarily
    // be legalized out, but handle it as a register just in case.
    assert(Offset == 0);
    IdxBaseReg = IdxReg;
  }

  ArrayRef<int16_t> SubRegs = TRI.getRegSplitParts(SuperRC, EltSize);

  // Skip out of bounds offsets, or else we would end up using an undefined
  // register.
  if (static_cast<unsigned>(Offset) >= SubRegs.size())
    return std::make_pair(IdxReg, SubRegs[0]);
  return std::make_pair(IdxBaseReg, SubRegs[Offset]);
}

bool AMDGPUInstructionSelector::selectG_EXTRACT_VECTOR_ELT(
  MachineInstr &MI) const {
  Register DstReg = MI.getOperand(0).getReg();
  Register SrcReg = MI.getOperand(1).getReg();
  Register IdxReg = MI.getOperand(2).getReg();

  LLT DstTy = MRI->getType(DstReg);
  LLT SrcTy = MRI->getType(SrcReg);

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

  // The index must be scalar. If it wasn't RegBankSelect should have moved this
  // into a waterfall loop.
  if (IdxRB->getID() != AMDGPU::SGPRRegBankID)
    return false;

  const TargetRegisterClass *SrcRC = TRI.getRegClassForTypeOnBank(SrcTy, *SrcRB,
                                                                  *MRI);
  const TargetRegisterClass *DstRC = TRI.getRegClassForTypeOnBank(DstTy, *DstRB,
                                                                  *MRI);
  if (!SrcRC || !DstRC)
    return false;
  if (!RBI.constrainGenericRegister(SrcReg, *SrcRC, *MRI) ||
      !RBI.constrainGenericRegister(DstReg, *DstRC, *MRI) ||
      !RBI.constrainGenericRegister(IdxReg, AMDGPU::SReg_32RegClass, *MRI))
    return false;

  MachineBasicBlock *BB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();
  const bool Is64 = DstTy.getSizeInBits() == 64;

  unsigned SubReg;
  std::tie(IdxReg, SubReg) = computeIndirectRegIndex(*MRI, TRI, SrcRC, IdxReg,
                                                     DstTy.getSizeInBits() / 8);

  if (SrcRB->getID() == AMDGPU::SGPRRegBankID) {
    if (DstTy.getSizeInBits() != 32 && !Is64)
      return false;

    BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
      .addReg(IdxReg);

    unsigned Opc = Is64 ? AMDGPU::S_MOVRELS_B64 : AMDGPU::S_MOVRELS_B32;
    BuildMI(*BB, &MI, DL, TII.get(Opc), DstReg)
      .addReg(SrcReg, 0, SubReg)
      .addReg(SrcReg, RegState::Implicit);
    MI.eraseFromParent();
    return true;
  }

  if (SrcRB->getID() != AMDGPU::VGPRRegBankID || DstTy.getSizeInBits() != 32)
    return false;

  if (!STI.useVGPRIndexMode()) {
    BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
      .addReg(IdxReg);
    BuildMI(*BB, &MI, DL, TII.get(AMDGPU::V_MOVRELS_B32_e32), DstReg)
      .addReg(SrcReg, 0, SubReg)
      .addReg(SrcReg, RegState::Implicit);
    MI.eraseFromParent();
    return true;
  }

  BuildMI(*BB, MI, DL, TII.get(AMDGPU::S_SET_GPR_IDX_ON))
    .addReg(IdxReg)
    .addImm(AMDGPU::VGPRIndexMode::SRC0_ENABLE);
  BuildMI(*BB, MI, DL, TII.get(AMDGPU::V_MOV_B32_e32), DstReg)
    .addReg(SrcReg, 0, SubReg)
    .addReg(SrcReg, RegState::Implicit)
    .addReg(AMDGPU::M0, RegState::Implicit);
  BuildMI(*BB, MI, DL, TII.get(AMDGPU::S_SET_GPR_IDX_OFF));

  MI.eraseFromParent();
  return true;
}

// TODO: Fold insert_vector_elt (extract_vector_elt) into movrelsd
bool AMDGPUInstructionSelector::selectG_INSERT_VECTOR_ELT(
  MachineInstr &MI) const {
  Register DstReg = MI.getOperand(0).getReg();
  Register VecReg = MI.getOperand(1).getReg();
  Register ValReg = MI.getOperand(2).getReg();
  Register IdxReg = MI.getOperand(3).getReg();

  LLT VecTy = MRI->getType(DstReg);
  LLT ValTy = MRI->getType(ValReg);
  unsigned VecSize = VecTy.getSizeInBits();
  unsigned ValSize = ValTy.getSizeInBits();

  const RegisterBank *VecRB = RBI.getRegBank(VecReg, *MRI, TRI);
  const RegisterBank *ValRB = RBI.getRegBank(ValReg, *MRI, TRI);
  const RegisterBank *IdxRB = RBI.getRegBank(IdxReg, *MRI, TRI);

  assert(VecTy.getElementType() == ValTy);

  // The index must be scalar. If it wasn't RegBankSelect should have moved this
  // into a waterfall loop.
  if (IdxRB->getID() != AMDGPU::SGPRRegBankID)
    return false;

  const TargetRegisterClass *VecRC = TRI.getRegClassForTypeOnBank(VecTy, *VecRB,
                                                                  *MRI);
  const TargetRegisterClass *ValRC = TRI.getRegClassForTypeOnBank(ValTy, *ValRB,
                                                                  *MRI);

  if (!RBI.constrainGenericRegister(VecReg, *VecRC, *MRI) ||
      !RBI.constrainGenericRegister(DstReg, *VecRC, *MRI) ||
      !RBI.constrainGenericRegister(ValReg, *ValRC, *MRI) ||
      !RBI.constrainGenericRegister(IdxReg, AMDGPU::SReg_32RegClass, *MRI))
    return false;

  if (VecRB->getID() == AMDGPU::VGPRRegBankID && ValSize != 32)
    return false;

  unsigned SubReg;
  std::tie(IdxReg, SubReg) = computeIndirectRegIndex(*MRI, TRI, VecRC, IdxReg,
                                                     ValSize / 8);

  const bool IndexMode = VecRB->getID() == AMDGPU::VGPRRegBankID &&
                         STI.useVGPRIndexMode();

  MachineBasicBlock *BB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  if (IndexMode) {
    BuildMI(*BB, MI, DL, TII.get(AMDGPU::S_SET_GPR_IDX_ON))
      .addReg(IdxReg)
      .addImm(AMDGPU::VGPRIndexMode::DST_ENABLE);
  } else {
    BuildMI(*BB, &MI, DL, TII.get(AMDGPU::COPY), AMDGPU::M0)
      .addReg(IdxReg);
  }

  const MCInstrDesc &RegWriteOp
    = TII.getIndirectRegWritePseudo(VecSize, ValSize,
                                    VecRB->getID() == AMDGPU::SGPRRegBankID);
  BuildMI(*BB, MI, DL, RegWriteOp, DstReg)
    .addReg(VecReg)
    .addReg(ValReg)
    .addImm(SubReg);

  if (IndexMode)
    BuildMI(*BB, MI, DL, TII.get(AMDGPU::S_SET_GPR_IDX_OFF));

  MI.eraseFromParent();
  return true;
}

static bool isZeroOrUndef(int X) {
  return X == 0 || X == -1;
}

static bool isOneOrUndef(int X) {
  return X == 1 || X == -1;
}

static bool isZeroOrOneOrUndef(int X) {
  return X == 0 || X == 1 || X == -1;
}

// Normalize a VOP3P shuffle mask to refer to the low/high half of a single
// 32-bit register.
static Register normalizeVOP3PMask(int NewMask[2], Register Src0, Register Src1,
                                   ArrayRef<int> Mask) {
  NewMask[0] = Mask[0];
  NewMask[1] = Mask[1];
  if (isZeroOrOneOrUndef(Mask[0]) && isZeroOrOneOrUndef(Mask[1]))
    return Src0;

  assert(NewMask[0] == 2 || NewMask[0] == 3 || NewMask[0] == -1);
  assert(NewMask[1] == 2 || NewMask[1] == 3 || NewMask[1] == -1);

  // Shift the mask inputs to be 0/1;
  NewMask[0] = NewMask[0] == -1 ? -1 : NewMask[0] - 2;
  NewMask[1] = NewMask[1] == -1 ? -1 : NewMask[1] - 2;
  return Src1;
}

