//===-- DAGCombiner.cpp - Implement a DAG node combiner -------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This pass combines dag nodes to form fewer, simpler DAG nodes.  It can be run
// both before and after the DAG is legalized.
//
// This pass is not a substitute for the LLVM IR instcombine pass. This pass is
// primarily intended to handle simplification opportunities that are implicit
// in the LLVM IR and exposed by the various codegen lowering phases.
//
//===----------------------------------------------------------------------===//

#define DEBUG_TYPE "dagcombine"
#include "llvm/CodeGen/SelectionDAG.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/DerivedTypes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetOptions.h"
#include "llvm/Target/TargetRegisterInfo.h"
#include "llvm/Target/TargetSubtargetInfo.h"
#include <algorithm>
using namespace llvm;

STATISTIC(NodesCombined   , "Number of dag nodes combined");
STATISTIC(PreIndexedNodes , "Number of pre-indexed nodes created");
STATISTIC(PostIndexedNodes, "Number of post-indexed nodes created");
STATISTIC(OpsNarrowed     , "Number of load/op/store narrowed");
STATISTIC(LdStFP2Int      , "Number of fp load/store pairs transformed to int");
STATISTIC(SlicedLoads, "Number of load sliced");

namespace {
  static cl::opt<bool>
    CombinerAA("combiner-alias-analysis", cl::Hidden,
               cl::desc("Turn on alias analysis during testing"));

  static cl::opt<bool>
    CombinerGlobalAA("combiner-global-alias-analysis", cl::Hidden,
               cl::desc("Include global information in alias analysis"));

  /// Hidden option to stress test load slicing, i.e., when this option
  /// is enabled, load slicing bypasses most of its profitability guards.
  static cl::opt<bool>
  StressLoadSlicing("combiner-stress-load-slicing", cl::Hidden,
                    cl::desc("Bypass the profitability model of load "
                             "slicing"),
                    cl::init(false));

//------------------------------ DAGCombiner ---------------------------------//

  class DAGCombiner {
    SelectionDAG &DAG;
    const TargetLowering &TLI;
    CombineLevel Level;
    CodeGenOpt::Level OptLevel;
    bool LegalOperations;
    bool LegalTypes;
    bool ForCodeSize;

    // Worklist of all of the nodes that need to be simplified.
    //
    // This has the semantics that when adding to the worklist,
    // the item added must be next to be processed. It should
    // also only appear once. The naive approach to this takes
    // linear time.
    //
    // To reduce the insert/remove time to logarithmic, we use
    // a set and a vector to maintain our worklist.
    //
    // The set contains the items on the worklist, but does not
    // maintain the order they should be visited.
    //
    // The vector maintains the order nodes should be visited, but may
    // contain duplicate or removed nodes. When choosing a node to
    // visit, we pop off the order stack until we find an item that is
    // also in the contents set. All operations are O(log N).
    SmallPtrSet<SDNode*, 64> WorkListContents;
    SmallVector<SDNode*, 64> WorkListOrder;

    // AA - Used for DAG load/store alias analysis.
    AliasAnalysis &AA;

    /// AddUsersToWorkList - When an instruction is simplified, add all users of
    /// the instruction to the work lists because they might get more simplified
    /// now.
    ///
    void AddUsersToWorkList(SDNode *N) {
      for (SDNode::use_iterator UI = N->use_begin(), UE = N->use_end();
           UI != UE; ++UI)
        AddToWorkList(*UI);
    }

    /// visit - call the node-specific routine that knows how to fold each
    /// particular type of node.
    SDValue visit(SDNode *N);

  public:
    /// AddToWorkList - Add to the work list making sure its instance is at the
    /// back (next to be processed.)
    void AddToWorkList(SDNode *N) {
      WorkListContents.insert(N);
      WorkListOrder.push_back(N);
    }

    /// removeFromWorkList - remove all instances of N from the worklist.
    ///
    void removeFromWorkList(SDNode *N) {
      WorkListContents.erase(N);
    }

    SDValue CombineTo(SDNode *N, const SDValue *To, unsigned NumTo,
                      bool AddTo = true);

    SDValue CombineTo(SDNode *N, SDValue Res, bool AddTo = true) {
      return CombineTo(N, &Res, 1, AddTo);
    }

    SDValue CombineTo(SDNode *N, SDValue Res0, SDValue Res1,
                      bool AddTo = true) {
      SDValue To[] = { Res0, Res1 };
      return CombineTo(N, To, 2, AddTo);
    }

    void CommitTargetLoweringOpt(const TargetLowering::TargetLoweringOpt &TLO);

  private:

    /// SimplifyDemandedBits - Check the specified integer node value to see if
    /// it can be simplified or if things it uses can be simplified by bit
    /// propagation.  If so, return true.
    bool SimplifyDemandedBits(SDValue Op) {
      unsigned BitWidth = Op.getValueType().getScalarType().getSizeInBits();
      APInt Demanded = APInt::getAllOnesValue(BitWidth);
      return SimplifyDemandedBits(Op, Demanded);
    }

    bool SimplifyDemandedBits(SDValue Op, const APInt &Demanded);

    bool CombineToPreIndexedLoadStore(SDNode *N);
    bool CombineToPostIndexedLoadStore(SDNode *N);
    bool SliceUpLoad(SDNode *N);

    void ReplaceLoadWithPromotedLoad(SDNode *Load, SDNode *ExtLoad);
    SDValue PromoteOperand(SDValue Op, EVT PVT, bool &Replace);
    SDValue SExtPromoteOperand(SDValue Op, EVT PVT);
    SDValue ZExtPromoteOperand(SDValue Op, EVT PVT);
    SDValue PromoteIntBinOp(SDValue Op);
    SDValue PromoteIntShiftOp(SDValue Op);
    SDValue PromoteExtend(SDValue Op);
    bool PromoteLoad(SDValue Op);

    void ExtendSetCCUses(const SmallVectorImpl<SDNode *> &SetCCs,
                         SDValue Trunc, SDValue ExtLoad, SDLoc DL,
                         ISD::NodeType ExtType);

    /// combine - call the node-specific routine that knows how to fold each
    /// particular type of node. If that doesn't do anything, try the
    /// target-specific DAG combines.
    SDValue combine(SDNode *N);

