bool AArch64FrameLowering::spillCalleeSavedRegisters(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MI,
const std::vector<CalleeSavedInfo> &CSI,
const TargetRegisterInfo *TRI) const {
MachineFunction &MF = *MBB.getParent();
const TargetInstrInfo &TII = *MF.getSubtarget().getInstrInfo();
DebugLoc DL;
SmallVector<RegPairInfo, 8> RegPairs;
computeCalleeSaveRegisterPairs(MF, CSI, TRI, RegPairs);
for (auto RPII = RegPairs.rbegin(), RPIE = RegPairs.rend(); RPII != RPIE;
++RPII) {
RegPairInfo RPI = *RPII;
unsigned Reg1 = RPI.Reg1;
unsigned Reg2 = RPI.Reg2;
unsigned StrOpc;
// Issue sequence of spills for cs regs. The first spill may be converted
// to a pre-decrement store later by emitPrologue if the callee-save stack
// area allocation can't be combined with the local stack area allocation.
// For example:
// stp x22, x21, [sp, #0] // addImm(+0)
// stp x20, x19, [sp, #16] // addImm(+2)
// stp fp, lr, [sp, #32] // addImm(+4)
// Rationale: This sequence saves uop updates compared to a sequence of
// pre-increment spills like stp xi,xj,[sp,#-16]!
// Note: Similar rationale and sequence for restores in epilog.
if (RPI.IsGPR)
StrOpc = RPI.isPaired() ? AArch64::STPXi : AArch64::STRXui;
else
StrOpc = RPI.isPaired() ? AArch64::STPDi : AArch64::STRDui;
DEBUG(dbgs() << "CSR spill: (" << TRI->getName(Reg1);
if (RPI.isPaired())
dbgs() << ", " << TRI->getName(Reg2);
dbgs() << ") -> fi#(" << RPI.FrameIdx;
if (RPI.isPaired())
dbgs() << ", " << RPI.FrameIdx+1;
dbgs() << ")\n");
MachineInstrBuilder MIB = BuildMI(MBB, MI, DL, TII.get(StrOpc));
MBB.addLiveIn(Reg1);
if (RPI.isPaired()) {
MBB.addLiveIn(Reg2);
MIB.addReg(Reg2, getPrologueDeath(MF, Reg2));
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx + 1),
MachineMemOperand::MOStore, 8, 8));
}
MIB.addReg(Reg1, getPrologueDeath(MF, Reg1))
.addReg(AArch64::SP)
.addImm(RPI.Offset) // [sp, #offset*8], where factor*8 is implicit
.setMIFlag(MachineInstr::FrameSetup);
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx),
MachineMemOperand::MOStore, 8, 8));
}
return true;
}
bool AArch64FrameLowering::restoreCalleeSavedRegisters(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MI,
const std::vector<CalleeSavedInfo> &CSI,
const TargetRegisterInfo *TRI) const {
MachineFunction &MF = *MBB.getParent();
const TargetInstrInfo &TII = *MF.getSubtarget().getInstrInfo();
DebugLoc DL;
SmallVector<RegPairInfo, 8> RegPairs;
if (MI != MBB.end())
DL = MI->getDebugLoc();
computeCalleeSaveRegisterPairs(MF, CSI, TRI, RegPairs);
for (auto RPII = RegPairs.begin(), RPIE = RegPairs.end(); RPII != RPIE;
++RPII) {
RegPairInfo RPI = *RPII;
unsigned Reg1 = RPI.Reg1;
unsigned Reg2 = RPI.Reg2;
// Issue sequence of restores for cs regs. The last restore may be converted
// to a post-increment load later by emitEpilogue if the callee-save stack
// area allocation can't be combined with the local stack area allocation.