// This is only legal with VOP3P instructions as an aid to op_sel matching.
bool AMDGPUInstructionSelector::selectG_SHUFFLE_VECTOR(
  MachineInstr &MI) const {
  Register DstReg = MI.getOperand(0).getReg();
  Register Src0Reg = MI.getOperand(1).getReg();
  Register Src1Reg = MI.getOperand(2).getReg();
  ArrayRef<int> ShufMask = MI.getOperand(3).getShuffleMask();

  const LLT V2S16 = LLT::vector(2, 16);
  if (MRI->getType(DstReg) != V2S16 || MRI->getType(Src0Reg) != V2S16)
    return false;

  if (!AMDGPU::isLegalVOP3PShuffleMask(ShufMask))
    return false;

  assert(ShufMask.size() == 2);
  assert(STI.hasSDWA() && "no target has VOP3P but not SDWA");

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  const RegisterBank *DstRB = RBI.getRegBank(DstReg, *MRI, TRI);
  const bool IsVALU = DstRB->getID() == AMDGPU::VGPRRegBankID;
  const TargetRegisterClass &RC = IsVALU ?
    AMDGPU::VGPR_32RegClass : AMDGPU::SReg_32RegClass;

  // Handle the degenerate case which should have folded out.
  if (ShufMask[0] == -1 && ShufMask[1] == -1) {
    BuildMI(*MBB, MI, DL, TII.get(AMDGPU::IMPLICIT_DEF), DstReg);

    MI.eraseFromParent();
    return RBI.constrainGenericRegister(DstReg, RC, *MRI);
  }

  // A legal VOP3P mask only reads one of the sources.
  int Mask[2];
  Register SrcVec = normalizeVOP3PMask(Mask, Src0Reg, Src1Reg, ShufMask);

  if (!RBI.constrainGenericRegister(DstReg, RC, *MRI) ||
      !RBI.constrainGenericRegister(SrcVec, RC, *MRI))
    return false;

  // TODO: This also should have been folded out
  if (isZeroOrUndef(Mask[0]) && isOneOrUndef(Mask[1])) {
    BuildMI(*MBB, MI, DL, TII.get(AMDGPU::COPY), DstReg)
      .addReg(SrcVec);

    MI.eraseFromParent();
    return true;
  }

  if (Mask[0] == 1 && Mask[1] == -1) {
    if (IsVALU) {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::V_LSHRREV_B32_e64), DstReg)
        .addImm(16)
        .addReg(SrcVec);
    } else {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_LSHR_B32), DstReg)
        .addReg(SrcVec)
        .addImm(16);
    }
  } else if (Mask[0] == -1 && Mask[1] == 0) {
    if (IsVALU) {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::V_LSHLREV_B32_e64), DstReg)
        .addImm(16)
        .addReg(SrcVec);
    } else {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_LSHL_B32), DstReg)
        .addReg(SrcVec)
        .addImm(16);
    }
  } else if (Mask[0] == 0 && Mask[1] == 0) {
    if (IsVALU) {
      // Write low half of the register into the high half.
      MachineInstr *MovSDWA =
        BuildMI(*MBB, MI, DL, TII.get(AMDGPU::V_MOV_B32_sdwa), DstReg)
        .addImm(0)                             // $src0_modifiers
        .addReg(SrcVec)                        // $src0
        .addImm(0)                             // $clamp
        .addImm(AMDGPU::SDWA::WORD_1)          // $dst_sel
        .addImm(AMDGPU::SDWA::UNUSED_PRESERVE) // $dst_unused
        .addImm(AMDGPU::SDWA::WORD_0)          // $src0_sel
        .addReg(SrcVec, RegState::Implicit);
      MovSDWA->tieOperands(0, MovSDWA->getNumOperands() - 1);
    } else {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_PACK_LL_B32_B16), DstReg)
        .addReg(SrcVec)
        .addReg(SrcVec);
    }
  } else if (Mask[0] == 1 && Mask[1] == 1) {
    if (IsVALU) {
      // Write high half of the register into the low half.
      MachineInstr *MovSDWA =
        BuildMI(*MBB, MI, DL, TII.get(AMDGPU::V_MOV_B32_sdwa), DstReg)
        .addImm(0)                             // $src0_modifiers
        .addReg(SrcVec)                        // $src0
        .addImm(0)                             // $clamp
        .addImm(AMDGPU::SDWA::WORD_0)          // $dst_sel
        .addImm(AMDGPU::SDWA::UNUSED_PRESERVE) // $dst_unused
        .addImm(AMDGPU::SDWA::WORD_1)          // $src0_sel
        .addReg(SrcVec, RegState::Implicit);
      MovSDWA->tieOperands(0, MovSDWA->getNumOperands() - 1);
    } else {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_PACK_HH_B32_B16), DstReg)
        .addReg(SrcVec)
        .addReg(SrcVec);
    }
  } else if (Mask[0] == 1 && Mask[1] == 0) {
    if (IsVALU) {
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::V_ALIGNBIT_B32), DstReg)
        .addReg(SrcVec)
        .addReg(SrcVec)
        .addImm(16);
    } else {
      Register TmpReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_LSHR_B32), TmpReg)
        .addReg(SrcVec)
        .addImm(16);
      BuildMI(*MBB, MI, DL, TII.get(AMDGPU::S_PACK_LL_B32_B16), DstReg)
        .addReg(TmpReg)
        .addReg(SrcVec);
    }
  } else
    llvm_unreachable("all shuffle masks should be handled");

  MI.eraseFromParent();
  return true;
}

bool AMDGPUInstructionSelector::selectAMDGPU_BUFFER_ATOMIC_FADD(
  MachineInstr &MI) const {

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  if (!MRI->use_nodbg_empty(MI.getOperand(0).getReg())) {
    Function &F = MBB->getParent()->getFunction();
    DiagnosticInfoUnsupported
      NoFpRet(F, "return versions of fp atomics not supported",
              MI.getDebugLoc(), DS_Error);
    F.getContext().diagnose(NoFpRet);
    return false;
  }

  // FIXME: This is only needed because tablegen requires number of dst operands
  // in match and replace pattern to be the same. Otherwise patterns can be
  // exported from SDag path.
  MachineOperand &VDataIn = MI.getOperand(1);
  MachineOperand &VIndex = MI.getOperand(3);
  MachineOperand &VOffset = MI.getOperand(4);
  MachineOperand &SOffset = MI.getOperand(5);
  int16_t Offset = MI.getOperand(6).getImm();

  bool HasVOffset = !isOperandImmEqual(VOffset, 0, *MRI);
  bool HasVIndex = !isOperandImmEqual(VIndex, 0, *MRI);

  unsigned Opcode;
  if (HasVOffset) {
    Opcode = HasVIndex ? AMDGPU::BUFFER_ATOMIC_ADD_F32_BOTHEN
                       : AMDGPU::BUFFER_ATOMIC_ADD_F32_OFFEN;
  } else {
    Opcode = HasVIndex ? AMDGPU::BUFFER_ATOMIC_ADD_F32_IDXEN
                       : AMDGPU::BUFFER_ATOMIC_ADD_F32_OFFSET;
  }

  if (MRI->getType(VDataIn.getReg()).isVector()) {
    switch (Opcode) {
    case AMDGPU::BUFFER_ATOMIC_ADD_F32_BOTHEN:
      Opcode = AMDGPU::BUFFER_ATOMIC_PK_ADD_F16_BOTHEN;
      break;
    case AMDGPU::BUFFER_ATOMIC_ADD_F32_OFFEN:
      Opcode = AMDGPU::BUFFER_ATOMIC_PK_ADD_F16_OFFEN;
      break;
    case AMDGPU::BUFFER_ATOMIC_ADD_F32_IDXEN:
      Opcode = AMDGPU::BUFFER_ATOMIC_PK_ADD_F16_IDXEN;
      break;
    case AMDGPU::BUFFER_ATOMIC_ADD_F32_OFFSET:
      Opcode = AMDGPU::BUFFER_ATOMIC_PK_ADD_F16_OFFSET;
      break;
    }
  }

  auto I = BuildMI(*MBB, MI, DL, TII.get(Opcode));
  I.add(VDataIn);

  if (Opcode == AMDGPU::BUFFER_ATOMIC_ADD_F32_BOTHEN ||
      Opcode == AMDGPU::BUFFER_ATOMIC_PK_ADD_F16_BOTHEN) {
    Register IdxReg = MRI->createVirtualRegister(&AMDGPU::VReg_64RegClass);
    BuildMI(*MBB, &*I, DL, TII.get(AMDGPU::REG_SEQUENCE), IdxReg)
      .addReg(VIndex.getReg())
      .addImm(AMDGPU::sub0)
      .addReg(VOffset.getReg())
      .addImm(AMDGPU::sub1);