    // Visitation implementation - Implement dag node combining for different
    // node types.  The semantics are as follows:
    // Return Value:
    //   SDValue.getNode() == 0 - No change was made
    //   SDValue.getNode() == N - N was replaced, is dead and has been handled.
    //   otherwise              - N should be replaced by the returned Operand.
    //
    SDValue visitTokenFactor(SDNode *N);
    SDValue visitMERGE_VALUES(SDNode *N);
    SDValue visitADD(SDNode *N);
    SDValue visitSUB(SDNode *N);
    SDValue visitADDC(SDNode *N);
    SDValue visitSUBC(SDNode *N);
    SDValue visitADDE(SDNode *N);
    SDValue visitSUBE(SDNode *N);
    SDValue visitMUL(SDNode *N);
    SDValue visitSDIV(SDNode *N);
    SDValue visitUDIV(SDNode *N);
    SDValue visitSREM(SDNode *N);
    SDValue visitUREM(SDNode *N);
    SDValue visitMULHU(SDNode *N);
    SDValue visitMULHS(SDNode *N);
    SDValue visitSMUL_LOHI(SDNode *N);
    SDValue visitUMUL_LOHI(SDNode *N);
    SDValue visitSMULO(SDNode *N);
    SDValue visitUMULO(SDNode *N);
    SDValue visitSDIVREM(SDNode *N);
    SDValue visitUDIVREM(SDNode *N);
    SDValue visitAND(SDNode *N);
    SDValue visitOR(SDNode *N);
    SDValue visitXOR(SDNode *N);
    SDValue SimplifyVBinOp(SDNode *N);
    SDValue SimplifyVUnaryOp(SDNode *N);
    SDValue visitSHL(SDNode *N);
    SDValue visitSRA(SDNode *N);
    SDValue visitSRL(SDNode *N);
    SDValue visitCTLZ(SDNode *N);
    SDValue visitCTLZ_ZERO_UNDEF(SDNode *N);
    SDValue visitCTTZ(SDNode *N);
    SDValue visitCTTZ_ZERO_UNDEF(SDNode *N);
    SDValue visitCTPOP(SDNode *N);
    SDValue visitSELECT(SDNode *N);
    SDValue visitVSELECT(SDNode *N);
    SDValue visitSELECT_CC(SDNode *N);
    SDValue visitSETCC(SDNode *N);
    SDValue visitSIGN_EXTEND(SDNode *N);
    SDValue visitZERO_EXTEND(SDNode *N);
    SDValue visitANY_EXTEND(SDNode *N);
    SDValue visitSIGN_EXTEND_INREG(SDNode *N);
    SDValue visitTRUNCATE(SDNode *N);
    SDValue visitBITCAST(SDNode *N);
    SDValue visitBUILD_PAIR(SDNode *N);
    SDValue visitFADD(SDNode *N);
    SDValue visitFSUB(SDNode *N);
    SDValue visitFMUL(SDNode *N);
    SDValue visitFMA(SDNode *N);
    SDValue visitFDIV(SDNode *N);
    SDValue visitFREM(SDNode *N);
    SDValue visitFCOPYSIGN(SDNode *N);
    SDValue visitSINT_TO_FP(SDNode *N);
    SDValue visitUINT_TO_FP(SDNode *N);
    SDValue visitFP_TO_SINT(SDNode *N);
    SDValue visitFP_TO_UINT(SDNode *N);
    SDValue visitFP_ROUND(SDNode *N);
    SDValue visitFP_ROUND_INREG(SDNode *N);
    SDValue visitFP_EXTEND(SDNode *N);
    SDValue visitFNEG(SDNode *N);
    SDValue visitFABS(SDNode *N);
    SDValue visitFCEIL(SDNode *N);
    SDValue visitFTRUNC(SDNode *N);
    SDValue visitFFLOOR(SDNode *N);
    SDValue visitBRCOND(SDNode *N);
    SDValue visitBR_CC(SDNode *N);
    SDValue visitLOAD(SDNode *N);
    SDValue visitSTORE(SDNode *N);
    SDValue visitINSERT_VECTOR_ELT(SDNode *N);
    SDValue visitEXTRACT_VECTOR_ELT(SDNode *N);
    SDValue visitBUILD_VECTOR(SDNode *N);
    SDValue visitCONCAT_VECTORS(SDNode *N);
    SDValue visitEXTRACT_SUBVECTOR(SDNode *N);
    SDValue visitVECTOR_SHUFFLE(SDNode *N);

    SDValue XformToShuffleWithZero(SDNode *N);
    SDValue ReassociateOps(unsigned Opc, SDLoc DL, SDValue LHS, SDValue RHS);

    SDValue visitShiftByConstant(SDNode *N, unsigned Amt);

    bool SimplifySelectOps(SDNode *SELECT, SDValue LHS, SDValue RHS);
    SDValue SimplifyBinOpWithSameOpcodeHands(SDNode *N);
    SDValue SimplifySelect(SDLoc DL, SDValue N0, SDValue N1, SDValue N2);
    SDValue SimplifySelectCC(SDLoc DL, SDValue N0, SDValue N1, SDValue N2,
                             SDValue N3, ISD::CondCode CC,
                             bool NotExtCompare = false);
    SDValue SimplifySetCC(EVT VT, SDValue N0, SDValue N1, ISD::CondCode Cond,
                          SDLoc DL, bool foldBooleans = true);
    SDValue SimplifyNodeWithTwoResults(SDNode *N, unsigned LoOp,
                                         unsigned HiOp);
    SDValue CombineConsecutiveLoads(SDNode *N, EVT VT);
    SDValue ConstantFoldBITCASTofBUILD_VECTOR(SDNode *, EVT);
    SDValue BuildSDIV(SDNode *N);
    SDValue BuildUDIV(SDNode *N);
    SDValue MatchBSwapHWordLow(SDNode *N, SDValue N0, SDValue N1,
                               bool DemandHighBits = true);
    SDValue MatchBSwapHWord(SDNode *N, SDValue N0, SDValue N1);
    SDNode *MatchRotate(SDValue LHS, SDValue RHS, SDLoc DL);
    SDValue ReduceLoadWidth(SDNode *N);
    SDValue ReduceLoadOpStoreWidth(SDNode *N);
    SDValue TransformFPLoadStorePair(SDNode *N);
    SDValue reduceBuildVecExtToExtBuildVec(SDNode *N);
    SDValue reduceBuildVecConvertToConvertBuildVec(SDNode *N);

    SDValue GetDemandedBits(SDValue V, const APInt &Mask);

    /// GatherAllAliases - Walk up chain skipping non-aliasing memory nodes,
    /// looking for aliasing nodes and adding them to the Aliases vector.
    void GatherAllAliases(SDNode *N, SDValue OriginalChain,
                          SmallVectorImpl<SDValue> &Aliases);

    /// isAlias - Return true if there is any possibility that the two addresses
    /// overlap.
    bool isAlias(SDValue Ptr1, int64_t Size1, bool IsVolatile1,
                 const Value *SrcValue1, int SrcValueOffset1,
                 unsigned SrcValueAlign1,
                 const MDNode *TBAAInfo1,
                 SDValue Ptr2, int64_t Size2, bool IsVolatile2,
                 const Value *SrcValue2, int SrcValueOffset2,
                 unsigned SrcValueAlign2,
                 const MDNode *TBAAInfo2) const;

    /// isAlias - Return true if there is any possibility that the two addresses
    /// overlap.
    bool isAlias(LSBaseSDNode *Op0, LSBaseSDNode *Op1);

    /// FindAliasInfo - Extracts the relevant alias information from the memory
    /// node.  Returns true if the operand was a load.
    bool FindAliasInfo(SDNode *N,
                       SDValue &Ptr, int64_t &Size, bool &IsVolatile,
                       const Value *&SrcValue, int &SrcValueOffset,
                       unsigned &SrcValueAlignment,
                       const MDNode *&TBAAInfo) const;

    /// FindBetterChain - Walk up chain skipping non-aliasing memory nodes,
    /// looking for a better chain (aliasing node.)
    SDValue FindBetterChain(SDNode *N, SDValue Chain);