// For example:
// ldp fp, lr, [sp, #32] // addImm(+4)
// ldp x20, x19, [sp, #16] // addImm(+2)
// ldp x22, x21, [sp, #0] // addImm(+0)
// Note: see comment in spillCalleeSavedRegisters()
unsigned LdrOpc;
if (RPI.IsGPR)
LdrOpc = RPI.isPaired() ? AArch64::LDPXi : AArch64::LDRXui;
else
LdrOpc = RPI.isPaired() ? AArch64::LDPDi : AArch64::LDRDui;
DEBUG(dbgs() << "CSR restore: (" << TRI->getName(Reg1);
if (RPI.isPaired())
dbgs() << ", " << TRI->getName(Reg2);
dbgs() << ") -> fi#(" << RPI.FrameIdx;
if (RPI.isPaired())
dbgs() << ", " << RPI.FrameIdx+1;
dbgs() << ")\n");
MachineInstrBuilder MIB = BuildMI(MBB, MI, DL, TII.get(LdrOpc));
if (RPI.isPaired()) {
MIB.addReg(Reg2, getDefRegState(true));
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx + 1),
MachineMemOperand::MOLoad, 8, 8));
}
MIB.addReg(Reg1, getDefRegState(true))
.addReg(AArch64::SP)
.addImm(RPI.Offset) // [sp, #offset*8] where the factor*8 is implicit
.setMIFlag(MachineInstr::FrameDestroy);
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx),
MachineMemOperand::MOLoad, 8, 8));
}
return true;
}
void ARMInstrInfo::expandLoadStackGuard(MachineBasicBlock::iterator MI) const {
MachineFunction &MF = *MI->getParent()->getParent();
const ARMSubtarget &Subtarget = MF.getSubtarget<ARMSubtarget>();
const TargetMachine &TM = MF.getTarget();
if (!Subtarget.useMovt(MF)) {
if (TM.isPositionIndependent())
expandLoadStackGuardBase(MI, ARM::LDRLIT_ga_pcrel, ARM::LDRi12);
else
expandLoadStackGuardBase(MI, ARM::LDRLIT_ga_abs, ARM::LDRi12);
return;
}
if (!TM.isPositionIndependent()) {
expandLoadStackGuardBase(MI, ARM::MOVi32imm, ARM::LDRi12);
return;
}
const GlobalValue *GV =
cast<GlobalValue>((*MI->memoperands_begin())->getValue());
if (!Subtarget.isGVIndirectSymbol(GV)) {
expandLoadStackGuardBase(MI, ARM::MOV_ga_pcrel, ARM::LDRi12);
return;
}
MachineBasicBlock &MBB = *MI->getParent();
DebugLoc DL = MI->getDebugLoc();
unsigned Reg = MI->getOperand(0).getReg();
MachineInstrBuilder MIB;
MIB = BuildMI(MBB, MI, DL, get(ARM::MOV_ga_pcrel_ldr), Reg)
.addGlobalAddress(GV, 0, ARMII::MO_NONLAZY);
auto Flags = MachineMemOperand::MOLoad |
MachineMemOperand::MODereferenceable |
MachineMemOperand::MOInvariant;
MachineMemOperand *MMO = MBB.getParent()->getMachineMemOperand(
MachinePointerInfo::getGOT(*MBB.getParent()), Flags, 4, 4);
MIB.addMemOperand(MMO);
BuildMI(MBB, MI, DL, get(ARM::LDRi12), Reg)
.addReg(Reg, RegState::Kill)
.addImm(0)
.cloneMemRefs(*MI)
.add(predOps(ARMCC::AL));
}
void WebAssemblyFastISel::addLoadStoreOperands(const Address &Addr,
const MachineInstrBuilder &MIB,
MachineMemOperand *MMO) {
if (const GlobalValue *GV = Addr.getGlobalValue())
MIB.addGlobalAddress(GV, Addr.getOffset());
else
MIB.addImm(Addr.getOffset());
if (Addr.isRegBase())
MIB.addReg(Addr.getReg());
else
MIB.addFrameIndex(Addr.getFI());
// Set the alignment operand (this is rewritten in SetP2AlignOperands).
// TODO: Disable SetP2AlignOperands for FastISel and just do it here.
