llvm/lib/Target/X86/X86ISelDAGToDAG.cpp

//===- X86ISelDAGToDAG.cpp - A DAG pattern matching inst selector for X86 -===//
//
//                     The LLVM Compiler Infrastructure
//
// This file was developed by the Evan Cheng and is distributed under
// the University of Illinois Open Source License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file defines a DAG pattern matching instruction selector for X86,
// converting from a legalized dag to a X86 dag.
//
//===----------------------------------------------------------------------===//

#define DEBUG_TYPE "x86-isel"
#include "X86.h"
#include "X86InstrBuilder.h"
#include "X86ISelLowering.h"
#include "X86RegisterInfo.h"
#include "X86Subtarget.h"
#include "X86TargetMachine.h"
#include "llvm/GlobalValue.h"
#include "llvm/Instructions.h"
#include "llvm/Intrinsics.h"
#include "llvm/Support/CFG.h"
#include "llvm/CodeGen/MachineConstantPool.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/SSARegMap.h"
#include "llvm/CodeGen/SelectionDAGISel.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/ADT/Statistic.h"
#include <iostream>
#include <queue>
#include <set>
using namespace llvm;

//===----------------------------------------------------------------------===//
//                      Pattern Matcher Implementation
//===----------------------------------------------------------------------===//

namespace {
  /// X86ISelAddressMode - This corresponds to X86AddressMode, but uses
  /// SDOperand's instead of register numbers for the leaves of the matched
  /// tree.
  struct X86ISelAddressMode {
    enum {
      RegBase,
      FrameIndexBase
    } BaseType;

    struct {            // This is really a union, discriminated by BaseType!
      SDOperand Reg;
      int FrameIndex;
    } Base;

    bool isRIPRel;     // RIP relative?
    unsigned Scale;
    SDOperand IndexReg; 
    unsigned Disp;
    GlobalValue *GV;
    Constant *CP;
    const char *ES;
    int JT;
    unsigned Align;    // CP alignment.

    X86ISelAddressMode()
      : BaseType(RegBase), isRIPRel(false), Scale(1), IndexReg(), Disp(0),
        GV(0), CP(0), ES(0), JT(-1), Align(0) {
    }
  };
}

namespace {
  Statistic<>
  NumFPKill("x86-codegen", "Number of FP_REG_KILL instructions added");

  Statistic<>
  NumLoadMoved("x86-codegen", "Number of loads moved below TokenFactor");

  //===--------------------------------------------------------------------===//
  /// ISel - X86 specific code to select X86 machine instructions for
  /// SelectionDAG operations.
  ///
  class VISIBILITY_HIDDEN X86DAGToDAGISel : public SelectionDAGISel {
    /// ContainsFPCode - Every instruction we select that uses or defines a FP
    /// register should set this to true.
    bool ContainsFPCode;

    /// FastISel - Enable fast(er) instruction selection.
    ///
    bool FastISel;

    /// TM - Keep a reference to X86TargetMachine.
    ///
    X86TargetMachine &TM;

    /// X86Lowering - This object fully describes how to lower LLVM code to an
    /// X86-specific SelectionDAG.
    X86TargetLowering X86Lowering;

    /// Subtarget - Keep a pointer to the X86Subtarget around so that we can
    /// make the right decision when generating code for different targets.
    const X86Subtarget *Subtarget;

    /// GlobalBaseReg - keeps track of the virtual register mapped onto global
    /// base register.
    unsigned GlobalBaseReg;

  public:
    X86DAGToDAGISel(X86TargetMachine &tm, bool fast)
      : SelectionDAGISel(X86Lowering),
        ContainsFPCode(false), FastISel(fast), TM(tm),
        X86Lowering(*TM.getTargetLowering()),
        Subtarget(&TM.getSubtarget<X86Subtarget>()) {}

    virtual bool runOnFunction(Function &Fn) {
      // Make sure we re-emit a set of the global base reg if necessary
      GlobalBaseReg = 0;
      return SelectionDAGISel::runOnFunction(Fn);
    }
   
    virtual const char *getPassName() const {
      return "X86 DAG->DAG Instruction Selection";
    }

    /// InstructionSelectBasicBlock - This callback is invoked by
    /// SelectionDAGISel when it has created a SelectionDAG for us to codegen.
    virtual void InstructionSelectBasicBlock(SelectionDAG &DAG);

    virtual void EmitFunctionEntryCode(Function &Fn, MachineFunction &MF);

    virtual bool CanBeFoldedBy(SDNode *N, SDNode *U);

// Include the pieces autogenerated from the target description.
#include "X86GenDAGISel.inc"

  private:
    SDNode *Select(SDOperand N);

    bool MatchAddress(SDOperand N, X86ISelAddressMode &AM, bool isRoot = true);
    bool SelectAddr(SDOperand N, SDOperand &Base, SDOperand &Scale,
                    SDOperand &Index, SDOperand &Disp);
    bool SelectLEAAddr(SDOperand N, SDOperand &Base, SDOperand &Scale,
                       SDOperand &Index, SDOperand &Disp);
    bool SelectScalarSSELoad(SDOperand N, SDOperand &Base, SDOperand &Scale,
                             SDOperand &Index, SDOperand &Disp,
                             SDOperand &InChain, SDOperand &OutChain);
    bool TryFoldLoad(SDOperand P, SDOperand N,
                     SDOperand &Base, SDOperand &Scale,
                     SDOperand &Index, SDOperand &Disp);
    void InstructionSelectPreprocess(SelectionDAG &DAG);