    /// Merge consecutive store operations into a wide store.
    /// This optimization uses wide integers or vectors when possible.
    /// \return True if some memory operations were changed.
    bool MergeConsecutiveStores(StoreSDNode *N);

  public:
    DAGCombiner(SelectionDAG &D, AliasAnalysis &A, CodeGenOpt::Level OL)
        : DAG(D), TLI(D.getTargetLoweringInfo()), Level(BeforeLegalizeTypes),
          OptLevel(OL), LegalOperations(false), LegalTypes(false), AA(A) {
      AttributeSet FnAttrs =
          DAG.getMachineFunction().getFunction()->getAttributes();
      ForCodeSize =
          FnAttrs.hasAttribute(AttributeSet::FunctionIndex,
                               Attribute::OptimizeForSize) ||
          FnAttrs.hasAttribute(AttributeSet::FunctionIndex, Attribute::MinSize);
    }

    /// Run - runs the dag combiner on all nodes in the work list
    void Run(CombineLevel AtLevel);

    SelectionDAG &getDAG() const { return DAG; }

    /// getShiftAmountTy - Returns a type large enough to hold any valid
    /// shift amount - before type legalization these can be huge.
    EVT getShiftAmountTy(EVT LHSTy) {
      assert(LHSTy.isInteger() && "Shift amount is not an integer type!");
      if (LHSTy.isVector())
        return LHSTy;
      return LegalTypes ? TLI.getScalarShiftAmountTy(LHSTy)
                        : TLI.getPointerTy();
    }

    /// isTypeLegal - This method returns true if we are running before type
    /// legalization or if the specified VT is legal.
    bool isTypeLegal(const EVT &VT) {
      if (!LegalTypes) return true;
      return TLI.isTypeLegal(VT);
    }

    /// getSetCCResultType - Convenience wrapper around
    /// TargetLowering::getSetCCResultType
    EVT getSetCCResultType(EVT VT) const {
      return TLI.getSetCCResultType(*DAG.getContext(), VT);
    }
  };
}


namespace {
/// WorkListRemover - This class is a DAGUpdateListener that removes any deleted
/// nodes from the worklist.
class WorkListRemover : public SelectionDAG::DAGUpdateListener {
  DAGCombiner &DC;
public:
  explicit WorkListRemover(DAGCombiner &dc)
    : SelectionDAG::DAGUpdateListener(dc.getDAG()), DC(dc) {}

  virtual void NodeDeleted(SDNode *N, SDNode *E) {
    DC.removeFromWorkList(N);
  }
};
}

//===----------------------------------------------------------------------===//
//  TargetLowering::DAGCombinerInfo implementation
//===----------------------------------------------------------------------===//

void TargetLowering::DAGCombinerInfo::AddToWorklist(SDNode *N) {
  ((DAGCombiner*)DC)->AddToWorkList(N);
}

void TargetLowering::DAGCombinerInfo::RemoveFromWorklist(SDNode *N) {
  ((DAGCombiner*)DC)->removeFromWorkList(N);
}

SDValue TargetLowering::DAGCombinerInfo::
CombineTo(SDNode *N, const std::vector<SDValue> &To, bool AddTo) {
  return ((DAGCombiner*)DC)->CombineTo(N, &To[0], To.size(), AddTo);
}

SDValue TargetLowering::DAGCombinerInfo::
CombineTo(SDNode *N, SDValue Res, bool AddTo) {
  return ((DAGCombiner*)DC)->CombineTo(N, Res, AddTo);
}


SDValue TargetLowering::DAGCombinerInfo::
CombineTo(SDNode *N, SDValue Res0, SDValue Res1, bool AddTo) {
  return ((DAGCombiner*)DC)->CombineTo(N, Res0, Res1, AddTo);
}

void TargetLowering::DAGCombinerInfo::
CommitTargetLoweringOpt(const TargetLowering::TargetLoweringOpt &TLO) {
  return ((DAGCombiner*)DC)->CommitTargetLoweringOpt(TLO);
}

//===----------------------------------------------------------------------===//
// Helper Functions
//===----------------------------------------------------------------------===//

/// isNegatibleForFree - Return 1 if we can compute the negated form of the
/// specified expression for the same cost as the expression itself, or 2 if we
/// can compute the negated form more cheaply than the expression itself.
static char isNegatibleForFree(SDValue Op, bool LegalOperations,
                               const TargetLowering &TLI,
                               const TargetOptions *Options,
                               unsigned Depth = 0) {
  // fneg is removable even if it has multiple uses.
  if (Op.getOpcode() == ISD::FNEG) return 2;

  // Don't allow anything with multiple uses.
  if (!Op.hasOneUse()) return 0;

  // Don't recurse exponentially.
  if (Depth > 6) return 0;

  switch (Op.getOpcode()) {
  default: return false;
  case ISD::ConstantFP:
    // Don't invert constant FP values after legalize.  The negated constant
    // isn't necessarily legal.
    return LegalOperations ? 0 : 1;
  case ISD::FADD:
    // FIXME: determine better conditions for this xform.
    if (!Options->UnsafeFPMath) return 0;

    // After operation legalization, it might not be legal to create new FSUBs.
    if (LegalOperations &&
        !TLI.isOperationLegalOrCustom(ISD::FSUB,  Op.getValueType()))
      return 0;

    // fold (fneg (fadd A, B)) -> (fsub (fneg A), B)
    if (char V = isNegatibleForFree(Op.getOperand(0), LegalOperations, TLI,
                                    Options, Depth + 1))
      return V;
    // fold (fneg (fadd A, B)) -> (fsub (fneg B), A)
    return isNegatibleForFree(Op.getOperand(1), LegalOperations, TLI, Options,
                              Depth + 1);
  case ISD::FSUB:
    // We can't turn -(A-B) into B-A when we honor signed zeros.
    if (!Options->UnsafeFPMath) return 0;

    // fold (fneg (fsub A, B)) -> (fsub B, A)
    return 1;

  case ISD::FMUL:
  case ISD::FDIV:
    if (Options->HonorSignDependentRoundingFPMath()) return 0;

    // fold (fneg (fmul X, Y)) -> (fmul (fneg X), Y) or (fmul X, (fneg Y))
    if (char V = isNegatibleForFree(Op.getOperand(0), LegalOperations, TLI,
                                    Options, Depth + 1))
      return V;

    return isNegatibleForFree(Op.getOperand(1), LegalOperations, TLI, Options,
                              Depth + 1);

  case ISD::FP_EXTEND:
  case ISD::FP_ROUND:
  case ISD::FSIN:
    return isNegatibleForFree(Op.getOperand(0), LegalOperations, TLI, Options,
                              Depth + 1);
  }
}

/// GetNegatedExpression - If isNegatibleForFree returns true, this function
/// returns the newly negated expression.
static SDValue GetNegatedExpression(SDValue Op, SelectionDAG &DAG,
                                    bool LegalOperations, unsigned Depth = 0) {
  // fneg is removable even if it has multiple uses.
  if (Op.getOpcode() == ISD::FNEG) return Op.getOperand(0);

