//===----------------------------------------------------------------------===//
#define DEBUG_TYPE "mips-lower"
-//#include <algorithm>
#include "MipsISelLowering.h"
#include "MipsMachineFunction.h"
#include "MipsTargetMachine.h"
#include "llvm/GlobalVariable.h"
#include "llvm/Intrinsics.h"
#include "llvm/CallingConv.h"
+#include "InstPrinter/MipsInstPrinter.h"
#include "llvm/CodeGen/CallingConvLower.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/Support/ErrorHandling.h"
using namespace llvm;
+// If I is a shifted mask, set the size (Size) and the first bit of the
+// mask (Pos), and return true.
+// For example, if I is 0x003ff800, (Pos, Size) = (11, 11).
+static bool IsShiftedMask(uint64_t I, uint64_t &Pos, uint64_t &Size) {
+ if (!isUInt<32>(I) || !isShiftedMask_32(I))
+ return false;
+
+ Size = CountPopulation_32(I);
+ Pos = CountTrailingZeros_32(I);
+ return true;
+}
+
const char *MipsTargetLowering::getTargetNodeName(unsigned Opcode) const {
switch (Opcode) {
case MipsISD::JmpLink: return "MipsISD::JmpLink";
case MipsISD::Hi: return "MipsISD::Hi";
case MipsISD::Lo: return "MipsISD::Lo";
case MipsISD::GPRel: return "MipsISD::GPRel";
+ case MipsISD::TlsGd: return "MipsISD::TlsGd";
+ case MipsISD::TprelHi: return "MipsISD::TprelHi";
+ case MipsISD::TprelLo: return "MipsISD::TprelLo";
+ case MipsISD::ThreadPointer: return "MipsISD::ThreadPointer";
case MipsISD::Ret: return "MipsISD::Ret";
case MipsISD::FPBrcond: return "MipsISD::FPBrcond";
case MipsISD::FPCmp: return "MipsISD::FPCmp";
case MipsISD::DivRemU: return "MipsISD::DivRemU";
case MipsISD::BuildPairF64: return "MipsISD::BuildPairF64";
case MipsISD::ExtractElementF64: return "MipsISD::ExtractElementF64";
- default: return NULL;
+ case MipsISD::WrapperPIC: return "MipsISD::WrapperPIC";
+ case MipsISD::DynAlloc: return "MipsISD::DynAlloc";
+ case MipsISD::Sync: return "MipsISD::Sync";
+ case MipsISD::Ext: return "MipsISD::Ext";
+ case MipsISD::Ins: return "MipsISD::Ins";
+ default: return NULL;
}
}
MipsTargetLowering(MipsTargetMachine &TM)
: TargetLowering(TM, new MipsTargetObjectFile()) {
Subtarget = &TM.getSubtarget<MipsSubtarget>();
+ bool HasMips64 = Subtarget->hasMips64();
// Mips does not have i1 type, so use i32 for
// setcc operations results (slt, sgt, ...).
setBooleanContents(ZeroOrOneBooleanContent);
+ setBooleanVectorContents(ZeroOrOneBooleanContent); // FIXME: Is this correct?
// Set up the register classes
addRegisterClass(MVT::i32, Mips::CPURegsRegisterClass);
addRegisterClass(MVT::f32, Mips::FGR32RegisterClass);
+ if (HasMips64)
+ addRegisterClass(MVT::i64, Mips::CPU64RegsRegisterClass);
+
// When dealing with single precision only, use libcalls
- if (!Subtarget->isSingleFloat())
- if (!Subtarget->isFP64bit())
+ if (!Subtarget->isSingleFloat()) {
+ if (HasMips64)
+ addRegisterClass(MVT::f64, Mips::FGR64RegisterClass);
+ else
addRegisterClass(MVT::f64, Mips::AFGR64RegisterClass);
+ }
// Load extented operations for i1 types must be promoted
setLoadExtAction(ISD::EXTLOAD, MVT::i1, Promote);
setOperationAction(ISD::SELECT, MVT::i32, Custom);
setOperationAction(ISD::BRCOND, MVT::Other, Custom);
setOperationAction(ISD::DYNAMIC_STACKALLOC, MVT::i32, Custom);
- setOperationAction(ISD::FP_TO_SINT, MVT::i32, Custom);
setOperationAction(ISD::VASTART, MVT::Other, Custom);
setOperationAction(ISD::SDIV, MVT::i32, Expand);
setOperationAction(ISD::CTTZ, MVT::i32, Expand);
setOperationAction(ISD::ROTL, MVT::i32, Expand);
- if (!Subtarget->isMips32r2())
+ if (!Subtarget->hasMips32r2())
setOperationAction(ISD::ROTR, MVT::i32, Expand);
setOperationAction(ISD::SHL_PARTS, MVT::i32, Expand);
setOperationAction(ISD::SRA_PARTS, MVT::i32, Expand);
setOperationAction(ISD::SRL_PARTS, MVT::i32, Expand);
- setOperationAction(ISD::FCOPYSIGN, MVT::f32, Expand);
- setOperationAction(ISD::FCOPYSIGN, MVT::f64, Expand);
+ setOperationAction(ISD::FCOPYSIGN, MVT::f32, Custom);
+ setOperationAction(ISD::FCOPYSIGN, MVT::f64, Custom);
setOperationAction(ISD::FSIN, MVT::f32, Expand);
setOperationAction(ISD::FSIN, MVT::f64, Expand);
setOperationAction(ISD::FCOS, MVT::f32, Expand);
setOperationAction(ISD::FLOG2, MVT::f32, Expand);
setOperationAction(ISD::FLOG10, MVT::f32, Expand);
setOperationAction(ISD::FEXP, MVT::f32, Expand);
+ setOperationAction(ISD::FMA, MVT::f32, Expand);
+ setOperationAction(ISD::FMA, MVT::f64, Expand);
- setOperationAction(ISD::EH_LABEL, MVT::Other, Expand);
+ setOperationAction(ISD::EXCEPTIONADDR, MVT::i32, Expand);
+ setOperationAction(ISD::EHSELECTION, MVT::i32, Expand);
setOperationAction(ISD::VAARG, MVT::Other, Expand);
setOperationAction(ISD::VACOPY, MVT::Other, Expand);
// Use the default for now
setOperationAction(ISD::STACKSAVE, MVT::Other, Expand);
setOperationAction(ISD::STACKRESTORE, MVT::Other, Expand);
- setOperationAction(ISD::MEMBARRIER, MVT::Other, Expand);
+
+ setOperationAction(ISD::MEMBARRIER, MVT::Other, Custom);
+ setOperationAction(ISD::ATOMIC_FENCE, MVT::Other, Custom);
+
+ setOperationAction(ISD::ATOMIC_LOAD, MVT::i32, Expand);
+ setOperationAction(ISD::ATOMIC_STORE, MVT::i32, Expand);
+
+ setInsertFencesForAtomic(true);
if (Subtarget->isSingleFloat())
setOperationAction(ISD::SELECT_CC, MVT::f64, Expand);
setTargetDAGCombine(ISD::SDIVREM);
setTargetDAGCombine(ISD::UDIVREM);
setTargetDAGCombine(ISD::SETCC);
+ setTargetDAGCombine(ISD::AND);
+ setTargetDAGCombine(ISD::OR);
setMinFunctionAlignment(2);
setStackPointerRegisterToSaveRestore(Mips::SP);
computeRegisterProperties();
+
+ setExceptionPointerRegister(Mips::A0);
+ setExceptionSelectorRegister(Mips::A1);
+}
+
+bool MipsTargetLowering::allowsUnalignedMemoryAccesses(EVT VT) const {
+ MVT::SimpleValueType SVT = VT.getSimpleVT().SimpleTy;
+ return SVT == MVT::i32 || SVT == MVT::i16;
}
-MVT::SimpleValueType MipsTargetLowering::getSetCCResultType(EVT VT) const {
+EVT MipsTargetLowering::getSetCCResultType(EVT VT) const {
return MVT::i32;
}
if (DCI.isBeforeLegalize())
return SDValue();
- if (Subtarget->isMips32() && SelectMadd(N, &DAG))
+ if (Subtarget->hasMips32() && SelectMadd(N, &DAG))
return SDValue(N, 0);
return SDValue();
if (DCI.isBeforeLegalize())
return SDValue();
- if (Subtarget->isMips32() && SelectMsub(N, &DAG))
+ if (Subtarget->hasMips32() && SelectMsub(N, &DAG))
return SDValue(N, 0);
return SDValue();
return CreateCMovFP(DAG, Cond, True, False, N->getDebugLoc());
}
+static SDValue PerformANDCombine(SDNode *N, SelectionDAG& DAG,
+ TargetLowering::DAGCombinerInfo &DCI,
+ const MipsSubtarget* Subtarget) {
+ // Pattern match EXT.
+ // $dst = and ((sra or srl) $src , pos), (2**size - 1)
+ // => ext $dst, $src, size, pos
+ if (DCI.isBeforeLegalizeOps() || !Subtarget->hasMips32r2())
+ return SDValue();
+
+ SDValue ShiftRight = N->getOperand(0), Mask = N->getOperand(1);
+
+ // Op's first operand must be a shift right.
+ if (ShiftRight.getOpcode() != ISD::SRA && ShiftRight.getOpcode() != ISD::SRL)
+ return SDValue();
+
+ // The second operand of the shift must be an immediate.
+ uint64_t Pos;
+ ConstantSDNode *CN;
+ if (!(CN = dyn_cast<ConstantSDNode>(ShiftRight.getOperand(1))))
+ return SDValue();
+
+ Pos = CN->getZExtValue();
+
+ uint64_t SMPos, SMSize;
+ // Op's second operand must be a shifted mask.
+ if (!(CN = dyn_cast<ConstantSDNode>(Mask)) ||
+ !IsShiftedMask(CN->getZExtValue(), SMPos, SMSize))
+ return SDValue();
+
+ // Return if the shifted mask does not start at bit 0 or the sum of its size
+ // and Pos exceeds the word's size.
+ if (SMPos != 0 || Pos + SMSize > 32)
+ return SDValue();
+
+ return DAG.getNode(MipsISD::Ext, N->getDebugLoc(), MVT::i32,
+ ShiftRight.getOperand(0),
+ DAG.getConstant(Pos, MVT::i32),
+ DAG.getConstant(SMSize, MVT::i32));
+}
+
+static SDValue PerformORCombine(SDNode *N, SelectionDAG& DAG,
+ TargetLowering::DAGCombinerInfo &DCI,
+ const MipsSubtarget* Subtarget) {
+ // Pattern match INS.
