#include "llvm/CodeGen/SelectionDAG.h"
#include "llvm/Constants.h"
-#include "llvm/GlobalValue.h"
+#include "llvm/GlobalVariable.h"
#include "llvm/Intrinsics.h"
+#include "llvm/DerivedTypes.h"
#include "llvm/Assembly/Writer.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineConstantPool.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Target/MRegisterInfo.h"
+#include "llvm/Target/TargetData.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetInstrInfo.h"
#include "llvm/Target/TargetMachine.h"
ID.AddDouble(cast<ConstantFPSDNode>(N)->getValue());
break;
case ISD::TargetGlobalAddress:
- case ISD::GlobalAddress: {
+ case ISD::GlobalAddress:
+ case ISD::TargetGlobalTLSAddress:
+ case ISD::GlobalTLSAddress: {
GlobalAddressSDNode *GA = cast<GlobalAddressSDNode>(N);
ID.AddPointer(GA->getGlobal());
ID.AddInteger(GA->getOffset());
// not subject to CSE.
if (!Erased && N->getValueType(N->getNumValues()-1) != MVT::Flag &&
!N->isTargetOpcode()) {
- N->dump();
+ N->dump(this);
cerr << "\n";
assert(0 && "Node is not in map!");
}
return 0; // Never CSE anything that produces a flag.
FoldingSetNodeID ID;
- AddNodeIDNode(ID, N->getOpcode(), N->getVTList(), 0, 0);
+ AddNodeIDNode(ID, N->getOpcode(), N->getVTList(), Ops, NumOps);
if (const LoadSDNode *LD = dyn_cast<LoadSDNode>(N)) {
ID.AddInteger(LD->getAddressingMode());
ID.AddInteger(ST->isVolatile());
}
- AddNodeIDOperands(ID, Ops, NumOps);
return CSEMap.FindNodeOrInsertPos(ID, InsertPos);
}
SDOperand SelectionDAG::getConstantFP(double Val, MVT::ValueType VT,
bool isTarget) {
assert(MVT::isFloatingPoint(VT) && "Cannot create integer FP constant!");
- if (VT == MVT::f32)
+ MVT::ValueType EltVT =
+ MVT::isVector(VT) ? MVT::getVectorElementType(VT) : VT;
+ if (EltVT == MVT::f32)
Val = (float)Val; // Mask out extra precision.
// Do the map lookup using the actual bit pattern for the floating point
// we don't have issues with SNANs.
unsigned Opc = isTarget ? ISD::TargetConstantFP : ISD::ConstantFP;
FoldingSetNodeID ID;
- AddNodeIDNode(ID, Opc, getVTList(VT), 0, 0);
+ AddNodeIDNode(ID, Opc, getVTList(EltVT), 0, 0);
ID.AddDouble(Val);
void *IP = 0;
- if (SDNode *E = CSEMap.FindNodeOrInsertPos(ID, IP))
- return SDOperand(E, 0);
- SDNode *N = new ConstantFPSDNode(isTarget, Val, VT);
- CSEMap.InsertNode(N, IP);
- AllNodes.push_back(N);
- return SDOperand(N, 0);
+ SDNode *N = NULL;
+ if ((N = CSEMap.FindNodeOrInsertPos(ID, IP)))
+ if (!MVT::isVector(VT))
+ return SDOperand(N, 0);
+ if (!N) {
+ N = new ConstantFPSDNode(isTarget, Val, EltVT);
+ CSEMap.InsertNode(N, IP);
+ AllNodes.push_back(N);
+ }
+
+ SDOperand Result(N, 0);
+ if (MVT::isVector(VT)) {
+ SmallVector<SDOperand, 8> Ops;
+ Ops.assign(MVT::getVectorNumElements(VT), Result);
+ Result = getNode(ISD::BUILD_VECTOR, VT, &Ops[0], Ops.size());
+ }
+ return Result;
}
SDOperand SelectionDAG::getGlobalAddress(const GlobalValue *GV,
MVT::ValueType VT, int Offset,
bool isTargetGA) {
- unsigned Opc = isTargetGA ? ISD::TargetGlobalAddress : ISD::GlobalAddress;
+ const GlobalVariable *GVar = dyn_cast<GlobalVariable>(GV);
+ unsigned Opc;
+ if (GVar && GVar->isThreadLocal())
+ Opc = isTargetGA ? ISD::TargetGlobalTLSAddress : ISD::GlobalTLSAddress;
+ else
+ Opc = isTargetGA ? ISD::TargetGlobalAddress : ISD::GlobalAddress;
FoldingSetNodeID ID;
AddNodeIDNode(ID, Opc, getVTList(VT), 0, 0);
ID.AddPointer(GV);
return SDOperand();
}
+/// MaskedValueIsZero - Return true if 'V & Mask' is known to be zero. We use
+/// this predicate to simplify operations downstream. Mask is known to be zero
+/// for bits that V cannot have.
+bool SelectionDAG::MaskedValueIsZero(SDOperand Op, uint64_t Mask,
+ unsigned Depth) const {
+ // The masks are not wide enough to represent this type! Should use APInt.
+ if (Op.getValueType() == MVT::i128)
+ return false;
+
+ uint64_t KnownZero, KnownOne;
+ ComputeMaskedBits(Op, Mask, KnownZero, KnownOne, Depth);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ return (KnownZero & Mask) == Mask;
+}
+
+/// ComputeMaskedBits - Determine which of the bits specified in Mask are
+/// known to be either zero or one and return them in the KnownZero/KnownOne
+/// bitsets. This code only analyzes bits in Mask, in order to short-circuit
+/// processing.
+void SelectionDAG::ComputeMaskedBits(SDOperand Op, uint64_t Mask,
+ uint64_t &KnownZero, uint64_t &KnownOne,
+ unsigned Depth) const {
+ KnownZero = KnownOne = 0; // Don't know anything.