  // Don't allow anything with multiple uses.
  assert(Op.hasOneUse() && "Unknown reuse!");

  assert(Depth <= 6 && "GetNegatedExpression doesn't match isNegatibleForFree");
  switch (Op.getOpcode()) {
  default: llvm_unreachable("Unknown code");
  case ISD::ConstantFP: {
    APFloat V = cast<ConstantFPSDNode>(Op)->getValueAPF();
    V.changeSign();
    return DAG.getConstantFP(V, Op.getValueType());
  }
  case ISD::FADD:
    // FIXME: determine better conditions for this xform.
    assert(DAG.getTarget().Options.UnsafeFPMath);

    // fold (fneg (fadd A, B)) -> (fsub (fneg A), B)
    if (isNegatibleForFree(Op.getOperand(0), LegalOperations,
                           DAG.getTargetLoweringInfo(),
                           &DAG.getTarget().Options, Depth+1))
      return DAG.getNode(ISD::FSUB, SDLoc(Op), Op.getValueType(),
                         GetNegatedExpression(Op.getOperand(0), DAG,
                                              LegalOperations, Depth+1),
                         Op.getOperand(1));
    // fold (fneg (fadd A, B)) -> (fsub (fneg B), A)
    return DAG.getNode(ISD::FSUB, SDLoc(Op), Op.getValueType(),
                       GetNegatedExpression(Op.getOperand(1), DAG,
                                            LegalOperations, Depth+1),
                       Op.getOperand(0));
  case ISD::FSUB:
    // We can't turn -(A-B) into B-A when we honor signed zeros.
    assert(DAG.getTarget().Options.UnsafeFPMath);

    // fold (fneg (fsub 0, B)) -> B
    if (ConstantFPSDNode *N0CFP = dyn_cast<ConstantFPSDNode>(Op.getOperand(0)))
      if (N0CFP->getValueAPF().isZero())
        return Op.getOperand(1);

    // fold (fneg (fsub A, B)) -> (fsub B, A)
    return DAG.getNode(ISD::FSUB, SDLoc(Op), Op.getValueType(),
                       Op.getOperand(1), Op.getOperand(0));

  case ISD::FMUL:
  case ISD::FDIV:
    assert(!DAG.getTarget().Options.HonorSignDependentRoundingFPMath());

    // fold (fneg (fmul X, Y)) -> (fmul (fneg X), Y)
    if (isNegatibleForFree(Op.getOperand(0), LegalOperations,
                           DAG.getTargetLoweringInfo(),
                           &DAG.getTarget().Options, Depth+1))
      return DAG.getNode(Op.getOpcode(), SDLoc(Op), Op.getValueType(),
                         GetNegatedExpression(Op.getOperand(0), DAG,
                                              LegalOperations, Depth+1),
                         Op.getOperand(1));

    // fold (fneg (fmul X, Y)) -> (fmul X, (fneg Y))
    return DAG.getNode(Op.getOpcode(), SDLoc(Op), Op.getValueType(),
                       Op.getOperand(0),
                       GetNegatedExpression(Op.getOperand(1), DAG,
                                            LegalOperations, Depth+1));

  case ISD::FP_EXTEND:
  case ISD::FSIN:
    return DAG.getNode(Op.getOpcode(), SDLoc(Op), Op.getValueType(),
                       GetNegatedExpression(Op.getOperand(0), DAG,
                                            LegalOperations, Depth+1));
  case ISD::FP_ROUND:
      return DAG.getNode(ISD::FP_ROUND, SDLoc(Op), Op.getValueType(),
                         GetNegatedExpression(Op.getOperand(0), DAG,
                                              LegalOperations, Depth+1),
                         Op.getOperand(1));
  }
}


// isSetCCEquivalent - Return true if this node is a setcc, or is a select_cc
// that selects between the values 1 and 0, making it equivalent to a setcc.
// Also, set the incoming LHS, RHS, and CC references to the appropriate
// nodes based on the type of node we are checking.  This simplifies life a
// bit for the callers.
static bool isSetCCEquivalent(SDValue N, SDValue &LHS, SDValue &RHS,
                              SDValue &CC) {
  if (N.getOpcode() == ISD::SETCC) {
    LHS = N.getOperand(0);
    RHS = N.getOperand(1);
    CC  = N.getOperand(2);
    return true;
  }
  if (N.getOpcode() == ISD::SELECT_CC &&
      N.getOperand(2).getOpcode() == ISD::Constant &&
      N.getOperand(3).getOpcode() == ISD::Constant &&
      cast<ConstantSDNode>(N.getOperand(2))->getAPIntValue() == 1 &&
      cast<ConstantSDNode>(N.getOperand(3))->isNullValue()) {
    LHS = N.getOperand(0);
    RHS = N.getOperand(1);
    CC  = N.getOperand(4);
    return true;
  }
  return false;
}

// isOneUseSetCC - Return true if this is a SetCC-equivalent operation with only
// one use.  If this is true, it allows the users to invert the operation for
// free when it is profitable to do so.
static bool isOneUseSetCC(SDValue N) {
  SDValue N0, N1, N2;
  if (isSetCCEquivalent(N, N0, N1, N2) && N.getNode()->hasOneUse())
    return true;
  return false;
}

SDValue DAGCombiner::ReassociateOps(unsigned Opc, SDLoc DL,
                                    SDValue N0, SDValue N1) {
  EVT VT = N0.getValueType();
  if (N0.getOpcode() == Opc && isa<ConstantSDNode>(N0.getOperand(1))) {
    if (isa<ConstantSDNode>(N1)) {
      // reassoc. (op (op x, c1), c2) -> (op x, (op c1, c2))
      SDValue OpNode =
        DAG.FoldConstantArithmetic(Opc, VT,
                                   cast<ConstantSDNode>(N0.getOperand(1)),
                                   cast<ConstantSDNode>(N1));
      return DAG.getNode(Opc, DL, VT, N0.getOperand(0), OpNode);
    }
    if (N0.hasOneUse()) {
      // reassoc. (op (op x, c1), y) -> (op (op x, y), c1) iff x+c1 has one use
      SDValue OpNode = DAG.getNode(Opc, SDLoc(N0), VT,
                                   N0.getOperand(0), N1);
      AddToWorkList(OpNode.getNode());
      return DAG.getNode(Opc, DL, VT, OpNode, N0.getOperand(1));
    }
  }

  if (N1.getOpcode() == Opc && isa<ConstantSDNode>(N1.getOperand(1))) {
    if (isa<ConstantSDNode>(N0)) {
      // reassoc. (op c2, (op x, c1)) -> (op x, (op c1, c2))
      SDValue OpNode =
        DAG.FoldConstantArithmetic(Opc, VT,
                                   cast<ConstantSDNode>(N1.getOperand(1)),
                                   cast<ConstantSDNode>(N0));
      return DAG.getNode(Opc, DL, VT, N1.getOperand(0), OpNode);
    }
    if (N1.hasOneUse()) {
      // reassoc. (op y, (op x, c1)) -> (op (op x, y), c1) iff x+c1 has one use
      SDValue OpNode = DAG.getNode(Opc, SDLoc(N0), VT,
                                   N1.getOperand(0), N0);
      AddToWorkList(OpNode.getNode());
      return DAG.getNode(Opc, DL, VT, OpNode, N1.getOperand(1));
    }
  }