+ // $dst = or (and $src1 , mask0), (and (shl $src, pos), mask1),
+ // where mask1 = (2**size - 1) << pos, mask0 = ~mask1
+ // => ins $dst, $src, size, pos, $src1
+ if (DCI.isBeforeLegalizeOps() || !Subtarget->hasMips32r2())
+ return SDValue();
+
+ SDValue And0 = N->getOperand(0), And1 = N->getOperand(1);
+ uint64_t SMPos0, SMSize0, SMPos1, SMSize1;
+ ConstantSDNode *CN;
+
+ // See if Op's first operand matches (and $src1 , mask0).
+ if (And0.getOpcode() != ISD::AND)
+ return SDValue();
+
+ if (!(CN = dyn_cast<ConstantSDNode>(And0.getOperand(1))) ||
+ !IsShiftedMask(~CN->getSExtValue(), SMPos0, SMSize0))
+ return SDValue();
+
+ // See if Op's second operand matches (and (shl $src, pos), mask1).
+ if (And1.getOpcode() != ISD::AND)
+ return SDValue();
+
+ if (!(CN = dyn_cast<ConstantSDNode>(And1.getOperand(1))) ||
+ !IsShiftedMask(CN->getZExtValue(), SMPos1, SMSize1))
+ return SDValue();
+
+ // The shift masks must have the same position and size.
+ if (SMPos0 != SMPos1 || SMSize0 != SMSize1)
+ return SDValue();
+
+ SDValue Shl = And1.getOperand(0);
+ if (Shl.getOpcode() != ISD::SHL)
+ return SDValue();
+
+ if (!(CN = dyn_cast<ConstantSDNode>(Shl.getOperand(1))))
+ return SDValue();
+
+ unsigned Shamt = CN->getZExtValue();
+
+ // Return if the shift amount and the first bit position of mask are not the
+ // same.
+ if (Shamt != SMPos0)
+ return SDValue();
+
+ return DAG.getNode(MipsISD::Ins, N->getDebugLoc(), MVT::i32,
+ Shl.getOperand(0),
+ DAG.getConstant(SMPos0, MVT::i32),
+ DAG.getConstant(SMSize0, MVT::i32),
+ And0.getOperand(0));
+}
+
SDValue MipsTargetLowering::PerformDAGCombine(SDNode *N, DAGCombinerInfo &DCI)
const {
SelectionDAG &DAG = DCI.DAG;
return PerformDivRemCombine(N, DAG, DCI, Subtarget);
case ISD::SETCC:
return PerformSETCCCombine(N, DAG, DCI, Subtarget);
+ case ISD::AND:
+ return PerformANDCombine(N, DAG, DCI, Subtarget);
+ case ISD::OR:
+ return PerformORCombine(N, DAG, DCI, Subtarget);
}
return SDValue();
case ISD::BRCOND: return LowerBRCOND(Op, DAG);
case ISD::ConstantPool: return LowerConstantPool(Op, DAG);
case ISD::DYNAMIC_STACKALLOC: return LowerDYNAMIC_STACKALLOC(Op, DAG);
- case ISD::FP_TO_SINT: return LowerFP_TO_SINT(Op, DAG);
case ISD::GlobalAddress: return LowerGlobalAddress(Op, DAG);
case ISD::BlockAddress: return LowerBlockAddress(Op, DAG);
case ISD::GlobalTLSAddress: return LowerGlobalTLSAddress(Op, DAG);
case ISD::JumpTable: return LowerJumpTable(Op, DAG);
case ISD::SELECT: return LowerSELECT(Op, DAG);
case ISD::VASTART: return LowerVASTART(Op, DAG);
+ case ISD::FCOPYSIGN: return LowerFCOPYSIGN(Op, DAG);
+ case ISD::FRAMEADDR: return LowerFRAMEADDR(Op, DAG);
+ case ISD::MEMBARRIER: return LowerMEMBARRIER(Op, DAG);
+ case ISD::ATOMIC_FENCE: return LowerATOMIC_FENCE(Op, DAG);
}
return SDValue();
}
return Mips::BRANCH_INVALID;
}
-MachineBasicBlock *
-MipsTargetLowering::EmitInstrWithCustomInserter(MachineInstr *MI,
- MachineBasicBlock *BB) const {
+static MachineBasicBlock* ExpandCondMov(MachineInstr *MI, MachineBasicBlock *BB,
+ DebugLoc dl,
+ const MipsSubtarget* Subtarget,
+ const TargetInstrInfo *TII,
+ bool isFPCmp, unsigned Opc) {
// There is no need to expand CMov instructions if target has
// conditional moves.
if (Subtarget->hasCondMov())
return BB;
- const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
- bool isFPCmp = false;
- DebugLoc dl = MI->getDebugLoc();
- unsigned Opc;
-
- switch (MI->getOpcode()) {
- default: assert(false && "Unexpected instr type to insert");
- case Mips::MOVT:
- case Mips::MOVT_S:
- case Mips::MOVT_D:
- isFPCmp = true;
- Opc = Mips::BC1F;
- break;
- case Mips::MOVF:
- case Mips::MOVF_S:
- case Mips::MOVF_D:
- isFPCmp = true;
- Opc = Mips::BC1T;
- break;
- case Mips::MOVZ_I:
- case Mips::MOVZ_S:
- case Mips::MOVZ_D:
- Opc = Mips::BNE;
- break;
- case Mips::MOVN_I:
- case Mips::MOVN_S:
- case Mips::MOVN_D:
- Opc = Mips::BEQ;
- break;
- }
-
// To "insert" a SELECT_CC instruction, we actually have to insert the
// diamond control-flow pattern. The incoming instruction knows the
// destination vreg to set, the condition code register to branch on, the
BuildMI(BB, dl, TII->get(Opc)).addReg(MI->getOperand(2).getReg())
.addReg(Mips::ZERO).addMBB(sinkMBB);
-
// copy0MBB:
// %FalseValue = ...
// # fallthrough to sinkMBB
return BB;
}
-//===----------------------------------------------------------------------===//
-// Misc Lower Operation implementation
-//===----------------------------------------------------------------------===//
+MachineBasicBlock *
+MipsTargetLowering::EmitInstrWithCustomInserter(MachineInstr *MI,
+ MachineBasicBlock *BB) const {
+ const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
+ DebugLoc dl = MI->getDebugLoc();
-SDValue MipsTargetLowering::
-LowerFP_TO_SINT(SDValue Op, SelectionDAG &DAG) const
-{
- if (!Subtarget->isMips1())
- return Op;
+ switch (MI->getOpcode()) {
+ default:
+ assert(false && "Unexpected instr type to insert");
+ return NULL;
+ case Mips::MOVT:
+ case Mips::MOVT_S:
+ case Mips::MOVT_D:
+ return ExpandCondMov(MI, BB, dl, Subtarget, TII, true, Mips::BC1F);
+ case Mips::MOVF:
+ case Mips::MOVF_S:
+ case Mips::MOVF_D:
+ return ExpandCondMov(MI, BB, dl, Subtarget, TII, true, Mips::BC1T);
+ case Mips::MOVZ_I:
+ case Mips::MOVZ_S:
+ case Mips::MOVZ_D:
+ return ExpandCondMov(MI, BB, dl, Subtarget, TII, false, Mips::BNE);
+ case Mips::MOVN_I:
+ case Mips::MOVN_S:
+ case Mips::MOVN_D:
+ return ExpandCondMov(MI, BB, dl, Subtarget, TII, false, Mips::BEQ);
+
+ case Mips::ATOMIC_LOAD_ADD_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, Mips::ADDu);
+ case Mips::ATOMIC_LOAD_ADD_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, Mips::ADDu);
+ case Mips::ATOMIC_LOAD_ADD_I32:
+ return EmitAtomicBinary(MI, BB, 4, Mips::ADDu);
+
+ case Mips::ATOMIC_LOAD_AND_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, Mips::AND);
+ case Mips::ATOMIC_LOAD_AND_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, Mips::AND);
+ case Mips::ATOMIC_LOAD_AND_I32:
+ return EmitAtomicBinary(MI, BB, 4, Mips::AND);
+
+ case Mips::ATOMIC_LOAD_OR_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, Mips::OR);
+ case Mips::ATOMIC_LOAD_OR_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, Mips::OR);
+ case Mips::ATOMIC_LOAD_OR_I32:
+ return EmitAtomicBinary(MI, BB, 4, Mips::OR);
+
+ case Mips::ATOMIC_LOAD_XOR_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, Mips::XOR);
+ case Mips::ATOMIC_LOAD_XOR_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, Mips::XOR);
+ case Mips::ATOMIC_LOAD_XOR_I32:
+ return EmitAtomicBinary(MI, BB, 4, Mips::XOR);
+
+ case Mips::ATOMIC_LOAD_NAND_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, 0, true);
+ case Mips::ATOMIC_LOAD_NAND_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, 0, true);
+ case Mips::ATOMIC_LOAD_NAND_I32:
+ return EmitAtomicBinary(MI, BB, 4, 0, true);
+
+ case Mips::ATOMIC_LOAD_SUB_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, Mips::SUBu);
+ case Mips::ATOMIC_LOAD_SUB_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, Mips::SUBu);
+ case Mips::ATOMIC_LOAD_SUB_I32:
+ return EmitAtomicBinary(MI, BB, 4, Mips::SUBu);
+
+ case Mips::ATOMIC_SWAP_I8:
+ return EmitAtomicBinaryPartword(MI, BB, 1, 0);
+ case Mips::ATOMIC_SWAP_I16:
+ return EmitAtomicBinaryPartword(MI, BB, 2, 0);
+ case Mips::ATOMIC_SWAP_I32:
+ return EmitAtomicBinary(MI, BB, 4, 0);
+
+ case Mips::ATOMIC_CMP_SWAP_I8:
+ return EmitAtomicCmpSwapPartword(MI, BB, 1);
+ case Mips::ATOMIC_CMP_SWAP_I16:
+ return EmitAtomicCmpSwapPartword(MI, BB, 2);
+ case Mips::ATOMIC_CMP_SWAP_I32:
+ return EmitAtomicCmpSwap(MI, BB, 4);
+ }
+}
- MachineFunction &MF = DAG.getMachineFunction();
- unsigned CCReg = AddLiveIn(MF, Mips::FCR31, Mips::CCRRegisterClass);
+// This function also handles Mips::ATOMIC_SWAP_I32 (when BinOpcode == 0), and
+// Mips::ATOMIC_LOAD_NAND_I32 (when Nand == true)
+MachineBasicBlock *
+MipsTargetLowering::EmitAtomicBinary(MachineInstr *MI, MachineBasicBlock *BB,
+ unsigned Size, unsigned BinOpcode,
+ bool Nand) const {
+ assert(Size == 4 && "Unsupported size for EmitAtomicBinary.");
+
+ MachineFunction *MF = BB->getParent();
+ MachineRegisterInfo &RegInfo = MF->getRegInfo();
+ const TargetRegisterClass *RC = getRegClassFor(MVT::i32);
+ const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
+ DebugLoc dl = MI->getDebugLoc();
- SDValue Chain = DAG.getEntryNode();
- DebugLoc dl = Op.getDebugLoc();
- SDValue Src = Op.getOperand(0);
-
- // Set the condition register
- SDValue CondReg = DAG.getCopyFromReg(Chain, dl, CCReg, MVT::i32);
- CondReg = DAG.getCopyToReg(Chain, dl, Mips::AT, CondReg);
- CondReg = DAG.getCopyFromReg(CondReg, dl, Mips::AT, MVT::i32);
-
- SDValue Cst = DAG.getConstant(3, MVT::i32);
- SDValue Or = DAG.getNode(ISD::OR, dl, MVT::i32, CondReg, Cst);
- Cst = DAG.getConstant(2, MVT::i32);
- SDValue Xor = DAG.getNode(ISD::XOR, dl, MVT::i32, Or, Cst);
-
- SDValue InFlag(0, 0);
- CondReg = DAG.getCopyToReg(Chain, dl, Mips::FCR31, Xor, InFlag);
-
- // Emit the round instruction and bit convert to integer
- SDValue Trunc = DAG.getNode(MipsISD::FPRound, dl, MVT::f32,
- Src, CondReg.getValue(1));
- SDValue BitCvt = DAG.getNode(ISD::BITCAST, dl, MVT::i32, Trunc);
- return BitCvt;
+ unsigned OldVal = MI->getOperand(0).getReg();
+ unsigned Ptr = MI->getOperand(1).getReg();
+ unsigned Incr = MI->getOperand(2).getReg();
+
+ unsigned StoreVal = RegInfo.createVirtualRegister(RC);
+ unsigned AndRes = RegInfo.createVirtualRegister(RC);
+ unsigned Success = RegInfo.createVirtualRegister(RC);
+
+ // insert new blocks after the current block
+ const BasicBlock *LLVM_BB = BB->getBasicBlock();
+ MachineBasicBlock *loopMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *exitMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineFunction::iterator It = BB;
+ ++It;
+ MF->insert(It, loopMBB);
+ MF->insert(It, exitMBB);
+
+ // Transfer the remainder of BB and its successor edges to exitMBB.