    /// SelectInlineAsmMemoryOperand - Implement addressing mode selection for
    /// inline asm expressions.
    virtual bool SelectInlineAsmMemoryOperand(const SDOperand &Op,
                                              char ConstraintCode,
                                              std::vector<SDOperand> &OutOps,
                                              SelectionDAG &DAG);
    
    void EmitSpecialCodeForMain(MachineBasicBlock *BB, MachineFrameInfo *MFI);

    inline void getAddressOperands(X86ISelAddressMode &AM, SDOperand &Base, 
                                   SDOperand &Scale, SDOperand &Index,
                                   SDOperand &Disp) {
      Base  = (AM.BaseType == X86ISelAddressMode::FrameIndexBase) ?
        CurDAG->getTargetFrameIndex(AM.Base.FrameIndex, TLI.getPointerTy()) :
        AM.Base.Reg;
      Scale = getI8Imm(AM.Scale);
      Index = AM.IndexReg;
      // These are 32-bit even in 64-bit mode since RIP relative offset
      // is 32-bit.
      if (AM.GV)
        Disp = CurDAG->getTargetGlobalAddress(AM.GV, MVT::i32, AM.Disp);
      else if (AM.CP)
        Disp = CurDAG->getTargetConstantPool(AM.CP, MVT::i32, AM.Align, AM.Disp);
      else if (AM.ES)
        Disp = CurDAG->getTargetExternalSymbol(AM.ES, MVT::i32);
      else if (AM.JT != -1)
        Disp = CurDAG->getTargetJumpTable(AM.JT, MVT::i32);
      else
        Disp = getI32Imm(AM.Disp);
    }

    /// getI8Imm - Return a target constant with the specified value, of type
    /// i8.
    inline SDOperand getI8Imm(unsigned Imm) {
      return CurDAG->getTargetConstant(Imm, MVT::i8);
    }

    /// getI16Imm - Return a target constant with the specified value, of type
    /// i16.
    inline SDOperand getI16Imm(unsigned Imm) {
      return CurDAG->getTargetConstant(Imm, MVT::i16);
    }

    /// getI32Imm - Return a target constant with the specified value, of type
    /// i32.
    inline SDOperand getI32Imm(unsigned Imm) {
      return CurDAG->getTargetConstant(Imm, MVT::i32);
    }

    /// getGlobalBaseReg - insert code into the entry mbb to materialize the PIC
    /// base register.  Return the virtual register that holds this value.
    SDNode *getGlobalBaseReg();

#ifndef NDEBUG
    unsigned Indent;
#endif
  };
}

static SDNode *findFlagUse(SDNode *N) {
  unsigned FlagResNo = N->getNumValues()-1;
  for (SDNode::use_iterator I = N->use_begin(), E = N->use_end(); I != E; ++I) {
    SDNode *User = *I;
    for (unsigned i = 0, e = User->getNumOperands(); i != e; ++i) {
      SDOperand Op = User->getOperand(i);
      if (Op.ResNo == FlagResNo)
        return User;
    }
  }
  return NULL;
}

static void findNonImmUse(SDNode* Use, SDNode* Def, SDNode *Ignore, bool &found,
                          std::set<SDNode *> &Visited) {
  if (found ||
      Use->getNodeId() > Def->getNodeId() ||
      !Visited.insert(Use).second)
    return;

  for (unsigned i = 0, e = Use->getNumOperands(); i != e; ++i) {
    SDNode *N = Use->getOperand(i).Val;
    if (N == Ignore)
      continue;
    if (N != Def) {
      findNonImmUse(N, Def, Ignore, found, Visited);
    } else {
      found = true;
      break;
    }
  }
}

static inline bool isNonImmUse(SDNode* Use, SDNode* Def, SDNode *Ignore=NULL) {
  std::set<SDNode *> Visited;
  bool found = false;
  for (unsigned i = 0, e = Use->getNumOperands(); i != e; ++i) {
    SDNode *N = Use->getOperand(i).Val;
    if (N != Def && N != Ignore) {
      findNonImmUse(N, Def, Ignore, found, Visited);
      if (found) break;
    }
  }

  if (!found && Ignore) {
    // We must be checking for reachability between Def and a flag use. Go down
    // recursively if Use also produces a flag.
    MVT::ValueType VT = Use->getValueType(Use->getNumValues()-1);
    if (VT == MVT::Flag && !Use->use_empty()) {
      SDNode *FU = findFlagUse(Use);
      if (FU)
        return !isNonImmUse(FU, Def, Use);
    }
  }
  return found;
}


bool X86DAGToDAGISel::CanBeFoldedBy(SDNode *N, SDNode *U) {
  // If U use can somehow reach N through another path then U can't fold N or
  // it will create a cycle. e.g. In the following diagram, U can reach N
  // through X. If N is folded into into U, then X is both a predecessor and
  // a successor of U.
  //
  //         [ N ]
  //         ^  ^
  //         |  |
  //        /   \---
  //      /        [X]
  //      |         ^
  //     [U]--------|
  if (!FastISel && !isNonImmUse(U, N)) {
    // If U produces a flag, then it gets (even more) interesting. Since it
    // would have been "glued" together with its flag use, we need to check if
    // it might reach N:
    //
    //      [ N ]
    //        ^ ^
    //        | |
    //       [U] \--
    //        ^   [TF]
    //        |    |
    //         \  /
    //          [FU]
    //
    // If FU (flag use) indirectly reach N (the load), and U fold N (call it
    // NU), then TF is a predecessor of FU and a successor of NU. But since
    // NU and FU are flagged together, this effectively creates a cycle.
    MVT::ValueType VT = U->getValueType(U->getNumValues()-1);
    if (VT == MVT::Flag && !U->use_empty()) {
      SDNode *FU = findFlagUse(U);
      if (FU)
        return !isNonImmUse(FU, N, U);
    }
    return true;
  }
  return false;
}

/// MoveBelowTokenFactor - Replace TokenFactor operand with load's chain operand
/// and move load below the TokenFactor. Replace store's chain operand with
/// load's chain result.
static void MoveBelowTokenFactor(SelectionDAG &DAG, SDOperand Load,
                                 SDOperand Store, SDOperand TF) {
  std::vector<SDOperand> Ops;
  for (unsigned i = 0, e = TF.Val->getNumOperands(); i != e; ++i)
    if (Load.Val == TF.Val->getOperand(i).Val)
      Ops.push_back(Load.Val->getOperand(0));
    else
      Ops.push_back(TF.Val->getOperand(i));
  DAG.UpdateNodeOperands(TF, &Ops[0], Ops.size());
  DAG.UpdateNodeOperands(Load, TF, Load.getOperand(1), Load.getOperand(2));
  DAG.UpdateNodeOperands(Store, Load.getValue(1), Store.getOperand(1),
                         Store.getOperand(2), Store.getOperand(3));
}