  return SDValue();
}

SDValue DAGCombiner::CombineTo(SDNode *N, const SDValue *To, unsigned NumTo,
                               bool AddTo) {
  assert(N->getNumValues() == NumTo && "Broken CombineTo call!");
  ++NodesCombined;
  DEBUG(dbgs() << "\nReplacing.1 ";
        N->dump(&DAG);
        dbgs() << "\nWith: ";
        To[0].getNode()->dump(&DAG);
        dbgs() << " and " << NumTo-1 << " other values\n";
        for (unsigned i = 0, e = NumTo; i != e; ++i)
          assert((!To[i].getNode() ||
                  N->getValueType(i) == To[i].getValueType()) &&
                 "Cannot combine value to value of different type!"));
  WorkListRemover DeadNodes(*this);
  DAG.ReplaceAllUsesWith(N, To);
  if (AddTo) {
    // Push the new nodes and any users onto the worklist
    for (unsigned i = 0, e = NumTo; i != e; ++i) {
      if (To[i].getNode()) {
        AddToWorkList(To[i].getNode());
        AddUsersToWorkList(To[i].getNode());
      }
    }
  }

  // Finally, if the node is now dead, remove it from the graph.  The node
  // may not be dead if the replacement process recursively simplified to
  // something else needing this node.
  if (N->use_empty()) {
    // Nodes can be reintroduced into the worklist.  Make sure we do not
    // process a node that has been replaced.
    removeFromWorkList(N);

    // Finally, since the node is now dead, remove it from the graph.
    DAG.DeleteNode(N);
  }
  return SDValue(N, 0);
}

void DAGCombiner::
CommitTargetLoweringOpt(const TargetLowering::TargetLoweringOpt &TLO) {
  // Replace all uses.  If any nodes become isomorphic to other nodes and
  // are deleted, make sure to remove them from our worklist.
  WorkListRemover DeadNodes(*this);
  DAG.ReplaceAllUsesOfValueWith(TLO.Old, TLO.New);

  // Push the new node and any (possibly new) users onto the worklist.
  AddToWorkList(TLO.New.getNode());
  AddUsersToWorkList(TLO.New.getNode());

  // Finally, if the node is now dead, remove it from the graph.  The node
  // may not be dead if the replacement process recursively simplified to
  // something else needing this node.
  if (TLO.Old.getNode()->use_empty()) {
    removeFromWorkList(TLO.Old.getNode());

    // If the operands of this node are only used by the node, they will now
    // be dead.  Make sure to visit them first to delete dead nodes early.
    for (unsigned i = 0, e = TLO.Old.getNode()->getNumOperands(); i != e; ++i)
      if (TLO.Old.getNode()->getOperand(i).getNode()->hasOneUse())
        AddToWorkList(TLO.Old.getNode()->getOperand(i).getNode());

    DAG.DeleteNode(TLO.Old.getNode());
  }
}

/// SimplifyDemandedBits - Check the specified integer node value to see if
/// it can be simplified or if things it uses can be simplified by bit
/// propagation.  If so, return true.
bool DAGCombiner::SimplifyDemandedBits(SDValue Op, const APInt &Demanded) {
  TargetLowering::TargetLoweringOpt TLO(DAG, LegalTypes, LegalOperations);
  APInt KnownZero, KnownOne;
  if (!TLI.SimplifyDemandedBits(Op, Demanded, KnownZero, KnownOne, TLO))
    return false;

  // Revisit the node.
  AddToWorkList(Op.getNode());

  // Replace the old value with the new one.
  ++NodesCombined;
  DEBUG(dbgs() << "\nReplacing.2 ";
        TLO.Old.getNode()->dump(&DAG);
        dbgs() << "\nWith: ";
        TLO.New.getNode()->dump(&DAG);
        dbgs() << '\n');

  CommitTargetLoweringOpt(TLO);
  return true;
}

void DAGCombiner::ReplaceLoadWithPromotedLoad(SDNode *Load, SDNode *ExtLoad) {
  SDLoc dl(Load);
  EVT VT = Load->getValueType(0);
  SDValue Trunc = DAG.getNode(ISD::TRUNCATE, dl, VT, SDValue(ExtLoad, 0));

  DEBUG(dbgs() << "\nReplacing.9 ";
        Load->dump(&DAG);
        dbgs() << "\nWith: ";
        Trunc.getNode()->dump(&DAG);
        dbgs() << '\n');
  WorkListRemover DeadNodes(*this);
  DAG.ReplaceAllUsesOfValueWith(SDValue(Load, 0), Trunc);
  DAG.ReplaceAllUsesOfValueWith(SDValue(Load, 1), SDValue(ExtLoad, 1));
  removeFromWorkList(Load);
  DAG.DeleteNode(Load);
  AddToWorkList(Trunc.getNode());
}

SDValue DAGCombiner::PromoteOperand(SDValue Op, EVT PVT, bool &Replace) {
  Replace = false;
  SDLoc dl(Op);
  if (LoadSDNode *LD = dyn_cast<LoadSDNode>(Op)) {
    EVT MemVT = LD->getMemoryVT();
    ISD::LoadExtType ExtType = ISD::isNON_EXTLoad(LD)
      ? (TLI.isLoadExtLegal(ISD::ZEXTLOAD, MemVT) ? ISD::ZEXTLOAD
                                                  : ISD::EXTLOAD)
      : LD->getExtensionType();
    Replace = true;
    return DAG.getExtLoad(ExtType, dl, PVT,
                          LD->getChain(), LD->getBasePtr(),
                          MemVT, LD->getMemOperand());
  }

  unsigned Opc = Op.getOpcode();
  switch (Opc) {
  default: break;
  case ISD::AssertSext:
    return DAG.getNode(ISD::AssertSext, dl, PVT,
                       SExtPromoteOperand(Op.getOperand(0), PVT),
                       Op.getOperand(1));
  case ISD::AssertZext:
    return DAG.getNode(ISD::AssertZext, dl, PVT,
                       ZExtPromoteOperand(Op.getOperand(0), PVT),
                       Op.getOperand(1));
  case ISD::Constant: {
    unsigned ExtOpc =
      Op.getValueType().isByteSized() ? ISD::SIGN_EXTEND : ISD::ZERO_EXTEND;
    return DAG.getNode(ExtOpc, dl, PVT, Op);
  }
  }

  if (!TLI.isOperationLegal(ISD::ANY_EXTEND, PVT))
    return SDValue();
  return DAG.getNode(ISD::ANY_EXTEND, dl, PVT, Op);
}

SDValue DAGCombiner::SExtPromoteOperand(SDValue Op, EVT PVT) {
  if (!TLI.isOperationLegal(ISD::SIGN_EXTEND_INREG, PVT))
    return SDValue();
  EVT OldVT = Op.getValueType();
  SDLoc dl(Op);
  bool Replace = false;
  SDValue NewOp = PromoteOperand(Op, PVT, Replace);
  if (NewOp.getNode() == 0)
    return SDValue();
  AddToWorkList(NewOp.getNode());

  if (Replace)
    ReplaceLoadWithPromotedLoad(Op.getNode(), NewOp.getNode());
  return DAG.getNode(ISD::SIGN_EXTEND_INREG, dl, NewOp.getValueType(), NewOp,
                     DAG.getValueType(OldVT));
}