    I.addReg(IdxReg);
  } else if (HasVIndex) {
    I.add(VIndex);
  } else if (HasVOffset) {
    I.add(VOffset);
  }

  I.add(MI.getOperand(2)); // rsrc
  I.add(SOffset);
  I.addImm(Offset);
  renderExtractSLC(I, MI, 7);
  I.cloneMemRefs(MI);

  MI.eraseFromParent();

  return true;
}

bool AMDGPUInstructionSelector::selectGlobalAtomicFaddIntrinsic(
  MachineInstr &MI) const{

  MachineBasicBlock *MBB = MI.getParent();
  const DebugLoc &DL = MI.getDebugLoc();

  if (!MRI->use_nodbg_empty(MI.getOperand(0).getReg())) {
    Function &F = MBB->getParent()->getFunction();
    DiagnosticInfoUnsupported
      NoFpRet(F, "return versions of fp atomics not supported",
              MI.getDebugLoc(), DS_Error);
    F.getContext().diagnose(NoFpRet);
    return false;
  }

  // FIXME: This is only needed because tablegen requires number of dst operands
  // in match and replace pattern to be the same. Otherwise patterns can be
  // exported from SDag path.
  auto Addr = selectFlatOffsetImpl<true>(MI.getOperand(2));

  Register Data = MI.getOperand(3).getReg();
  const unsigned Opc = MRI->getType(Data).isVector() ?
    AMDGPU::GLOBAL_ATOMIC_PK_ADD_F16 : AMDGPU::GLOBAL_ATOMIC_ADD_F32;
  auto MIB = BuildMI(*MBB, &MI, DL, TII.get(Opc))
    .addReg(Addr.first)
    .addReg(Data)
    .addImm(Addr.second)
    .addImm(0) // SLC
    .cloneMemRefs(MI);

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

bool AMDGPUInstructionSelector::selectBVHIntrinsic(MachineInstr &MI) const{
  MI.setDesc(TII.get(MI.getOperand(1).getImm()));
  MI.RemoveOperand(1);
  MI.addImplicitDefUseOperands(*MI.getParent()->getParent());
  return true;
}

bool AMDGPUInstructionSelector::select(MachineInstr &I) {
  if (I.isPHI())
    return selectPHI(I);

  if (!I.isPreISelOpcode()) {
    if (I.isCopy())
      return selectCOPY(I);
    return true;
  }

  switch (I.getOpcode()) {
  case TargetOpcode::G_AND:
  case TargetOpcode::G_OR:
  case TargetOpcode::G_XOR:
    if (selectImpl(I, *CoverageInfo))
      return true;
    return selectG_AND_OR_XOR(I);
  case TargetOpcode::G_ADD:
  case TargetOpcode::G_SUB:
    if (selectImpl(I, *CoverageInfo))
      return true;
    return selectG_ADD_SUB(I);
  case TargetOpcode::G_UADDO:
  case TargetOpcode::G_USUBO:
  case TargetOpcode::G_UADDE:
  case TargetOpcode::G_USUBE:
    return selectG_UADDO_USUBO_UADDE_USUBE(I);
  case TargetOpcode::G_INTTOPTR:
  case TargetOpcode::G_BITCAST:
  case TargetOpcode::G_PTRTOINT:
    return selectCOPY(I);
  case TargetOpcode::G_CONSTANT:
  case TargetOpcode::G_FCONSTANT:
    return selectG_CONSTANT(I);
  case TargetOpcode::G_FNEG:
    if (selectImpl(I, *CoverageInfo))
      return true;
    return selectG_FNEG(I);
  case TargetOpcode::G_FABS:
    if (selectImpl(I, *CoverageInfo))
      return true;
    return selectG_FABS(I);
  case TargetOpcode::G_EXTRACT:
    return selectG_EXTRACT(I);
  case TargetOpcode::G_MERGE_VALUES:
  case TargetOpcode::G_BUILD_VECTOR:
  case TargetOpcode::G_CONCAT_VECTORS:
    return selectG_MERGE_VALUES(I);
  case TargetOpcode::G_UNMERGE_VALUES:
    return selectG_UNMERGE_VALUES(I);
  case TargetOpcode::G_BUILD_VECTOR_TRUNC:
    return selectG_BUILD_VECTOR_TRUNC(I);
  case TargetOpcode::G_PTR_ADD:
    return selectG_PTR_ADD(I);
  case TargetOpcode::G_IMPLICIT_DEF:
    return selectG_IMPLICIT_DEF(I);
  case TargetOpcode::G_FREEZE:
    return selectCOPY(I);
  case TargetOpcode::G_INSERT:
    return selectG_INSERT(I);
  case TargetOpcode::G_INTRINSIC:
    return selectG_INTRINSIC(I);
  case TargetOpcode::G_INTRINSIC_W_SIDE_EFFECTS:
    return selectG_INTRINSIC_W_SIDE_EFFECTS(I);
  case TargetOpcode::G_ICMP:
    if (selectG_ICMP(I))
      return true;
    return selectImpl(I, *CoverageInfo);
  case TargetOpcode::G_LOAD:
  case TargetOpcode::G_STORE:
  case TargetOpcode::G_ATOMIC_CMPXCHG:
  case TargetOpcode::G_ATOMICRMW_XCHG:
  case TargetOpcode::G_ATOMICRMW_ADD:
  case TargetOpcode::G_ATOMICRMW_SUB:
  case TargetOpcode::G_ATOMICRMW_AND:
  case TargetOpcode::G_ATOMICRMW_OR:
  case TargetOpcode::G_ATOMICRMW_XOR:
  case TargetOpcode::G_ATOMICRMW_MIN:
  case TargetOpcode::G_ATOMICRMW_MAX:
  case TargetOpcode::G_ATOMICRMW_UMIN:
  case TargetOpcode::G_ATOMICRMW_UMAX:
  case TargetOpcode::G_ATOMICRMW_FADD:
  case AMDGPU::G_AMDGPU_ATOMIC_INC:
  case AMDGPU::G_AMDGPU_ATOMIC_DEC:
  case AMDGPU::G_AMDGPU_ATOMIC_FMIN:
  case AMDGPU::G_AMDGPU_ATOMIC_FMAX:
    return selectG_LOAD_STORE_ATOMICRMW(I);
  case AMDGPU::G_AMDGPU_ATOMIC_CMPXCHG:
    return selectG_AMDGPU_ATOMIC_CMPXCHG(I);
  case TargetOpcode::G_SELECT:
    return selectG_SELECT(I);
  case TargetOpcode::G_TRUNC:
    return selectG_TRUNC(I);
  case TargetOpcode::G_SEXT:
  case TargetOpcode::G_ZEXT:
  case TargetOpcode::G_ANYEXT:
  case TargetOpcode::G_SEXT_INREG:
    if (selectImpl(I, *CoverageInfo))
      return true;
    return selectG_SZA_EXT(I);
  case TargetOpcode::G_BRCOND:
    return selectG_BRCOND(I);
  case TargetOpcode::G_GLOBAL_VALUE:
    return selectG_GLOBAL_VALUE(I);
  case TargetOpcode::G_PTRMASK:
    return selectG_PTRMASK(I);
  case TargetOpcode::G_EXTRACT_VECTOR_ELT:
    return selectG_EXTRACT_VECTOR_ELT(I);
  case TargetOpcode::G_INSERT_VECTOR_ELT:
    return selectG_INSERT_VECTOR_ELT(I);
  case TargetOpcode::G_SHUFFLE_VECTOR:
    return selectG_SHUFFLE_VECTOR(I);
  case AMDGPU::G_AMDGPU_INTRIN_IMAGE_LOAD:
  case AMDGPU::G_AMDGPU_INTRIN_IMAGE_STORE: {
    const AMDGPU::ImageDimIntrinsicInfo *Intr
      = AMDGPU::getImageDimIntrinsicInfo(I.getIntrinsicID());
    assert(Intr && "not an image intrinsic with image pseudo");
    return selectImageIntrinsic(I, Intr);
  }
  case AMDGPU::G_AMDGPU_INTRIN_BVH_INTERSECT_RAY:
    return selectBVHIntrinsic(I);
  case AMDGPU::G_AMDGPU_BUFFER_ATOMIC_FADD:
    return selectAMDGPU_BUFFER_ATOMIC_FADD(I);
  default:
    return selectImpl(I, *CoverageInfo);
  }
  return false;
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVCSRC(MachineOperand &Root) const {
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.add(Root); }
  }};