MIB.addImm(0);
MIB.addMemOperand(MMO);
}
void ARMInstrInfo::expandLoadStackGuard(MachineBasicBlock::iterator MI,
Reloc::Model RM) const {
MachineFunction &MF = *MI->getParent()->getParent();
const ARMSubtarget &Subtarget = MF.getSubtarget<ARMSubtarget>();
if (!Subtarget.useMovt(MF)) {
if (RM == Reloc::PIC_)
expandLoadStackGuardBase(MI, ARM::LDRLIT_ga_pcrel, ARM::LDRi12, RM);
else
expandLoadStackGuardBase(MI, ARM::LDRLIT_ga_abs, ARM::LDRi12, RM);
return;
}
if (RM != Reloc::PIC_) {
expandLoadStackGuardBase(MI, ARM::MOVi32imm, ARM::LDRi12, RM);
return;
}
const GlobalValue *GV =
cast<GlobalValue>((*MI->memoperands_begin())->getValue());
if (!Subtarget.GVIsIndirectSymbol(GV, RM)) {
expandLoadStackGuardBase(MI, ARM::MOV_ga_pcrel, ARM::LDRi12, RM);
return;
}
MachineBasicBlock &MBB = *MI->getParent();
DebugLoc DL = MI->getDebugLoc();
unsigned Reg = MI->getOperand(0).getReg();
MachineInstrBuilder MIB;
MIB = BuildMI(MBB, MI, DL, get(ARM::MOV_ga_pcrel_ldr), Reg)
.addGlobalAddress(GV, 0, ARMII::MO_NONLAZY);
unsigned Flag = MachineMemOperand::MOLoad | MachineMemOperand::MOInvariant;
MachineMemOperand *MMO = MBB.getParent()->getMachineMemOperand(
MachinePointerInfo::getGOT(*MBB.getParent()), Flag, 4, 4);
MIB.addMemOperand(MMO);
MIB = BuildMI(MBB, MI, DL, get(ARM::LDRi12), Reg);
MIB.addReg(Reg, RegState::Kill).addImm(0);
MIB.setMemRefs(MI->memoperands_begin(), MI->memoperands_end());
AddDefaultPred(MIB);
}
bool ARMInstructionSelector::selectGlobal(MachineInstrBuilder &MIB,
MachineRegisterInfo &MRI) const {
if ((STI.isROPI() || STI.isRWPI()) && !STI.isTargetELF()) {
LLVM_DEBUG(dbgs() << "ROPI and RWPI only supported for ELF\n");
return false;
}
auto GV = MIB->getOperand(1).getGlobal();
if (GV->isThreadLocal()) {
LLVM_DEBUG(dbgs() << "TLS variables not supported yet\n");
return false;
}
auto &MBB = *MIB->getParent();
auto &MF = *MBB.getParent();
bool UseMovt = STI.useMovt(MF);
unsigned Size = TM.getPointerSize(0);
unsigned Alignment = 4;
auto addOpsForConstantPoolLoad = [&MF, Alignment,
Size](MachineInstrBuilder &MIB,
const GlobalValue *GV, bool IsSBREL) {
assert(MIB->getOpcode() == ARM::LDRi12 && "Unsupported instruction");
auto ConstPool = MF.getConstantPool();
auto CPIndex =
// For SB relative entries we need a target-specific constant pool.
// Otherwise, just use a regular constant pool entry.
IsSBREL
? ConstPool->getConstantPoolIndex(
ARMConstantPoolConstant::Create(GV, ARMCP::SBREL), Alignment)
: ConstPool->getConstantPoolIndex(GV, Alignment);
MIB.addConstantPoolIndex(CPIndex, /*Offset*/ 0, /*TargetFlags*/ 0)
.addMemOperand(
MF.getMachineMemOperand(MachinePointerInfo::getConstantPool(MF),
MachineMemOperand::MOLoad, Size, Alignment))
.addImm(0)
.add(predOps(ARMCC::AL));
};
if (TM.isPositionIndependent()) {
bool Indirect = STI.isGVIndirectSymbol(GV);
// FIXME: Taking advantage of MOVT for ELF is pretty involved, so we don't
// support it yet. See PR28229.
unsigned Opc =
UseMovt && !STI.isTargetELF()
? (Indirect ? ARM::MOV_ga_pcrel_ldr : ARM::MOV_ga_pcrel)
: (Indirect ? ARM::LDRLIT_ga_pcrel_ldr : ARM::LDRLIT_ga_pcrel);
MIB->setDesc(TII.get(Opc));
int TargetFlags = ARMII::MO_NO_FLAG;
if (STI.isTargetDarwin())
TargetFlags |= ARMII::MO_NONLAZY;
if (STI.isGVInGOT(GV))
TargetFlags |= ARMII::MO_GOT;
MIB->getOperand(1).setTargetFlags(TargetFlags);
if (Indirect)
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getGOT(MF), MachineMemOperand::MOLoad,
TM.getProgramPointerSize(), Alignment));
return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}
bool isReadOnly = STI.getTargetLowering()->isReadOnly(GV);
if (STI.isROPI() && isReadOnly) {
unsigned Opc = UseMovt ? ARM::MOV_ga_pcrel : ARM::LDRLIT_ga_pcrel;
MIB->setDesc(TII.get(Opc));
return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}
if (STI.isRWPI() && !isReadOnly) {
auto Offset = MRI.createVirtualRegister(&ARM::GPRRegClass);
MachineInstrBuilder OffsetMIB;
if (UseMovt) {
OffsetMIB = BuildMI(MBB, *MIB, MIB->getDebugLoc(),
TII.get(ARM::MOVi32imm), Offset);
OffsetMIB.addGlobalAddress(GV, /*Offset*/ 0, ARMII::MO_SBREL);
} else {
// Load the offset from the constant pool.