+ exitMBB->splice(exitMBB->begin(), BB,
+ llvm::next(MachineBasicBlock::iterator(MI)),
+ BB->end());
+ exitMBB->transferSuccessorsAndUpdatePHIs(BB);
+
+ // thisMBB:
+ // ...
+ // fallthrough --> loopMBB
+ BB->addSuccessor(loopMBB);
+ loopMBB->addSuccessor(loopMBB);
+ loopMBB->addSuccessor(exitMBB);
+
+ // loopMBB:
+ // ll oldval, 0(ptr)
+ // <binop> storeval, oldval, incr
+ // sc success, storeval, 0(ptr)
+ // beq success, $0, loopMBB
+ BB = loopMBB;
+ BuildMI(BB, dl, TII->get(Mips::LL), OldVal).addReg(Ptr).addImm(0);
+ if (Nand) {
+ // and andres, oldval, incr
+ // nor storeval, $0, andres
+ BuildMI(BB, dl, TII->get(Mips::AND), AndRes).addReg(OldVal).addReg(Incr);
+ BuildMI(BB, dl, TII->get(Mips::NOR), StoreVal)
+ .addReg(Mips::ZERO).addReg(AndRes);
+ } else if (BinOpcode) {
+ // <binop> storeval, oldval, incr
+ BuildMI(BB, dl, TII->get(BinOpcode), StoreVal).addReg(OldVal).addReg(Incr);
+ } else {
+ StoreVal = Incr;
+ }
+ BuildMI(BB, dl, TII->get(Mips::SC), Success)
+ .addReg(StoreVal).addReg(Ptr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::BEQ))
+ .addReg(Success).addReg(Mips::ZERO).addMBB(loopMBB);
+
+ MI->eraseFromParent(); // The instruction is gone now.
+
+ return exitMBB;
+}
+
+MachineBasicBlock *
+MipsTargetLowering::EmitAtomicBinaryPartword(MachineInstr *MI,
+ MachineBasicBlock *BB,
+ unsigned Size, unsigned BinOpcode,
+ bool Nand) const {
+ assert((Size == 1 || Size == 2) &&
+ "Unsupported size for EmitAtomicBinaryPartial.");
+
+ MachineFunction *MF = BB->getParent();
+ MachineRegisterInfo &RegInfo = MF->getRegInfo();
+ const TargetRegisterClass *RC = getRegClassFor(MVT::i32);
+ const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
+ DebugLoc dl = MI->getDebugLoc();
+
+ unsigned Dest = MI->getOperand(0).getReg();
+ unsigned Ptr = MI->getOperand(1).getReg();
+ unsigned Incr = MI->getOperand(2).getReg();
+
+ unsigned AlignedAddr = RegInfo.createVirtualRegister(RC);
+ unsigned ShiftAmt = RegInfo.createVirtualRegister(RC);
+ unsigned Mask = RegInfo.createVirtualRegister(RC);
+ unsigned Mask2 = RegInfo.createVirtualRegister(RC);
+ unsigned NewVal = RegInfo.createVirtualRegister(RC);
+ unsigned OldVal = RegInfo.createVirtualRegister(RC);
+ unsigned Incr2 = RegInfo.createVirtualRegister(RC);
+ unsigned MaskLSB2 = RegInfo.createVirtualRegister(RC);
+ unsigned PtrLSB2 = RegInfo.createVirtualRegister(RC);
+ unsigned MaskUpper = RegInfo.createVirtualRegister(RC);
+ unsigned AndRes = RegInfo.createVirtualRegister(RC);
+ unsigned BinOpRes = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedOldVal0 = RegInfo.createVirtualRegister(RC);
+ unsigned StoreVal = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedOldVal1 = RegInfo.createVirtualRegister(RC);
+ unsigned SrlRes = RegInfo.createVirtualRegister(RC);
+ unsigned SllRes = RegInfo.createVirtualRegister(RC);
+ unsigned Success = RegInfo.createVirtualRegister(RC);
+
+ // insert new blocks after the current block
+ const BasicBlock *LLVM_BB = BB->getBasicBlock();
+ MachineBasicBlock *loopMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *sinkMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *exitMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineFunction::iterator It = BB;
+ ++It;
+ MF->insert(It, loopMBB);
+ MF->insert(It, sinkMBB);
+ MF->insert(It, exitMBB);
+
+ // Transfer the remainder of BB and its successor edges to exitMBB.
+ exitMBB->splice(exitMBB->begin(), BB,
+ llvm::next(MachineBasicBlock::iterator(MI)),
+ BB->end());
+ exitMBB->transferSuccessorsAndUpdatePHIs(BB);
+
+ BB->addSuccessor(loopMBB);
+ loopMBB->addSuccessor(loopMBB);
+ loopMBB->addSuccessor(sinkMBB);
+ sinkMBB->addSuccessor(exitMBB);
+
+ // thisMBB:
+ // addiu masklsb2,$0,-4 # 0xfffffffc
+ // and alignedaddr,ptr,masklsb2
+ // andi ptrlsb2,ptr,3
+ // sll shiftamt,ptrlsb2,3
+ // ori maskupper,$0,255 # 0xff
+ // sll mask,maskupper,shiftamt
+ // nor mask2,$0,mask
+ // sll incr2,incr,shiftamt
+
+ int64_t MaskImm = (Size == 1) ? 255 : 65535;
+ BuildMI(BB, dl, TII->get(Mips::ADDiu), MaskLSB2)
+ .addReg(Mips::ZERO).addImm(-4);
+ BuildMI(BB, dl, TII->get(Mips::AND), AlignedAddr)
+ .addReg(Ptr).addReg(MaskLSB2);
+ BuildMI(BB, dl, TII->get(Mips::ANDi), PtrLSB2).addReg(Ptr).addImm(3);
+ BuildMI(BB, dl, TII->get(Mips::SLL), ShiftAmt).addReg(PtrLSB2).addImm(3);
+ BuildMI(BB, dl, TII->get(Mips::ORi), MaskUpper)
+ .addReg(Mips::ZERO).addImm(MaskImm);
+ BuildMI(BB, dl, TII->get(Mips::SLLV), Mask)
+ .addReg(ShiftAmt).addReg(MaskUpper);
+ BuildMI(BB, dl, TII->get(Mips::NOR), Mask2).addReg(Mips::ZERO).addReg(Mask);
+ BuildMI(BB, dl, TII->get(Mips::SLLV), Incr2).addReg(ShiftAmt).addReg(Incr);
+
+
+ // atomic.load.binop
+ // loopMBB:
+ // ll oldval,0(alignedaddr)
+ // binop binopres,oldval,incr2
+ // and newval,binopres,mask
+ // and maskedoldval0,oldval,mask2
+ // or storeval,maskedoldval0,newval
+ // sc success,storeval,0(alignedaddr)
+ // beq success,$0,loopMBB
+
+ // atomic.swap
+ // loopMBB:
+ // ll oldval,0(alignedaddr)
+ // and newval,incr2,mask
+ // and maskedoldval0,oldval,mask2
+ // or storeval,maskedoldval0,newval
+ // sc success,storeval,0(alignedaddr)
+ // beq success,$0,loopMBB
+
+ BB = loopMBB;
+ BuildMI(BB, dl, TII->get(Mips::LL), OldVal).addReg(AlignedAddr).addImm(0);
+ if (Nand) {
+ // and andres, oldval, incr2
+ // nor binopres, $0, andres
+ // and newval, binopres, mask
+ BuildMI(BB, dl, TII->get(Mips::AND), AndRes).addReg(OldVal).addReg(Incr2);
+ BuildMI(BB, dl, TII->get(Mips::NOR), BinOpRes)
+ .addReg(Mips::ZERO).addReg(AndRes);
+ BuildMI(BB, dl, TII->get(Mips::AND), NewVal).addReg(BinOpRes).addReg(Mask);
+ } else if (BinOpcode) {
+ // <binop> binopres, oldval, incr2
+ // and newval, binopres, mask
+ BuildMI(BB, dl, TII->get(BinOpcode), BinOpRes).addReg(OldVal).addReg(Incr2);
+ BuildMI(BB, dl, TII->get(Mips::AND), NewVal).addReg(BinOpRes).addReg(Mask);
+ } else {// atomic.swap
+ // and newval, incr2, mask
+ BuildMI(BB, dl, TII->get(Mips::AND), NewVal).addReg(Incr2).addReg(Mask);
+ }
+
+ BuildMI(BB, dl, TII->get(Mips::AND), MaskedOldVal0)
+ .addReg(OldVal).addReg(Mask2);
+ BuildMI(BB, dl, TII->get(Mips::OR), StoreVal)
+ .addReg(MaskedOldVal0).addReg(NewVal);
+ BuildMI(BB, dl, TII->get(Mips::SC), Success)
+ .addReg(StoreVal).addReg(AlignedAddr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::BEQ))
+ .addReg(Success).addReg(Mips::ZERO).addMBB(loopMBB);
+
+ // sinkMBB:
+ // and maskedoldval1,oldval,mask
+ // srl srlres,maskedoldval1,shiftamt
+ // sll sllres,srlres,24
+ // sra dest,sllres,24
+ BB = sinkMBB;
+ int64_t ShiftImm = (Size == 1) ? 24 : 16;
+
+ BuildMI(BB, dl, TII->get(Mips::AND), MaskedOldVal1)
+ .addReg(OldVal).addReg(Mask);
+ BuildMI(BB, dl, TII->get(Mips::SRLV), SrlRes)
+ .addReg(ShiftAmt).addReg(MaskedOldVal1);
+ BuildMI(BB, dl, TII->get(Mips::SLL), SllRes)
+ .addReg(SrlRes).addImm(ShiftImm);
+ BuildMI(BB, dl, TII->get(Mips::SRA), Dest)
+ .addReg(SllRes).addImm(ShiftImm);
+
+ MI->eraseFromParent(); // The instruction is gone now.