/// InstructionSelectPreprocess - Preprocess the DAG to allow the instruction
/// selector to pick more load-modify-store instructions. This is a common
/// case:
///
///     [Load chain]
///         ^
///         |
///       [Load]
///       ^    ^
///       |    |
///      /      \-
///     /         |
/// [TokenFactor] [Op]
///     ^          ^
///     |          |
///      \        /
///       \      /
///       [Store]
///
/// The fact the store's chain operand != load's chain will prevent the
/// (store (op (load))) instruction from being selected. We can transform it to:
///
///     [Load chain]
///         ^
///         |
///    [TokenFactor]
///         ^
///         |
///       [Load]
///       ^    ^
///       |    |
///       |     \- 
///       |       | 
///       |     [Op]
///       |       ^
///       |       |
///       \      /
///        \    /
///       [Store]
void X86DAGToDAGISel::InstructionSelectPreprocess(SelectionDAG &DAG) {
  for (SelectionDAG::allnodes_iterator I = DAG.allnodes_begin(),
         E = DAG.allnodes_end(); I != E; ++I) {
    if (I->getOpcode() != ISD::STORE)
      continue;
    SDOperand Chain = I->getOperand(0);
    if (Chain.Val->getOpcode() != ISD::TokenFactor)
      continue;

    SDOperand N1 = I->getOperand(1);
    SDOperand N2 = I->getOperand(2);
    if (MVT::isFloatingPoint(N1.getValueType()) ||
        MVT::isVector(N1.getValueType()) ||
        !N1.hasOneUse())
      continue;

    bool RModW = false;
    SDOperand Load;
    unsigned Opcode = N1.Val->getOpcode();
    switch (Opcode) {
      case ISD::ADD:
      case ISD::MUL:
      case ISD::AND:
      case ISD::OR:
      case ISD::XOR:
      case ISD::ADDC:
      case ISD::ADDE: {
        SDOperand N10 = N1.getOperand(0);
        SDOperand N11 = N1.getOperand(1);
        if (ISD::isNON_EXTLoad(N10.Val))
          RModW = true;
        else if (ISD::isNON_EXTLoad(N11.Val)) {
          RModW = true;
          std::swap(N10, N11);
        }
        RModW = RModW && N10.Val->isOperand(Chain.Val) && N10.hasOneUse() &&
          (N10.getOperand(1) == N2) &&
          (N10.Val->getValueType(0) == N1.getValueType());
        if (RModW)
          Load = N10;
        break;
      }
      case ISD::SUB:
      case ISD::SHL:
      case ISD::SRA:
      case ISD::SRL:
      case ISD::ROTL:
      case ISD::ROTR:
      case ISD::SUBC:
      case ISD::SUBE:
      case X86ISD::SHLD:
      case X86ISD::SHRD: {
        SDOperand N10 = N1.getOperand(0);
        if (ISD::isNON_EXTLoad(N10.Val))
          RModW = N10.Val->isOperand(Chain.Val) && N10.hasOneUse() &&
            (N10.getOperand(1) == N2) &&
            (N10.Val->getValueType(0) == N1.getValueType());
        if (RModW)
          Load = N10;
        break;
      }
    }

    if (RModW) {
      MoveBelowTokenFactor(DAG, Load, SDOperand(I, 0), Chain);
      ++NumLoadMoved;
    }
  }
}

/// InstructionSelectBasicBlock - This callback is invoked by SelectionDAGISel
/// when it has created a SelectionDAG for us to codegen.
void X86DAGToDAGISel::InstructionSelectBasicBlock(SelectionDAG &DAG) {
  DEBUG(BB->dump());
  MachineFunction::iterator FirstMBB = BB;

  if (!FastISel)
    InstructionSelectPreprocess(DAG);

  // Codegen the basic block.
#ifndef NDEBUG
  DEBUG(std::cerr << "===== Instruction selection begins:\n");
  Indent = 0;
#endif
  DAG.setRoot(SelectRoot(DAG.getRoot()));
#ifndef NDEBUG
  DEBUG(std::cerr << "===== Instruction selection ends:\n");
#endif

  DAG.RemoveDeadNodes();

  // Emit machine code to BB. 
  ScheduleAndEmitDAG(DAG);
  
  // If we are emitting FP stack code, scan the basic block to determine if this
  // block defines any FP values.  If so, put an FP_REG_KILL instruction before
  // the terminator of the block.
  if (!Subtarget->hasSSE2()) {
    // Note that FP stack instructions *are* used in SSE code when returning
    // values, but these are not live out of the basic block, so we don't need
    // an FP_REG_KILL in this case either.
    bool ContainsFPCode = false;
    
    // Scan all of the machine instructions in these MBBs, checking for FP
    // stores.
    MachineFunction::iterator MBBI = FirstMBB;
    do {
      for (MachineBasicBlock::iterator I = MBBI->begin(), E = MBBI->end();
           !ContainsFPCode && I != E; ++I) {
        for (unsigned op = 0, e = I->getNumOperands(); op != e; ++op) {
          if (I->getOperand(op).isRegister() && I->getOperand(op).isDef() &&
              MRegisterInfo::isVirtualRegister(I->getOperand(op).getReg()) &&
              RegMap->getRegClass(I->getOperand(0).getReg()) == 
                X86::RFPRegisterClass) {
            ContainsFPCode = true;
            break;
          }
        }
      }
    } while (!ContainsFPCode && &*(MBBI++) != BB);
    
    // Check PHI nodes in successor blocks.  These PHI's will be lowered to have
    // a copy of the input value in this block.
    if (!ContainsFPCode) {
      // Final check, check LLVM BB's that are successors to the LLVM BB
      // corresponding to BB for FP PHI nodes.
      const BasicBlock *LLVMBB = BB->getBasicBlock();
      const PHINode *PN;
      for (succ_const_iterator SI = succ_begin(LLVMBB), E = succ_end(LLVMBB);
           !ContainsFPCode && SI != E; ++SI) {
        for (BasicBlock::const_iterator II = SI->begin();
             (PN = dyn_cast<PHINode>(II)); ++II) {
          if (PN->getType()->isFloatingPoint()) {
            ContainsFPCode = true;
            break;
          }
        }
      }
    }