OffsetMIB =
BuildMI(MBB, *MIB, MIB->getDebugLoc(), TII.get(ARM::LDRi12), Offset);
addOpsForConstantPoolLoad(OffsetMIB, GV, /*IsSBREL*/ true);
}
if (!constrainSelectedInstRegOperands(*OffsetMIB, TII, TRI, RBI))
return false;
// Add the offset to the SB register.
MIB->setDesc(TII.get(ARM::ADDrr));
MIB->RemoveOperand(1);
MIB.addReg(ARM::R9) // FIXME: don't hardcode R9
.addReg(Offset)
.add(predOps(ARMCC::AL))
.add(condCodeOp());
return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}
if (STI.isTargetELF()) {
if (UseMovt) {
MIB->setDesc(TII.get(ARM::MOVi32imm));
} else {
// Load the global's address from the constant pool.
MIB->setDesc(TII.get(ARM::LDRi12));
MIB->RemoveOperand(1);
addOpsForConstantPoolLoad(MIB, GV, /*IsSBREL*/ false);
}
} else if (STI.isTargetMachO()) {
if (UseMovt)
MIB->setDesc(TII.get(ARM::MOVi32imm));
else
MIB->setDesc(TII.get(ARM::LDRLIT_ga_abs));
} else {
LLVM_DEBUG(dbgs() << "Object format not supported yet\n");
return false;
}
return constrainSelectedInstRegOperands(*MIB, TII, TRI, RBI);
}
unsigned Disassembler::decodeInstruction(unsigned Address,
MachineBasicBlock *Block) {
// Disassemble instruction
const MCDisassembler *DA = MC->getMCDisassembler();
uint64_t InstSize;
MCInst *Inst = new MCInst();
StringRef Bytes;
if (!(DA->getInstruction(*Inst, InstSize, *CurSectionMemory, Address,
nulls(), nulls()))) {
printError("Unknown instruction encountered, instruction decode failed!");
return 1;
// Instructions[Address] = NULL;
// Block->push_back(NULL);
// TODO: Replace with default size for each target.
// return 1;
// outs() << format("%8" PRIx64 ":\t", SectAddr + Index);
// Dism->rawBytesToString(StringRef(Bytes.data() + Index, Size));
// outs() << " unkn\n";
}
Instructions[Address] = Inst;
// Recover Instruction information
const MCInstrInfo *MII = MC->getMCInstrInfo();
MCInstrDesc *MCID = new MCInstrDesc(MII->get(Inst->getOpcode()));
MCID->Size = InstSize;
// Check if the instruction can load to program counter and mark it as a Ret
// FIXME: Better analysis would be to see if the PC value references memory
// sent as a parameter or set locally in the function, but that would need to
// happen after decompilation. In either case, this is definitely a BB
// terminator or branch!
if (MCID->mayLoad()
&& MCID->mayAffectControlFlow(*Inst, *MC->getMCRegisterInfo())) {
MCID->Flags |= (1 << MCID::Return);
MCID->Flags |= (1 << MCID::Terminator);
}
// Recover MachineInstr representation
DebugLoc *Location = setDebugLoc(Address);
MachineInstrBuilder MIB = BuildMI(Block, *Location, *MCID);
unsigned int numDefs = MCID->getNumDefs();
for (unsigned int i = 0; i < Inst->getNumOperands(); i++) {
MCOperand MCO = Inst->getOperand(i);
// FIXME: This hack is a workaround for the assert in MachineInstr.cpp:653,
// where OpNo >= MCID->getNumOperands()...
if (i >= MCID->getNumOperands() && !(MCID->isVariadic()))
break;
if (MCO.isReg()) {
unsigned flags = 0;
// Defs always start at the beginning of the operands list,
// unfortunately BuildMI doesn't set default define flags so we have
// to do it manually here.
// NOTE: This should always be true, but might not be if operands list
// is not populated correctly by the MC Backend for the target.
if (i < numDefs) {
flags |= RegState::Define;
}
// NOTE: No need to worry about imp defs and uses, as these are already
// specificed in the MCID attached to the MachineInst object.