}

std::pair<Register, unsigned>
AMDGPUInstructionSelector::selectVOP3ModsImpl(MachineOperand &Root,
                                              bool AllowAbs) const {
  Register Src = Root.getReg();
  Register OrigSrc = Src;
  unsigned Mods = 0;
  MachineInstr *MI = getDefIgnoringCopies(Src, *MRI);

  if (MI && MI->getOpcode() == AMDGPU::G_FNEG) {
    Src = MI->getOperand(1).getReg();
    Mods |= SISrcMods::NEG;
    MI = getDefIgnoringCopies(Src, *MRI);
  }

  if (AllowAbs && MI && MI->getOpcode() == AMDGPU::G_FABS) {
    Src = MI->getOperand(1).getReg();
    Mods |= SISrcMods::ABS;
  }

  if (Mods != 0 &&
      RBI.getRegBank(Src, *MRI, TRI)->getID() != AMDGPU::VGPRRegBankID) {
    MachineInstr *UseMI = Root.getParent();

    // If we looked through copies to find source modifiers on an SGPR operand,
    // we now have an SGPR register source. To avoid potentially violating the
    // constant bus restriction, we need to insert a copy to a VGPR.
    Register VGPRSrc = MRI->cloneVirtualRegister(OrigSrc);
    BuildMI(*UseMI->getParent(), UseMI, UseMI->getDebugLoc(),
            TII.get(AMDGPU::COPY), VGPRSrc)
      .addReg(Src);
    Src = VGPRSrc;
  }

  return std::make_pair(Src, Mods);
}

///
/// This will select either an SGPR or VGPR operand and will save us from
/// having to write an extra tablegen pattern.
InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVSRC0(MachineOperand &Root) const {
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.add(Root); }
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3Mods0(MachineOperand &Root) const {
  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3ModsImpl(Root);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); }, // src0_mods
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); },    // clamp
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); }     // omod
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3BMods0(MachineOperand &Root) const {
  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3ModsImpl(Root, /* AllowAbs */ false);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); }, // src0_mods
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); },    // clamp
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); }     // omod
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3OMods(MachineOperand &Root) const {
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.add(Root); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); }, // clamp
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); }  // omod
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3Mods(MachineOperand &Root) const {
  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3ModsImpl(Root);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); }  // src_mods
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3BMods(MachineOperand &Root) const {
  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3ModsImpl(Root, /* AllowAbs */ false);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); } // src_mods
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3NoMods(MachineOperand &Root) const {
  Register Reg = Root.getReg();
  const MachineInstr *Def = getDefIgnoringCopies(Reg, *MRI);
  if (Def && (Def->getOpcode() == AMDGPU::G_FNEG ||
              Def->getOpcode() == AMDGPU::G_FABS))
    return {};
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Reg); },
  }};
}

std::pair<Register, unsigned>
AMDGPUInstructionSelector::selectVOP3PModsImpl(
  Register Src, const MachineRegisterInfo &MRI) const {
  unsigned Mods = 0;
  MachineInstr *MI = MRI.getVRegDef(Src);

  if (MI && MI->getOpcode() == AMDGPU::G_FNEG &&
      // It's possible to see an f32 fneg here, but unlikely.
      // TODO: Treat f32 fneg as only high bit.
      MRI.getType(Src) == LLT::vector(2, 16)) {
    Mods ^= (SISrcMods::NEG | SISrcMods::NEG_HI);
    Src = MI->getOperand(1).getReg();
    MI = MRI.getVRegDef(Src);
  }

  // TODO: Match op_sel through g_build_vector_trunc and g_shuffle_vector.

  // Packed instructions do not have abs modifiers.
  Mods |= SISrcMods::OP_SEL_1;

  return std::make_pair(Src, Mods);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3PMods(MachineOperand &Root) const {
  MachineRegisterInfo &MRI
    = Root.getParent()->getParent()->getParent()->getRegInfo();

  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3PModsImpl(Root.getReg(), MRI);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); }  // src_mods
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3Mods_nnan(MachineOperand &Root) const {
  Register Src;
  unsigned Mods;
  std::tie(Src, Mods) = selectVOP3ModsImpl(Root);
  if (!isKnownNeverNaN(Src, *MRI))
    return None;

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Src); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Mods); }  // src_mods
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectVOP3OpSelMods(MachineOperand &Root) const {
  // FIXME: Handle op_sel
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Root.getReg()); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(0); } // src_mods
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectSmrdImm(MachineOperand &Root) const {
  SmallVector<GEPInfo, 4> AddrInfo;
  getAddrModeInfo(*Root.getParent(), *MRI, AddrInfo);

  if (AddrInfo.empty() || AddrInfo[0].SgprParts.size() != 1)
    return None;

  const GEPInfo &GEPInfo = AddrInfo[0];
  Optional<int64_t> EncodedImm =
      AMDGPU::getSMRDEncodedOffset(STI, GEPInfo.Imm, false);
  if (!EncodedImm)
    return None;

  unsigned PtrReg = GEPInfo.SgprParts[0];
  return {{
    [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrReg); },
    [=](MachineInstrBuilder &MIB) { MIB.addImm(*EncodedImm); }
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectSmrdImm32(MachineOperand &Root) const {
  SmallVector<GEPInfo, 4> AddrInfo;
  getAddrModeInfo(*Root.getParent(), *MRI, AddrInfo);

  if (AddrInfo.empty() || AddrInfo[0].SgprParts.size() != 1)
    return None;

  const GEPInfo &GEPInfo = AddrInfo[0];
  Register PtrReg = GEPInfo.SgprParts[0];
  Optional<int64_t> EncodedImm =
      AMDGPU::getSMRDEncodedLiteralOffset32(STI, GEPInfo.Imm);
  if (!EncodedImm)
    return None;

  return {{
    [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrReg); },
    [=](MachineInstrBuilder &MIB) { MIB.addImm(*EncodedImm); }
  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectSmrdSgpr(MachineOperand &Root) const {
  MachineInstr *MI = Root.getParent();
  MachineBasicBlock *MBB = MI->getParent();

  SmallVector<GEPInfo, 4> AddrInfo;
  getAddrModeInfo(*MI, *MRI, AddrInfo);

  // FIXME: We should shrink the GEP if the offset is known to be <= 32-bits,
  // then we can select all ptr + 32-bit offsets not just immediate offsets.
  if (AddrInfo.empty() || AddrInfo[0].SgprParts.size() != 1)
    return None;

  const GEPInfo &GEPInfo = AddrInfo[0];
  // SGPR offset is unsigned.
  if (!GEPInfo.Imm || GEPInfo.Imm < 0 || !isUInt<32>(GEPInfo.Imm))
    return None;

  // If we make it this far we have a load with an 32-bit immediate offset.
  // It is OK to select this using a sgpr offset, because we have already
  // failed trying to select this load into one of the _IMM variants since
  // the _IMM Patterns are considered before the _SGPR patterns.
  Register PtrReg = GEPInfo.SgprParts[0];
  Register OffsetReg = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  BuildMI(*MBB, MI, MI->getDebugLoc(), TII.get(AMDGPU::S_MOV_B32), OffsetReg)
          .addImm(GEPInfo.Imm);
  return {{
    [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrReg); },
    [=](MachineInstrBuilder &MIB) { MIB.addReg(OffsetReg); }
  }};
}

template <bool Signed>
std::pair<Register, int>
AMDGPUInstructionSelector::selectFlatOffsetImpl(MachineOperand &Root) const {
  MachineInstr *MI = Root.getParent();

  auto Default = std::make_pair(Root.getReg(), 0);

  if (!STI.hasFlatInstOffsets())
    return Default;

  const MachineInstr *OpDef = MRI->getVRegDef(Root.getReg());
  if (!OpDef || OpDef->getOpcode() != AMDGPU::G_PTR_ADD)
    return Default;