    // Finally, if we found any FP code, emit the FP_REG_KILL instruction.
    if (ContainsFPCode) {
      BuildMI(*BB, BB->getFirstTerminator(), X86::FP_REG_KILL, 0);
      ++NumFPKill;
    }
  }
}

/// EmitSpecialCodeForMain - Emit any code that needs to be executed only in
/// the main function.
void X86DAGToDAGISel::EmitSpecialCodeForMain(MachineBasicBlock *BB,
                                             MachineFrameInfo *MFI) {
  if (Subtarget->isTargetCygwin())
    BuildMI(BB, X86::CALLpcrel32, 1).addExternalSymbol("__main");

  // Switch the FPU to 64-bit precision mode for better compatibility and speed.
  int CWFrameIdx = MFI->CreateStackObject(2, 2);
  addFrameReference(BuildMI(BB, X86::FNSTCW16m, 4), CWFrameIdx);

  // Set the high part to be 64-bit precision.
  addFrameReference(BuildMI(BB, X86::MOV8mi, 5),
                    CWFrameIdx, 1).addImm(2);

  // Reload the modified control word now.
  addFrameReference(BuildMI(BB, X86::FLDCW16m, 4), CWFrameIdx);
}

void X86DAGToDAGISel::EmitFunctionEntryCode(Function &Fn, MachineFunction &MF) {
  // If this is main, emit special code for main.
  MachineBasicBlock *BB = MF.begin();
  if (Fn.hasExternalLinkage() && Fn.getName() == "main")
    EmitSpecialCodeForMain(BB, MF.getFrameInfo());
}

/// MatchAddress - Add the specified node to the specified addressing mode,
/// returning true if it cannot be done.  This just pattern matches for the
/// addressing mode
bool X86DAGToDAGISel::MatchAddress(SDOperand N, X86ISelAddressMode &AM,
                                   bool isRoot) {
  // RIP relative addressing: %rip + 32-bit displacement!
  if (AM.isRIPRel) {
    if (!AM.ES && AM.JT != -1 && N.getOpcode() == ISD::Constant) {
      int64_t Val = cast<ConstantSDNode>(N)->getSignExtended();
      if (isInt32(AM.Disp + Val)) {
        AM.Disp += Val;
        return false;
      }
    }
    return true;
  }

  int id = N.Val->getNodeId();
  bool Available = isSelected(id);

  switch (N.getOpcode()) {
  default: break;
  case ISD::Constant: {
    int64_t Val = cast<ConstantSDNode>(N)->getSignExtended();
    if (isInt32(AM.Disp + Val)) {
      AM.Disp += Val;
      return false;
    }
    break;
  }

  case X86ISD::Wrapper:
    // If value is available in a register both base and index components have
    // been picked, we can't fit the result available in the register in the
    // addressing mode. Duplicate GlobalAddress or ConstantPool as displacement.

    // Can't fit GV or CP in addressing mode for X86-64 medium or large code
    // model since the displacement field is 32-bit. Ok for small code model.

    // For X86-64 PIC code, only allow GV / CP + displacement so we can use RIP
    // relative addressing mode.
    if ((!Subtarget->is64Bit() || TM.getCodeModel() == CodeModel::Small) &&
        (!Available || (AM.Base.Reg.Val && AM.IndexReg.Val))) {
      bool isRIP = Subtarget->is64Bit();
      if (isRIP && (AM.Base.Reg.Val || AM.Scale > 1 || AM.IndexReg.Val ||
                    AM.BaseType == X86ISelAddressMode::FrameIndexBase))
        break;
      if (ConstantPoolSDNode *CP =
          dyn_cast<ConstantPoolSDNode>(N.getOperand(0))) {
        if (AM.CP == 0) {
          AM.CP = CP->getConstVal();
          AM.Align = CP->getAlignment();
          AM.Disp += CP->getOffset();
          if (isRIP)
            AM.isRIPRel = true;
          return false;
        }
      } else if (GlobalAddressSDNode *G =
                 dyn_cast<GlobalAddressSDNode>(N.getOperand(0))) {
        if (AM.GV == 0) {
          AM.GV = G->getGlobal();
          AM.Disp += G->getOffset();
          if (isRIP)
            AM.isRIPRel = true;
          return false;
        }
      } else if (isRoot && isRIP) {
        if (ExternalSymbolSDNode *S =
            dyn_cast<ExternalSymbolSDNode>(N.getOperand(0))) {
          AM.ES = S->getSymbol();
          AM.isRIPRel = true;
          return false;
        } else if (JumpTableSDNode *J =
                   dyn_cast<JumpTableSDNode>(N.getOperand(0))) {
          AM.JT = J->getIndex();
          AM.isRIPRel = true;
          return false;
        }
      }
    }
    break;

  case ISD::FrameIndex:
    if (AM.BaseType == X86ISelAddressMode::RegBase && AM.Base.Reg.Val == 0) {
      AM.BaseType = X86ISelAddressMode::FrameIndexBase;
      AM.Base.FrameIndex = cast<FrameIndexSDNode>(N)->getIndex();
      return false;
    }
    break;

  case ISD::SHL:
    if (!Available && AM.IndexReg.Val == 0 && AM.Scale == 1)
      if (ConstantSDNode *CN = dyn_cast<ConstantSDNode>(N.Val->getOperand(1))) {
        unsigned Val = CN->getValue();
        if (Val == 1 || Val == 2 || Val == 3) {
          AM.Scale = 1 << Val;
          SDOperand ShVal = N.Val->getOperand(0);