SDValue DAGCombiner::ZExtPromoteOperand(SDValue Op, EVT PVT) {
  EVT OldVT = Op.getValueType();
  SDLoc dl(Op);
  bool Replace = false;
  SDValue NewOp = PromoteOperand(Op, PVT, Replace);
  if (NewOp.getNode() == 0)
    return SDValue();
  AddToWorkList(NewOp.getNode());

  if (Replace)
    ReplaceLoadWithPromotedLoad(Op.getNode(), NewOp.getNode());
  return DAG.getZeroExtendInReg(NewOp, dl, OldVT);
}

/// PromoteIntBinOp - Promote the specified integer binary operation if the
/// target indicates it is beneficial. e.g. On x86, it's usually better to
/// promote i16 operations to i32 since i16 instructions are longer.
SDValue DAGCombiner::PromoteIntBinOp(SDValue Op) {
  if (!LegalOperations)
    return SDValue();

  EVT VT = Op.getValueType();
  if (VT.isVector() || !VT.isInteger())
    return SDValue();

  // If operation type is 'undesirable', e.g. i16 on x86, consider
  // promoting it.
  unsigned Opc = Op.getOpcode();
  if (TLI.isTypeDesirableForOp(Opc, VT))
    return SDValue();

  EVT PVT = VT;
  // Consult target whether it is a good idea to promote this operation and
  // what's the right type to promote it to.
  if (TLI.IsDesirableToPromoteOp(Op, PVT)) {
    assert(PVT != VT && "Don't know what type to promote to!");

    bool Replace0 = false;
    SDValue N0 = Op.getOperand(0);
    SDValue NN0 = PromoteOperand(N0, PVT, Replace0);
    if (NN0.getNode() == 0)
      return SDValue();

    bool Replace1 = false;
    SDValue N1 = Op.getOperand(1);
    SDValue NN1;
    if (N0 == N1)
      NN1 = NN0;
    else {
      NN1 = PromoteOperand(N1, PVT, Replace1);
      if (NN1.getNode() == 0)
        return SDValue();
    }

    AddToWorkList(NN0.getNode());
    if (NN1.getNode())
      AddToWorkList(NN1.getNode());

    if (Replace0)
      ReplaceLoadWithPromotedLoad(N0.getNode(), NN0.getNode());
    if (Replace1)
      ReplaceLoadWithPromotedLoad(N1.getNode(), NN1.getNode());

    DEBUG(dbgs() << "\nPromoting ";
          Op.getNode()->dump(&DAG));
    SDLoc dl(Op);
    return DAG.getNode(ISD::TRUNCATE, dl, VT,
                       DAG.getNode(Opc, dl, PVT, NN0, NN1));
  }
  return SDValue();
}

/// PromoteIntShiftOp - Promote the specified integer shift operation if the
/// target indicates it is beneficial. e.g. On x86, it's usually better to
/// promote i16 operations to i32 since i16 instructions are longer.
SDValue DAGCombiner::PromoteIntShiftOp(SDValue Op) {
  if (!LegalOperations)
    return SDValue();

  EVT VT = Op.getValueType();
  if (VT.isVector() || !VT.isInteger())
    return SDValue();

  // If operation type is 'undesirable', e.g. i16 on x86, consider
  // promoting it.
  unsigned Opc = Op.getOpcode();
  if (TLI.isTypeDesirableForOp(Opc, VT))
    return SDValue();

  EVT PVT = VT;
  // Consult target whether it is a good idea to promote this operation and
  // what's the right type to promote it to.
  if (TLI.IsDesirableToPromoteOp(Op, PVT)) {
    assert(PVT != VT && "Don't know what type to promote to!");

    bool Replace = false;
    SDValue N0 = Op.getOperand(0);
    if (Opc == ISD::SRA)
      N0 = SExtPromoteOperand(Op.getOperand(0), PVT);
    else if (Opc == ISD::SRL)
      N0 = ZExtPromoteOperand(Op.getOperand(0), PVT);
    else
      N0 = PromoteOperand(N0, PVT, Replace);
    if (N0.getNode() == 0)
      return SDValue();

    AddToWorkList(N0.getNode());
    if (Replace)
      ReplaceLoadWithPromotedLoad(Op.getOperand(0).getNode(), N0.getNode());

    DEBUG(dbgs() << "\nPromoting ";
          Op.getNode()->dump(&DAG));
    SDLoc dl(Op);
    return DAG.getNode(ISD::TRUNCATE, dl, VT,
                       DAG.getNode(Opc, dl, PVT, N0, Op.getOperand(1)));
  }
  return SDValue();
}

SDValue DAGCombiner::PromoteExtend(SDValue Op) {
  if (!LegalOperations)
    return SDValue();

  EVT VT = Op.getValueType();
  if (VT.isVector() || !VT.isInteger())
    return SDValue();

  // If operation type is 'undesirable', e.g. i16 on x86, consider
  // promoting it.
  unsigned Opc = Op.getOpcode();
  if (TLI.isTypeDesirableForOp(Opc, VT))
    return SDValue();

  EVT PVT = VT;
  // Consult target whether it is a good idea to promote this operation and
  // what's the right type to promote it to.
  if (TLI.IsDesirableToPromoteOp(Op, PVT)) {
    assert(PVT != VT && "Don't know what type to promote to!");
    // fold (aext (aext x)) -> (aext x)
    // fold (aext (zext x)) -> (zext x)
    // fold (aext (sext x)) -> (sext x)
    DEBUG(dbgs() << "\nPromoting ";
          Op.getNode()->dump(&DAG));
    return DAG.getNode(Op.getOpcode(), SDLoc(Op), VT, Op.getOperand(0));
  }
  return SDValue();
}

bool DAGCombiner::PromoteLoad(SDValue Op) {
  if (!LegalOperations)
    return false;

  EVT VT = Op.getValueType();
  if (VT.isVector() || !VT.isInteger())
    return false;

  // If operation type is 'undesirable', e.g. i16 on x86, consider
  // promoting it.
  unsigned Opc = Op.getOpcode();
  if (TLI.isTypeDesirableForOp(Opc, VT))
    return false;

  EVT PVT = VT;
  // Consult target whether it is a good idea to promote this operation and
  // what's the right type to promote it to.
  if (TLI.IsDesirableToPromoteOp(Op, PVT)) {
    assert(PVT != VT && "Don't know what type to promote to!");

    SDLoc dl(Op);
    SDNode *N = Op.getNode();
    LoadSDNode *LD = cast<LoadSDNode>(N);
    EVT MemVT = LD->getMemoryVT();
    ISD::LoadExtType ExtType = ISD::isNON_EXTLoad(LD)
      ? (TLI.isLoadExtLegal(ISD::ZEXTLOAD, MemVT) ? ISD::ZEXTLOAD
                                                  : ISD::EXTLOAD)
      : LD->getExtensionType();
    SDValue NewLD = DAG.getExtLoad(ExtType, dl, PVT,
                                   LD->getChain(), LD->getBasePtr(),
                                   MemVT, LD->getMemOperand());
    SDValue Result = DAG.getNode(ISD::TRUNCATE, dl, VT, NewLD);