  Optional<int64_t> Offset =
    getConstantVRegVal(OpDef->getOperand(2).getReg(), *MRI);
  if (!Offset.hasValue())
    return Default;

  unsigned AddrSpace = (*MI->memoperands_begin())->getAddrSpace();
  if (!TII.isLegalFLATOffset(Offset.getValue(), AddrSpace, Signed))
    return Default;

  Register BasePtr = OpDef->getOperand(1).getReg();

  return std::make_pair(BasePtr, Offset.getValue());
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectFlatOffset(MachineOperand &Root) const {
  auto PtrWithOffset = selectFlatOffsetImpl<false>(Root);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrWithOffset.first); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(PtrWithOffset.second); },
    }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectFlatOffsetSigned(MachineOperand &Root) const {
  auto PtrWithOffset = selectFlatOffsetImpl<true>(Root);

  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrWithOffset.first); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(PtrWithOffset.second); },
    }};
}

/// Match a zero extend from a 32-bit value to 64-bits.
static Register matchZeroExtendFromS32(MachineRegisterInfo &MRI, Register Reg) {
  Register ZExtSrc;
  if (mi_match(Reg, MRI, m_GZExt(m_Reg(ZExtSrc))))
    return MRI.getType(ZExtSrc) == LLT::scalar(32) ? ZExtSrc : Register();

  // Match legalized form %zext = G_MERGE_VALUES (s32 %x), (s32 0)
  const MachineInstr *Def = getDefIgnoringCopies(Reg, MRI);
  if (Def->getOpcode() != AMDGPU::G_MERGE_VALUES)
    return false;

  if (mi_match(Def->getOperand(2).getReg(), MRI, m_ZeroInt())) {
    return Def->getOperand(1).getReg();
  }

  return Register();
}

// Match (64-bit SGPR base) + (zext vgpr offset) + sext(imm offset)
InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectGlobalSAddr(MachineOperand &Root) const {
  Register Addr = Root.getReg();
  Register PtrBase;
  int64_t ConstOffset;
  int64_t ImmOffset = 0;

  // Match the immediate offset first, which canonically is moved as low as
  // possible.
  std::tie(PtrBase, ConstOffset) = getPtrBaseWithConstantOffset(Addr, *MRI);

  if (ConstOffset != 0) {
    if (TII.isLegalFLATOffset(ConstOffset, AMDGPUAS::GLOBAL_ADDRESS, true)) {
      Addr = PtrBase;
      ImmOffset = ConstOffset;
    } else if (ConstOffset > 0) {
      auto PtrBaseDef = getDefSrcRegIgnoringCopies(PtrBase, *MRI);
      if (!PtrBaseDef)
        return None;

      if (isSGPR(PtrBaseDef->Reg)) {
        // Offset is too large.
        //
        // saddr + large_offset -> saddr + (voffset = large_offset & ~MaxOffset)
        //                         + (large_offset & MaxOffset);
        int64_t SplitImmOffset, RemainderOffset;
        std::tie(SplitImmOffset, RemainderOffset)
          = TII.splitFlatOffset(ConstOffset, AMDGPUAS::GLOBAL_ADDRESS, true);

        if (isUInt<32>(RemainderOffset)) {
          MachineInstr *MI = Root.getParent();
          MachineBasicBlock *MBB = MI->getParent();
          Register HighBits
            = MRI->createVirtualRegister(&AMDGPU::VGPR_32RegClass);

          BuildMI(*MBB, MI, MI->getDebugLoc(), TII.get(AMDGPU::V_MOV_B32_e32),
                  HighBits)
            .addImm(RemainderOffset);

          return {{
            [=](MachineInstrBuilder &MIB) { MIB.addReg(PtrBase); },  // saddr
            [=](MachineInstrBuilder &MIB) { MIB.addReg(HighBits); }, // voffset
            [=](MachineInstrBuilder &MIB) { MIB.addImm(SplitImmOffset); },
          }};
        }
      }
    }
  }

  auto AddrDef = getDefSrcRegIgnoringCopies(Addr, *MRI);
  if (!AddrDef)
    return None;

  // Match the variable offset.
  if (AddrDef->MI->getOpcode() != AMDGPU::G_PTR_ADD) {
    // FIXME: We should probably have folded COPY (G_IMPLICIT_DEF) earlier, and
    // drop this.
    if (AddrDef->MI->getOpcode() == AMDGPU::G_IMPLICIT_DEF ||
        AddrDef->MI->getOpcode() == AMDGPU::G_CONSTANT)
      return None;

    // It's cheaper to materialize a single 32-bit zero for vaddr than the two
    // moves required to copy a 64-bit SGPR to VGPR.
    const Register SAddr = AddrDef->Reg;
    if (!isSGPR(SAddr))
      return None;

    MachineInstr *MI = Root.getParent();
    MachineBasicBlock *MBB = MI->getParent();
    Register VOffset = MRI->createVirtualRegister(&AMDGPU::VGPR_32RegClass);

    BuildMI(*MBB, MI, MI->getDebugLoc(), TII.get(AMDGPU::V_MOV_B32_e32),
            VOffset)
      .addImm(0);

    return {{
        [=](MachineInstrBuilder &MIB) { MIB.addReg(SAddr); },    // saddr
        [=](MachineInstrBuilder &MIB) { MIB.addReg(VOffset); },  // voffset
        [=](MachineInstrBuilder &MIB) { MIB.addImm(ImmOffset); } // offset
    }};
  }

  // Look through the SGPR->VGPR copy.
  Register SAddr =
    getSrcRegIgnoringCopies(AddrDef->MI->getOperand(1).getReg(), *MRI);
  if (!SAddr || !isSGPR(SAddr))
    return None;

  Register PtrBaseOffset = AddrDef->MI->getOperand(2).getReg();

  // It's possible voffset is an SGPR here, but the copy to VGPR will be
  // inserted later.
  Register VOffset = matchZeroExtendFromS32(*MRI, PtrBaseOffset);
  if (!VOffset)
    return None;

  return {{[=](MachineInstrBuilder &MIB) { // saddr
             MIB.addReg(SAddr);
           },
           [=](MachineInstrBuilder &MIB) { // voffset
             MIB.addReg(VOffset);
           },
           [=](MachineInstrBuilder &MIB) { // offset
             MIB.addImm(ImmOffset);
           }}};
}

static bool isStackPtrRelative(const MachinePointerInfo &PtrInfo) {
  auto PSV = PtrInfo.V.dyn_cast<const PseudoSourceValue *>();
  return PSV && PSV->isStack();
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFScratchOffen(MachineOperand &Root) const {
  MachineInstr *MI = Root.getParent();
  MachineBasicBlock *MBB = MI->getParent();
  MachineFunction *MF = MBB->getParent();
  const SIMachineFunctionInfo *Info = MF->getInfo<SIMachineFunctionInfo>();

  int64_t Offset = 0;
  if (mi_match(Root.getReg(), *MRI, m_ICst(Offset)) &&
      Offset != TM.getNullPointerValue(AMDGPUAS::PRIVATE_ADDRESS)) {
    Register HighBits = MRI->createVirtualRegister(&AMDGPU::VGPR_32RegClass);

    // TODO: Should this be inside the render function? The iterator seems to
    // move.
    BuildMI(*MBB, MI, MI->getDebugLoc(), TII.get(AMDGPU::V_MOV_B32_e32),
            HighBits)
      .addImm(Offset & ~4095);

    return {{[=](MachineInstrBuilder &MIB) { // rsrc
               MIB.addReg(Info->getScratchRSrcReg());
             },
             [=](MachineInstrBuilder &MIB) { // vaddr
               MIB.addReg(HighBits);
             },
             [=](MachineInstrBuilder &MIB) { // soffset
               const MachineMemOperand *MMO = *MI->memoperands_begin();
               const MachinePointerInfo &PtrInfo = MMO->getPointerInfo();

               if (isStackPtrRelative(PtrInfo))
                 MIB.addReg(Info->getStackPtrOffsetReg());
               else
                 MIB.addImm(0);
             },
             [=](MachineInstrBuilder &MIB) { // offset
               MIB.addImm(Offset & 4095);
             }}};
  }

  assert(Offset == 0 || Offset == -1);

