X-Git-Url: http://plrg.eecs.uci.edu/git/?a=blobdiff_plain;f=lib%2FTarget%2FX86%2FX86MCInstLower.cpp;h=b423e1e7d19154061c5a3e4b5a7833895cb7edd3;hb=8cfa23f93aeb8d3596a1ddc53b458953e487b8cf;hp=03c059528e5470cca77f7dbc5218a96ac353e30d;hpb=de42e5c09b1e1aa597de7b3e09775e26a0f68aa6;p=oota-llvm.git diff --git a/lib/Target/X86/X86MCInstLower.cpp b/lib/Target/X86/X86MCInstLower.cpp index 03c059528e5..2b30ce6d362 100644 --- a/lib/Target/X86/X86MCInstLower.cpp +++ b/lib/Target/X86/X86MCInstLower.cpp @@ -12,135 +12,225 @@ // //===----------------------------------------------------------------------===// -#include "InstPrinter/X86ATTInstPrinter.h" -#include "X86MCInstLower.h" #include "X86AsmPrinter.h" -#include "X86COFFMachineModuleInfo.h" -#include "X86MCAsmInfo.h" +#include "X86RegisterInfo.h" +#include "InstPrinter/X86ATTInstPrinter.h" +#include "MCTargetDesc/X86BaseInfo.h" +#include "Utils/X86ShuffleDecode.h" +#include "llvm/ADT/Optional.h" +#include "llvm/ADT/SmallString.h" +#include "llvm/CodeGen/MachineFunction.h" +#include "llvm/CodeGen/MachineConstantPool.h" +#include "llvm/CodeGen/MachineOperand.h" #include "llvm/CodeGen/MachineModuleInfoImpls.h" +#include "llvm/CodeGen/StackMaps.h" +#include "llvm/IR/DataLayout.h" +#include "llvm/IR/GlobalValue.h" +#include "llvm/IR/Mangler.h" +#include "llvm/MC/MCAsmInfo.h" +#include "llvm/MC/MCCodeEmitter.h" #include "llvm/MC/MCContext.h" #include "llvm/MC/MCExpr.h" +#include "llvm/MC/MCFixup.h" #include "llvm/MC/MCInst.h" +#include "llvm/MC/MCInstBuilder.h" #include "llvm/MC/MCStreamer.h" #include "llvm/MC/MCSymbol.h" -#include "llvm/Target/Mangler.h" -#include "llvm/Support/FormattedStream.h" -#include "llvm/ADT/SmallString.h" -#include "llvm/Type.h" +#include "llvm/Support/TargetRegistry.h" using namespace llvm; -X86MCInstLower::X86MCInstLower(Mangler *mang, const MachineFunction &mf, +namespace { + +/// X86MCInstLower - This class is used to lower an MachineInstr into an MCInst. +class X86MCInstLower { + MCContext &Ctx; + const MachineFunction &MF; + const TargetMachine &TM; + const MCAsmInfo &MAI; + X86AsmPrinter &AsmPrinter; +public: + X86MCInstLower(const MachineFunction &MF, X86AsmPrinter &asmprinter); + + Optional LowerMachineOperand(const MachineInstr *MI, + const MachineOperand &MO) const; + void Lower(const MachineInstr *MI, MCInst &OutMI) const; + + MCSymbol *GetSymbolFromOperand(const MachineOperand &MO) const; + MCOperand LowerSymbolOperand(const MachineOperand &MO, MCSymbol *Sym) const; + +private: + MachineModuleInfoMachO &getMachOMMI() const; + Mangler *getMang() const { + return AsmPrinter.Mang; + } +}; + +} // end anonymous namespace + +// Emit a minimal sequence of nops spanning NumBytes bytes. +static void EmitNops(MCStreamer &OS, unsigned NumBytes, bool Is64Bit, + const MCSubtargetInfo &STI); + +namespace llvm { + X86AsmPrinter::StackMapShadowTracker::StackMapShadowTracker(TargetMachine &TM) + : TM(TM), InShadow(false), RequiredShadowSize(0), CurrentShadowSize(0) {} + + X86AsmPrinter::StackMapShadowTracker::~StackMapShadowTracker() {} + + void + X86AsmPrinter::StackMapShadowTracker::startFunction(MachineFunction &F) { + MF = &F; + CodeEmitter.reset(TM.getTarget().createMCCodeEmitter( + *MF->getSubtarget().getInstrInfo(), + *MF->getSubtarget().getRegisterInfo(), MF->getContext())); + } + + void X86AsmPrinter::StackMapShadowTracker::count(MCInst &Inst, + const MCSubtargetInfo &STI) { + if (InShadow) { + SmallString<256> Code; + SmallVector Fixups; + raw_svector_ostream VecOS(Code); + CodeEmitter->encodeInstruction(Inst, VecOS, Fixups, STI); + VecOS.flush(); + CurrentShadowSize += Code.size(); + if (CurrentShadowSize >= RequiredShadowSize) + InShadow = false; // The shadow is big enough. Stop counting. + } + } + + void X86AsmPrinter::StackMapShadowTracker::emitShadowPadding( + MCStreamer &OutStreamer, const MCSubtargetInfo &STI) { + if (InShadow && CurrentShadowSize < RequiredShadowSize) { + InShadow = false; + EmitNops(OutStreamer, RequiredShadowSize - CurrentShadowSize, + MF->getSubtarget().is64Bit(), STI); + } + } + + void X86AsmPrinter::EmitAndCountInstruction(MCInst &Inst) { + OutStreamer->EmitInstruction(Inst, getSubtargetInfo()); + SMShadowTracker.count(Inst, getSubtargetInfo()); + } +} // end llvm namespace + +X86MCInstLower::X86MCInstLower(const MachineFunction &mf, X86AsmPrinter &asmprinter) -: Ctx(mf.getContext()), Mang(mang), MF(mf), TM(mf.getTarget()), - MAI(*TM.getMCAsmInfo()), AsmPrinter(asmprinter) {} + : Ctx(mf.getContext()), MF(mf), TM(mf.getTarget()), MAI(*TM.getMCAsmInfo()), + AsmPrinter(asmprinter) {} MachineModuleInfoMachO &X86MCInstLower::getMachOMMI() const { return MF.getMMI().getObjFileInfo(); } -MCSymbol *X86MCInstLower::GetPICBaseSymbol() const { - return static_cast(TM.getTargetLowering())-> - getPICBaseSymbol(&MF, Ctx); -} - /// GetSymbolFromOperand - Lower an MO_GlobalAddress or MO_ExternalSymbol /// operand to an MCSymbol. MCSymbol *X86MCInstLower:: GetSymbolFromOperand(const MachineOperand &MO) const { - assert((MO.isGlobal() || MO.isSymbol()) && "Isn't a symbol reference"); + const DataLayout &DL = MF.getDataLayout(); + assert((MO.isGlobal() || MO.isSymbol() || MO.isMBB()) && "Isn't a symbol reference"); + MCSymbol *Sym = nullptr; SmallString<128> Name; - - if (!MO.isGlobal()) { - assert(MO.isSymbol()); - Name += MAI.getGlobalPrefix(); - Name += MO.getSymbolName(); - } else { + StringRef Suffix; + + switch (MO.getTargetFlags()) { + case X86II::MO_DLLIMPORT: + // Handle dllimport linkage. + Name += "__imp_"; + break; + case X86II::MO_DARWIN_STUB: + Suffix = "$stub"; + break; + case X86II::MO_DARWIN_NONLAZY: + case X86II::MO_DARWIN_NONLAZY_PIC_BASE: + case X86II::MO_DARWIN_HIDDEN_NONLAZY_PIC_BASE: + Suffix = "$non_lazy_ptr"; + break; + } + + if (!Suffix.empty()) + Name += DL.getPrivateGlobalPrefix(); + + unsigned PrefixLen = Name.size(); + + if (MO.isGlobal()) { const GlobalValue *GV = MO.getGlobal(); - bool isImplicitlyPrivate = false; - if (MO.getTargetFlags() == X86II::MO_DARWIN_STUB || - MO.getTargetFlags() == X86II::MO_DARWIN_NONLAZY || - MO.getTargetFlags() == X86II::MO_DARWIN_NONLAZY_PIC_BASE || - MO.getTargetFlags() == X86II::MO_DARWIN_HIDDEN_NONLAZY_PIC_BASE) - isImplicitlyPrivate = true; - - Mang->getNameWithPrefix(Name, GV, isImplicitlyPrivate); + AsmPrinter.getNameWithPrefix(Name, GV); + } else if (MO.isSymbol()) { + Mangler::getNameWithPrefix(Name, MO.getSymbolName(), DL); + } else if (MO.isMBB()) { + assert(Suffix.empty()); + Sym = MO.getMBB()->getSymbol(); } + unsigned OrigLen = Name.size() - PrefixLen; + + Name += Suffix; + if (!Sym) + Sym = Ctx.getOrCreateSymbol(Name); + + StringRef OrigName = StringRef(Name).substr(PrefixLen, OrigLen); // If the target flags on the operand changes the name of the symbol, do that // before we return the symbol. switch (MO.getTargetFlags()) { default: break; - case X86II::MO_DLLIMPORT: { - // Handle dllimport linkage. - const char *Prefix = "__imp_"; - Name.insert(Name.begin(), Prefix, Prefix+strlen(Prefix)); - break; - } case X86II::MO_DARWIN_NONLAZY: case X86II::MO_DARWIN_NONLAZY_PIC_BASE: { - Name += "$non_lazy_ptr"; - MCSymbol *Sym = Ctx.GetOrCreateSymbol(Name.str()); - MachineModuleInfoImpl::StubValueTy &StubSym = getMachOMMI().getGVStubEntry(Sym); - if (StubSym.getPointer() == 0) { + if (!StubSym.getPointer()) { assert(MO.isGlobal() && "Extern symbol not handled yet"); StubSym = MachineModuleInfoImpl:: - StubValueTy(Mang->getSymbol(MO.getGlobal()), + StubValueTy(AsmPrinter.getSymbol(MO.getGlobal()), !MO.getGlobal()->hasInternalLinkage()); } - return Sym; + break; } case X86II::MO_DARWIN_HIDDEN_NONLAZY_PIC_BASE: { - Name += "$non_lazy_ptr"; - MCSymbol *Sym = Ctx.GetOrCreateSymbol(Name.str()); MachineModuleInfoImpl::StubValueTy &StubSym = getMachOMMI().getHiddenGVStubEntry(Sym); - if (StubSym.getPointer() == 0) { + if (!StubSym.getPointer()) { assert(MO.isGlobal() && "Extern symbol not handled yet"); StubSym = MachineModuleInfoImpl:: - StubValueTy(Mang->getSymbol(MO.getGlobal()), + StubValueTy(AsmPrinter.getSymbol(MO.getGlobal()), !MO.getGlobal()->hasInternalLinkage()); } - return Sym; + break; } case X86II::MO_DARWIN_STUB: { - Name += "$stub"; - MCSymbol *Sym = Ctx.GetOrCreateSymbol(Name.str()); MachineModuleInfoImpl::StubValueTy &StubSym = getMachOMMI().getFnStubEntry(Sym); if (StubSym.getPointer()) return Sym; - + if (MO.isGlobal()) { StubSym = MachineModuleInfoImpl:: - StubValueTy(Mang->getSymbol(MO.getGlobal()), + StubValueTy(AsmPrinter.getSymbol(MO.getGlobal()), !MO.getGlobal()->hasInternalLinkage()); } else { - Name.erase(Name.end()-5, Name.end()); StubSym = MachineModuleInfoImpl:: - StubValueTy(Ctx.GetOrCreateSymbol(Name.str()), false); + StubValueTy(Ctx.getOrCreateSymbol(OrigName), false); } - return Sym; + break; } } - return Ctx.GetOrCreateSymbol(Name.str()); + return Sym; } MCOperand X86MCInstLower::LowerSymbolOperand(const MachineOperand &MO, MCSymbol *Sym) const { // FIXME: We would like an efficient form for this, so we don't have to do a // lot of extra uniquing. - const MCExpr *Expr = 0; + const MCExpr *Expr = nullptr; MCSymbolRefExpr::VariantKind RefKind = MCSymbolRefExpr::VK_None; - + switch (MO.getTargetFlags()) { default: llvm_unreachable("Unknown target flag on GV operand"); case X86II::MO_NO_FLAG: // No flag. @@ -149,21 +239,26 @@ MCOperand X86MCInstLower::LowerSymbolOperand(const MachineOperand &MO, case X86II::MO_DLLIMPORT: case X86II::MO_DARWIN_STUB: break; - + case X86II::MO_TLVP: RefKind = MCSymbolRefExpr::VK_TLVP; break; case X86II::MO_TLVP_PIC_BASE: - Expr = MCSymbolRefExpr::Create(Sym, MCSymbolRefExpr::VK_TLVP, Ctx); + Expr = MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_TLVP, Ctx); // Subtract the pic base. - Expr = MCBinaryExpr::CreateSub(Expr, - MCSymbolRefExpr::Create(GetPICBaseSymbol(), + Expr = MCBinaryExpr::createSub(Expr, + MCSymbolRefExpr::create(MF.getPICBaseSymbol(), Ctx), Ctx); break; + case X86II::MO_SECREL: RefKind = MCSymbolRefExpr::VK_SECREL; break; case X86II::MO_TLSGD: RefKind = MCSymbolRefExpr::VK_TLSGD; break; + case X86II::MO_TLSLD: RefKind = MCSymbolRefExpr::VK_TLSLD; break; + case X86II::MO_TLSLDM: RefKind = MCSymbolRefExpr::VK_TLSLDM; break; case X86II::MO_GOTTPOFF: RefKind = MCSymbolRefExpr::VK_GOTTPOFF; break; case X86II::MO_INDNTPOFF: RefKind = MCSymbolRefExpr::VK_INDNTPOFF; break; case X86II::MO_TPOFF: RefKind = MCSymbolRefExpr::VK_TPOFF; break; + case X86II::MO_DTPOFF: RefKind = MCSymbolRefExpr::VK_DTPOFF; break; case X86II::MO_NTPOFF: RefKind = MCSymbolRefExpr::VK_NTPOFF; break; + case X86II::MO_GOTNTPOFF: RefKind = MCSymbolRefExpr::VK_GOTNTPOFF; break; case X86II::MO_GOTPCREL: RefKind = MCSymbolRefExpr::VK_GOTPCREL; break; case X86II::MO_GOT: RefKind = MCSymbolRefExpr::VK_GOT; break; case X86II::MO_GOTOFF: RefKind = MCSymbolRefExpr::VK_GOTOFF; break; @@ -171,71 +266,41 @@ MCOperand X86MCInstLower::LowerSymbolOperand(const MachineOperand &MO, case X86II::MO_PIC_BASE_OFFSET: case X86II::MO_DARWIN_NONLAZY_PIC_BASE: case X86II::MO_DARWIN_HIDDEN_NONLAZY_PIC_BASE: - Expr = MCSymbolRefExpr::Create(Sym, Ctx); + Expr = MCSymbolRefExpr::create(Sym, Ctx); // Subtract the pic base. - Expr = MCBinaryExpr::CreateSub(Expr, - MCSymbolRefExpr::Create(GetPICBaseSymbol(), Ctx), + Expr = MCBinaryExpr::createSub(Expr, + MCSymbolRefExpr::create(MF.getPICBaseSymbol(), Ctx), Ctx); - if (MO.isJTI() && MAI.hasSetDirective()) { + if (MO.isJTI()) { + assert(MAI.doesSetDirectiveSuppressesReloc()); // If .set directive is supported, use it to reduce the number of // relocations the assembler will generate for differences between // local labels. This is only safe when the symbols are in the same // section so we are restricting it to jumptable references. - MCSymbol *Label = Ctx.CreateTempSymbol(); - AsmPrinter.OutStreamer.EmitAssignment(Label, Expr); - Expr = MCSymbolRefExpr::Create(Label, Ctx); + MCSymbol *Label = Ctx.createTempSymbol(); + AsmPrinter.OutStreamer->EmitAssignment(Label, Expr); + Expr = MCSymbolRefExpr::create(Label, Ctx); } break; } - - if (Expr == 0) - Expr = MCSymbolRefExpr::Create(Sym, RefKind, Ctx); - - if (!MO.isJTI() && MO.getOffset()) - Expr = MCBinaryExpr::CreateAdd(Expr, - MCConstantExpr::Create(MO.getOffset(), Ctx), - Ctx); - return MCOperand::CreateExpr(Expr); -} - - -static void lower_subreg32(MCInst *MI, unsigned OpNo) { - // Convert registers in the addr mode according to subreg32. - unsigned Reg = MI->getOperand(OpNo).getReg(); - if (Reg != 0) - MI->getOperand(OpNo).setReg(getX86SubSuperRegister(Reg, MVT::i32)); -} + if (!Expr) + Expr = MCSymbolRefExpr::create(Sym, RefKind, Ctx); -static void lower_lea64_32mem(MCInst *MI, unsigned OpNo) { - // Convert registers in the addr mode according to subreg64. - for (unsigned i = 0; i != 4; ++i) { - if (!MI->getOperand(OpNo+i).isReg()) continue; - - unsigned Reg = MI->getOperand(OpNo+i).getReg(); - if (Reg == 0) continue; - - MI->getOperand(OpNo+i).setReg(getX86SubSuperRegister(Reg, MVT::i64)); - } + if (!MO.isJTI() && !MO.isMBB() && MO.getOffset()) + Expr = MCBinaryExpr::createAdd(Expr, + MCConstantExpr::create(MO.getOffset(), Ctx), + Ctx); + return MCOperand::createExpr(Expr); } -/// LowerSubReg32_Op0 - Things like MOVZX16rr8 -> MOVZX32rr8. -static void LowerSubReg32_Op0(MCInst &OutMI, unsigned NewOpc) { - OutMI.setOpcode(NewOpc); - lower_subreg32(&OutMI, 0); -} -/// LowerUnaryToTwoAddr - R = setb -> R = sbb R, R -static void LowerUnaryToTwoAddr(MCInst &OutMI, unsigned NewOpc) { - OutMI.setOpcode(NewOpc); - OutMI.addOperand(OutMI.getOperand(0)); - OutMI.addOperand(OutMI.getOperand(0)); -} /// \brief Simplify FOO $imm, %{al,ax,eax,rax} to FOO $imm, for instruction with /// a short fixed-register form. static void SimplifyShortImmForm(MCInst &Inst, unsigned Opcode) { unsigned ImmOp = Inst.getNumOperands() - 1; - assert(Inst.getOperand(0).isReg() && Inst.getOperand(ImmOp).isImm() && + assert(Inst.getOperand(0).isReg() && + (Inst.getOperand(ImmOp).isImm() || Inst.getOperand(ImmOp).isExpr()) && ((Inst.getNumOperands() == 3 && Inst.getOperand(1).isReg() && Inst.getOperand(0).getReg() == Inst.getOperand(1).getReg()) || Inst.getNumOperands() == 2) && "Unexpected instruction!"); @@ -252,6 +317,34 @@ static void SimplifyShortImmForm(MCInst &Inst, unsigned Opcode) { Inst.addOperand(Saved); } +/// \brief If a movsx instruction has a shorter encoding for the used register +/// simplify the instruction to use it instead. +static void SimplifyMOVSX(MCInst &Inst) { + unsigned NewOpcode = 0; + unsigned Op0 = Inst.getOperand(0).getReg(), Op1 = Inst.getOperand(1).getReg(); + switch (Inst.getOpcode()) { + default: + llvm_unreachable("Unexpected instruction!"); + case X86::MOVSX16rr8: // movsbw %al, %ax --> cbtw + if (Op0 == X86::AX && Op1 == X86::AL) + NewOpcode = X86::CBW; + break; + case X86::MOVSX32rr16: // movswl %ax, %eax --> cwtl + if (Op0 == X86::EAX && Op1 == X86::AX) + NewOpcode = X86::CWDE; + break; + case X86::MOVSX64rr32: // movslq %eax, %rax --> cltq + if (Op0 == X86::RAX && Op1 == X86::EAX) + NewOpcode = X86::CDQE; + break; + } + + if (NewOpcode != 0) { + Inst = MCInst(); + Inst.setOpcode(NewOpcode); + } +} + /// \brief Simplify things like MOV32rm to MOV32o32a. static void SimplifyShortMoveForm(X86AsmPrinter &Printer, MCInst &Inst, unsigned Opcode) { @@ -265,12 +358,12 @@ static void SimplifyShortMoveForm(X86AsmPrinter &Printer, MCInst &Inst, unsigned RegOp = IsStore ? 0 : 5; unsigned AddrOp = AddrBase + 3; assert(Inst.getNumOperands() == 6 && Inst.getOperand(RegOp).isReg() && - Inst.getOperand(AddrBase + 0).isReg() && // base - Inst.getOperand(AddrBase + 1).isImm() && // scale - Inst.getOperand(AddrBase + 2).isReg() && // index register - (Inst.getOperand(AddrOp).isExpr() || // address - Inst.getOperand(AddrOp).isImm())&& - Inst.getOperand(AddrBase + 4).isReg() && // segment + Inst.getOperand(AddrBase + X86::AddrBaseReg).isReg() && + Inst.getOperand(AddrBase + X86::AddrScaleAmt).isImm() && + Inst.getOperand(AddrBase + X86::AddrIndexReg).isReg() && + Inst.getOperand(AddrBase + X86::AddrSegmentReg).isReg() && + (Inst.getOperand(AddrOp).isExpr() || + Inst.getOperand(AddrOp).isImm()) && "Unexpected instruction!"); // Check whether the destination register can be fixed. @@ -279,7 +372,7 @@ static void SimplifyShortMoveForm(X86AsmPrinter &Printer, MCInst &Inst, return; // Check whether this is an absolute address. - // FIXME: We know TLVP symbol refs aren't, but there should be a better way + // FIXME: We know TLVP symbol refs aren't, but there should be a better way // to do this here. bool Absolute = true; if (Inst.getOperand(AddrOp).isExpr()) { @@ -288,71 +381,70 @@ static void SimplifyShortMoveForm(X86AsmPrinter &Printer, MCInst &Inst, if (SRE->getKind() == MCSymbolRefExpr::VK_TLVP) Absolute = false; } - + if (Absolute && - (Inst.getOperand(AddrBase + 0).getReg() != 0 || - Inst.getOperand(AddrBase + 2).getReg() != 0 || - Inst.getOperand(AddrBase + 4).getReg() != 0 || - Inst.getOperand(AddrBase + 1).getImm() != 1)) + (Inst.getOperand(AddrBase + X86::AddrBaseReg).getReg() != 0 || + Inst.getOperand(AddrBase + X86::AddrScaleAmt).getImm() != 1 || + Inst.getOperand(AddrBase + X86::AddrIndexReg).getReg() != 0)) return; // If so, rewrite the instruction. MCOperand Saved = Inst.getOperand(AddrOp); + MCOperand Seg = Inst.getOperand(AddrBase + X86::AddrSegmentReg); Inst = MCInst(); Inst.setOpcode(Opcode); Inst.addOperand(Saved); + Inst.addOperand(Seg); +} + +static unsigned getRetOpcode(const X86Subtarget &Subtarget) { + return Subtarget.is64Bit() ? X86::RETQ : X86::RETL; +} + +Optional +X86MCInstLower::LowerMachineOperand(const MachineInstr *MI, + const MachineOperand &MO) const { + switch (MO.getType()) { + default: + MI->dump(); + llvm_unreachable("unknown operand type"); + case MachineOperand::MO_Register: + // Ignore all implicit register