+ if (Depth == 6 || Mask == 0)
+ return; // Limit search depth.
+
+ // The masks are not wide enough to represent this type! Should use APInt.
+ if (Op.getValueType() == MVT::i128)
+ return;
+
+ uint64_t KnownZero2, KnownOne2;
+
+ switch (Op.getOpcode()) {
+ case ISD::Constant:
+ // We know all of the bits for a constant!
+ KnownOne = cast<ConstantSDNode>(Op)->getValue() & Mask;
+ KnownZero = ~KnownOne & Mask;
+ return;
+ case ISD::AND:
+ // If either the LHS or the RHS are Zero, the result is zero.
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero, KnownOne, Depth+1);
+ Mask &= ~KnownZero;
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Output known-1 bits are only known if set in both the LHS & RHS.
+ KnownOne &= KnownOne2;
+ // Output known-0 are known to be clear if zero in either the LHS | RHS.
+ KnownZero |= KnownZero2;
+ return;
+ case ISD::OR:
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero, KnownOne, Depth+1);
+ Mask &= ~KnownOne;
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Output known-0 bits are only known if clear in both the LHS & RHS.
+ KnownZero &= KnownZero2;
+ // Output known-1 are known to be set if set in either the LHS | RHS.
+ KnownOne |= KnownOne2;
+ return;
+ case ISD::XOR: {
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero, KnownOne, Depth+1);
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Output known-0 bits are known if clear or set in both the LHS & RHS.
+ uint64_t KnownZeroOut = (KnownZero & KnownZero2) | (KnownOne & KnownOne2);
+ // Output known-1 are known to be set if set in only one of the LHS, RHS.
+ KnownOne = (KnownZero & KnownOne2) | (KnownOne & KnownZero2);
+ KnownZero = KnownZeroOut;
+ return;
+ }
+ case ISD::SELECT:
+ ComputeMaskedBits(Op.getOperand(2), Mask, KnownZero, KnownOne, Depth+1);
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Only known if known in both the LHS and RHS.
+ KnownOne &= KnownOne2;
+ KnownZero &= KnownZero2;
+ return;
+ case ISD::SELECT_CC:
+ ComputeMaskedBits(Op.getOperand(3), Mask, KnownZero, KnownOne, Depth+1);
+ ComputeMaskedBits(Op.getOperand(2), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Only known if known in both the LHS and RHS.
+ KnownOne &= KnownOne2;
+ KnownZero &= KnownZero2;
+ return;
+ case ISD::SETCC:
+ // If we know the result of a setcc has the top bits zero, use this info.
+ if (TLI.getSetCCResultContents() == TargetLowering::ZeroOrOneSetCCResult)
+ KnownZero |= (MVT::getIntVTBitMask(Op.getValueType()) ^ 1ULL);
+ return;
+ case ISD::SHL:
+ // (shl X, C1) & C2 == 0 iff (X & C2 >>u C1) == 0
+ if (ConstantSDNode *SA = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ ComputeMaskedBits(Op.getOperand(0), Mask >> SA->getValue(),
+ KnownZero, KnownOne, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ KnownZero <<= SA->getValue();
+ KnownOne <<= SA->getValue();
+ KnownZero |= (1ULL << SA->getValue())-1; // low bits known zero.
+ }
+ return;
+ case ISD::SRL:
+ // (ushr X, C1) & C2 == 0 iff (-1 >> C1) & C2 == 0
+ if (ConstantSDNode *SA = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ MVT::ValueType VT = Op.getValueType();
+ unsigned ShAmt = SA->getValue();
+
+ uint64_t TypeMask = MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op.getOperand(0), (Mask << ShAmt) & TypeMask,
+ KnownZero, KnownOne, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ KnownZero &= TypeMask;
+ KnownOne &= TypeMask;
+ KnownZero >>= ShAmt;
+ KnownOne >>= ShAmt;
+
+ uint64_t HighBits = (1ULL << ShAmt)-1;
+ HighBits <<= MVT::getSizeInBits(VT)-ShAmt;
+ KnownZero |= HighBits; // High bits known zero.
+ }
+ return;
+ case ISD::SRA:
+ if (ConstantSDNode *SA = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ MVT::ValueType VT = Op.getValueType();
+ unsigned ShAmt = SA->getValue();
+
+ // Compute the new bits that are at the top now.
+ uint64_t TypeMask = MVT::getIntVTBitMask(VT);
+
+ uint64_t InDemandedMask = (Mask << ShAmt) & TypeMask;
+ // If any of the demanded bits are produced by the sign extension, we also
+ // demand the input sign bit.
+ uint64_t HighBits = (1ULL << ShAmt)-1;
+ HighBits <<= MVT::getSizeInBits(VT) - ShAmt;
+ if (HighBits & Mask)
+ InDemandedMask |= MVT::getIntVTSignBit(VT);
+
+ ComputeMaskedBits(Op.getOperand(0), InDemandedMask, KnownZero, KnownOne,
+ Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ KnownZero &= TypeMask;
+ KnownOne &= TypeMask;
+ KnownZero >>= ShAmt;
+ KnownOne >>= ShAmt;
+
+ // Handle the sign bits.
+ uint64_t SignBit = MVT::getIntVTSignBit(VT);
+ SignBit >>= ShAmt; // Adjust to where it is now in the mask.
+
+ if (KnownZero & SignBit) {
+ KnownZero |= HighBits; // New bits are known zero.
+ } else if (KnownOne & SignBit) {
+ KnownOne |= HighBits; // New bits are known one.
+ }
+ }
+ return;
+ case ISD::SIGN_EXTEND_INREG: {
+ MVT::ValueType EVT = cast<VTSDNode>(Op.getOperand(1))->getVT();
+
+ // Sign extension. Compute the demanded bits in the result that are not
+ // present in the input.