MIB.addReg(MCO.getReg(), flags);
continue;
}
if (MCO.isImm()) {
MIB.addImm(MCO.getImm());
continue;
}
//else if (MCO.isFPImm()) MIB.addFPImm(MCO.getFPImm());
if (MCO.isExpr()) {
MCOperandInfo MCOpInfo = MCID->OpInfo[i];
switch (MCOpInfo.OperandType) {
case MCOI::OPERAND_MEMORY:
case MCOI::OPERAND_PCREL:
case MCOI::OPERAND_UNKNOWN:
default:
printError("Unknown how to handle this Expression at this time.");
}
}
printError("Unknown how to handle Operand!");
}
// NOTE: I tried MCOpInfo here, and it appearst o be NULL
// ... at least for ARM.
unsigned flags = 0;
if (MCID->mayLoad())
flags |= MachineMemOperand::MOLoad;
if (MCID->mayStore())
flags |= MachineMemOperand::MOStore;
if (flags != 0) {
// Constant* cInt = ConstantInt::get(Type::getInt64Ty(ctx), MCO.getImm());
// Value *Val = ConstantExpr::getIntToPtr(cInt,
// PointerType::getUnqual(Type::getInt32Ty(ctx)));
// FIXME: note size of 4 is known to be bad for
// some targets
//Copy & paste set getImm to zero
MachineMemOperand* MMO = new MachineMemOperand(
MachinePointerInfo(), flags, 4, 0); //MCO.getImm()
MIB.addMemOperand(MMO);
//outs() << "Name: " << MII->getName(Inst->getOpcode()) << " Flags: " << flags << "\n";
}
// Note: I don't know why they decided instruction size needed to be 64 bits,
// but the following conversion shouldn't be an issue.
return ((unsigned)InstSize);
}
bool AArch64FrameLowering::restoreCalleeSavedRegisters(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MI,
const std::vector<CalleeSavedInfo> &CSI,
const TargetRegisterInfo *TRI) const {
MachineFunction &MF = *MBB.getParent();
const TargetInstrInfo &TII = *MF.getSubtarget().getInstrInfo();
DebugLoc DL;
SmallVector<RegPairInfo, 8> RegPairs;
if (MI != MBB.end())
DL = MI->getDebugLoc();
computeCalleeSaveRegisterPairs(MF, CSI, TRI, RegPairs);
for (auto RPII = RegPairs.begin(), RPIE = RegPairs.end(); RPII != RPIE;
++RPII) {
RegPairInfo RPI = *RPII;
unsigned Reg1 = RPI.Reg1;
unsigned Reg2 = RPI.Reg2;
// Issue sequence of non-sp increment and sp-pi restores for cs regs. Only
// the last load is sp-pi post-increment and de-allocates the stack:
// For example:
// ldp fp, lr, [sp, #32] // addImm(+4)
// ldp x20, x19, [sp, #16] // addImm(+2)
// ldp x22, x21, [sp], #48 // addImm(+6)
// Note: see comment in spillCalleeSavedRegisters()
unsigned LdrOpc;
bool BumpSP = RPII == std::prev(RegPairs.end());
if (RPI.IsGPR) {
if (BumpSP)
LdrOpc = RPI.isPaired() ? AArch64::LDPXpost : AArch64::LDRXpost;
else
LdrOpc = RPI.isPaired() ? AArch64::LDPXi : AArch64::LDRXui;
} else {
if (BumpSP)
LdrOpc = RPI.isPaired() ? AArch64::LDPDpost : AArch64::LDRDpost;
else
LdrOpc = RPI.isPaired() ? AArch64::LDPDi : AArch64::LDRDui;
}
DEBUG(dbgs() << "CSR restore: (" << TRI->getName(Reg1);
if (RPI.isPaired())
dbgs() << ", " << TRI->getName(Reg2);
dbgs() << ") -> fi#(" << RPI.FrameIdx;
if (RPI.isPaired())
dbgs() << ", " << RPI.FrameIdx+1;
dbgs() << ")\n");
const int Offset = RPI.Offset;
MachineInstrBuilder MIB = BuildMI(MBB, MI, DL, TII.get(LdrOpc));
if (BumpSP)
MIB.addReg(AArch64::SP, RegState::Define);
if (RPI.isPaired()) {
MIB.addReg(Reg2, getDefRegState(true))
.addReg(Reg1, getDefRegState(true))