operands. + if (MO.isImplicit()) + return None; + return MCOperand::createReg(MO.getReg()); + case MachineOperand::MO_Immediate: + return MCOperand::createImm(MO.getImm()); + case MachineOperand::MO_MachineBasicBlock: + case MachineOperand::MO_GlobalAddress: + case MachineOperand::MO_ExternalSymbol: + return LowerSymbolOperand(MO, GetSymbolFromOperand(MO)); + case MachineOperand::MO_MCSymbol: + return LowerSymbolOperand(MO, MO.getMCSymbol()); + case MachineOperand::MO_JumpTableIndex: + return LowerSymbolOperand(MO, AsmPrinter.GetJTISymbol(MO.getIndex())); + case MachineOperand::MO_ConstantPoolIndex: + return LowerSymbolOperand(MO, AsmPrinter.GetCPISymbol(MO.getIndex())); + case MachineOperand::MO_BlockAddress: + return LowerSymbolOperand( + MO, AsmPrinter.GetBlockAddressSymbol(MO.getBlockAddress())); + case MachineOperand::MO_RegisterMask: + // Ignore call clobbers. + return None; + } } void X86MCInstLower::Lower(const MachineInstr *MI, MCInst &OutMI) const { OutMI.setOpcode(MI->getOpcode()); - - for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) { - const MachineOperand &MO = MI->getOperand(i); - - MCOperand MCOp; - switch (MO.getType()) { - default: - MI->dump(); - llvm_unreachable("unknown operand type"); - case MachineOperand::MO_Register: - // Ignore all implicit register operands. - if (MO.isImplicit()) continue; - MCOp = MCOperand::CreateReg(MO.getReg()); - break; - case MachineOperand::MO_Immediate: - MCOp = MCOperand::CreateImm(MO.getImm()); - break; - case MachineOperand::MO_MachineBasicBlock: - MCOp = MCOperand::CreateExpr(MCSymbolRefExpr::Create( - MO.getMBB()->getSymbol(), Ctx)); - break; - case MachineOperand::MO_GlobalAddress: - MCOp = LowerSymbolOperand(MO, GetSymbolFromOperand(MO)); - break; - case MachineOperand::MO_ExternalSymbol: - MCOp = LowerSymbolOperand(MO, GetSymbolFromOperand(MO)); - break; - case MachineOperand::MO_JumpTableIndex: - MCOp = LowerSymbolOperand(MO, AsmPrinter.GetJTISymbol(MO.getIndex())); - break; - case MachineOperand::MO_ConstantPoolIndex: - MCOp = LowerSymbolOperand(MO, AsmPrinter.GetCPISymbol(MO.getIndex())); - break; - case MachineOperand::MO_BlockAddress: - MCOp = LowerSymbolOperand(MO, - AsmPrinter.GetBlockAddressSymbol(MO.getBlockAddress())); - break; - } - - OutMI.addOperand(MCOp); - } - + + for (const MachineOperand &MO : MI->operands()) + if (auto MaybeMCOp = LowerMachineOperand(MI, MO)) + OutMI.addOperand(MaybeMCOp.getValue()); + // Handle a few special cases to eliminate operand modifiers. ReSimplify: switch (OutMI.getOpcode()) { - case X86::LEA64_32r: // Handle 'subreg rewriting' for the lea64_32mem operand. - lower_lea64_32mem(&OutMI, 1); - // FALL THROUGH. + case X86::LEA64_32r: case X86::LEA64r: case X86::LEA16r: case X86::LEA32r: @@ -362,52 +454,69 @@ ReSimplify: assert(OutMI.getOperand(1+X86::AddrSegmentReg).getReg() == 0 && "LEA has segment specified!"); break; - case X86::MOVZX16rr8: LowerSubReg32_Op0(OutMI, X86::MOVZX32rr8); break; - case X86::MOVZX16rm8: LowerSubReg32_Op0(OutMI, X86::MOVZX32rm8); break; - case X86::MOVSX16rr8: LowerSubReg32_Op0(OutMI, X86::MOVSX32rr8); break; - case X86::MOVSX16rm8: LowerSubReg32_Op0(OutMI, X86::MOVSX32rm8); break; - case X86::MOVZX64rr32: LowerSubReg32_Op0(OutMI, X86::MOV32rr); break; - case X86::MOVZX64rm32: LowerSubReg32_Op0(OutMI, X86::MOV32rm); break; - case X86::MOV64ri64i32: LowerSubReg32_Op0(OutMI, X86::MOV32ri); break; - case X86::MOVZX64rr8: LowerSubReg32_Op0(OutMI, X86::MOVZX32rr8); break; - case X86::MOVZX64rm8: LowerSubReg32_Op0(OutMI, X86::MOVZX32rm8); break; - case X86::MOVZX64rr16: LowerSubReg32_Op0(OutMI, X86::MOVZX32rr16); break; - case X86::MOVZX64rm16: LowerSubReg32_Op0(OutMI, X86::MOVZX32rm16); break; - case X86::SETB_C8r: LowerUnaryToTwoAddr(OutMI, X86::SBB8rr); break; - case X86::SETB_C16r: LowerUnaryToTwoAddr(OutMI, X86::SBB16rr); break; - case X86::SETB_C32r: LowerUnaryToTwoAddr(OutMI, X86::SBB32rr); break; - case X86::SETB_C64r: LowerUnaryToTwoAddr(OutMI, X86::SBB64rr); break; - case X86::MOV8r0: LowerUnaryToTwoAddr(OutMI, X86::XOR8rr); break; - case X86::MOV32r0: LowerUnaryToTwoAddr(OutMI, X86::XOR32rr); break; - case X86::FsFLD0SS: LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break; - case X86::FsFLD0SD: LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break; - case X86::V_SET0PS: LowerUnaryToTwoAddr(OutMI, X86::XORPSrr); break; - case X86::V_SET0PD: LowerUnaryToTwoAddr(OutMI, X86::XORPDrr); break; - case