+
+ return exitMBB;
+}
+
+MachineBasicBlock *
+MipsTargetLowering::EmitAtomicCmpSwap(MachineInstr *MI,
+ MachineBasicBlock *BB,
+ unsigned Size) const {
+ assert(Size == 4 && "Unsupported size for EmitAtomicCmpSwap.");
+
+ MachineFunction *MF = BB->getParent();
+ MachineRegisterInfo &RegInfo = MF->getRegInfo();
+ const TargetRegisterClass *RC = getRegClassFor(MVT::i32);
+ const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
+ DebugLoc dl = MI->getDebugLoc();
+
+ unsigned Dest = MI->getOperand(0).getReg();
+ unsigned Ptr = MI->getOperand(1).getReg();
+ unsigned OldVal = MI->getOperand(2).getReg();
+ unsigned NewVal = MI->getOperand(3).getReg();
+
+ unsigned Success = RegInfo.createVirtualRegister(RC);
+
+ // insert new blocks after the current block
+ const BasicBlock *LLVM_BB = BB->getBasicBlock();
+ MachineBasicBlock *loop1MBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *loop2MBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *exitMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineFunction::iterator It = BB;
+ ++It;
+ MF->insert(It, loop1MBB);
+ MF->insert(It, loop2MBB);
+ MF->insert(It, exitMBB);
+
+ // Transfer the remainder of BB and its successor edges to exitMBB.
+ exitMBB->splice(exitMBB->begin(), BB,
+ llvm::next(MachineBasicBlock::iterator(MI)),
+ BB->end());
+ exitMBB->transferSuccessorsAndUpdatePHIs(BB);
+
+ // thisMBB:
+ // ...
+ // fallthrough --> loop1MBB
+ BB->addSuccessor(loop1MBB);
+ loop1MBB->addSuccessor(exitMBB);
+ loop1MBB->addSuccessor(loop2MBB);
+ loop2MBB->addSuccessor(loop1MBB);
+ loop2MBB->addSuccessor(exitMBB);
+
+ // loop1MBB:
+ // ll dest, 0(ptr)
+ // bne dest, oldval, exitMBB
+ BB = loop1MBB;
+ BuildMI(BB, dl, TII->get(Mips::LL), Dest).addReg(Ptr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::BNE))
+ .addReg(Dest).addReg(OldVal).addMBB(exitMBB);
+
+ // loop2MBB:
+ // sc success, newval, 0(ptr)
+ // beq success, $0, loop1MBB
+ BB = loop2MBB;
+ BuildMI(BB, dl, TII->get(Mips::SC), Success)
+ .addReg(NewVal).addReg(Ptr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::BEQ))
+ .addReg(Success).addReg(Mips::ZERO).addMBB(loop1MBB);
+
+ MI->eraseFromParent(); // The instruction is gone now.
+
+ return exitMBB;
+}
+
+MachineBasicBlock *
+MipsTargetLowering::EmitAtomicCmpSwapPartword(MachineInstr *MI,
+ MachineBasicBlock *BB,
+ unsigned Size) const {
+ assert((Size == 1 || Size == 2) &&
+ "Unsupported size for EmitAtomicCmpSwapPartial.");
+
+ MachineFunction *MF = BB->getParent();
+ MachineRegisterInfo &RegInfo = MF->getRegInfo();
+ const TargetRegisterClass *RC = getRegClassFor(MVT::i32);
+ const TargetInstrInfo *TII = getTargetMachine().getInstrInfo();
+ DebugLoc dl = MI->getDebugLoc();
+
+ unsigned Dest = MI->getOperand(0).getReg();
+ unsigned Ptr = MI->getOperand(1).getReg();
+ unsigned CmpVal = MI->getOperand(2).getReg();
+ unsigned NewVal = MI->getOperand(3).getReg();
+
+ unsigned AlignedAddr = RegInfo.createVirtualRegister(RC);
+ unsigned ShiftAmt = RegInfo.createVirtualRegister(RC);
+ unsigned Mask = RegInfo.createVirtualRegister(RC);
+ unsigned Mask2 = RegInfo.createVirtualRegister(RC);
+ unsigned ShiftedCmpVal = RegInfo.createVirtualRegister(RC);
+ unsigned OldVal = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedOldVal0 = RegInfo.createVirtualRegister(RC);
+ unsigned ShiftedNewVal = RegInfo.createVirtualRegister(RC);
+ unsigned MaskLSB2 = RegInfo.createVirtualRegister(RC);
+ unsigned PtrLSB2 = RegInfo.createVirtualRegister(RC);
+ unsigned MaskUpper = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedCmpVal = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedNewVal = RegInfo.createVirtualRegister(RC);
+ unsigned MaskedOldVal1 = RegInfo.createVirtualRegister(RC);
+ unsigned StoreVal = RegInfo.createVirtualRegister(RC);
+ unsigned SrlRes = RegInfo.createVirtualRegister(RC);
+ unsigned SllRes = RegInfo.createVirtualRegister(RC);
+ unsigned Success = RegInfo.createVirtualRegister(RC);
+
+ // insert new blocks after the current block
+ const BasicBlock *LLVM_BB = BB->getBasicBlock();
+ MachineBasicBlock *loop1MBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *loop2MBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *sinkMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineBasicBlock *exitMBB = MF->CreateMachineBasicBlock(LLVM_BB);
+ MachineFunction::iterator It = BB;
+ ++It;
+ MF->insert(It, loop1MBB);
+ MF->insert(It, loop2MBB);
+ MF->insert(It, sinkMBB);
+ MF->insert(It, exitMBB);
+
+ // Transfer the remainder of BB and its successor edges to exitMBB.
+ exitMBB->splice(exitMBB->begin(), BB,
+ llvm::next(MachineBasicBlock::iterator(MI)),
+ BB->end());
+ exitMBB->transferSuccessorsAndUpdatePHIs(BB);
+
+ BB->addSuccessor(loop1MBB);
+ loop1MBB->addSuccessor(sinkMBB);
+ loop1MBB->addSuccessor(loop2MBB);
+ loop2MBB->addSuccessor(loop1MBB);
+ loop2MBB->addSuccessor(sinkMBB);
+ sinkMBB->addSuccessor(exitMBB);
+
+ // FIXME: computation of newval2 can be moved to loop2MBB.