.addReg(AArch64::SP)
.addImm(Offset) // [sp], #offset * 8 or [sp, #offset * 8]
// where the factor * 8 is implicit
.setMIFlag(MachineInstr::FrameDestroy);
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx + 1),
MachineMemOperand::MOLoad, 8, 8));
} else {
MIB.addReg(Reg1, getDefRegState(true))
.addReg(AArch64::SP)
.addImm(BumpSP ? Offset * 8 : Offset) // post-dec version is unscaled
.setMIFlag(MachineInstr::FrameDestroy);
}
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx),
MachineMemOperand::MOLoad, 8, 8));
}
return true;
}
bool AArch64FrameLowering::spillCalleeSavedRegisters(
MachineBasicBlock &MBB, MachineBasicBlock::iterator MI,
const std::vector<CalleeSavedInfo> &CSI,
const TargetRegisterInfo *TRI) const {
MachineFunction &MF = *MBB.getParent();
const TargetInstrInfo &TII = *MF.getSubtarget().getInstrInfo();
DebugLoc DL;
SmallVector<RegPairInfo, 8> RegPairs;
computeCalleeSaveRegisterPairs(MF, CSI, TRI, RegPairs);
for (auto RPII = RegPairs.rbegin(), RPIE = RegPairs.rend(); RPII != RPIE;
++RPII) {
RegPairInfo RPI = *RPII;
unsigned Reg1 = RPI.Reg1;
unsigned Reg2 = RPI.Reg2;
unsigned StrOpc;
// Issue sequence of non-sp increment and pi sp spills for cs regs. The
// first spill is a pre-increment that allocates the stack.
// For example:
// stp x22, x21, [sp, #-48]! // addImm(-6)
// stp x20, x19, [sp, #16] // addImm(+2)
// stp fp, lr, [sp, #32] // addImm(+4)
// Rationale: This sequence saves uop updates compared to a sequence of
// pre-increment spills like stp xi,xj,[sp,#-16]!
// Note: Similar rationale and sequence for restores in epilog.
bool BumpSP = RPII == RegPairs.rbegin();
if (RPI.IsGPR) {
// For first spill use pre-increment store.
if (BumpSP)
StrOpc = RPI.isPaired() ? AArch64::STPXpre : AArch64::STRXpre;
else
StrOpc = RPI.isPaired() ? AArch64::STPXi : AArch64::STRXui;
} else {
// For first spill use pre-increment store.
if (BumpSP)
StrOpc = RPI.isPaired() ? AArch64::STPDpre : AArch64::STRDpre;
else
StrOpc = RPI.isPaired() ? AArch64::STPDi : AArch64::STRDui;
}
DEBUG(dbgs() << "CSR spill: (" << TRI->getName(Reg1);
if (RPI.isPaired())
dbgs() << ", " << TRI->getName(Reg2);
dbgs() << ") -> fi#(" << RPI.FrameIdx;
if (RPI.isPaired())
dbgs() << ", " << RPI.FrameIdx+1;
dbgs() << ")\n");
const int Offset = BumpSP ? -RPI.Offset : RPI.Offset;
MachineInstrBuilder MIB = BuildMI(MBB, MI, DL, TII.get(StrOpc));
if (BumpSP)
MIB.addReg(AArch64::SP, RegState::Define);
if (RPI.isPaired()) {
MBB.addLiveIn(Reg1);
MBB.addLiveIn(Reg2);
MIB.addReg(Reg2, getPrologueDeath(MF, Reg2))
.addReg(Reg1, getPrologueDeath(MF, Reg1))
.addReg(AArch64::SP)
.addImm(Offset) // [sp, #offset * 8], where factor * 8 is implicit
.setMIFlag(MachineInstr::FrameSetup);
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx + 1),
MachineMemOperand::MOStore, 8, 8));
} else {
MBB.addLiveIn(Reg1);
MIB.addReg(Reg1, getPrologueDeath(MF, Reg1))
.addReg(AArch64::SP)
.addImm(BumpSP ? Offset * 8 : Offset) // pre-inc version is unscaled
.setMIFlag(MachineInstr::FrameSetup);
}
MIB.addMemOperand(MF.getMachineMemOperand(
MachinePointerInfo::getFixedStack(MF, RPI.FrameIdx),
MachineMemOperand::MOStore, 8, 8));
}
return true;
}