+ uint64_t NewBits = ~MVT::getIntVTBitMask(EVT) & Mask;
+
+ uint64_t InSignBit = MVT::getIntVTSignBit(EVT);
+ int64_t InputDemandedBits = Mask & MVT::getIntVTBitMask(EVT);
+
+ // If the sign extended bits are demanded, we know that the sign
+ // bit is demanded.
+ if (NewBits)
+ InputDemandedBits |= InSignBit;
+
+ ComputeMaskedBits(Op.getOperand(0), InputDemandedBits,
+ KnownZero, KnownOne, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+
+ // If the sign bit of the input is known set or clear, then we know the
+ // top bits of the result.
+ if (KnownZero & InSignBit) { // Input sign bit known clear
+ KnownZero |= NewBits;
+ KnownOne &= ~NewBits;
+ } else if (KnownOne & InSignBit) { // Input sign bit known set
+ KnownOne |= NewBits;
+ KnownZero &= ~NewBits;
+ } else { // Input sign bit unknown
+ KnownZero &= ~NewBits;
+ KnownOne &= ~NewBits;
+ }
+ return;
+ }
+ case ISD::CTTZ:
+ case ISD::CTLZ:
+ case ISD::CTPOP: {
+ MVT::ValueType VT = Op.getValueType();
+ unsigned LowBits = Log2_32(MVT::getSizeInBits(VT))+1;
+ KnownZero = ~((1ULL << LowBits)-1) & MVT::getIntVTBitMask(VT);
+ KnownOne = 0;
+ return;
+ }
+ case ISD::LOAD: {
+ if (ISD::isZEXTLoad(Op.Val)) {
+ LoadSDNode *LD = cast<LoadSDNode>(Op);
+ MVT::ValueType VT = LD->getLoadedVT();
+ KnownZero |= ~MVT::getIntVTBitMask(VT) & Mask;
+ }
+ return;
+ }
+ case ISD::ZERO_EXTEND: {
+ uint64_t InMask = MVT::getIntVTBitMask(Op.getOperand(0).getValueType());
+ uint64_t NewBits = (~InMask) & Mask;
+ ComputeMaskedBits(Op.getOperand(0), Mask & InMask, KnownZero,
+ KnownOne, Depth+1);
+ KnownZero |= NewBits & Mask;
+ KnownOne &= ~NewBits;
+ return;
+ }
+ case ISD::SIGN_EXTEND: {
+ MVT::ValueType InVT = Op.getOperand(0).getValueType();
+ unsigned InBits = MVT::getSizeInBits(InVT);
+ uint64_t InMask = MVT::getIntVTBitMask(InVT);
+ uint64_t InSignBit = 1ULL << (InBits-1);
+ uint64_t NewBits = (~InMask) & Mask;
+ uint64_t InDemandedBits = Mask & InMask;
+
+ // If any of the sign extended bits are demanded, we know that the sign
+ // bit is demanded.
+ if (NewBits & Mask)
+ InDemandedBits |= InSignBit;
+
+ ComputeMaskedBits(Op.getOperand(0), InDemandedBits, KnownZero,
+ KnownOne, Depth+1);
+ // If the sign bit is known zero or one, the top bits match.
+ if (KnownZero & InSignBit) {
+ KnownZero |= NewBits;
+ KnownOne &= ~NewBits;
+ } else if (KnownOne & InSignBit) {
+ KnownOne |= NewBits;
+ KnownZero &= ~NewBits;
+ } else { // Otherwise, top bits aren't known.
+ KnownOne &= ~NewBits;
+ KnownZero &= ~NewBits;
+ }
+ return;
+ }
+ case ISD::ANY_EXTEND: {
+ MVT::ValueType VT = Op.getOperand(0).getValueType();
+ ComputeMaskedBits(Op.getOperand(0), Mask & MVT::getIntVTBitMask(VT),
+ KnownZero, KnownOne, Depth+1);
+ return;
+ }
+ case ISD::TRUNCATE: {
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero, KnownOne, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ uint64_t OutMask = MVT::getIntVTBitMask(Op.getValueType());
+ KnownZero &= OutMask;
+ KnownOne &= OutMask;
+ break;
+ }
+ case ISD::AssertZext: {
+ MVT::ValueType VT = cast<VTSDNode>(Op.getOperand(1))->getVT();
+ uint64_t InMask = MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op.getOperand(0), Mask & InMask, KnownZero,
+ KnownOne, Depth+1);
+ KnownZero |= (~InMask) & Mask;
+ return;
+ }
+ case ISD::ADD: {
+ // If either the LHS or the RHS are Zero, the result is zero.
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero, KnownOne, Depth+1);
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero2, KnownOne2, Depth+1);
+ assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
+ assert((KnownZero2 & KnownOne2) == 0 && "Bits known to be one AND zero?");
+
+ // Output known-0 bits are known if clear or set in both the low clear bits
+ // common to both LHS & RHS. For example, 8+(X<<3) is known to have the
+ // low 3 bits clear.
+ uint64_t KnownZeroOut = std::min(CountTrailingZeros_64(~KnownZero),
+ CountTrailingZeros_64(~KnownZero2));
+
+ KnownZero = (1ULL << KnownZeroOut) - 1;
+ KnownOne = 0;
+ return;
+ }
+ case ISD::SUB: {
+ ConstantSDNode *CLHS = dyn_cast<ConstantSDNode>(Op.getOperand(0));
+ if (!CLHS) return;
+
+ // We know that the top bits of C-X are clear if X contains less bits
+ // than C (i.e. no wrap-around can happen). For example, 20-X is
+ // positive if we can prove that X is >= 0 and < 16.