+ // thisMBB:
+ // addiu masklsb2,$0,-4 # 0xfffffffc
+ // and alignedaddr,ptr,masklsb2
+ // andi ptrlsb2,ptr,3
+ // sll shiftamt,ptrlsb2,3
+ // ori maskupper,$0,255 # 0xff
+ // sll mask,maskupper,shiftamt
+ // nor mask2,$0,mask
+ // andi maskedcmpval,cmpval,255
+ // sll shiftedcmpval,maskedcmpval,shiftamt
+ // andi maskednewval,newval,255
+ // sll shiftednewval,maskednewval,shiftamt
+ int64_t MaskImm = (Size == 1) ? 255 : 65535;
+ BuildMI(BB, dl, TII->get(Mips::ADDiu), MaskLSB2)
+ .addReg(Mips::ZERO).addImm(-4);
+ BuildMI(BB, dl, TII->get(Mips::AND), AlignedAddr)
+ .addReg(Ptr).addReg(MaskLSB2);
+ BuildMI(BB, dl, TII->get(Mips::ANDi), PtrLSB2).addReg(Ptr).addImm(3);
+ BuildMI(BB, dl, TII->get(Mips::SLL), ShiftAmt).addReg(PtrLSB2).addImm(3);
+ BuildMI(BB, dl, TII->get(Mips::ORi), MaskUpper)
+ .addReg(Mips::ZERO).addImm(MaskImm);
+ BuildMI(BB, dl, TII->get(Mips::SLLV), Mask)
+ .addReg(ShiftAmt).addReg(MaskUpper);
+ BuildMI(BB, dl, TII->get(Mips::NOR), Mask2).addReg(Mips::ZERO).addReg(Mask);
+ BuildMI(BB, dl, TII->get(Mips::ANDi), MaskedCmpVal)
+ .addReg(CmpVal).addImm(MaskImm);
+ BuildMI(BB, dl, TII->get(Mips::SLLV), ShiftedCmpVal)
+ .addReg(ShiftAmt).addReg(MaskedCmpVal);
+ BuildMI(BB, dl, TII->get(Mips::ANDi), MaskedNewVal)
+ .addReg(NewVal).addImm(MaskImm);
+ BuildMI(BB, dl, TII->get(Mips::SLLV), ShiftedNewVal)
+ .addReg(ShiftAmt).addReg(MaskedNewVal);
+
+ // loop1MBB:
+ // ll oldval,0(alginedaddr)
+ // and maskedoldval0,oldval,mask
+ // bne maskedoldval0,shiftedcmpval,sinkMBB
+ BB = loop1MBB;
+ BuildMI(BB, dl, TII->get(Mips::LL), OldVal).addReg(AlignedAddr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::AND), MaskedOldVal0)
+ .addReg(OldVal).addReg(Mask);
+ BuildMI(BB, dl, TII->get(Mips::BNE))
+ .addReg(MaskedOldVal0).addReg(ShiftedCmpVal).addMBB(sinkMBB);
+
+ // loop2MBB:
+ // and maskedoldval1,oldval,mask2
+ // or storeval,maskedoldval1,shiftednewval
+ // sc success,storeval,0(alignedaddr)
+ // beq success,$0,loop1MBB
+ BB = loop2MBB;
+ BuildMI(BB, dl, TII->get(Mips::AND), MaskedOldVal1)
+ .addReg(OldVal).addReg(Mask2);
+ BuildMI(BB, dl, TII->get(Mips::OR), StoreVal)
+ .addReg(MaskedOldVal1).addReg(ShiftedNewVal);
+ BuildMI(BB, dl, TII->get(Mips::SC), Success)
+ .addReg(StoreVal).addReg(AlignedAddr).addImm(0);
+ BuildMI(BB, dl, TII->get(Mips::BEQ))
+ .addReg(Success).addReg(Mips::ZERO).addMBB(loop1MBB);
+
+ // sinkMBB:
+ // srl srlres,maskedoldval0,shiftamt
+ // sll sllres,srlres,24
+ // sra dest,sllres,24
+ BB = sinkMBB;
+ int64_t ShiftImm = (Size == 1) ? 24 : 16;
+
+ BuildMI(BB, dl, TII->get(Mips::SRLV), SrlRes)
+ .addReg(ShiftAmt).addReg(MaskedOldVal0);
+ BuildMI(BB, dl, TII->get(Mips::SLL), SllRes)
+ .addReg(SrlRes).addImm(ShiftImm);
+ BuildMI(BB, dl, TII->get(Mips::SRA), Dest)
+ .addReg(SllRes).addImm(ShiftImm);
+
+ MI->eraseFromParent(); // The instruction is gone now.
+
+ return exitMBB;
}
+//===----------------------------------------------------------------------===//
+// Misc Lower Operation implementation
+//===----------------------------------------------------------------------===//
SDValue MipsTargetLowering::
LowerDYNAMIC_STACKALLOC(SDValue Op, SelectionDAG &DAG) const
{
- unsigned StackAlignment =
- getTargetMachine().getFrameLowering()->getStackAlignment();
- assert(StackAlignment >=
+ MachineFunction &MF = DAG.getMachineFunction();
+ MipsFunctionInfo *MipsFI = MF.getInfo<MipsFunctionInfo>();
+
+ assert(getTargetMachine().getFrameLowering()->getStackAlignment() >=
cast<ConstantSDNode>(Op.getOperand(2).getNode())->getZExtValue() &&
"Cannot lower if the alignment of the allocated space is larger than \
that of the stack.");
// must be placed in the stack pointer register.
Chain = DAG.getCopyToReg(StackPointer.getValue(1), dl, Mips::SP, Sub,
SDValue());
- SDValue NewSP = DAG.getCopyFromReg(Chain, dl, Mips::SP, MVT::i32,
- Chain.getValue(1));
-
- // Align the allocated space.
- MachineFunction &MF = DAG.getMachineFunction();
- MipsFunctionInfo *MipsFI = MF.getInfo<MipsFunctionInfo>();
- unsigned SPOffset = (MipsFI->getMaxCallFrameSize() + StackAlignment - 1) /
- StackAlignment * StackAlignment;
- SDValue AllocPtr = DAG.getNode(ISD::ADD, dl, MVT::i32, NewSP,
- DAG.getConstant(SPOffset, MVT::i32));
// This node always has two return values: a new stack pointer
// value and a chain
- SDValue Ops[2] = { AllocPtr, NewSP.getValue(1) };
- return DAG.getMergeValues(Ops, 2, dl);
+ SDVTList VTLs = DAG.getVTList(MVT::i32, MVT::Other);
+ SDValue Ptr = DAG.getFrameIndex(MipsFI->getDynAllocFI(), getPointerTy());
+ SDValue Ops[] = { Chain, Ptr, Chain.getValue(1) };
+
+ return DAG.getNode(MipsISD::DynAlloc, dl, VTLs, Ops, 3);
}
SDValue MipsTargetLowering::
SDValue HiPart = DAG.getNode(MipsISD::Hi, dl, VTs, &GAHi, 1);
SDValue Lo = DAG.getNode(MipsISD::Lo, dl, MVT::i32, GALo);
return DAG.getNode(ISD::ADD, dl, MVT::i32, HiPart, Lo);
- } else {
- SDValue GA = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
- MipsII::MO_GOT);
- SDValue ResNode = DAG.getLoad(MVT::i32, dl,
- DAG.getEntryNode(), GA, MachinePointerInfo(),
- false, false, 0);
- // On functions and global targets not internal linked only
- // a load from got/GP is necessary for PIC to work.
- if (!GV->hasInternalLinkage() &&
- (!GV->hasLocalLinkage() || isa<Function>(GV)))
- return ResNode;
- SDValue GALo = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
- MipsII::MO_ABS_LO);
- SDValue Lo = DAG.getNode(MipsISD::Lo, dl, MVT::i32, GALo);
- return DAG.getNode(ISD::ADD, dl, MVT::i32, ResNode, Lo);
}
- llvm_unreachable("Dont know how to handle GlobalAddress");
- return SDValue(0,0);
+ SDValue GA = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
+ MipsII::MO_GOT);
+ GA = DAG.getNode(MipsISD::WrapperPIC, dl, MVT::i32, GA);
+ SDValue ResNode = DAG.getLoad(MVT::i32, dl,
+ DAG.getEntryNode(), GA, MachinePointerInfo(),
+ false, false, 0);
+ // On functions and global targets not internal linked only
+ // a load from got/GP is necessary for PIC to work.
+ if (!GV->hasInternalLinkage() &&
+ (!GV->hasLocalLinkage() || isa<Function>(GV)))
+ return ResNode;
+ SDValue GALo = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
+ MipsII::MO_ABS_LO);
+ SDValue Lo = DAG.getNode(MipsISD::Lo, dl, MVT::i32, GALo);
+ return DAG.getNode(ISD::ADD, dl, MVT::i32, ResNode, Lo);
}
SDValue MipsTargetLowering::LowerBlockAddress(SDValue Op,
SDValue BAGOTOffset = DAG.getBlockAddress(BA, MVT::i32, true,
MipsII::MO_GOT);
+ BAGOTOffset = DAG.getNode(MipsISD::WrapperPIC, dl, MVT::i32, BAGOTOffset);
SDValue BALOOffset = DAG.getBlockAddress(BA, MVT::i32, true,
MipsII::MO_ABS_LO);
SDValue Load = DAG.getLoad(MVT::i32, dl,
SDValue MipsTargetLowering::
LowerGlobalTLSAddress(SDValue Op, SelectionDAG &DAG) const
{
- llvm_unreachable("TLS not implemented for MIPS.");
- return SDValue(); // Not reached
+ // If the relocation model is PIC, use the General Dynamic TLS Model,
+ // otherwise use the Initial Exec or Local Exec TLS Model.
+ // TODO: implement Local Dynamic TLS model
+
+ GlobalAddressSDNode *GA = cast<GlobalAddressSDNode>(Op);
+ DebugLoc dl = GA->getDebugLoc();
+ const GlobalValue *GV = GA->getGlobal();
+ EVT PtrVT = getPointerTy();
+
+ if (getTargetMachine().getRelocationModel() == Reloc::PIC_) {
+ // General Dynamic TLS Model
+ SDValue TGA = DAG.getTargetGlobalAddress(GV, dl, MVT::i32,
+ 0, MipsII::MO_TLSGD);
+ SDValue Tlsgd = DAG.getNode(MipsISD::TlsGd, dl, MVT::i32, TGA);
+ SDValue GP = DAG.getRegister(Mips::GP, MVT::i32);
+ SDValue Argument = DAG.getNode(ISD::ADD, dl, MVT::i32, GP, Tlsgd);
+
+ ArgListTy Args;
+ ArgListEntry Entry;
+ Entry.Node = Argument;
+ Entry.Ty = (Type *) Type::getInt32Ty(*DAG.getContext());
+ Args.push_back(Entry);
+ std::pair<SDValue, SDValue> CallResult =
+ LowerCallTo(DAG.getEntryNode(),
+ (Type *) Type::getInt32Ty(*DAG.getContext()),
+ false, false, false, false, 0, CallingConv::C, false, true,
+ DAG.getExternalSymbol("__tls_get_addr", PtrVT), Args, DAG,
+ dl);
+
+ return CallResult.first;
+ }
+
+ SDValue Offset;
+ if (GV->isDeclaration()) {
+ // Initial Exec TLS Model
+ SDValue TGA = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
+ MipsII::MO_GOTTPREL);
+ Offset = DAG.getLoad(MVT::i32, dl,
+ DAG.getEntryNode(), TGA, MachinePointerInfo(),
+ false, false, 0);
+ } else {
+ // Local Exec TLS Model
+ SDVTList VTs = DAG.getVTList(MVT::i32);
+ SDValue TGAHi = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
+ MipsII::MO_TPREL_HI);
+ SDValue TGALo = DAG.getTargetGlobalAddress(GV, dl, MVT::i32, 0,
+ MipsII::MO_TPREL_LO);
+ SDValue Hi = DAG.getNode(MipsISD::TprelHi, dl, VTs, &TGAHi, 1);
+ SDValue Lo = DAG.getNode(MipsISD::TprelLo, dl, MVT::i32, TGALo);
+ Offset = DAG.getNode(ISD::ADD, dl, MVT::i32, Hi, Lo);
+ }
+
+ SDValue ThreadPointer = DAG.getNode(MipsISD::ThreadPointer, dl, PtrVT);
+ return DAG.getNode(ISD::ADD, dl, PtrVT, ThreadPointer, Offset);
}
SDValue MipsTargetLowering::
if (!IsPIC) {
SDValue Ops[] = { JTI };
HiPart = DAG.getNode(MipsISD::Hi, dl, DAG.getVTList(MVT::i32), Ops, 1);
- } else // Emit Load from Global Pointer
+ } else {// Emit Load from Global Pointer
+ JTI = DAG.getNode(MipsISD::WrapperPIC, dl, MVT::i32, JTI);
HiPart = DAG.getLoad(MVT::i32, dl, DAG.getEntryNode(), JTI,
MachinePointerInfo(),
false, false, 0);
+ }
SDValue JTILo = DAG.getTargetJumpTable(JT->getIndex(), PtrVT,
MipsII::MO_ABS_LO);
} else {
SDValue CP = DAG.getTargetConstantPool(C, MVT::i32, N->getAlignment(),
N->getOffset(), MipsII::MO_GOT);
+ CP = DAG.getNode(MipsISD::WrapperPIC, dl, MVT::i32, CP);
SDValue Load = DAG.getLoad(MVT::i32, dl, DAG.getEntryNode(),
CP, MachinePointerInfo::getConstantPool(),
false, false, 0);
false, false, 0);
}
+static SDValue LowerFCOPYSIGN32(SDValue Op, SelectionDAG &DAG) {
+ // FIXME: Use ext/ins instructions if target architecture is Mips32r2.