          // Okay, we know that we have a scale by now.  However, if the scaled
          // value is an add of something and a constant, we can fold the
          // constant into the disp field here.
          if (ShVal.Val->getOpcode() == ISD::ADD && ShVal.hasOneUse() &&
              isa<ConstantSDNode>(ShVal.Val->getOperand(1))) {
            AM.IndexReg = ShVal.Val->getOperand(0);
            ConstantSDNode *AddVal =
              cast<ConstantSDNode>(ShVal.Val->getOperand(1));
            uint64_t Disp = AM.Disp + AddVal->getValue() << Val;
            if (isInt32(Disp))
              AM.Disp = Disp;
            else
              AM.IndexReg = ShVal;
          } else {
            AM.IndexReg = ShVal;
          }
          return false;
        }
      }
    break;

  case ISD::MUL:
    // X*[3,5,9] -> X+X*[2,4,8]
    if (!Available &&
        AM.BaseType == X86ISelAddressMode::RegBase &&
        AM.Base.Reg.Val == 0 &&
        AM.IndexReg.Val == 0)
      if (ConstantSDNode *CN = dyn_cast<ConstantSDNode>(N.Val->getOperand(1)))
        if (CN->getValue() == 3 || CN->getValue() == 5 || CN->getValue() == 9) {
          AM.Scale = unsigned(CN->getValue())-1;

          SDOperand MulVal = N.Val->getOperand(0);
          SDOperand Reg;

          // Okay, we know that we have a scale by now.  However, if the scaled
          // value is an add of something and a constant, we can fold the
          // constant into the disp field here.
          if (MulVal.Val->getOpcode() == ISD::ADD && MulVal.hasOneUse() &&
              isa<ConstantSDNode>(MulVal.Val->getOperand(1))) {
            Reg = MulVal.Val->getOperand(0);
            ConstantSDNode *AddVal =
              cast<ConstantSDNode>(MulVal.Val->getOperand(1));
            uint64_t Disp = AM.Disp + AddVal->getValue() * CN->getValue();
            if (isInt32(Disp))
              AM.Disp = Disp;
            else
              Reg = N.Val->getOperand(0);
          } else {
            Reg = N.Val->getOperand(0);
          }

          AM.IndexReg = AM.Base.Reg = Reg;
          return false;
        }
    break;

  case ISD::ADD: {
    if (!Available) {
      X86ISelAddressMode Backup = AM;
      if (!MatchAddress(N.Val->getOperand(0), AM, false) &&
          !MatchAddress(N.Val->getOperand(1), AM, false))
        return false;
      AM = Backup;
      if (!MatchAddress(N.Val->getOperand(1), AM, false) &&
          !MatchAddress(N.Val->getOperand(0), AM, false))
        return false;
      AM = Backup;
    }
    break;
  }

  case ISD::OR: {
    if (!Available) {
      X86ISelAddressMode Backup = AM;
      // Look for (x << c1) | c2 where (c2 < c1)
      ConstantSDNode *CN = dyn_cast<ConstantSDNode>(N.Val->getOperand(0));
      if (CN && !MatchAddress(N.Val->getOperand(1), AM, false)) {
        if (AM.GV == NULL && AM.Disp == 0 && CN->getValue() < AM.Scale) {
          AM.Disp = CN->getValue();
          return false;
        }
      }
      AM = Backup;
      CN = dyn_cast<ConstantSDNode>(N.Val->getOperand(1));
      if (CN && !MatchAddress(N.Val->getOperand(0), AM, false)) {
        if (AM.GV == NULL && AM.Disp == 0 && CN->getValue() < AM.Scale) {
          AM.Disp = CN->getValue();
          return false;
        }
      }
      AM = Backup;
    }
    break;
  }
  }

  // Is the base register already occupied?
  if (AM.BaseType != X86ISelAddressMode::RegBase || AM.Base.Reg.Val) {
    // If so, check to see if the scale index register is set.
    if (AM.IndexReg.Val == 0) {
      AM.IndexReg = N;
      AM.Scale = 1;
      return false;
    }

    // Otherwise, we cannot select it.
    return true;
  }

  // Default, generate it as a register.
  AM.BaseType = X86ISelAddressMode::RegBase;
  AM.Base.Reg = N;
  return false;
}

/// SelectAddr - returns true if it is able pattern match an addressing mode.
/// It returns the operands which make up the maximal addressing mode it can
/// match by reference.
bool X86DAGToDAGISel::SelectAddr(SDOperand N, SDOperand &Base, SDOperand &Scale,
                                 SDOperand &Index, SDOperand &Disp) {
  X86ISelAddressMode AM;
  if (MatchAddress(N, AM))
    return false;

  MVT::ValueType VT = N.getValueType();
  if (AM.BaseType == X86ISelAddressMode::RegBase) {
    if (!AM.Base.Reg.Val)
      AM.Base.Reg = CurDAG->getRegister(0, VT);
  }

  if (!AM.IndexReg.Val)
    AM.IndexReg = CurDAG->getRegister(0, VT);

  getAddressOperands(AM, Base, Scale, Index, Disp);
  return true;
}

/// SelectScalarSSELoad - Match a scalar SSE load.  In particular, we want to
/// match a load whose top elements are either undef or zeros.  The load flavor
/// is derived from the type of N, which is either v4f32 or v2f64.
bool X86DAGToDAGISel::SelectScalarSSELoad(SDOperand N, SDOperand &Base,
                                          SDOperand &Scale, SDOperand &Index,
                                          SDOperand &Disp, SDOperand &InChain,
                                          SDOperand &OutChain) {
  if (N.getOpcode() == ISD::SCALAR_TO_VECTOR) {
    InChain  = N.getOperand(0);
    if (ISD::isNON_EXTLoad(InChain.Val)) {
      LoadSDNode *LD = cast<LoadSDNode>(InChain);
      SDOperand LoadAddr = LD->getBasePtr();
      if (!SelectAddr(LoadAddr, Base, Scale, Index, Disp))
        return false;
      OutChain = LD->getChain();
      return true;
    }
  }
  // TODO: Also handle the case where we explicitly require zeros in the top
  // elements.  This is a vector shuffle from the zero vector.
  
  return false;
}


/// SelectLEAAddr - it calls SelectAddr and determines if the maximal addressing
/// mode it matches can be cost effectively emitted as an LEA instruction.
bool X86DAGToDAGISel::SelectLEAAddr(SDOperand N, SDOperand &Base,
                                    SDOperand &Scale,
                                    SDOperand &Index, SDOperand &Disp) {
  X86ISelAddressMode AM;
  if (MatchAddress(N, AM))
    return false;