+ MVT::ValueType VT = CLHS->getValueType(0);
+ if ((CLHS->getValue() & MVT::getIntVTSignBit(VT)) == 0) { // sign bit clear
+ unsigned NLZ = CountLeadingZeros_64(CLHS->getValue()+1);
+ uint64_t MaskV = (1ULL << (63-NLZ))-1; // NLZ can't be 64 with no sign bit
+ MaskV = ~MaskV & MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op.getOperand(1), MaskV, KnownZero, KnownOne, Depth+1);
+
+ // If all of the MaskV bits are known to be zero, then we know the output
+ // top bits are zero, because we now know that the output is from [0-C].
+ if ((KnownZero & MaskV) == MaskV) {
+ unsigned NLZ2 = CountLeadingZeros_64(CLHS->getValue());
+ KnownZero = ~((1ULL << (64-NLZ2))-1) & Mask; // Top bits known zero.
+ KnownOne = 0; // No one bits known.
+ } else {
+ KnownZero = KnownOne = 0; // Otherwise, nothing known.
+ }
+ }
+ return;
+ }
+ default:
+ // Allow the target to implement this method for its nodes.
+ if (Op.getOpcode() >= ISD::BUILTIN_OP_END) {
+ case ISD::INTRINSIC_WO_CHAIN:
+ case ISD::INTRINSIC_W_CHAIN:
+ case ISD::INTRINSIC_VOID:
+ TLI.computeMaskedBitsForTargetNode(Op, Mask, KnownZero, KnownOne, *this);
+ }
+ return;
+ }
+}
+
+/// ComputeNumSignBits - Return the number of times the sign bit of the
+/// register is replicated into the other bits. We know that at least 1 bit
+/// is always equal to the sign bit (itself), but other cases can give us
+/// information. For example, immediately after an "SRA X, 2", we know that
+/// the top 3 bits are all equal to each other, so we return 3.
+unsigned SelectionDAG::ComputeNumSignBits(SDOperand Op, unsigned Depth) const{
+ MVT::ValueType VT = Op.getValueType();
+ assert(MVT::isInteger(VT) && "Invalid VT!");
+ unsigned VTBits = MVT::getSizeInBits(VT);
+ unsigned Tmp, Tmp2;
+
+ if (Depth == 6)
+ return 1; // Limit search depth.
+
+ switch (Op.getOpcode()) {
+ default: break;
+ case ISD::AssertSext:
+ Tmp = MVT::getSizeInBits(cast<VTSDNode>(Op.getOperand(1))->getVT());
+ return VTBits-Tmp+1;
+ case ISD::AssertZext:
+ Tmp = MVT::getSizeInBits(cast<VTSDNode>(Op.getOperand(1))->getVT());
+ return VTBits-Tmp;
+
+ case ISD::Constant: {
+ uint64_t Val = cast<ConstantSDNode>(Op)->getValue();
+ // If negative, invert the bits, then look at it.
+ if (Val & MVT::getIntVTSignBit(VT))
+ Val = ~Val;
+
+ // Shift the bits so they are the leading bits in the int64_t.
+ Val <<= 64-VTBits;
+
+ // Return # leading zeros. We use 'min' here in case Val was zero before
+ // shifting. We don't want to return '64' as for an i32 "0".
+ return std::min(VTBits, CountLeadingZeros_64(Val));
+ }
+
+ case ISD::SIGN_EXTEND:
+ Tmp = VTBits-MVT::getSizeInBits(Op.getOperand(0).getValueType());
+ return ComputeNumSignBits(Op.getOperand(0), Depth+1) + Tmp;
+
+ case ISD::SIGN_EXTEND_INREG:
+ // Max of the input and what this extends.
+ Tmp = MVT::getSizeInBits(cast<VTSDNode>(Op.getOperand(1))->getVT());
+ Tmp = VTBits-Tmp+1;
+
+ Tmp2 = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ return std::max(Tmp, Tmp2);
+
+ case ISD::SRA:
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ // SRA X, C -> adds C sign bits.
+ if (ConstantSDNode *C = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ Tmp += C->getValue();
+ if (Tmp > VTBits) Tmp = VTBits;
+ }
+ return Tmp;
+ case ISD::SHL:
+ if (ConstantSDNode *C = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ // shl destroys sign bits.
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (C->getValue() >= VTBits || // Bad shift.
+ C->getValue() >= Tmp) break; // Shifted all sign bits out.
+ return Tmp - C->getValue();
+ }
+ break;
+ case ISD::AND:
+ case ISD::OR:
+ case ISD::XOR: // NOT is handled here.
+ // Logical binary ops preserve the number of sign bits.
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (Tmp == 1) return 1; // Early out.
+ Tmp2 = ComputeNumSignBits(Op.getOperand(1), Depth+1);
+ return std::min(Tmp, Tmp2);
+
+ case ISD::SELECT:
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (Tmp == 1) return 1; // Early out.
+ Tmp2 = ComputeNumSignBits(Op.getOperand(1), Depth+1);
+ return std::min(Tmp, Tmp2);
+
+ case ISD::SETCC:
+ // If setcc returns 0/-1, all bits are sign bits.
+ if (TLI.getSetCCResultContents() ==
+ TargetLowering::ZeroOrNegativeOneSetCCResult)
+ return VTBits;
+ break;
+ case ISD::ROTL:
+ case ISD::ROTR:
+ if (ConstantSDNode *C = dyn_cast<ConstantSDNode>(Op.getOperand(1))) {
+ unsigned RotAmt = C->getValue() & (VTBits-1);
+
+ // Handle rotate right by N like a rotate left by 32-N.
+ if (Op.getOpcode() == ISD::ROTR)
+ RotAmt = (VTBits-RotAmt) & (VTBits-1);
+
+ // If we aren't rotating out all of the known-in sign bits, return the
+ // number that are left. This handles rotl(sext(x), 1) for example.
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (Tmp > RotAmt+1) return Tmp-RotAmt;
+ }
+ break;
+ case ISD::ADD:
+ // Add can have at most one carry bit. Thus we know that the output
+ // is, at worst, one more bit than the inputs.