+ DebugLoc dl = Op.getDebugLoc();
+ SDValue Op0 = DAG.getNode(ISD::BITCAST, dl, MVT::i32, Op.getOperand(0));
+ SDValue Op1 = DAG.getNode(ISD::BITCAST, dl, MVT::i32, Op.getOperand(1));
+ SDValue And0 = DAG.getNode(ISD::AND, dl, MVT::i32, Op0,
+ DAG.getConstant(0x7fffffff, MVT::i32));
+ SDValue And1 = DAG.getNode(ISD::AND, dl, MVT::i32, Op1,
+ DAG.getConstant(0x80000000, MVT::i32));
+ SDValue Result = DAG.getNode(ISD::OR, dl, MVT::i32, And0, And1);
+ return DAG.getNode(ISD::BITCAST, dl, MVT::f32, Result);
+}
+
+static SDValue LowerFCOPYSIGN64(SDValue Op, SelectionDAG &DAG, bool isLittle) {
+ // FIXME:
+ // Use ext/ins instructions if target architecture is Mips32r2.
+ // Eliminate redundant mfc1 and mtc1 instructions.
+ unsigned LoIdx = 0, HiIdx = 1;
+
+ if (!isLittle)
+ std::swap(LoIdx, HiIdx);
+
+ DebugLoc dl = Op.getDebugLoc();
+ SDValue Word0 = DAG.getNode(MipsISD::ExtractElementF64, dl, MVT::i32,
+ Op.getOperand(0),
+ DAG.getConstant(LoIdx, MVT::i32));
+ SDValue Hi0 = DAG.getNode(MipsISD::ExtractElementF64, dl, MVT::i32,
+ Op.getOperand(0), DAG.getConstant(HiIdx, MVT::i32));
+ SDValue Hi1 = DAG.getNode(MipsISD::ExtractElementF64, dl, MVT::i32,
+ Op.getOperand(1), DAG.getConstant(HiIdx, MVT::i32));
+ SDValue And0 = DAG.getNode(ISD::AND, dl, MVT::i32, Hi0,
+ DAG.getConstant(0x7fffffff, MVT::i32));
+ SDValue And1 = DAG.getNode(ISD::AND, dl, MVT::i32, Hi1,
+ DAG.getConstant(0x80000000, MVT::i32));
+ SDValue Word1 = DAG.getNode(ISD::OR, dl, MVT::i32, And0, And1);
+
+ if (!isLittle)
+ std::swap(Word0, Word1);
+
+ return DAG.getNode(MipsISD::BuildPairF64, dl, MVT::f64, Word0, Word1);
+}
+
+SDValue MipsTargetLowering::LowerFCOPYSIGN(SDValue Op, SelectionDAG &DAG)
+ const {
+ EVT Ty = Op.getValueType();
+
+ assert(Ty == MVT::f32 || Ty == MVT::f64);
+
+ if (Ty == MVT::f32)
+ return LowerFCOPYSIGN32(Op, DAG);
+ else
+ return LowerFCOPYSIGN64(Op, DAG, Subtarget->isLittle());
+}
+
+SDValue MipsTargetLowering::
+LowerFRAMEADDR(SDValue Op, SelectionDAG &DAG) const {
+ // check the depth
+ assert((cast<ConstantSDNode>(Op.getOperand(0))->getZExtValue() == 0) &&
+ "Frame address can only be determined for current frame.");
+
+ MachineFrameInfo *MFI = DAG.getMachineFunction().getFrameInfo();
+ MFI->setFrameAddressIsTaken(true);
+ EVT VT = Op.getValueType();
+ DebugLoc dl = Op.getDebugLoc();
+ SDValue FrameAddr = DAG.getCopyFromReg(DAG.getEntryNode(), dl, Mips::FP, VT);
+ return FrameAddr;
+}
+
+// TODO: set SType according to the desired memory barrier behavior.
+SDValue MipsTargetLowering::LowerMEMBARRIER(SDValue Op,
+ SelectionDAG& DAG) const {
+ unsigned SType = 0;
+ DebugLoc dl = Op.getDebugLoc();
+ return DAG.getNode(MipsISD::Sync, dl, MVT::Other, Op.getOperand(0),
+ DAG.getConstant(SType, MVT::i32));
+}
+
+SDValue MipsTargetLowering::LowerATOMIC_FENCE(SDValue Op,
+ SelectionDAG& DAG) const {
+ // FIXME: Need pseudo-fence for 'singlethread' fences
+ // FIXME: Set SType for weaker fences where supported/appropriate.
+ unsigned SType = 0;
+ DebugLoc dl = Op.getDebugLoc();
+ return DAG.getNode(MipsISD::Sync, dl, MVT::Other, Op.getOperand(0),
+ DAG.getConstant(SType, MVT::i32));
+}
+
//===----------------------------------------------------------------------===//
// Calling Convention Implementation
//===----------------------------------------------------------------------===//
Mips::A0, Mips::A1, Mips::A2, Mips::A3
};
+// Return next O32 integer argument register.
+static unsigned getNextIntArgReg(unsigned Reg) {
+ assert((Reg == Mips::A0) || (Reg == Mips::A2));
+ return (Reg == Mips::A0) ? Mips::A1 : Mips::A3;
+}
+
// Write ByVal Arg to arg registers and stack.
static void
-WriteByValArg(SDValue& Chain, DebugLoc dl,
+WriteByValArg(SDValue& ByValChain, SDValue Chain, DebugLoc dl,
SmallVector<std::pair<unsigned, SDValue>, 16>& RegsToPass,
SmallVector<SDValue, 8>& MemOpChains, int& LastFI,
MachineFrameInfo *MFI, SelectionDAG &DAG, SDValue Arg,
- const CCValAssign &VA, const ISD::ArgFlagsTy& Flags, MVT PtrType) {
- unsigned FirstWord = VA.getLocMemOffset() / 4;
- unsigned NumWords = (Flags.getByValSize() + 3) / 4;
- unsigned LastWord = FirstWord + NumWords;
- unsigned CurWord;
-
- // copy the first 4 words of byval arg to registers A0 - A3
- for (CurWord = FirstWord; CurWord < std::min(LastWord, O32IntRegsSize);
- ++CurWord) {
+ const CCValAssign &VA, const ISD::ArgFlagsTy& Flags,
+ MVT PtrType, bool isLittle) {
+ unsigned LocMemOffset = VA.getLocMemOffset();
+ unsigned Offset = 0;
+ uint32_t RemainingSize = Flags.getByValSize();
+ unsigned ByValAlign = Flags.getByValAlign();
+
+ // Copy the first 4 words of byval arg to registers A0 - A3.
+ // FIXME: Use a stricter alignment if it enables better optimization in passes
+ // run later.
+ for (; RemainingSize >= 4 && LocMemOffset < 4 * 4;
+ Offset += 4, RemainingSize -= 4, LocMemOffset += 4) {
SDValue LoadPtr = DAG.getNode(ISD::ADD, dl, MVT::i32, Arg,
- DAG.getConstant((CurWord - FirstWord) * 4,
- MVT::i32));
+ DAG.getConstant(Offset, MVT::i32));
SDValue LoadVal = DAG.getLoad(MVT::i32, dl, Chain, LoadPtr,
MachinePointerInfo(),
- false, false, 0);
+ false, false, std::min(ByValAlign,
+ (unsigned )4));
MemOpChains.push_back(LoadVal.getValue(1));
- unsigned DstReg = O32IntRegs[CurWord];
+ unsigned DstReg = O32IntRegs[LocMemOffset / 4];
RegsToPass.push_back(std::make_pair(DstReg, LoadVal));
}
- // copy remaining part of byval arg to stack.
- if (CurWord < LastWord) {
- unsigned SizeInBytes = (LastWord - CurWord) * 4;
- SDValue Src = DAG.getNode(ISD::ADD, dl, MVT::i32, Arg,
- DAG.getConstant((CurWord - FirstWord) * 4,
- MVT::i32));
- LastFI = MFI->CreateFixedObject(SizeInBytes, CurWord * 4, true);
- SDValue Dst = DAG.getFrameIndex(LastFI, PtrType);
- Chain = DAG.getMemcpy(Chain, dl, Dst, Src,
- DAG.getConstant(SizeInBytes, MVT::i32),
- /*Align*/4,
- /*isVolatile=*/false, /*AlwaysInline=*/false,
- MachinePointerInfo(0), MachinePointerInfo(0));
- MemOpChains.push_back(Chain);
+ if (RemainingSize == 0)
+ return;
+
+ // If there still is a register available for argument passing, write the
+ // remaining part of the structure to it using subword loads and shifts.