  // Try to fold a frame index directly into the MUBUF vaddr field, and any
  // offsets.
  Optional<int> FI;
  Register VAddr = Root.getReg();
  if (const MachineInstr *RootDef = MRI->getVRegDef(Root.getReg())) {
    if (isBaseWithConstantOffset(Root, *MRI)) {
      const MachineOperand &LHS = RootDef->getOperand(1);
      const MachineOperand &RHS = RootDef->getOperand(2);
      const MachineInstr *LHSDef = MRI->getVRegDef(LHS.getReg());
      const MachineInstr *RHSDef = MRI->getVRegDef(RHS.getReg());
      if (LHSDef && RHSDef) {
        int64_t PossibleOffset =
            RHSDef->getOperand(1).getCImm()->getSExtValue();
        if (SIInstrInfo::isLegalMUBUFImmOffset(PossibleOffset) &&
            (!STI.privateMemoryResourceIsRangeChecked() ||
             KnownBits->signBitIsZero(LHS.getReg()))) {
          if (LHSDef->getOpcode() == AMDGPU::G_FRAME_INDEX)
            FI = LHSDef->getOperand(1).getIndex();
          else
            VAddr = LHS.getReg();
          Offset = PossibleOffset;
        }
      }
    } else if (RootDef->getOpcode() == AMDGPU::G_FRAME_INDEX) {
      FI = RootDef->getOperand(1).getIndex();
    }
  }

  return {{[=](MachineInstrBuilder &MIB) { // rsrc
             MIB.addReg(Info->getScratchRSrcReg());
           },
           [=](MachineInstrBuilder &MIB) { // vaddr
             if (FI.hasValue())
               MIB.addFrameIndex(FI.getValue());
             else
               MIB.addReg(VAddr);
           },
           [=](MachineInstrBuilder &MIB) { // soffset
             // If we don't know this private access is a local stack object, it
             // needs to be relative to the entry point's scratch wave offset.
             // TODO: Should split large offsets that don't fit like above.
             // TODO: Don't use scratch wave offset just because the offset
             // didn't fit.
             if (!Info->isEntryFunction() && FI.hasValue())
               MIB.addReg(Info->getStackPtrOffsetReg());
             else
               MIB.addImm(0);
           },
           [=](MachineInstrBuilder &MIB) { // offset
             MIB.addImm(Offset);
           }}};
}

bool AMDGPUInstructionSelector::isDSOffsetLegal(Register Base,
                                                int64_t Offset) const {
  if (!isUInt<16>(Offset))
    return false;

  if (STI.hasUsableDSOffset() || STI.unsafeDSOffsetFoldingEnabled())
    return true;

  // On Southern Islands instruction with a negative base value and an offset
  // don't seem to work.
  return KnownBits->signBitIsZero(Base);
}

bool AMDGPUInstructionSelector::isDSOffset2Legal(Register Base, int64_t Offset0,
                                                 int64_t Offset1,
                                                 unsigned Size) const {
  if (Offset0 % Size != 0 || Offset1 % Size != 0)
    return false;
  if (!isUInt<8>(Offset0 / Size) || !isUInt<8>(Offset1 / Size))
    return false;

  if (STI.hasUsableDSOffset() || STI.unsafeDSOffsetFoldingEnabled())
    return true;

  // On Southern Islands instruction with a negative base value and an offset
  // don't seem to work.
  return KnownBits->signBitIsZero(Base);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFScratchOffset(
    MachineOperand &Root) const {
  MachineInstr *MI = Root.getParent();
  MachineBasicBlock *MBB = MI->getParent();

  int64_t Offset = 0;
  if (!mi_match(Root.getReg(), *MRI, m_ICst(Offset)) ||
      !SIInstrInfo::isLegalMUBUFImmOffset(Offset))
    return {};

  const MachineFunction *MF = MBB->getParent();
  const SIMachineFunctionInfo *Info = MF->getInfo<SIMachineFunctionInfo>();
  const MachineMemOperand *MMO = *MI->memoperands_begin();
  const MachinePointerInfo &PtrInfo = MMO->getPointerInfo();

  return {{
      [=](MachineInstrBuilder &MIB) { // rsrc
        MIB.addReg(Info->getScratchRSrcReg());
      },
      [=](MachineInstrBuilder &MIB) { // soffset
        if (isStackPtrRelative(PtrInfo))
          MIB.addReg(Info->getStackPtrOffsetReg());
        else
          MIB.addImm(0);
      },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset); } // offset
  }};
}

std::pair<Register, unsigned>
AMDGPUInstructionSelector::selectDS1Addr1OffsetImpl(MachineOperand &Root) const {
  const MachineInstr *RootDef = MRI->getVRegDef(Root.getReg());
  if (!RootDef)
    return std::make_pair(Root.getReg(), 0);

  int64_t ConstAddr = 0;

  Register PtrBase;
  int64_t Offset;
  std::tie(PtrBase, Offset) =
    getPtrBaseWithConstantOffset(Root.getReg(), *MRI);

  if (Offset) {
    if (isDSOffsetLegal(PtrBase, Offset)) {
      // (add n0, c0)
      return std::make_pair(PtrBase, Offset);
    }
  } else if (RootDef->getOpcode() == AMDGPU::G_SUB) {
    // TODO


  } else if (mi_match(Root.getReg(), *MRI, m_ICst(ConstAddr))) {
    // TODO

  }

  return std::make_pair(Root.getReg(), 0);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectDS1Addr1Offset(MachineOperand &Root) const {
  Register Reg;
  unsigned Offset;
  std::tie(Reg, Offset) = selectDS1Addr1OffsetImpl(Root);
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Reg); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset); }
    }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectDS64Bit4ByteAligned(MachineOperand &Root) const {
  return selectDSReadWrite2(Root, 4);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectDS128Bit8ByteAligned(MachineOperand &Root) const {
  return selectDSReadWrite2(Root, 8);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectDSReadWrite2(MachineOperand &Root,
                                              unsigned Size) const {
  Register Reg;
  unsigned Offset;
  std::tie(Reg, Offset) = selectDSReadWrite2Impl(Root, Size);
  return {{
      [=](MachineInstrBuilder &MIB) { MIB.addReg(Reg); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset); },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset+1); }
    }};
}

std::pair<Register, unsigned>
AMDGPUInstructionSelector::selectDSReadWrite2Impl(MachineOperand &Root,
                                                  unsigned Size) const {
  const MachineInstr *RootDef = MRI->getVRegDef(Root.getReg());
  if (!RootDef)
    return std::make_pair(Root.getReg(), 0);

  int64_t ConstAddr = 0;

  Register PtrBase;
  int64_t Offset;
  std::tie(PtrBase, Offset) =
    getPtrBaseWithConstantOffset(Root.getReg(), *MRI);

  if (Offset) {
    int64_t OffsetValue0 = Offset;
    int64_t OffsetValue1 = Offset + Size;
    if (isDSOffset2Legal(PtrBase, OffsetValue0, OffsetValue1, Size)) {
      // (add n0, c0)
      return std::make_pair(PtrBase, OffsetValue0 / Size);
    }
  } else if (RootDef->getOpcode() == AMDGPU::G_SUB) {
    // TODO

  } else if (mi_match(Root.getReg(), *MRI, m_ICst(ConstAddr))) {
    // TODO

  }

  return std::make_pair(Root.getReg(), 0);
}

/// If \p Root is a G_PTR_ADD with a G_CONSTANT on the right hand side, return
/// the base value with the constant offset. There may be intervening copies
/// between \p Root and the identified constant. Returns \p Root, 0 if this does
/// not match the pattern.
std::pair<Register, int64_t>
AMDGPUInstructionSelector::getPtrBaseWithConstantOffset(
  Register Root, const MachineRegisterInfo &MRI) const {
  MachineInstr *RootI = getDefIgnoringCopies(Root, MRI);
  if (RootI->getOpcode() != TargetOpcode::G_PTR_ADD)
    return {Root, 0};

  MachineOperand &RHS = RootI->getOperand(2);
  Optional<ValueAndVReg> MaybeOffset
    = getConstantVRegValWithLookThrough(RHS.getReg(), MRI, true);
  if (!MaybeOffset)
    return {Root, 0};
  return {RootI->getOperand(1).getReg(), MaybeOffset->Value};
}

static void addZeroImm(MachineInstrBuilder &MIB) {
  MIB.addImm(0);
}

/// Return a resource descriptor for use with an arbitrary 64-bit pointer. If \p
/// BasePtr is not valid, a null base pointer will be used.
static Register buildRSRC(MachineIRBuilder &B, MachineRegisterInfo &MRI,
                          uint32_t FormatLo, uint32_t FormatHi,
                          Register BasePtr) {
  Register RSrc2 = MRI.createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register RSrc3 = MRI.createVirtualRegister(&AMDGPU::SReg_32RegClass);
  Register RSrcHi = MRI.createVirtualRegister(&AMDGPU::SReg_64RegClass);
  Register RSrc = MRI.createVirtualRegister(&AMDGPU::SGPR_128RegClass);