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (Tmp == 1) return 1; // Early out.
+
+ // Special case decrementing a value (ADD X, -1):
+ if (ConstantSDNode *CRHS = dyn_cast<ConstantSDNode>(Op.getOperand(0)))
+ if (CRHS->isAllOnesValue()) {
+ uint64_t KnownZero, KnownOne;
+ uint64_t Mask = MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op.getOperand(0), Mask, KnownZero, KnownOne, Depth+1);
+
+ // If the input is known to be 0 or 1, the output is 0/-1, which is all
+ // sign bits set.
+ if ((KnownZero|1) == Mask)
+ return VTBits;
+
+ // If we are subtracting one from a positive number, there is no carry
+ // out of the result.
+ if (KnownZero & MVT::getIntVTSignBit(VT))
+ return Tmp;
+ }
+
+ Tmp2 = ComputeNumSignBits(Op.getOperand(1), Depth+1);
+ if (Tmp2 == 1) return 1;
+ return std::min(Tmp, Tmp2)-1;
+ break;
+
+ case ISD::SUB:
+ Tmp2 = ComputeNumSignBits(Op.getOperand(1), Depth+1);
+ if (Tmp2 == 1) return 1;
+
+ // Handle NEG.
+ if (ConstantSDNode *CLHS = dyn_cast<ConstantSDNode>(Op.getOperand(0)))
+ if (CLHS->getValue() == 0) {
+ uint64_t KnownZero, KnownOne;
+ uint64_t Mask = MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op.getOperand(1), Mask, KnownZero, KnownOne, Depth+1);
+ // If the input is known to be 0 or 1, the output is 0/-1, which is all
+ // sign bits set.
+ if ((KnownZero|1) == Mask)
+ return VTBits;
+
+ // If the input is known to be positive (the sign bit is known clear),
+ // the output of the NEG has the same number of sign bits as the input.
+ if (KnownZero & MVT::getIntVTSignBit(VT))
+ return Tmp2;
+
+ // Otherwise, we treat this like a SUB.
+ }
+
+ // Sub can have at most one carry bit. Thus we know that the output
+ // is, at worst, one more bit than the inputs.
+ Tmp = ComputeNumSignBits(Op.getOperand(0), Depth+1);
+ if (Tmp == 1) return 1; // Early out.
+ return std::min(Tmp, Tmp2)-1;
+ break;
+ case ISD::TRUNCATE:
+ // FIXME: it's tricky to do anything useful for this, but it is an important
+ // case for targets like X86.
+ break;
+ }
+
+ // Handle LOADX separately here. EXTLOAD case will fallthrough.
+ if (Op.getOpcode() == ISD::LOAD) {
+ LoadSDNode *LD = cast<LoadSDNode>(Op);
+ unsigned ExtType = LD->getExtensionType();
+ switch (ExtType) {
+ default: break;
+ case ISD::SEXTLOAD: // '17' bits known
+ Tmp = MVT::getSizeInBits(LD->getLoadedVT());
+ return VTBits-Tmp+1;
+ case ISD::ZEXTLOAD: // '16' bits known
+ Tmp = MVT::getSizeInBits(LD->getLoadedVT());
+ return VTBits-Tmp;
+ }
+ }
+
+ // Allow the target to implement this method for its nodes.
+ if (Op.getOpcode() >= ISD::BUILTIN_OP_END ||
+ Op.getOpcode() == ISD::INTRINSIC_WO_CHAIN ||
+ Op.getOpcode() == ISD::INTRINSIC_W_CHAIN ||
+ Op.getOpcode() == ISD::INTRINSIC_VOID) {
+ unsigned NumBits = TLI.ComputeNumSignBitsForTargetNode(Op, Depth);
+ if (NumBits > 1) return NumBits;
+ }
+
+ // Finally, if we can prove that the top bits of the result are 0's or 1's,
+ // use this information.
+ uint64_t KnownZero, KnownOne;
+ uint64_t Mask = MVT::getIntVTBitMask(VT);
+ ComputeMaskedBits(Op, Mask, KnownZero, KnownOne, Depth);
+
+ uint64_t SignBit = MVT::getIntVTSignBit(VT);
+ if (KnownZero & SignBit) { // SignBit is 0
+ Mask = KnownZero;
+ } else if (KnownOne & SignBit) { // SignBit is 1;
+ Mask = KnownOne;
+ } else {
+ // Nothing known.
+ return 1;
+ }
+
+ // Okay, we know that the sign bit in Mask is set. Use CLZ to determine
+ // the number of identical bits in the top of the input value.
+ Mask ^= ~0ULL;
+ Mask <<= 64-VTBits;
+ // Return # leading zeros. We use 'min' here in case Val was zero before
+ // shifting. We don't want to return '64' as for an i32 "0".
+ return std::min(VTBits, CountLeadingZeros_64(Mask));
+}
+
/// getNode - Gets or creates the specified node.
///
switch (Opcode) {
case ISD::TokenFactor:
return Operand; // Factor of one node? No factor.