+ if (LocMemOffset < 4 * 4) {
+ assert(RemainingSize <= 3 && RemainingSize >= 1 &&
+ "There must be one to three bytes remaining.");
+ unsigned LoadSize = (RemainingSize == 3 ? 2 : RemainingSize);
+ SDValue LoadPtr = DAG.getNode(ISD::ADD, dl, MVT::i32, Arg,
+ DAG.getConstant(Offset, MVT::i32));
+ unsigned Alignment = std::min(ByValAlign, (unsigned )4);
+ SDValue LoadVal = DAG.getExtLoad(ISD::ZEXTLOAD, dl, MVT::i32, Chain,
+ LoadPtr, MachinePointerInfo(),
+ MVT::getIntegerVT(LoadSize * 8), false,
+ false, Alignment);
+ MemOpChains.push_back(LoadVal.getValue(1));
+
+ // If target is big endian, shift it to the most significant half-word or
+ // byte.
+ if (!isLittle)
+ LoadVal = DAG.getNode(ISD::SHL, dl, MVT::i32, LoadVal,
+ DAG.getConstant(32 - LoadSize * 8, MVT::i32));
+
+ Offset += LoadSize;
+ RemainingSize -= LoadSize;
+
+ // Read second subword if necessary.
+ if (RemainingSize != 0) {
+ assert(RemainingSize == 1 && "There must be one byte remaining.");
+ LoadPtr = DAG.getNode(ISD::ADD, dl, MVT::i32, Arg,
+ DAG.getConstant(Offset, MVT::i32));
+ unsigned Alignment = std::min(ByValAlign, (unsigned )2);
+ SDValue Subword = DAG.getExtLoad(ISD::ZEXTLOAD, dl, MVT::i32, Chain,
+ LoadPtr, MachinePointerInfo(),
+ MVT::i8, false, false, Alignment);
+ MemOpChains.push_back(Subword.getValue(1));
+ // Insert the loaded byte to LoadVal.
+ // FIXME: Use INS if supported by target.
+ unsigned ShiftAmt = isLittle ? 16 : 8;
+ SDValue Shift = DAG.getNode(ISD::SHL, dl, MVT::i32, Subword,
+ DAG.getConstant(ShiftAmt, MVT::i32));
+ LoadVal = DAG.getNode(ISD::OR, dl, MVT::i32, LoadVal, Shift);
+ }
+
+ unsigned DstReg = O32IntRegs[LocMemOffset / 4];
+ RegsToPass.push_back(std::make_pair(DstReg, LoadVal));
+ return;
}
+
+ // Create a fixed object on stack at offset LocMemOffset and copy
+ // remaining part of byval arg to it using memcpy.
+ SDValue Src = DAG.getNode(ISD::ADD, dl, MVT::i32, Arg,
+ DAG.getConstant(Offset, MVT::i32));
+ LastFI = MFI->CreateFixedObject(RemainingSize, LocMemOffset, true);
+ SDValue Dst = DAG.getFrameIndex(LastFI, PtrType);
+ ByValChain = DAG.getMemcpy(ByValChain, dl, Dst, Src,
+ DAG.getConstant(RemainingSize, MVT::i32),
+ std::min(ByValAlign, (unsigned)4),
+ /*isVolatile=*/false, /*AlwaysInline=*/false,
+ MachinePointerInfo(0), MachinePointerInfo(0));
}
/// LowerCall - functions arguments are copied from virtual regs to
/// (physical regs)/(stack frame), CALLSEQ_START and CALLSEQ_END are emitted.
/// TODO: isTailCall.
SDValue
-MipsTargetLowering::LowerCall(SDValue Chain, SDValue Callee,
+MipsTargetLowering::LowerCall(SDValue InChain, SDValue Callee,
CallingConv::ID CallConv, bool isVarArg,
bool &isTailCall,
const SmallVectorImpl<ISD::OutputArg> &Outs,
// Analyze operands of the call, assigning locations to each operand.
SmallVector<CCValAssign, 16> ArgLocs;
- CCState CCInfo(CallConv, isVarArg, getTargetMachine(), ArgLocs,
- *DAG.getContext());
+ CCState CCInfo(CallConv, isVarArg, DAG.getMachineFunction(),
+ getTargetMachine(), ArgLocs, *DAG.getContext());
if (Subtarget->isABI_O32())
CCInfo.AnalyzeCallOperands(Outs, CC_MipsO32);
CCInfo.AnalyzeCallOperands(Outs, CC_Mips);
// Get a count of how many bytes are to be pushed on the stack.
- unsigned NumBytes = CCInfo.getNextStackOffset();
- Chain = DAG.getCALLSEQ_START(Chain, DAG.getIntPtrConstant(NumBytes, true));
+ unsigned NextStackOffset = CCInfo.getNextStackOffset();
+
+ // Chain is the output chain of the last Load/Store or CopyToReg node.
+ // ByValChain is the output chain of the last Memcpy node created for copying
+ // byval arguments to the stack.
+ SDValue Chain, CallSeqStart, ByValChain;
+ SDValue NextStackOffsetVal = DAG.getIntPtrConstant(NextStackOffset, true);
+ Chain = CallSeqStart = DAG.getCALLSEQ_START(InChain, NextStackOffsetVal);
+ ByValChain = InChain;
+
+ // If this is the first call, create a stack frame object that points to
+ // a location to which .cprestore saves $gp.
+ if (IsPIC && !MipsFI->getGPFI())
+ MipsFI->setGPFI(MFI->CreateFixedObject(4, 0, true));
+
+ // Get the frame index of the stack frame object that points to the location
+ // of dynamically allocated area on the stack.
+ int DynAllocFI = MipsFI->getDynAllocFI();
+
+ // Update size of the maximum argument space.
+ // For O32, a minimum of four words (16 bytes) of argument space is
+ // allocated.
+ if (Subtarget->isABI_O32())
+ NextStackOffset = std::max(NextStackOffset, (unsigned)16);
+
+ unsigned MaxCallFrameSize = MipsFI->getMaxCallFrameSize();
+
+ if (MaxCallFrameSize < NextStackOffset) {
+ MipsFI->setMaxCallFrameSize(NextStackOffset);
+
+ // Set the offsets relative to $sp of the $gp restore slot and dynamically
+ // allocated stack space. These offsets must be aligned to a boundary
+ // determined by the stack alignment of the ABI.
+ unsigned StackAlignment = TFL->getStackAlignment();
+ NextStackOffset = (NextStackOffset + StackAlignment - 1) /
+ StackAlignment * StackAlignment;
+
+ if (IsPIC)
+ MFI->setObjectOffset(MipsFI->getGPFI(), NextStackOffset);
+
+ MFI->setObjectOffset(DynAllocFI, NextStackOffset);
+ }
// With EABI is it possible to have 16 args on registers.
SmallVector<std::pair<unsigned, SDValue>, 16> RegsToPass;
SmallVector<SDValue, 8> MemOpChains;
- MipsFI->setHasCall();
-
- // Create GP frame object if this is the first call.
- // SPOffset will be updated after call frame size is known.
- if (IsPIC && !MipsFI->getGPFI())
- MipsFI->setGPFI(MFI->CreateFixedObject(4, 0, true));
-
- int FirstFI = -MFI->getNumFixedObjects() - 1, LastFI = 0;
+ int FirstFI = -MFI->getNumFixedObjects() - 1, LastFI = 0;
// Walk the register/memloc assignments, inserting copies/loads.
for (unsigned i = 0, e = ArgLocs.size(); i != e; ++i) {
Arg, DAG.getConstant(1, MVT::i32));
if (!Subtarget->isLittle())
std::swap(Lo, Hi);
- RegsToPass.push_back(std::make_pair(VA.getLocReg(), Lo));
- RegsToPass.push_back(std::make_pair(VA.getLocReg()+1, Hi));
+ unsigned LocRegLo = VA.getLocReg();
+ unsigned LocRegHigh = getNextIntArgReg(LocRegLo);
+ RegsToPass.push_back(std::make_pair(LocRegLo, Lo));
+ RegsToPass.push_back(std::make_pair(LocRegHigh, Hi));
continue;
}
}
// Register can't get to this point...
assert(VA.isMemLoc());
- // ByVal Arg.
+ // ByVal Arg.
ISD::ArgFlagsTy Flags = Outs[i].Flags;
if (Flags.isByVal()) {
assert(Subtarget->isABI_O32() &&
"No support for ByVal args by ABIs other than O32 yet.");
assert(Flags.getByValSize() &&
"ByVal args of size 0 should have been ignored by front-end.");
- WriteByValArg(Chain, dl, RegsToPass, MemOpChains, LastFI, MFI, DAG, Arg,
- VA, Flags, getPointerTy());
+ WriteByValArg(ByValChain, Chain, dl, RegsToPass, MemOpChains, LastFI, MFI,
+ DAG, Arg, VA, Flags, getPointerTy(), Subtarget->isLittle());
continue;
}
// Create the frame index object for this incoming parameter
- // This guarantees that when allocating Local Area the firsts
- // 16 bytes which are alwayes reserved won't be overwritten
- // if O32 ABI is used. For EABI the first address is zero.
- LastFI = MFI->CreateFixedObject(VA.getValVT().getSizeInBits()/8,
+ LastFI = MFI->CreateFixedObject(VA.getValVT().getSizeInBits()/8,
VA.getLocMemOffset(), true);
SDValue PtrOff = DAG.getFrameIndex(LastFI, getPointerTy());
false, false, 0));
}
+ // Extend range of indices of frame objects for outgoing arguments that were
+ // created during this function call. Skip this step if no such objects were
+ // created.
+ if (LastFI)
+ MipsFI->extendOutArgFIRange(FirstFI, LastFI);
+
+ // If a memcpy has been created to copy a byval arg to a stack, replace the
+ // chain input of CallSeqStart with ByValChain.
+ if (InChain != ByValChain)
+ DAG.UpdateNodeOperands(CallSeqStart.getNode(), ByValChain,
+ NextStackOffsetVal);
+
// Transform all store nodes into one single node because all store
// nodes are independent of each other.
if (!MemOpChains.empty())
if (IsPIC) {
if (LoadSymAddr) {
// Load callee address
- SDValue LoadValue = DAG.getLoad(MVT::i32, dl, Chain, Callee,
+ Callee = DAG.getNode(MipsISD::WrapperPIC, dl, MVT::i32, Callee);
+ SDValue LoadValue = DAG.getLoad(MVT::i32, dl, DAG.getEntryNode(), Callee,
MachinePointerInfo::getGOT(),
false, false, 0);
Callee = DAG.getNode(ISD::ADD, dl, MVT::i32, LoadValue, Lo);
} else
Callee = LoadValue;
-
- // Use chain output from LoadValue
- Chain = LoadValue.getValue(1);
}
// copy to T9
Chain = DAG.getNode(MipsISD::JmpLink, dl, NodeTys, &Ops[0], Ops.size());
InFlag = Chain.getValue(1);
- // Create a stack location to hold GP when PIC is used. This stack
- // location is used on function prologue to save GP and also after all
- // emitted CALL's to restore GP.