  MVT::ValueType VT = N.getValueType();
  unsigned Complexity = 0;
  if (AM.BaseType == X86ISelAddressMode::RegBase)
    if (AM.Base.Reg.Val)
      Complexity = 1;
    else
      AM.Base.Reg = CurDAG->getRegister(0, VT);
  else if (AM.BaseType == X86ISelAddressMode::FrameIndexBase)
    Complexity = 4;

  if (AM.IndexReg.Val)
    Complexity++;
  else
    AM.IndexReg = CurDAG->getRegister(0, VT);

  if (AM.Scale > 2) 
    Complexity += 2;
  // Don't match just leal(,%reg,2). It's cheaper to do addl %reg, %reg
  else if (AM.Scale > 1)
    Complexity++;

  // FIXME: We are artificially lowering the criteria to turn ADD %reg, $GA
  // to a LEA. This is determined with some expermentation but is by no means
  // optimal (especially for code size consideration). LEA is nice because of
  // its three-address nature. Tweak the cost function again when we can run
  // convertToThreeAddress() at register allocation time.
  if (AM.GV || AM.CP || AM.ES || AM.JT != -1) {
    // For X86-64, we should always use lea to materialize RIP relative
    // addresses.
    if (Subtarget->is64Bit())
      Complexity = 4;
    else
      Complexity += 2;
  }

  if (AM.Disp && (AM.Base.Reg.Val || AM.IndexReg.Val))
    Complexity++;

  if (Complexity > 2) {
    getAddressOperands(AM, Base, Scale, Index, Disp);
    return true;
  }
  return false;
}

bool X86DAGToDAGISel::TryFoldLoad(SDOperand P, SDOperand N,
                                  SDOperand &Base, SDOperand &Scale,
                                  SDOperand &Index, SDOperand &Disp) {
  if (ISD::isNON_EXTLoad(N.Val) &&
      N.hasOneUse() &&
      CanBeFoldedBy(N.Val, P.Val))
    return SelectAddr(N.getOperand(1), Base, Scale, Index, Disp);
  return false;
}

static bool isRegister0(SDOperand Op) {
  if (RegisterSDNode *R = dyn_cast<RegisterSDNode>(Op))
    return (R->getReg() == 0);
  return false;
}

/// getGlobalBaseReg - Output the instructions required to put the
/// base address to use for accessing globals into a register.
///
SDNode *X86DAGToDAGISel::getGlobalBaseReg() {
  assert(!Subtarget->is64Bit() && "X86-64 PIC uses RIP relative addressing");
  if (!GlobalBaseReg) {
    // Insert the set of GlobalBaseReg into the first MBB of the function
    MachineBasicBlock &FirstMBB = BB->getParent()->front();
    MachineBasicBlock::iterator MBBI = FirstMBB.begin();
    SSARegMap *RegMap = BB->getParent()->getSSARegMap();
    // FIXME: when we get to LP64, we will need to create the appropriate
    // type of register here.
    GlobalBaseReg = RegMap->createVirtualRegister(X86::GR32RegisterClass);
    BuildMI(FirstMBB, MBBI, X86::MovePCtoStack, 0);
    BuildMI(FirstMBB, MBBI, X86::POP32r, 1, GlobalBaseReg);
  }
  return CurDAG->getRegister(GlobalBaseReg, TLI.getPointerTy()).Val;
}

static SDNode *FindCallStartFromCall(SDNode *Node) {
  if (Node->getOpcode() == ISD::CALLSEQ_START) return Node;
    assert(Node->getOperand(0).getValueType() == MVT::Other &&
         "Node doesn't have a token chain argument!");
  return FindCallStartFromCall(Node->getOperand(0).Val);
}

SDNode *X86DAGToDAGISel::Select(SDOperand N) {
  SDNode *Node = N.Val;
  MVT::ValueType NVT = Node->getValueType(0);
  unsigned Opc, MOpc;
  unsigned Opcode = Node->getOpcode();

#ifndef NDEBUG
  DEBUG(std::cerr << std::string(Indent, ' '));
  DEBUG(std::cerr << "Selecting: ");
  DEBUG(Node->dump(CurDAG));
  DEBUG(std::cerr << "\n");
  Indent += 2;
#endif

  if (Opcode >= ISD::BUILTIN_OP_END && Opcode < X86ISD::FIRST_NUMBER) {
#ifndef NDEBUG
    DEBUG(std::cerr << std::string(Indent-2, ' '));
    DEBUG(std::cerr << "== ");
    DEBUG(Node->dump(CurDAG));
    DEBUG(std::cerr << "\n");
    Indent -= 2;
#endif
    return NULL;   // Already selected.
  }

  switch (Opcode) {
    default: break;
    case X86ISD::GlobalBaseReg: 
      return getGlobalBaseReg();

    case ISD::ADD: {
      // Turn ADD X, c to MOV32ri X+c. This cannot be done with tblgen'd
      // code and is matched first so to prevent it from being turned into
      // LEA32r X+c.
      // In 64-bit mode, use LEA to take advantage of RIP-relative addressing.
      MVT::ValueType PtrVT = TLI.getPointerTy();
      SDOperand N0 = N.getOperand(0);
      SDOperand N1 = N.getOperand(1);
      if (N.Val->getValueType(0) == PtrVT &&
          N0.getOpcode() == X86ISD::Wrapper &&
          N1.getOpcode() == ISD::Constant) {
        unsigned Offset = (unsigned)cast<ConstantSDNode>(N1)->getValue();
        SDOperand C(0, 0);
        // TODO: handle ExternalSymbolSDNode.
        if (GlobalAddressSDNode *G =
            dyn_cast<GlobalAddressSDNode>(N0.getOperand(0))) {
          C = CurDAG->getTargetGlobalAddress(G->getGlobal(), PtrVT,
                                             G->getOffset() + Offset);
        } else if (ConstantPoolSDNode *CP =
                   dyn_cast<ConstantPoolSDNode>(N0.getOperand(0))) {
          C = CurDAG->getTargetConstantPool(CP->getConstVal(), PtrVT,
                                            CP->getAlignment(),
                                            CP->getOffset()+Offset);
        }

        if (C.Val) {
          if (Subtarget->is64Bit()) {
            SDOperand Ops[] = { CurDAG->getRegister(0, PtrVT), getI8Imm(1),
                                CurDAG->getRegister(0, PtrVT), C };
            return CurDAG->SelectNodeTo(N.Val, X86::LEA64r, MVT::i64, Ops, 4);
          } else
            return CurDAG->SelectNodeTo(N.Val, X86::MOV32ri, PtrVT, C);
        }
      }