+ case ISD::FP_ROUND:
+ case ISD::FP_EXTEND:
+ assert(MVT::isFloatingPoint(VT) &&
+ MVT::isFloatingPoint(Operand.getValueType()) && "Invalid FP cast!");
+ break;
case ISD::SIGN_EXTEND:
+ assert(MVT::isInteger(VT) && MVT::isInteger(Operand.getValueType()) &&
+ "Invalid SIGN_EXTEND!");
if (Operand.getValueType() == VT) return Operand; // noop extension
assert(Operand.getValueType() < VT && "Invalid sext node, dst < src!");
if (OpOpcode == ISD::SIGN_EXTEND || OpOpcode == ISD::ZERO_EXTEND)
return getNode(OpOpcode, VT, Operand.Val->getOperand(0));
break;
case ISD::ZERO_EXTEND:
+ assert(MVT::isInteger(VT) && MVT::isInteger(Operand.getValueType()) &&
+ "Invalid ZERO_EXTEND!");
if (Operand.getValueType() == VT) return Operand; // noop extension
assert(Operand.getValueType() < VT && "Invalid zext node, dst < src!");
if (OpOpcode == ISD::ZERO_EXTEND) // (zext (zext x)) -> (zext x)
return getNode(ISD::ZERO_EXTEND, VT, Operand.Val->getOperand(0));
break;
case ISD::ANY_EXTEND:
+ assert(MVT::isInteger(VT) && MVT::isInteger(Operand.getValueType()) &&
+ "Invalid ANY_EXTEND!");
if (Operand.getValueType() == VT) return Operand; // noop extension
assert(Operand.getValueType() < VT && "Invalid anyext node, dst < src!");
if (OpOpcode == ISD::ZERO_EXTEND || OpOpcode == ISD::SIGN_EXTEND)
return getNode(OpOpcode, VT, Operand.Val->getOperand(0));
break;
case ISD::TRUNCATE:
+ assert(MVT::isInteger(VT) && MVT::isInteger(Operand.getValueType()) &&
+ "Invalid TRUNCATE!");
if (Operand.getValueType() == VT) return Operand; // noop truncate
assert(Operand.getValueType() > VT && "Invalid truncate node, src < dst!");
if (OpOpcode == ISD::TRUNCATE)
break;
case ISD::SCALAR_TO_VECTOR:
assert(MVT::isVector(VT) && !MVT::isVector(Operand.getValueType()) &&
- MVT::getVectorBaseType(VT) == Operand.getValueType() &&
+ MVT::getVectorElementType(VT) == Operand.getValueType() &&
"Illegal SCALAR_TO_VECTOR node!");
break;
case ISD::FNEG:
double F;
uint64_t I;
} u1;
- union {
- double F;
- int64_t I;
- } u2;
u1.F = C1;
- u2.F = C2;
- if (u2.I < 0) // Sign bit of RHS set?
+ if (int64_t(DoubleToBits(C2)) < 0) // Sign bit of RHS set?
u1.I |= 1ULL << 63; // Set the sign bit of the LHS.
else
u1.I &= (1ULL << 63)-1; // Clear the sign bit of the LHS.
case ISD::SREM:
case ISD::SRL:
case ISD::SHL:
- return getConstant(0, VT); // fold op(undef, arg2) -> 0
+ if (!MVT::isVector(VT))
+ return getConstant(0, VT); // fold op(undef, arg2) -> 0
+ // For vectors, we can't easily build an all zero vector, just return
+ // the LHS.
+ return N2;
}
}
}
case ISD::AND:
case ISD::SRL:
case ISD::SHL:
- return getConstant(0, VT); // fold op(arg1, undef) -> 0
+ if (!MVT::isVector(VT))
+ return getConstant(0, VT); // fold op(arg1, undef) -> 0
+ // For vectors, we can't easily build an all zero vector, just return
+ // the LHS.
+ return N1;
case ISD::OR:
- return getConstant(MVT::getIntVTBitMask(VT), VT);
+ if (!MVT::isVector(VT))
+ return getConstant(MVT::getIntVTBitMask(VT), VT);
+ // For vectors, we can't easily build an all one vector, just return
+ // the LHS.
+ return N1;
case ISD::SRA:
return N1;
}
if (EVT == VT) return N1; // Not actually extending
break;
}
+ case ISD::EXTRACT_VECTOR_ELT:
+ assert(N2C && "Bad EXTRACT_VECTOR_ELT!");
+
+ // EXTRACT_VECTOR_ELT of BUILD_PAIR is often formed while lowering is
+ // expanding copies of large vectors from registers.
+ if (N1.getOpcode() == ISD::BUILD_PAIR) {
+ unsigned NewNumElts = MVT::getVectorNumElements(N1.getValueType()) / 2;
+ bool Low = N2C->getValue() < NewNumElts;
+ return getNode(ISD::EXTRACT_VECTOR_ELT, VT, N1.getOperand(!Low),
+ Low ? N2 : getConstant(N2C->getValue() - NewNumElts,
+ N2.getValueType()));
+ }
+ // EXTRACT_VECTOR_ELT of BUILD_VECTOR is often formed while lowering is
+ // expanding large vector constants.
+ if (N1.getOpcode() == ISD::BUILD_VECTOR)
+ return N1.getOperand(N2C->getValue());
+ break;
case ISD::EXTRACT_ELEMENT:
assert(N2C && (unsigned)N2C->getValue() < 2 && "Bad EXTRACT_ELEMENT!");
MVT::getVectorNumElements(VT) == N3.getNumOperands() &&
"Illegal VECTOR_SHUFFLE node!");
break;
+ case ISD::BIT_CONVERT:
+ // Fold bit_convert nodes from a type to themselves.
+ if (N1.getValueType() == VT)
+ return N1;
+ break;
}
// Memoize node if it doesn't produce a flag.
SDOperand SelectionDAG::getLoad(MVT::ValueType VT,
SDOperand Chain, SDOperand Ptr,
const Value *SV, int SVOffset,
- bool isVolatile) {
- // FIXME: Alignment == 1 for now.