  B.buildInstr(AMDGPU::S_MOV_B32)
    .addDef(RSrc2)
    .addImm(FormatLo);
  B.buildInstr(AMDGPU::S_MOV_B32)
    .addDef(RSrc3)
    .addImm(FormatHi);

  // Build the half of the subregister with the constants before building the
  // full 128-bit register. If we are building multiple resource descriptors,
  // this will allow CSEing of the 2-component register.
  B.buildInstr(AMDGPU::REG_SEQUENCE)
    .addDef(RSrcHi)
    .addReg(RSrc2)
    .addImm(AMDGPU::sub0)
    .addReg(RSrc3)
    .addImm(AMDGPU::sub1);

  Register RSrcLo = BasePtr;
  if (!BasePtr) {
    RSrcLo = MRI.createVirtualRegister(&AMDGPU::SReg_64RegClass);
    B.buildInstr(AMDGPU::S_MOV_B64)
      .addDef(RSrcLo)
      .addImm(0);
  }

  B.buildInstr(AMDGPU::REG_SEQUENCE)
    .addDef(RSrc)
    .addReg(RSrcLo)
    .addImm(AMDGPU::sub0_sub1)
    .addReg(RSrcHi)
    .addImm(AMDGPU::sub2_sub3);

  return RSrc;
}

static Register buildAddr64RSrc(MachineIRBuilder &B, MachineRegisterInfo &MRI,
                                const SIInstrInfo &TII, Register BasePtr) {
  uint64_t DefaultFormat = TII.getDefaultRsrcDataFormat();

  // FIXME: Why are half the "default" bits ignored based on the addressing
  // mode?
  return buildRSRC(B, MRI, 0, Hi_32(DefaultFormat), BasePtr);
}

static Register buildOffsetSrc(MachineIRBuilder &B, MachineRegisterInfo &MRI,
                               const SIInstrInfo &TII, Register BasePtr) {
  uint64_t DefaultFormat = TII.getDefaultRsrcDataFormat();

  // FIXME: Why are half the "default" bits ignored based on the addressing
  // mode?
  return buildRSRC(B, MRI, -1, Hi_32(DefaultFormat), BasePtr);
}

AMDGPUInstructionSelector::MUBUFAddressData
AMDGPUInstructionSelector::parseMUBUFAddress(Register Src) const {
  MUBUFAddressData Data;
  Data.N0 = Src;

  Register PtrBase;
  int64_t Offset;

  std::tie(PtrBase, Offset) = getPtrBaseWithConstantOffset(Src, *MRI);
  if (isUInt<32>(Offset)) {
    Data.N0 = PtrBase;
    Data.Offset = Offset;
  }

  if (MachineInstr *InputAdd
      = getOpcodeDef(TargetOpcode::G_PTR_ADD, Data.N0, *MRI)) {
    Data.N2 = InputAdd->getOperand(1).getReg();
    Data.N3 = InputAdd->getOperand(2).getReg();

    // FIXME: Need to fix extra SGPR->VGPRcopies inserted
    // FIXME: Don't know this was defined by operand 0
    //
    // TODO: Remove this when we have copy folding optimizations after
    // RegBankSelect.
    Data.N2 = getDefIgnoringCopies(Data.N2, *MRI)->getOperand(0).getReg();
    Data.N3 = getDefIgnoringCopies(Data.N3, *MRI)->getOperand(0).getReg();
  }

  return Data;
}

/// Return if the addr64 mubuf mode should be used for the given address.
bool AMDGPUInstructionSelector::shouldUseAddr64(MUBUFAddressData Addr) const {
  // (ptr_add N2, N3) -> addr64, or
  // (ptr_add (ptr_add N2, N3), C1) -> addr64
  if (Addr.N2)
    return true;

  const RegisterBank *N0Bank = RBI.getRegBank(Addr.N0, *MRI, TRI);
  return N0Bank->getID() == AMDGPU::VGPRRegBankID;
}

/// Split an immediate offset \p ImmOffset depending on whether it fits in the
/// immediate field. Modifies \p ImmOffset and sets \p SOffset to the variable
/// component.
void AMDGPUInstructionSelector::splitIllegalMUBUFOffset(
  MachineIRBuilder &B, Register &SOffset, int64_t &ImmOffset) const {
  if (SIInstrInfo::isLegalMUBUFImmOffset(ImmOffset))
    return;

  // Illegal offset, store it in soffset.
  SOffset = MRI->createVirtualRegister(&AMDGPU::SReg_32RegClass);
  B.buildInstr(AMDGPU::S_MOV_B32)
    .addDef(SOffset)
    .addImm(ImmOffset);
  ImmOffset = 0;
}

bool AMDGPUInstructionSelector::selectMUBUFAddr64Impl(
  MachineOperand &Root, Register &VAddr, Register &RSrcReg,
  Register &SOffset, int64_t &Offset) const {
  // FIXME: Predicates should stop this from reaching here.
  // addr64 bit was removed for volcanic islands.
  if (!STI.hasAddr64() || STI.useFlatForGlobal())
    return false;

  MUBUFAddressData AddrData = parseMUBUFAddress(Root.getReg());
  if (!shouldUseAddr64(AddrData))
    return false;

  Register N0 = AddrData.N0;
  Register N2 = AddrData.N2;
  Register N3 = AddrData.N3;
  Offset = AddrData.Offset;

  // Base pointer for the SRD.
  Register SRDPtr;

  if (N2) {
    if (RBI.getRegBank(N2, *MRI, TRI)->getID() == AMDGPU::VGPRRegBankID) {
      assert(N3);
      if (RBI.getRegBank(N3, *MRI, TRI)->getID() == AMDGPU::VGPRRegBankID) {
        // Both N2 and N3 are divergent. Use N0 (the result of the add) as the
        // addr64, and construct the default resource from a 0 address.
        VAddr = N0;
      } else {
        SRDPtr = N3;
        VAddr = N2;
      }
    } else {
      // N2 is not divergent.
      SRDPtr = N2;
      VAddr = N3;
    }
  } else if (RBI.getRegBank(N0, *MRI, TRI)->getID() == AMDGPU::VGPRRegBankID) {
    // Use the default null pointer in the resource
    VAddr = N0;
  } else {
    // N0 -> offset, or
    // (N0 + C1) -> offset
    SRDPtr = N0;
  }

  MachineIRBuilder B(*Root.getParent());
  RSrcReg = buildAddr64RSrc(B, *MRI, TII, SRDPtr);
  splitIllegalMUBUFOffset(B, SOffset, Offset);
  return true;
}

bool AMDGPUInstructionSelector::selectMUBUFOffsetImpl(
  MachineOperand &Root, Register &RSrcReg, Register &SOffset,
  int64_t &Offset) const {

  // FIXME: Pattern should not reach here.
  if (STI.useFlatForGlobal())
    return false;

  MUBUFAddressData AddrData = parseMUBUFAddress(Root.getReg());
  if (shouldUseAddr64(AddrData))
    return false;

  // N0 -> offset, or
  // (N0 + C1) -> offset
  Register SRDPtr = AddrData.N0;
  Offset = AddrData.Offset;

  // TODO: Look through extensions for 32-bit soffset.
  MachineIRBuilder B(*Root.getParent());

  RSrcReg = buildOffsetSrc(B, *MRI, TII, SRDPtr);
  splitIllegalMUBUFOffset(B, SOffset, Offset);
  return true;
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFAddr64(MachineOperand &Root) const {
  Register VAddr;
  Register RSrcReg;
  Register SOffset;
  int64_t Offset = 0;

  if (!selectMUBUFAddr64Impl(Root, VAddr, RSrcReg, SOffset, Offset))
    return {};