      // Other cases are handled by auto-generated code.
      break;
    }

    case ISD::MULHU:
    case ISD::MULHS: {
      if (Opcode == ISD::MULHU)
        switch (NVT) {
        default: assert(0 && "Unsupported VT!");
        case MVT::i8:  Opc = X86::MUL8r;  MOpc = X86::MUL8m;  break;
        case MVT::i16: Opc = X86::MUL16r; MOpc = X86::MUL16m; break;
        case MVT::i32: Opc = X86::MUL32r; MOpc = X86::MUL32m; break;
        case MVT::i64: Opc = X86::MUL64r; MOpc = X86::MUL64m; break;
        }
      else
        switch (NVT) {
        default: assert(0 && "Unsupported VT!");
        case MVT::i8:  Opc = X86::IMUL8r;  MOpc = X86::IMUL8m;  break;
        case MVT::i16: Opc = X86::IMUL16r; MOpc = X86::IMUL16m; break;
        case MVT::i32: Opc = X86::IMUL32r; MOpc = X86::IMUL32m; break;
        case MVT::i64: Opc = X86::IMUL64r; MOpc = X86::IMUL64m; break;
        }

      unsigned LoReg, HiReg;
      switch (NVT) {
      default: assert(0 && "Unsupported VT!");
      case MVT::i8:  LoReg = X86::AL;  HiReg = X86::AH;  break;
      case MVT::i16: LoReg = X86::AX;  HiReg = X86::DX;  break;
      case MVT::i32: LoReg = X86::EAX; HiReg = X86::EDX; break;
      case MVT::i64: LoReg = X86::RAX; HiReg = X86::RDX; break;
      }

      SDOperand N0 = Node->getOperand(0);
      SDOperand N1 = Node->getOperand(1);

      bool foldedLoad = false;
      SDOperand Tmp0, Tmp1, Tmp2, Tmp3;
      foldedLoad = TryFoldLoad(N, N1, Tmp0, Tmp1, Tmp2, Tmp3);
      // MULHU and MULHS are commmutative
      if (!foldedLoad) {
        foldedLoad = TryFoldLoad(N, N0, Tmp0, Tmp1, Tmp2, Tmp3);
        if (foldedLoad) {
          N0 = Node->getOperand(1);
          N1 = Node->getOperand(0);
        }
      }

      SDOperand Chain;
      if (foldedLoad) {
        Chain = N1.getOperand(0);
        AddToISelQueue(Chain);
      } else
        Chain = CurDAG->getEntryNode();

      SDOperand InFlag(0, 0);
      AddToISelQueue(N0);
      Chain  = CurDAG->getCopyToReg(Chain, CurDAG->getRegister(LoReg, NVT),
                                    N0, InFlag);
      InFlag = Chain.getValue(1);

      if (foldedLoad) {
        AddToISelQueue(Tmp0);
        AddToISelQueue(Tmp1);
        AddToISelQueue(Tmp2);
        AddToISelQueue(Tmp3);
        SDOperand Ops[] = { Tmp0, Tmp1, Tmp2, Tmp3, Chain, InFlag };
        SDNode *CNode =
          CurDAG->getTargetNode(MOpc, MVT::Other, MVT::Flag, Ops, 6);
        Chain  = SDOperand(CNode, 0);
        InFlag = SDOperand(CNode, 1);
      } else {
        AddToISelQueue(N1);
        InFlag =
          SDOperand(CurDAG->getTargetNode(Opc, MVT::Flag, N1, InFlag), 0);
      }

      SDOperand Result = CurDAG->getCopyFromReg(Chain, HiReg, NVT, InFlag);
      ReplaceUses(N.getValue(0), Result);
      if (foldedLoad)
        ReplaceUses(N1.getValue(1), Result.getValue(1));

#ifndef NDEBUG
      DEBUG(std::cerr << std::string(Indent-2, ' '));
      DEBUG(std::cerr << "=> ");
      DEBUG(Result.Val->dump(CurDAG));
      DEBUG(std::cerr << "\n");
      Indent -= 2;
#endif
      return NULL;
    }
      
    case ISD::SDIV:
    case ISD::UDIV:
    case ISD::SREM:
    case ISD::UREM: {
      bool isSigned = Opcode == ISD::SDIV || Opcode == ISD::SREM;
      bool isDiv    = Opcode == ISD::SDIV || Opcode == ISD::UDIV;
      if (!isSigned)
        switch (NVT) {
        default: assert(0 && "Unsupported VT!");
        case MVT::i8:  Opc = X86::DIV8r;  MOpc = X86::DIV8m;  break;
        case MVT::i16: Opc = X86::DIV16r; MOpc = X86::DIV16m; break;
        case MVT::i32: Opc = X86::DIV32r; MOpc = X86::DIV32m; break;
        case MVT::i64: Opc = X86::DIV64r; MOpc = X86::DIV64m; break;
        }
      else
        switch (NVT) {
        default: assert(0 && "Unsupported VT!");
        case MVT::i8:  Opc = X86::IDIV8r;  MOpc = X86::IDIV8m;  break;
        case MVT::i16: Opc = X86::IDIV16r; MOpc = X86::IDIV16m; break;
        case MVT::i32: Opc = X86::IDIV32r; MOpc = X86::IDIV32m; break;
        case MVT::i64: Opc = X86::IDIV64r; MOpc = X86::IDIV64m; break;
        }

      unsigned LoReg, HiReg;
      unsigned ClrOpcode, SExtOpcode;
      switch (NVT) {
      default: assert(0 && "Unsupported VT!");
      case MVT::i8:
        LoReg = X86::AL;  HiReg = X86::AH;
        ClrOpcode  = X86::MOV8r0;
        SExtOpcode = X86::CBW;
        break;
      case MVT::i16:
        LoReg = X86::AX;  HiReg = X86::DX;
        ClrOpcode  = X86::MOV16r0;
        SExtOpcode = X86::CWD;
        break;
      case MVT::i32:
        LoReg = X86::EAX; HiReg = X86::EDX;
        ClrOpcode  = X86::MOV32r0;
        SExtOpcode = X86::CDQ;
        break;
      case MVT::i64:
        LoReg = X86::RAX; HiReg = X86::RDX;
        ClrOpcode  = X86::MOV64r0;
        SExtOpcode = X86::CQO;
        break;
      }