X86::V_SET0PI: LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break; - case X86::V_SETALLONES: LowerUnaryToTwoAddr(OutMI, X86::PCMPEQDrr); break; - case X86::AVX_SET0PS: LowerUnaryToTwoAddr(OutMI, X86::VXORPSrr); break; - case X86::AVX_SET0PSY: LowerUnaryToTwoAddr(OutMI, X86::VXORPSYrr); break; - case X86::AVX_SET0PD: LowerUnaryToTwoAddr(OutMI, X86::VXORPDrr); break; - case X86::AVX_SET0PDY: LowerUnaryToTwoAddr(OutMI, X86::VXORPDYrr); break; - case X86::AVX_SET0PI: LowerUnaryToTwoAddr(OutMI, X86::VPXORrr); break; - - case X86::MOV16r0: - LowerSubReg32_Op0(OutMI, X86::MOV32r0); // MOV16r0 -> MOV32r0 - LowerUnaryToTwoAddr(OutMI, X86::XOR32rr); // MOV32r0 -> XOR32rr + + case X86::MOV32ri64: + OutMI.setOpcode(X86::MOV32ri); break; - case X86::MOV64r0: - LowerSubReg32_Op0(OutMI, X86::MOV32r0); // MOV64r0 -> MOV32r0 - LowerUnaryToTwoAddr(OutMI, X86::XOR32rr); // MOV32r0 -> XOR32rr + + // Commute operands to get a smaller encoding by using VEX.R instead of VEX.B + // if one of the registers is extended, but other isn't. + case X86::VMOVAPDrr: + case X86::VMOVAPDYrr: + case X86::VMOVAPSrr: + case X86::VMOVAPSYrr: + case X86::VMOVDQArr: + case X86::VMOVDQAYrr: + case X86::VMOVDQUrr: + case X86::VMOVDQUYrr: + case X86::VMOVUPDrr: + case X86::VMOVUPDYrr: + case X86::VMOVUPSrr: + case X86::VMOVUPSYrr: { + if (!X86II::isX86_64ExtendedReg(OutMI.getOperand(0).getReg()) && + X86II::isX86_64ExtendedReg(OutMI.getOperand(1).getReg())) { + unsigned NewOpc; + switch (OutMI.getOpcode()) { + default: llvm_unreachable("Invalid opcode"); + case X86::VMOVAPDrr: NewOpc = X86::VMOVAPDrr_REV; break; + case X86::VMOVAPDYrr: NewOpc = X86::VMOVAPDYrr_REV; break; + case X86::VMOVAPSrr: NewOpc = X86::VMOVAPSrr_REV; break; + case X86::VMOVAPSYrr: NewOpc = X86::VMOVAPSYrr_REV; break; + case X86::VMOVDQArr: NewOpc = X86::VMOVDQArr_REV; break; + case X86::VMOVDQAYrr: NewOpc = X86::VMOVDQAYrr_REV; break; + case X86::VMOVDQUrr: NewOpc = X86::VMOVDQUrr_REV; break; + case X86::VMOVDQUYrr: NewOpc = X86::VMOVDQUYrr_REV; break; + case X86::VMOVUPDrr: NewOpc = X86::VMOVUPDrr_REV; break; + case X86::VMOVUPDYrr: NewOpc = X86::VMOVUPDYrr_REV; break; + case X86::VMOVUPSrr: NewOpc = X86::VMOVUPSrr_REV; break; + case X86::VMOVUPSYrr: NewOpc = X86::VMOVUPSYrr_REV; break; + } + OutMI.setOpcode(NewOpc); + } break; + } + case X86::VMOVSDrr: + case X86::VMOVSSrr: { + if (!X86II::isX86_64ExtendedReg(OutMI.getOperand(0).getReg()) && + X86II::isX86_64ExtendedReg(OutMI.getOperand(2).getReg())) { + unsigned NewOpc; + switch (OutMI.getOpcode()) { + default: llvm_unreachable("Invalid opcode"); + case X86::VMOVSDrr: NewOpc = X86::VMOVSDrr_REV; break; + case X86::VMOVSSrr: NewOpc = X86::VMOVSSrr_REV; break; + } + OutMI.setOpcode(NewOpc); + } + break; + } - // TAILJMPr64, [WIN]CALL64r, [WIN]CALL64pcrel32 - These instructions have - // register inputs modeled as normal uses instead of implicit uses. As such, - // truncate off all but the first operand (the callee). FIXME: Change isel. + // TAILJMPr64, CALL64r, CALL64pcrel32 - These instructions have register + // inputs modeled as normal uses instead of implicit uses. As such, truncate + // off all but the first operand (the callee). FIXME: Change isel. case X86::TAILJMPr64: + case X86::TAILJMPr64_REX: case X86::CALL64r: - case X86::CALL64pcrel32: - case X86::WINCALL64r: - case X86::WINCALL64pcrel32: { + case X86::CALL64pcrel32: { unsigned Opcode = OutMI.getOpcode(); MCOperand Saved = OutMI.getOperand(0); OutMI = MCInst(); @@ -419,7 +528,7 @@ ReSimplify: case X86::EH_RETURN: case X86::EH_RETURN64: { OutMI = MCInst(); - OutMI.setOpcode(X86::RET); + OutMI.setOpcode(getRetOpcode(AsmPrinter.getSubtarget())); break; } @@ -429,12 +538,12 @@ ReSimplify: case X86::TAILJMPd64: { unsigned Opcode; switch (OutMI.getOpcode()) { - default: assert(0 && "Invalid opcode"); + default: llvm_unreachable("Invalid opcode"); case X86::TAILJMPr: Opcode = X86::JMP32r; break; case X86::TAILJMPd: case X86::TAILJMPd64: Opcode = X86::JMP_1; break; } - + MCOperand Saved = OutMI.getOperand(0); OutMI = MCInst(); OutMI.setOpcode(Opcode); @@ -442,6 +551,24 @@ ReSimplify: break; } + case X86::DEC16r: + case X86::DEC32r: + case X86::INC16r: + case X86::INC32r: + // If we aren't in 64-bit mode we can use the 1-byte inc/dec instructions. + if (!AsmPrinter.getSubtarget().is64Bit()) { + unsigned Opcode; + switch (OutMI.getOpcode()) { + default: llvm_unreachable("Invalid opcode"); + case X86::DEC16r: Opcode = X86::DEC16r_alt; break; + case X86::DEC32r: Opcode = X86::DEC32r_alt; break; + case X86::INC16r: Opcode = X86::INC16r_alt; break; + case X86::INC32r: Opcode = X86::INC32r_alt; break; + } + OutMI.setOpcode(Opcode); + } + break; + // These are pseudo-ops for OR to help with the OR->ADD transformation. We do // this with an ugly goto in case the resultant OR uses EAX and needs the // short form. @@ -454,28 +581,54 @@ ReSimplify: case X86::ADD16ri8_DB: OutMI.setOpcode(X86::OR16ri8); goto ReSimplify; case X86::ADD32ri8_DB: OutMI.setOpcode(X86::OR32ri8); goto ReSimplify; case X86::ADD64ri8_DB: OutMI.setOpcode(X86::OR64ri8); goto ReSimplify; - - // The assembler backend wants to see branches in their small form and relax - // them to their large form. The JIT can only handle the large form because - // it does not do relaxation. For now, translate the large form to the - // small one here. - case X86::JMP_4: OutMI.setOpcode(X86::JMP_1); break; - case X86::JO_4: OutMI.setOpcode(X86::JO_1); break; - case X86::JNO_4: OutMI.setOpcode(X86::JNO_1); break; - case X86::JB_4: OutMI.setOpcode(X86::JB_1); break; - case X86::JAE_4: OutMI.setOpcode(X86::JAE_1); break; - case X86::JE_4: OutMI.setOpcode(X86::JE_1); break; - case X86::JNE_4: OutMI.setOpcode(X86::JNE_1); break; - case X86::JBE_4: OutMI.setOpcode(X86::JBE_1); break; - case X86::JA_4: OutMI.setOpcode(X86::JA_1); break; - case X86::JS_4: OutMI.setOpcode(X86::JS_1); break; - case X86::JNS_4: OutMI.setOpcode(X86::JNS_1); break; - case X86::JP_4: OutMI.setOpcode(X86::JP_1); break; - case X86::JNP_4: OutMI.setOpcode(X86::JNP_1); break; - case X86::JL_4: OutMI.setOpcode(X86::JL_1); break; - case X86::JGE_4: OutMI.setOpcode(X86::JGE_1); break; - case X86::JLE_4: OutMI.setOpcode(X86::JLE_1); break; - case X86::JG_4: OutMI.setOpcode(X86::JG_1); break; + + // Atomic load and store require a separate pseudo-inst because Acquire + // implies mayStore and Release implies mayLoad; fix these to regular MOV + // instructions here + case X86::ACQUIRE_MOV8rm: OutMI.setOpcode(X86::MOV8rm); goto ReSimplify; + case X86::ACQUIRE_MOV16rm: OutMI.setOpcode(X86::MOV16rm); goto ReSimplify; + case X86::ACQUIRE_MOV32rm: OutMI.setOpcode(X86::MOV32rm); goto ReSimplify; + case X86::ACQUIRE_MOV64rm: OutMI.setOpcode(X86::MOV64rm); goto ReSimplify; + case X86::RELEASE_MOV8mr: OutMI.setOpcode(X86::MOV8mr); goto ReSimplify; + case X86::RELEASE_MOV16mr: OutMI.setOpcode(X86::MOV16mr); goto ReSimplify; + case X86::RELEASE_MOV32mr: OutMI.setOpcode(X86::MOV32mr); goto ReSimplify; + case X86::RELEASE_MOV64mr: OutMI.setOpcode(X86::MOV64mr); goto ReSimplify; + case X86::RELEASE_MOV8mi: OutMI.setOpcode(X86::MOV8mi); goto ReSimplify; + case X86::RELEASE_MOV16mi: OutMI.setOpcode(X86::MOV16mi); goto ReSimplify; + case X86::RELEASE_MOV32mi: OutMI.setOpcode(X86::MOV32mi); goto ReSimplify; + case X86::RELEASE_MOV64mi32: OutMI.setOpcode(X86::MOV64mi32); goto ReSimplify; + case X86::RELEASE_ADD8mi: OutMI.setOpcode(X86::ADD8mi); goto ReSimplify; + case X86::RELEASE_ADD8mr: OutMI.setOpcode(X86::ADD8mr); goto ReSimplify; + case X86::RELEASE_ADD32mi: OutMI.setOpcode(X86::ADD32mi); goto ReSimplify; + case X86::RELEASE_ADD32mr: OutMI.setOpcode(X86::ADD32mr); goto ReSimplify; + case X86::RELEASE_ADD64mi32: OutMI.setOpcode(X86::ADD64mi32); goto ReSimplify; + case X86::RELEASE_ADD64mr: OutMI.setOpcode(X86::ADD64mr); goto ReSimplify; + case X86::RELEASE_AND8mi: OutMI.setOpcode(X86::AND8mi); goto ReSimplify; + case X86::RELEASE_AND8mr: OutMI.setOpcode(X86::AND8mr); goto ReSimplify; + case X86::RELEASE_AND32mi: OutMI.setOpcode(X86::AND32mi); goto ReSimplify; + case X86::RELEASE_AND32mr: OutMI.setOpcode(X86::AND32mr); goto ReSimplify; + case X86::RELEASE_AND64mi32: OutMI.setOpcode(X86::AND64mi32); goto ReSimplify; + case X86::RELEASE_AND64mr: OutMI.setOpcode(X86::AND64mr); goto ReSimplify; + case X86::RELEASE_OR8mi: OutMI.setOpcode(X86::OR8mi); goto ReSimplify; + case X86::RELEASE_OR8mr: OutMI.setOpcode(X86::OR8mr); goto ReSimplify; + case X86::RELEASE_OR32mi: OutMI.setOpcode(X86::OR32mi); goto ReSimplify; + case X86::RELEASE_OR32mr: OutMI.setOpcode(X86::OR32mr); goto ReSimplify; + case X86::RELEASE_OR64mi32: OutMI.setOpcode(X86::OR64mi32); goto ReSimplify; + case X86::RELEASE_OR64mr: OutMI.setOpcode(X86::OR64mr); goto ReSimplify; + case X86::RELEASE_XOR8mi: OutMI.setOpcode(X86::XOR8mi); goto ReSimplify; + case X86::RELEASE_XOR8mr: OutMI.setOpcode(X86::XOR8mr); goto