    DEBUG(dbgs() << "\nPromoting ";
          N->dump(&DAG);
          dbgs() << "\nTo: ";
          Result.getNode()->dump(&DAG);
          dbgs() << '\n');
    WorkListRemover DeadNodes(*this);
    DAG.ReplaceAllUsesOfValueWith(SDValue(N, 0), Result);
    DAG.ReplaceAllUsesOfValueWith(SDValue(N, 1), NewLD.getValue(1));
    removeFromWorkList(N);
    DAG.DeleteNode(N);
    AddToWorkList(Result.getNode());
    return true;
  }
  return false;
}


//===----------------------------------------------------------------------===//
//  Main DAG Combiner implementation
//===----------------------------------------------------------------------===//

void DAGCombiner::Run(CombineLevel AtLevel) {
  // set the instance variables, so that the various visit routines may use it.
  Level = AtLevel;
  LegalOperations = Level >= AfterLegalizeVectorOps;
  LegalTypes = Level >= AfterLegalizeTypes;

  // Add all the dag nodes to the worklist.
  for (SelectionDAG::allnodes_iterator I = DAG.allnodes_begin(),
       E = DAG.allnodes_end(); I != E; ++I)
    AddToWorkList(I);

  // Create a dummy node (which is not added to allnodes), that adds a reference
  // to the root node, preventing it from being deleted, and tracking any
  // changes of the root.
  HandleSDNode Dummy(DAG.getRoot());

  // The root of the dag may dangle to deleted nodes until the dag combiner is
  // done.  Set it to null to avoid confusion.
  DAG.setRoot(SDValue());

  // while the worklist isn't empty, find a node and
  // try and combine it.
  while (!WorkListContents.empty()) {
    SDNode *N;
    // The WorkListOrder holds the SDNodes in order, but it may contain
    // duplicates.
    // In order to avoid a linear scan, we use a set (O(log N)) to hold what the
    // worklist *should* contain, and check the node we want to visit is should
    // actually be visited.
    do {
      N = WorkListOrder.pop_back_val();
    } while (!WorkListContents.erase(N));

    // If N has no uses, it is dead.  Make sure to revisit all N's operands once
    // N is deleted from the DAG, since they too may now be dead or may have a
    // reduced number of uses, allowing other xforms.
    if (N->use_empty() && N != &Dummy) {
      for (unsigned i = 0, e = N->getNumOperands(); i != e; ++i)
        AddToWorkList(N->getOperand(i).getNode());

      DAG.DeleteNode(N);
      continue;
    }

    SDValue RV = combine(N);

    if (RV.getNode() == 0)
      continue;

    ++NodesCombined;

    // If we get back the same node we passed in, rather than a new node or
    // zero, we know that the node must have defined multiple values and
    // CombineTo was used.  Since CombineTo takes care of the worklist
    // mechanics for us, we have no work to do in this case.
    if (RV.getNode() == N)
      continue;

    assert(N->getOpcode() != ISD::DELETED_NODE &&
           RV.getNode()->getOpcode() != ISD::DELETED_NODE &&
           "Node was deleted but visit returned new node!");

    DEBUG(dbgs() << "\nReplacing.3 ";
          N->dump(&DAG);
          dbgs() << "\nWith: ";
          RV.getNode()->dump(&DAG);
          dbgs() << '\n');

    // Transfer debug value.
    DAG.TransferDbgValues(SDValue(N, 0), RV);
    WorkListRemover DeadNodes(*this);
    if (N->getNumValues() == RV.getNode()->getNumValues())
      DAG.ReplaceAllUsesWith(N, RV.getNode());
    else {
      assert(N->getValueType(0) == RV.getValueType() &&
             N->getNumValues() == 1 && "Type mismatch");
      SDValue OpV = RV;
      DAG.ReplaceAllUsesWith(N, &OpV);
    }

    // Push the new node and any users onto the worklist
    AddToWorkList(RV.getNode());
    AddUsersToWorkList(RV.getNode());

    // Add any uses of the old node to the worklist in case this node is the
    // last one that uses them.  They may become dead after this node is
    // deleted.
    for (unsigned i = 0, e = N->getNumOperands(); i != e; ++i)
      AddToWorkList(N->getOperand(i).getNode());

    // Finally, if the node is now dead, remove it from the graph.  The node
    // may not be dead if the replacement process recursively simplified to
    // something else needing this node.
    if (N->use_empty()) {
      // Nodes can be reintroduced into the worklist.  Make sure we do not
      // process a node that has been replaced.
      removeFromWorkList(N);

      // Finally, since the node is now dead, remove it from the graph.
      DAG.DeleteNode(N);
    }
  }

  // If the root changed (e.g. it was a dead load, update the root).
  DAG.setRoot(Dummy.getValue());
  DAG.RemoveDeadNodes();
}