      SDOperand N0 = Node->getOperand(0);
      SDOperand N1 = Node->getOperand(1);

      bool foldedLoad = false;
      SDOperand Tmp0, Tmp1, Tmp2, Tmp3;
      foldedLoad = TryFoldLoad(N, N1, Tmp0, Tmp1, Tmp2, Tmp3);
      SDOperand Chain;
      if (foldedLoad) {
        Chain = N1.getOperand(0);
        AddToISelQueue(Chain);
      } else
        Chain = CurDAG->getEntryNode();

      SDOperand InFlag(0, 0);
      AddToISelQueue(N0);
      Chain  = CurDAG->getCopyToReg(Chain, CurDAG->getRegister(LoReg, NVT),
                                    N0, InFlag);
      InFlag = Chain.getValue(1);

      if (isSigned) {
        // Sign extend the low part into the high part.
        InFlag =
          SDOperand(CurDAG->getTargetNode(SExtOpcode, MVT::Flag, InFlag), 0);
      } else {
        // Zero out the high part, effectively zero extending the input.
        SDOperand ClrNode = SDOperand(CurDAG->getTargetNode(ClrOpcode, NVT), 0);
        Chain  = CurDAG->getCopyToReg(Chain, CurDAG->getRegister(HiReg, NVT),
                                      ClrNode, InFlag);
        InFlag = Chain.getValue(1);
      }

      if (foldedLoad) {
        AddToISelQueue(Tmp0);
        AddToISelQueue(Tmp1);
        AddToISelQueue(Tmp2);
        AddToISelQueue(Tmp3);
        SDOperand Ops[] = { Tmp0, Tmp1, Tmp2, Tmp3, Chain, InFlag };
        SDNode *CNode =
          CurDAG->getTargetNode(MOpc, MVT::Other, MVT::Flag, Ops, 6);
        Chain  = SDOperand(CNode, 0);
        InFlag = SDOperand(CNode, 1);
      } else {
        AddToISelQueue(N1);
        InFlag =
          SDOperand(CurDAG->getTargetNode(Opc, MVT::Flag, N1, InFlag), 0);
      }

      SDOperand Result = CurDAG->getCopyFromReg(Chain, isDiv ? LoReg : HiReg,
                                                NVT, InFlag);
      ReplaceUses(N.getValue(0), Result);
      if (foldedLoad)
        ReplaceUses(N1.getValue(1), Result.getValue(1));

#ifndef NDEBUG
      DEBUG(std::cerr << std::string(Indent-2, ' '));
      DEBUG(std::cerr << "=> ");
      DEBUG(Result.Val->dump(CurDAG));
      DEBUG(std::cerr << "\n");
      Indent -= 2;
#endif

      return NULL;
    }

    case ISD::TRUNCATE: {
      if (!Subtarget->is64Bit() && NVT == MVT::i8) {
        unsigned Opc2;
        MVT::ValueType VT;
        switch (Node->getOperand(0).getValueType()) {
        default: assert(0 && "Unknown truncate!");
        case MVT::i16:
          Opc = X86::MOV16to16_;
          VT = MVT::i16;
          Opc2 = X86::TRUNC_16_to8;
          break;
        case MVT::i32:
          Opc = X86::MOV32to32_;
          VT = MVT::i32;
          Opc2 = X86::TRUNC_32_to8;
          break;
        }

        AddToISelQueue(Node->getOperand(0));
        SDOperand Tmp =
          SDOperand(CurDAG->getTargetNode(Opc, VT, Node->getOperand(0)), 0);
        SDNode *ResNode = CurDAG->getTargetNode(Opc2, NVT, Tmp);
      
#ifndef NDEBUG
        DEBUG(std::cerr << std::string(Indent-2, ' '));
        DEBUG(std::cerr << "=> ");
        DEBUG(ResNode->dump(CurDAG));
        DEBUG(std::cerr << "\n");
        Indent -= 2;
#endif
        return ResNode;
      }

      break;
    }
  }

  SDNode *ResNode = SelectCode(N);

#ifndef NDEBUG
  DEBUG(std::cerr << std::string(Indent-2, ' '));
  DEBUG(std::cerr << "=> ");
  if (ResNode == NULL || ResNode == N.Val)
    DEBUG(N.Val->dump(CurDAG));
  else
    DEBUG(ResNode->dump(CurDAG));
  DEBUG(std::cerr << "\n");
  Indent -= 2;
#endif

  return ResNode;
}

bool X86DAGToDAGISel::
SelectInlineAsmMemoryOperand(const SDOperand &Op, char ConstraintCode,
                             std::vector<SDOperand> &OutOps, SelectionDAG &DAG){
  SDOperand Op0, Op1, Op2, Op3;
  switch (ConstraintCode) {
  case 'o':   // offsetable        ??
  case 'v':   // not offsetable    ??
  default: return true;
  case 'm':   // memory
    if (!SelectAddr(Op, Op0, Op1, Op2, Op3))
      return true;
    break;
  }
  
  OutOps.push_back(Op0);
  OutOps.push_back(Op1);
  OutOps.push_back(Op2);
  OutOps.push_back(Op3);
  AddToISelQueue(Op0);
  AddToISelQueue(Op1);
  AddToISelQueue(Op2);
  AddToISelQueue(Op3);
  return false;
}

/// createX86ISelDag - This pass converts a legalized DAG into a 
/// X86-specific DAG, ready for instruction scheduling.
///
FunctionPass *llvm::createX86ISelDag(X86TargetMachine &TM, bool Fast) {
  return new X86DAGToDAGISel(TM, Fast);
}