- unsigned Alignment = 1;
+ bool isVolatile, unsigned Alignment) {
+ if (Alignment == 0) { // Ensure that codegen never sees alignment 0
+ const Type *Ty = 0;
+ if (VT != MVT::iPTR) {
+ Ty = MVT::getTypeForValueType(VT);
+ } else if (SV) {
+ const PointerType *PT = dyn_cast<PointerType>(SV->getType());
+ assert(PT && "Value for load must be a pointer");
+ Ty = PT->getElementType();
+ }
+ assert(Ty && "Could not get type information for load");
+ Alignment = TLI.getTargetData()->getABITypeAlignment(Ty);
+ }
SDVTList VTs = getVTList(VT, MVT::Other);
SDOperand Undef = getNode(ISD::UNDEF, Ptr.getValueType());
SDOperand Ops[] = { Chain, Ptr, Undef };
SDOperand Chain, SDOperand Ptr,
const Value *SV,
int SVOffset, MVT::ValueType EVT,
- bool isVolatile) {
+ bool isVolatile, unsigned Alignment) {
// If they are asking for an extending load from/to the same thing, return a
// normal load.
if (VT == EVT)
ExtType = ISD::NON_EXTLOAD;
if (MVT::isVector(VT))
- assert(EVT == MVT::getVectorBaseType(VT) && "Invalid vector extload!");
+ assert(EVT == MVT::getVectorElementType(VT) && "Invalid vector extload!");
else
assert(EVT < VT && "Should only be an extending load, not truncating!");
assert((ExtType == ISD::EXTLOAD || MVT::isInteger(VT)) &&
assert(MVT::isInteger(VT) == MVT::isInteger(EVT) &&
"Cannot convert from FP to Int or Int -> FP!");
- // FIXME: Alignment == 1 for now.
- unsigned Alignment = 1;
+ if (Alignment == 0) { // Ensure that codegen never sees alignment 0
+ const Type *Ty = 0;
+ if (VT != MVT::iPTR) {
+ Ty = MVT::getTypeForValueType(VT);
+ } else if (SV) {
+ const PointerType *PT = dyn_cast<PointerType>(SV->getType());
+ assert(PT && "Value for load must be a pointer");
+ Ty = PT->getElementType();
+ }
+ assert(Ty && "Could not get type information for load");
+ Alignment = TLI.getTargetData()->getABITypeAlignment(Ty);
+ }
SDVTList VTs = getVTList(VT, MVT::Other);
SDOperand Undef = getNode(ISD::UNDEF, Ptr.getValueType());
SDOperand Ops[] = { Chain, Ptr, Undef };
return SDOperand(N, 0);
}
-SDOperand SelectionDAG::getVecLoad(unsigned Count, MVT::ValueType EVT,
- SDOperand Chain, SDOperand Ptr,
- SDOperand SV) {
- SDOperand Ops[] = { Chain, Ptr, SV, getConstant(Count, MVT::i32),
- getValueType(EVT) };
- return getNode(ISD::VLOAD, getVTList(MVT::Vector, MVT::Other), Ops, 5);
-}
-
SDOperand SelectionDAG::getStore(SDOperand Chain, SDOperand Val,
SDOperand Ptr, const Value *SV, int SVOffset,
- bool isVolatile) {
+ bool isVolatile, unsigned Alignment) {
MVT::ValueType VT = Val.getValueType();
- // FIXME: Alignment == 1 for now.
- unsigned Alignment = 1;
+ if (Alignment == 0) { // Ensure that codegen never sees alignment 0
+ const Type *Ty = 0;
+ if (VT != MVT::iPTR) {
+ Ty = MVT::getTypeForValueType(VT);
+ } else if (SV) {
+ const PointerType *PT = dyn_cast<PointerType>(SV->getType());
+ assert(PT && "Value for store must be a pointer");
+ Ty = PT->getElementType();
+ }
+ assert(Ty && "Could not get type information for store");
+ Alignment = TLI.getTargetData()->getABITypeAlignment(Ty);
+ }
SDVTList VTs = getVTList(MVT::Other);
SDOperand Undef = getNode(ISD::UNDEF, Ptr.getValueType());
SDOperand Ops[] = { Chain, Val, Ptr, Undef };
SDOperand SelectionDAG::getTruncStore(SDOperand Chain, SDOperand Val,
SDOperand Ptr, const Value *SV,
int SVOffset, MVT::ValueType SVT,
- bool isVolatile) {
+ bool isVolatile, unsigned Alignment) {
MVT::ValueType VT = Val.getValueType();
bool isTrunc = VT != SVT;
assert(MVT::isInteger(VT) == MVT::isInteger(SVT) &&
"Can't do FP-INT conversion!");
- // FIXME: Alignment == 1 for now.
- unsigned Alignment = 1;
+ if (Alignment == 0) { // Ensure that codegen never sees alignment 0
+ const Type *Ty = 0;
+ if (VT != MVT::iPTR) {
+ Ty = MVT::getTypeForValueType(VT);
+ } else if (SV) {
+ const PointerType *PT = dyn_cast<PointerType>(SV->getType());
+ assert(PT && "Value for store must be a pointer");
+ Ty = PT->getElementType();
+ }
+ assert(Ty && "Could not get type information for store");
+ Alignment = TLI.getTargetData()->getABITypeAlignment(Ty);
+ }
SDVTList VTs = getVTList(MVT::Other);
SDOperand Undef = getNode(ISD::UNDEF, Ptr.getValueType());
SDOperand Ops[] = { Chain, Val, Ptr, Undef };
}
SDVTList SelectionDAG::getVTList(MVT::ValueType VT) {
- return makeVTList(SDNode::getValueTypeList(VT), 1);
+ if (!MVT::isExtendedVT(VT))
+ return makeVTList(SDNode::getValueTypeList(VT), 1);
+
+ for (std::list<std::vector<MVT::ValueType> >::iterator I = VTList.begin(),
+ E = VTList.end(); I != E; ++I) {
+ if (I->size() == 1 && (*I)[0] == VT)
+ return makeVTList(&(*I)[0], 1);
+ }
+ std::vector<MVT::ValueType> V;
+ V.push_back(VT);
+ VTList.push_front(V);
+ return makeVTList(&(*VTList.begin())[0], 1);
}
SDVTList SelectionDAG::getVTList(MVT::ValueType VT1, MVT::ValueType VT2) {
SDVTList SelectionDAG::getVTList(const MVT::ValueType *VTs, unsigned NumVTs) {
switch (NumVTs) {
case 0: assert(0 && "Cannot have nodes without results!");
- case 1: return makeVTList(SDNode::getValueTypeList(VTs[0]), 1);
+ case 1: return getVTList(VTs[0]);
case 2: return getVTList(VTs[0], VTs[1]);
case 3: return getVTList(VTs[0], VTs[1], VTs[2]);
default: break;
SDOperand Ops[] = { Op1, Op2 };
return getNode(ISD::BUILTIN_OP_END+Opcode, VTs, 3, Ops, 2).Val;
}
+SDNode *SelectionDAG::getTargetNode(unsigned Opcode, MVT::ValueType VT1,
+ MVT::ValueType VT2, MVT::ValueType VT3,
+ SDOperand Op1, SDOperand Op2,
+ SDOperand Op3) {
+ const MVT::ValueType *VTs = getNodeValueTypes(VT1, VT2, VT3);
+ SDOperand Ops[] = { Op1, Op2, Op3 };
+ return getNode(ISD::BUILTIN_OP_END+Opcode, VTs, 3, Ops, 3).Val;
+}
SDNode *SelectionDAG::getTargetNode(unsigned Opcode, MVT::ValueType VT1,
MVT::ValueType VT2, MVT::ValueType VT3,
const SDOperand *Ops, unsigned NumOps) {
MorphNodeTo(ISD::HANDLENODE, VTs, 0, 0); // Drops operand uses.