  // FIXME: Use defaulted operands for trailing 0s and remove from the complex
  // pattern.
  return {{
      [=](MachineInstrBuilder &MIB) {  // rsrc
        MIB.addReg(RSrcReg);
      },
      [=](MachineInstrBuilder &MIB) { // vaddr
        MIB.addReg(VAddr);
      },
      [=](MachineInstrBuilder &MIB) { // soffset
        if (SOffset)
          MIB.addReg(SOffset);
        else
          MIB.addImm(0);
      },
      [=](MachineInstrBuilder &MIB) { // offset
        MIB.addImm(Offset);
      },
      addZeroImm, //  glc
      addZeroImm, //  slc
      addZeroImm, //  tfe
      addZeroImm, //  dlc
      addZeroImm  //  swz
    }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFOffset(MachineOperand &Root) const {
  Register RSrcReg;
  Register SOffset;
  int64_t Offset = 0;

  if (!selectMUBUFOffsetImpl(Root, RSrcReg, SOffset, Offset))
    return {};

  return {{
      [=](MachineInstrBuilder &MIB) {  // rsrc
        MIB.addReg(RSrcReg);
      },
      [=](MachineInstrBuilder &MIB) { // soffset
        if (SOffset)
          MIB.addReg(SOffset);
        else
          MIB.addImm(0);
      },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset); }, // offset
      addZeroImm, //  glc
      addZeroImm, //  slc
      addZeroImm, //  tfe
      addZeroImm, //  dlc
      addZeroImm  //  swz
    }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFAddr64Atomic(MachineOperand &Root) const {
  Register VAddr;
  Register RSrcReg;
  Register SOffset;
  int64_t Offset = 0;

  if (!selectMUBUFAddr64Impl(Root, VAddr, RSrcReg, SOffset, Offset))
    return {};

  // FIXME: Use defaulted operands for trailing 0s and remove from the complex
  // pattern.
  return {{
      [=](MachineInstrBuilder &MIB) {  // rsrc
        MIB.addReg(RSrcReg);
      },
      [=](MachineInstrBuilder &MIB) { // vaddr
        MIB.addReg(VAddr);
      },
      [=](MachineInstrBuilder &MIB) { // soffset
        if (SOffset)
          MIB.addReg(SOffset);
        else
          MIB.addImm(0);
      },
      [=](MachineInstrBuilder &MIB) { // offset
        MIB.addImm(Offset);
      },
      addZeroImm //  slc
    }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectMUBUFOffsetAtomic(MachineOperand &Root) const {
  Register RSrcReg;
  Register SOffset;
  int64_t Offset = 0;

  if (!selectMUBUFOffsetImpl(Root, RSrcReg, SOffset, Offset))
    return {};

  return {{
      [=](MachineInstrBuilder &MIB) {  // rsrc
        MIB.addReg(RSrcReg);
      },
      [=](MachineInstrBuilder &MIB) { // soffset
        if (SOffset)
          MIB.addReg(SOffset);
        else
          MIB.addImm(0);
      },
      [=](MachineInstrBuilder &MIB) { MIB.addImm(Offset); }, // offset
      addZeroImm //  slc
    }};
}

/// Get an immediate that must be 32-bits, and treated as zero extended.
static Optional<uint64_t> getConstantZext32Val(Register Reg,
                                               const MachineRegisterInfo &MRI) {
  // getConstantVRegVal sexts any values, so see if that matters.
  Optional<int64_t> OffsetVal = getConstantVRegVal(Reg, MRI);
  if (!OffsetVal || !isInt<32>(*OffsetVal))
    return None;
  return Lo_32(*OffsetVal);
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectSMRDBufferImm(MachineOperand &Root) const {
  Optional<uint64_t> OffsetVal = getConstantZext32Val(Root.getReg(), *MRI);
  if (!OffsetVal)
    return {};

  Optional<int64_t> EncodedImm =
      AMDGPU::getSMRDEncodedOffset(STI, *OffsetVal, true);
  if (!EncodedImm)
    return {};

  return {{ [=](MachineInstrBuilder &MIB) { MIB.addImm(*EncodedImm); }  }};
}

InstructionSelector::ComplexRendererFns
AMDGPUInstructionSelector::selectSMRDBufferImm32(MachineOperand &Root) const {
  assert(STI.getGeneration() == AMDGPUSubtarget::SEA_ISLANDS);

  Optional<uint64_t> OffsetVal = getConstantZext32Val(Root.getReg(), *MRI);
  if (!OffsetVal)
    return {};

  Optional<int64_t> EncodedImm
    = AMDGPU::getSMRDEncodedLiteralOffset32(STI, *OffsetVal);
  if (!EncodedImm)
    return {};

  return {{ [=](MachineInstrBuilder &MIB) { MIB.addImm(*EncodedImm); }  }};
}

void AMDGPUInstructionSelector::renderTruncImm32(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(MI.getOpcode() == TargetOpcode::G_CONSTANT && OpIdx == -1 &&
         "Expected G_CONSTANT");
  MIB.addImm(MI.getOperand(1).getCImm()->getSExtValue());
}

void AMDGPUInstructionSelector::renderNegateImm(MachineInstrBuilder &MIB,
                                                const MachineInstr &MI,
                                                int OpIdx) const {
  assert(MI.getOpcode() == TargetOpcode::G_CONSTANT && OpIdx == -1 &&
         "Expected G_CONSTANT");
  MIB.addImm(-MI.getOperand(1).getCImm()->getSExtValue());
}

void AMDGPUInstructionSelector::renderBitcastImm(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(OpIdx == -1);

  const MachineOperand &Op = MI.getOperand(1);
  if (MI.getOpcode() == TargetOpcode::G_FCONSTANT)
    MIB.addImm(Op.getFPImm()->getValueAPF().bitcastToAPInt().getZExtValue());
  else {
    assert(MI.getOpcode() == TargetOpcode::G_CONSTANT && "Expected G_CONSTANT");
    MIB.addImm(Op.getCImm()->getSExtValue());
  }
}

void AMDGPUInstructionSelector::renderPopcntImm(MachineInstrBuilder &MIB,
                                                const MachineInstr &MI,
                                                int OpIdx) const {
  assert(MI.getOpcode() == TargetOpcode::G_CONSTANT && OpIdx == -1 &&
         "Expected G_CONSTANT");
  MIB.addImm(MI.getOperand(1).getCImm()->getValue().countPopulation());
}

/// This only really exists to satisfy DAG type checking machinery, so is a
/// no-op here.
void AMDGPUInstructionSelector::renderTruncTImm(MachineInstrBuilder &MIB,
                                                const MachineInstr &MI,
                                                int OpIdx) const {
  MIB.addImm(MI.getOperand(OpIdx).getImm());
}

void AMDGPUInstructionSelector::renderExtractGLC(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(OpIdx >= 0 && "expected to match an immediate operand");
  MIB.addImm(MI.getOperand(OpIdx).getImm() & 1);
}

void AMDGPUInstructionSelector::renderExtractSLC(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(OpIdx >= 0 && "expected to match an immediate operand");
  MIB.addImm((MI.getOperand(OpIdx).getImm() >> 1) & 1);
}

void AMDGPUInstructionSelector::renderExtractDLC(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(OpIdx >= 0 && "expected to match an immediate operand");
  MIB.addImm((MI.getOperand(OpIdx).getImm() >> 2) & 1);
}

void AMDGPUInstructionSelector::renderExtractSWZ(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  assert(OpIdx >= 0 && "expected to match an immediate operand");
  MIB.addImm((MI.getOperand(OpIdx).getImm() >> 3) & 1);
}

void AMDGPUInstructionSelector::renderFrameIndex(MachineInstrBuilder &MIB,
                                                 const MachineInstr &MI,
                                                 int OpIdx) const {
  MIB.addFrameIndex((MI.getOperand(1).getIndex()));
}

bool AMDGPUInstructionSelector::isInlineImmediate16(int64_t Imm) const {
  return AMDGPU::isInlinableLiteral16(Imm, STI.hasInv2PiInlineImm());
}

bool AMDGPUInstructionSelector::isInlineImmediate32(int64_t Imm) const {
  return AMDGPU::isInlinableLiteral32(Imm, STI.hasInv2PiInlineImm());
}

bool AMDGPUInstructionSelector::isInlineImmediate64(int64_t Imm) const {
  return AMDGPU::isInlinableLiteral64(Imm, STI.hasInv2PiInlineImm());
}

bool AMDGPUInstructionSelector::isInlineImmediate(const APFloat &Imm) const {
  return TII.isInlineConstant(Imm);
}