- if (IsPIC) {
- // Function can have an arbitrary number of calls, so
- // hold the LastArgStackLoc with the biggest offset.
- int MaxCallFrameSize = MipsFI->getMaxCallFrameSize();
- unsigned NextStackOffset = CCInfo.getNextStackOffset();
-
- // For O32, a minimum of four words (16 bytes) of argument space is
- // allocated.
- if (Subtarget->isABI_O32())
- NextStackOffset = std::max(NextStackOffset, (unsigned)16);
-
- if (MaxCallFrameSize < (int)NextStackOffset) {
- MipsFI->setMaxCallFrameSize(NextStackOffset);
-
- // $gp restore slot must be aligned.
- unsigned StackAlignment = TFL->getStackAlignment();
- NextStackOffset = (NextStackOffset + StackAlignment - 1) /
- StackAlignment * StackAlignment;
- int GPFI = MipsFI->getGPFI();
- MFI->setObjectOffset(GPFI, NextStackOffset);
- }
- }
-
- // Extend range of indices of frame objects for outgoing arguments that were
- // created during this function call. Skip this step if no such objects were
- // created.
- if (LastFI)
- MipsFI->extendOutArgFIRange(FirstFI, LastFI);
-
// Create the CALLSEQ_END node.
- Chain = DAG.getCALLSEQ_END(Chain, DAG.getIntPtrConstant(NumBytes, true),
+ Chain = DAG.getCALLSEQ_END(Chain,
+ DAG.getIntPtrConstant(NextStackOffset, true),
DAG.getIntPtrConstant(0, true), InFlag);
InFlag = Chain.getValue(1);
const SmallVectorImpl<ISD::InputArg> &Ins,
DebugLoc dl, SelectionDAG &DAG,
SmallVectorImpl<SDValue> &InVals) const {
-
// Assign locations to each value returned by this call.
SmallVector<CCValAssign, 16> RVLocs;
- CCState CCInfo(CallConv, isVarArg, getTargetMachine(),
- RVLocs, *DAG.getContext());
+ CCState CCInfo(CallConv, isVarArg, DAG.getMachineFunction(),
+ getTargetMachine(), RVLocs, *DAG.getContext());
CCInfo.AnalyzeCallResult(Ins, RetCC_Mips);
DebugLoc dl, SelectionDAG &DAG,
SmallVectorImpl<SDValue> &InVals)
const {
-
MachineFunction &MF = DAG.getMachineFunction();
MachineFrameInfo *MFI = MF.getFrameInfo();
MipsFunctionInfo *MipsFI = MF.getInfo<MipsFunctionInfo>();
// Assign locations to all of the incoming arguments.
SmallVector<CCValAssign, 16> ArgLocs;
- CCState CCInfo(CallConv, isVarArg, getTargetMachine(),
- ArgLocs, *DAG.getContext());
+ CCState CCInfo(CallConv, isVarArg, DAG.getMachineFunction(),
+ getTargetMachine(), ArgLocs, *DAG.getContext());
if (Subtarget->isABI_O32())
CCInfo.AnalyzeFormalArguments(Ins, CC_MipsO32);
if (RegVT == MVT::i32)
RC = Mips::CPURegsRegisterClass;
+ else if (RegVT == MVT::i64)
+ RC = Mips::CPU64RegsRegisterClass;
else if (RegVT == MVT::f32)
RC = Mips::FGR32RegisterClass;
else if (RegVT == MVT::f64) {
ArgValue = DAG.getNode(ISD::BITCAST, dl, MVT::f32, ArgValue);
if (RegVT == MVT::i32 && VA.getValVT() == MVT::f64) {
unsigned Reg2 = AddLiveIn(DAG.getMachineFunction(),
- VA.getLocReg()+1, RC);
+ getNextIntArgReg(ArgReg), RC);
SDValue ArgValue2 = DAG.getCopyFromReg(Chain, dl, Reg2, RegVT);
if (!Subtarget->isLittle())
std::swap(ArgValue, ArgValue2);
}
// The stack pointer offset is relative to the caller stack frame.
- // Since the real stack size is unknown here, a negative SPOffset
- // is used so there's a way to adjust these offsets when the stack
- // size get known (on EliminateFrameIndex). A dummy SPOffset is
- // used instead of a direct negative address (which is recorded to
- // be used on emitPrologue) to avoid mis-calc of the first stack
- // offset on PEI::calculateFrameObjectOffsets.
LastFI = MFI->CreateFixedObject(VA.getValVT().getSizeInBits()/8,
VA.getLocMemOffset(), true);
Chain = DAG.getNode(ISD::TokenFactor, dl, MVT::Other, Copy, Chain);
}
- // To meet ABI, when VARARGS are passed on registers, the registers
- // must have their values written to the caller stack frame. If the last
- // argument was placed in the stack, there's no need to save any register.
if (isVarArg && Subtarget->isABI_O32()) {
// Record the frame index of the first variable argument
- // which is a value necessary to VASTART.
+ // which is a value necessary to VASTART.
unsigned NextStackOffset = CCInfo.getNextStackOffset();
assert(NextStackOffset % 4 == 0 &&
"NextStackOffset must be aligned to 4-byte boundaries.");
LastFI = MFI->CreateFixedObject(4, NextStackOffset, true);
MipsFI->setVarArgsFrameIndex(LastFI);
-
- // Copy variable arguments passed in registers to stack.
+
+ // If NextStackOffset is smaller than o32's 16-byte reserved argument area,
+ // copy the integer registers that have not been used for argument passing
+ // to the caller's stack frame.
for (; NextStackOffset < 16; NextStackOffset += 4) {
TargetRegisterClass *RC = Mips::CPURegsRegisterClass;
unsigned Idx = NextStackOffset / 4;
SmallVector<CCValAssign, 16> RVLocs;
// CCState - Info about the registers and stack slot.
- CCState CCInfo(CallConv, isVarArg, getTargetMachine(),
- RVLocs, *DAG.getContext());
+ CCState CCInfo(CallConv, isVarArg, DAG.getMachineFunction(),
+ getTargetMachine(), RVLocs, *DAG.getContext());
// Analize return values.
CCInfo.AnalyzeReturn(Outs, RetCC_Mips);
// but allow it at the lowest weight.
if (CallOperandVal == NULL)
return CW_Default;
- const Type *type = CallOperandVal->getType();
+ Type *type = CallOperandVal->getType();
// Look at the constraint type.
switch (*constraint) {
default:
return weight;
}
-/// getRegClassForInlineAsmConstraint - Given a constraint letter (e.g. "r"),
-/// return a list of registers that can be used to satisfy the constraint.
-/// This should only be used for C_RegisterClass constraints.
+/// Given a register class constraint, like 'r', if this corresponds directly
+/// to an LLVM register class, return a register of 0 and the register class
+/// pointer.
std::pair<unsigned, const TargetRegisterClass*> MipsTargetLowering::
getRegForInlineAsmConstraint(const std::string &Constraint, EVT VT) const
{
if (Constraint.size() == 1) {
switch (Constraint[0]) {
+ case 'd': // Address register. Same as 'r' unless generating MIPS16 code.
+ case 'y': // Same as 'r'. Exists for compatibility.
case 'r':
return std::make_pair(0U, Mips::CPURegsRegisterClass);
case 'f':
if (VT == MVT::f64)
if ((!Subtarget->isSingleFloat()) && (!Subtarget->isFP64bit()))
return std::make_pair(0U, Mips::AFGR64RegisterClass);
+ break;
}
}
return TargetLowering::getRegForInlineAsmConstraint(Constraint, VT);
}
-/// Given a register class constraint, like 'r', if this corresponds directly
-/// to an LLVM register class, return a register of 0 and the register class
-/// pointer.
-std::vector<unsigned> MipsTargetLowering::
-getRegClassForInlineAsmConstraint(const std::string &Constraint,
- EVT VT) const
-{
- if (Constraint.size() != 1)
- return std::vector<unsigned>();
-
- switch (Constraint[0]) {
- default : break;
- case 'r':
- // GCC Mips Constraint Letters
- case 'd':
- case 'y':
- return make_vector<unsigned>(Mips::T0, Mips::T1, Mips::T2, Mips::T3,
- Mips::T4, Mips::T5, Mips::T6, Mips::T7, Mips::S0, Mips::S1,
- Mips::S2, Mips::S3, Mips::S4, Mips::S5, Mips::S6, Mips::S7,
- Mips::T8, 0);
-
- case 'f':
- if (VT == MVT::f32) {
- if (Subtarget->isSingleFloat())
- return make_vector<unsigned>(Mips::F2, Mips::F3, Mips::F4, Mips::F5,
- Mips::F6, Mips::F7, Mips::F8, Mips::F9, Mips::F10, Mips::F11,
- Mips::F20, Mips::F21, Mips::F22, Mips::F23, Mips::F24,
- Mips::F25, Mips::F26, Mips::F27, Mips::F28, Mips::F29,
- Mips::F30, Mips::F31, 0);
- else
- return make_vector<unsigned>(Mips::F2, Mips::F4, Mips::F6, Mips::F8,
- Mips::F10, Mips::F20, Mips::F22, Mips::F24, Mips::F26,
- Mips::F28, Mips::F30, 0);
- }
-
- if (VT == MVT::f64)
- if ((!Subtarget->isSingleFloat()) && (!Subtarget->isFP64bit()))
- return make_vector<unsigned>(Mips::D1, Mips::D2, Mips::D3, Mips::D4,
- Mips::D5, Mips::D10, Mips::D11, Mips::D12, Mips::D13,
- Mips::D14, Mips::D15, 0);
- }
- return std::vector<unsigned>();
-}
-
bool
MipsTargetLowering::isOffsetFoldingLegal(const GlobalAddressSDNode *GA) const {
// The Mips target isn't yet aware of offsets.
projects / oota-llvm.git / blobdiff