}
+GlobalAddressSDNode::GlobalAddressSDNode(bool isTarget, const GlobalValue *GA,
+ MVT::ValueType VT, int o)
+ : SDNode(isa<GlobalVariable>(GA) &&
+ dyn_cast<GlobalVariable>(GA)->isThreadLocal() ?
+ // Thread Local
+ (isTarget ? ISD::TargetGlobalTLSAddress : ISD::GlobalTLSAddress) :
+ // Non Thread Local
+ (isTarget ? ISD::TargetGlobalAddress : ISD::GlobalAddress),
+ getSDVTList(VT)), Offset(o) {
+ TheGlobal = const_cast<GlobalValue*>(GA);
+}
/// Profile - Gather unique data for the node.
///
case ISD::Constant: return "Constant";
case ISD::ConstantFP: return "ConstantFP";
case ISD::GlobalAddress: return "GlobalAddress";
+ case ISD::GlobalTLSAddress: return "GlobalTLSAddress";
case ISD::FrameIndex: return "FrameIndex";
case ISD::JumpTable: return "JumpTable";
case ISD::GLOBAL_OFFSET_TABLE: return "GLOBAL_OFFSET_TABLE";
case ISD::TargetConstant: return "TargetConstant";
case ISD::TargetConstantFP:return "TargetConstantFP";
case ISD::TargetGlobalAddress: return "TargetGlobalAddress";
+ case ISD::TargetGlobalTLSAddress: return "TargetGlobalTLSAddress";
case ISD::TargetFrameIndex: return "TargetFrameIndex";
case ISD::TargetJumpTable: return "TargetJumpTable";
case ISD::TargetConstantPool: return "TargetConstantPool";
case ISD::CopyToReg: return "CopyToReg";
case ISD::CopyFromReg: return "CopyFromReg";
case ISD::UNDEF: return "undef";
- case ISD::MERGE_VALUES: return "mergevalues";
+ case ISD::MERGE_VALUES: return "merge_values";
case ISD::INLINEASM: return "inlineasm";
case ISD::LABEL: return "label";
case ISD::HANDLENODE: return "handlenode";
case ISD::FDIV: return "fdiv";
case ISD::FREM: return "frem";
case ISD::FCOPYSIGN: return "fcopysign";
- case ISD::VADD: return "vadd";
- case ISD::VSUB: return "vsub";
- case ISD::VMUL: return "vmul";
- case ISD::VSDIV: return "vsdiv";
- case ISD::VUDIV: return "vudiv";
- case ISD::VAND: return "vand";
- case ISD::VOR: return "vor";
- case ISD::VXOR: return "vxor";
case ISD::SETCC: return "setcc";
case ISD::SELECT: return "select";
case ISD::SELECT_CC: return "select_cc";
- case ISD::VSELECT: return "vselect";
case ISD::INSERT_VECTOR_ELT: return "insert_vector_elt";
- case ISD::VINSERT_VECTOR_ELT: return "vinsert_vector_elt";
case ISD::EXTRACT_VECTOR_ELT: return "extract_vector_elt";
- case ISD::VEXTRACT_VECTOR_ELT: return "vextract_vector_elt";
+ case ISD::CONCAT_VECTORS: return "concat_vectors";
+ case ISD::EXTRACT_SUBVECTOR: return "extract_subvector";
case ISD::SCALAR_TO_VECTOR: return "scalar_to_vector";
- case ISD::VBUILD_VECTOR: return "vbuild_vector";
case ISD::VECTOR_SHUFFLE: return "vector_shuffle";
- case ISD::VVECTOR_SHUFFLE: return "vvector_shuffle";
- case ISD::VBIT_CONVERT: return "vbit_convert";
case ISD::CARRY_FALSE: return "carry_false";
case ISD::ADDC: return "addc";
case ISD::ADDE: return "adde";
// Other operators
case ISD::LOAD: return "load";
case ISD::STORE: return "store";
- case ISD::VLOAD: return "vload";
case ISD::VAARG: return "vaarg";
case ISD::VACOPY: return "vacopy";
case ISD::VAEND: return "vaend";
else
cerr << "<null:" << M->getOffset() << ">";
} else if (const VTSDNode *N = dyn_cast<VTSDNode>(this)) {
- cerr << ":" << getValueTypeString(N->getVT());
+ cerr << ":" << MVT::getValueTypeString(N->getVT());
} else if (const LoadSDNode *LD = dyn_cast<LoadSDNode>(this)) {
bool doExt = true;
switch (LD->getExtensionType()) {