SDValue DAGCombiner::visit(SDNode *N) {
  switch (N->getOpcode()) {
  default: break;
  case ISD::TokenFactor:        return visitTokenFactor(N);
  case ISD::MERGE_VALUES:       return visitMERGE_VALUES(N);
  case ISD::ADD:                return visitADD(N);
  case ISD::SUB:                return visitSUB(N);
  case ISD::ADDC:               return visitADDC(N);
  case ISD::SUBC:               return visitSUBC(N);
  case ISD::ADDE:               return visitADDE(N);
  case ISD::SUBE:               return visitSUBE(N);
  case ISD::MUL:                return visitMUL(N);
  case ISD::SDIV:               return visitSDIV(N);
  case ISD::UDIV:               return visitUDIV(N);
  case ISD::SREM:               return visitSREM(N);
  case ISD::UREM:               return visitUREM(N);
  case ISD::MULHU:              return visitMULHU(N);
  case ISD::MULHS:              return visitMULHS(N);
  case ISD::SMUL_LOHI:          return visitSMUL_LOHI(N);
  case ISD::UMUL_LOHI:          return visitUMUL_LOHI(N);
  case ISD::SMULO:              return visitSMULO(N);
  case ISD::UMULO:              return visitUMULO(N);
  case ISD::SDIVREM:            return visitSDIVREM(N);
  case ISD::UDIVREM:            return visitUDIVREM(N);
  case ISD::AND:                return visitAND(N);
  case ISD::OR:                 return visitOR(N);
  case ISD::XOR:                return visitXOR(N);
  case ISD::SHL:                return visitSHL(N);
  case ISD::SRA:                return visitSRA(N);
  case ISD::SRL:                return visitSRL(N);
  case ISD::CTLZ:               return visitCTLZ(N);
  case ISD::CTLZ_ZERO_UNDEF:    return visitCTLZ_ZERO_UNDEF(N);
  case ISD::CTTZ:               return visitCTTZ(N);
  case ISD::CTTZ_ZERO_UNDEF:    return visitCTTZ_ZERO_UNDEF(N);
  case ISD::CTPOP:              return visitCTPOP(N);
  case ISD::SELECT:             return visitSELECT(N);
  case ISD::VSELECT:            return visitVSELECT(N);
  case ISD::SELECT_CC:          return visitSELECT_CC(N);
  case ISD::SETCC:              return visitSETCC(N);
  case ISD::SIGN_EXTEND:        return visitSIGN_EXTEND(N);
  case ISD::ZERO_EXTEND:        return visitZERO_EXTEND(N);
  case ISD::ANY_EXTEND:         return visitANY_EXTEND(N);
  case ISD::SIGN_EXTEND_INREG:  return visitSIGN_EXTEND_INREG(N);
  case ISD::TRUNCATE:           return visitTRUNCATE(N);
  case ISD::BITCAST:            return visitBITCAST(N);
  case ISD::BUILD_PAIR:         return visitBUILD_PAIR(N);
  case ISD::FADD:               return visitFADD(N);
  case ISD::FSUB:               return visitFSUB(N);
  case ISD::FMUL:               return visitFMUL(N);
  case ISD::FMA:                return visitFMA(N);
  case ISD::FDIV:               return visitFDIV(N);
  case ISD::FREM:               return visitFREM(N);
  case ISD::FCOPYSIGN:          return visitFCOPYSIGN(N);
  case ISD::SINT_TO_FP:         return visitSINT_TO_FP(N);
  case ISD::UINT_TO_FP:         return visitUINT_TO_FP(N);
  case ISD::FP_TO_SINT:         return visitFP_TO_SINT(N);
  case ISD::FP_TO_UINT:         return visitFP_TO_UINT(N);
  case ISD::FP_ROUND:           return visitFP_ROUND(N);
  case ISD::FP_ROUND_INREG:     return visitFP_ROUND_INREG(N);
  case ISD::FP_EXTEND:          return visitFP_EXTEND(N);
  case ISD::FNEG:               return visitFNEG(N);
  case ISD::FABS:               return visitFABS(N);
  case ISD::FFLOOR:             return visitFFLOOR(N);
  case ISD::FCEIL:              return visitFCEIL(N);
  case ISD::FTRUNC:             return visitFTRUNC(N);
  case ISD::BRCOND:             return visitBRCOND(N);
  case ISD::BR_CC:              return visitBR_CC(N);
  case ISD::LOAD:               return visitLOAD(N);
  case ISD::STORE:              return visitSTORE(N);
  case ISD::INSERT_VECTOR_ELT:  return visitINSERT_VECTOR_ELT(N);
  case ISD::EXTRACT_VECTOR_ELT: return visitEXTRACT_VECTOR_ELT(N);
  case ISD::BUILD_VECTOR:       return visitBUILD_VECTOR(N);
  case ISD::CONCAT_VECTORS:     return visitCONCAT_VECTORS(N);
  case ISD::EXTRACT_SUBVECTOR:  return visitEXTRACT_SUBVECTOR(N);
  case ISD::VECTOR_SHUFFLE:     return visitVECTOR_SHUFFLE(N);
  }
  return SDValue();
}

SDValue DAGCombiner::combine(SDNode *N) {
  SDValue RV = visit(N);

  // If nothing happened, try a target-specific DAG combine.
  if (RV.getNode() == 0) {
    assert(N->getOpcode() != ISD::DELETED_NODE &&
           "Node was deleted but visit returned NULL!");

    if (N->getOpcode() >= ISD::BUILTIN_OP_END ||
        TLI.hasTargetDAGCombine((ISD::NodeType)N->getOpcode())) {

      // Expose the DAG combiner to the target combiner impls.
      TargetLowering::DAGCombinerInfo
        DagCombineInfo(DAG, Level, false, this);

      RV = TLI.PerformDAGCombine(N, DagCombineInfo);
    }
  }

  // If nothing happened still, try promoting the operation.
  if (RV.getNode() == 0) {
    switch (N->getOpcode()) {
    default: break;
    case ISD::ADD:
    case ISD::SUB:
    case ISD::MUL:
    case ISD::AND:
    case ISD::OR:
    case ISD::XOR:
      RV = PromoteIntBinOp(SDValue(N, 0));
      break;
    case ISD::SHL:
    case ISD::SRA:
    case ISD::SRL:
      RV = PromoteIntShiftOp(SDValue(N, 0));
      break;
    case ISD::SIGN_EXTEND:
    case ISD::ZERO_EXTEND:
    case ISD::ANY_EXTEND:
      RV = PromoteExtend(SDValue(N, 0));
      break;
    case ISD::LOAD:
      if (PromoteLoad(SDValue(N, 0)))
        RV = SDValue(N, 0);
      break;
    }
  }

  // If N is a commutative binary node, try commuting it to enable more
  // sdisel CSE.
  if (RV.getNode() == 0 &&
      SelectionDAG::isCommutativeBinOp(N->getOpcode()) &&
      N->getNumValues() == 1) {
    SDValue N0 = N->getOperand(0);
    SDValue N1 = N->getOperand(1);

    // Constant operands are canonicalized to RHS.
    if (isa<ConstantSDNode>(N0) || !isa<ConstantSDNode>(N1)) {
      SDValue Ops[] = { N1, N0 };
      SDNode *CSENode = DAG.getNodeIfExists(N->getOpcode(), N->getVTList(),
                                            Ops, 2);
      if (CSENode)
        return SDValue(CSENode, 0);
    }
  }

  return RV;
}

/// getInputChainForNode - Given a node, return its input chain if it has one,
/// otherwise return a null sd operand.
static SDValue getInputChainForNode(SDNode *N) {
  if (unsigned NumOps = N->getNumOperands()) {
    if (N->getOperand(0).getValueType() == MVT::Other)
      return N->getOperand(0);
    if (N->getOperand(NumOps-1).getValueType() == MVT::Other)
      return N->getOperand(NumOps-1);
    for (unsigned i = 1; i < NumOps-1; ++i)
      if (N->getOperand(i).getValueType() == MVT::Other)
        return N->getOperand(i);
  }
  return SDValue();
}

SDValue DAGCombiner::visitTokenFactor(SDNode *N) {
  // If N has two operands, where one has an input chain equal to the other,
  // the 'other' chain is redundant.
  if (N->getNumOperands() == 2) {
    if (getInputChainForNode(N->getOperand(0).getNode()) == N->getOperand(1))
      return N->getOperand(0);
    if (getInputChainForNode(N->getOperand(1).getNode()) == N->getOperand(0))
      return N->getOperand(1);
  }

  SmallVector<SDNode *, 8> TFs;     // List of token factors to visit.
  SmallVector<SDValue, 8> Ops;    // Ops for replacing token factor.
  SmallPtrSet<SDNode*, 16> SeenOps;
  bool Changed = false;             // If we should replace this token factor.

  // Start out with this token factor.
  TFs.push_back(N);

  // Iterate through token factors.  The TFs grows when new token factors are
  // encountered.
  for (unsigned i = 0; i < TFs.size(); ++i) {
    SDNode *TF = TFs[i];

    // Check each of the operands.
    for (unsigned i = 0, ie = TF->getNumOperands(); i != ie; ++i) {
      SDValue Op = TF->getOperand(i);

      switch (Op.getOpcode()) {
      case ISD::EntryToken:
        // Entry tokens don't need to be added to the list. They are
        // rededundant.
        Changed = true;
        break;

      case ISD::TokenFactor:
        if (Op.hasOneUse() &&
            std::find(TFs.begin(), TFs.end(), Op.getNode()) == TFs.end()) {
          // Q…