ReSimplify; + case X86::RELEASE_XOR32mi: OutMI.setOpcode(X86::XOR32mi); goto ReSimplify; + case X86::RELEASE_XOR32mr: OutMI.setOpcode(X86::XOR32mr); goto ReSimplify; + case X86::RELEASE_XOR64mi32: OutMI.setOpcode(X86::XOR64mi32); goto ReSimplify; + case X86::RELEASE_XOR64mr: OutMI.setOpcode(X86::XOR64mr); goto ReSimplify; + case X86::RELEASE_INC8m: OutMI.setOpcode(X86::INC8m); goto ReSimplify; + case X86::RELEASE_INC16m: OutMI.setOpcode(X86::INC16m); goto ReSimplify; + case X86::RELEASE_INC32m: OutMI.setOpcode(X86::INC32m); goto ReSimplify; + case X86::RELEASE_INC64m: OutMI.setOpcode(X86::INC64m); goto ReSimplify; + case X86::RELEASE_DEC8m: OutMI.setOpcode(X86::DEC8m); goto ReSimplify; + case X86::RELEASE_DEC16m: OutMI.setOpcode(X86::DEC16m); goto ReSimplify; + case X86::RELEASE_DEC32m: OutMI.setOpcode(X86::DEC32m); goto ReSimplify; + case X86::RELEASE_DEC64m: OutMI.setOpcode(X86::DEC64m); goto ReSimplify; // We don't currently select the correct instruction form for instructions // which have a short %eax, etc. form. Handle this by custom lowering, for @@ -485,13 +638,13 @@ ReSimplify: // MOV64ao8, MOV64o8a // XCHG16ar, XCHG32ar, XCHG64ar case X86::MOV8mr_NOREX: - case X86::MOV8mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV8ao8); break; + case X86::MOV8mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV8o32a); break; case X86::MOV8rm_NOREX: - case X86::MOV8rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV8o8a); break; - case X86::MOV16mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV16ao16); break; - case X86::MOV16rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV16o16a); break; - case X86::MOV32mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV32ao32); break; - case X86::MOV32rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV32o32a); break; + case X86::MOV8rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV8ao32); break; + case X86::MOV16mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV16o32a); break; + case X86::MOV16rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV16ao32); break; + case X86::MOV32mr: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV32o32a); break; + case X86::MOV32rm: SimplifyShortMoveForm(AsmPrinter, OutMI, X86::MOV32ao32); break; case X86::ADC8ri: SimplifyShortImmForm(OutMI, X86::ADC8i8); break; case X86::ADC16ri: SimplifyShortImmForm(OutMI, X86::ADC16i16); break; @@ -529,109 +682,674 @@ ReSimplify: case X86::XOR16ri: SimplifyShortImmForm(OutMI, X86::XOR16i16); break; case X86::XOR32ri: SimplifyShortImmForm(OutMI, X86::XOR32i32); break; case X86::XOR64ri32: SimplifyShortImmForm(OutMI, X86::XOR64i32); break; + + // Try to shrink some forms of movsx. + case X86::MOVSX16rr8: + case X86::MOVSX32rr16: + case X86::MOVSX64rr32: + SimplifyMOVSX(OutMI); + break; + } +} + +void X86AsmPrinter::LowerTlsAddr(X86MCInstLower &MCInstLowering, + const MachineInstr &MI) { + + bool is64Bits = MI.getOpcode() == X86::TLS_addr64 || + MI.getOpcode() == X86::TLS_base_addr64; + + bool needsPadding = MI.getOpcode() == X86::TLS_addr64; + + MCContext &context = OutStreamer->getContext(); + + if (needsPadding) + EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX)); + + MCSymbolRefExpr::VariantKind SRVK; + switch (MI.getOpcode()) { + case X86::TLS_addr32: + case X86::TLS_addr64: + SRVK = MCSymbolRefExpr::VK_TLSGD; + break; + case X86::TLS_base_addr32: + SRVK = MCSymbolRefExpr::VK_TLSLDM; + break; + case X86::TLS_base_addr64: + SRVK = MCSymbolRefExpr::VK_TLSLD; + break; + default: + llvm_unreachable("unexpected opcode"); + } + + MCSymbol *sym = MCInstLowering.GetSymbolFromOperand(MI.getOperand(3)); + const MCSymbolRefExpr *symRef = MCSymbolRefExpr::create(sym, SRVK, context); + + MCInst LEA; + if (is64Bits) { + LEA.setOpcode(X86::LEA64r); + LEA.addOperand(MCOperand::createReg(X86::RDI)); // dest + LEA.addOperand(MCOperand::createReg(X86::RIP)); // base + LEA.addOperand(MCOperand::createImm(1)); // scale + LEA.addOperand(MCOperand::createReg(0)); // index + LEA.addOperand(MCOperand::createExpr(symRef)); // disp + LEA.addOperand(MCOperand::createReg(0)); // seg + } else if (SRVK == MCSymbolRefExpr::VK_TLSLDM) { + LEA.setOpcode(X86::LEA32r); + LEA.addOperand(MCOperand::createReg(X86::EAX)); // dest + LEA.addOperand(MCOperand::createReg(X86::EBX)); // base + LEA.addOperand(MCOperand::createImm(1)); // scale + LEA.addOperand(MCOperand::createReg(0)); // index + LEA.addOperand(MCOperand::createExpr(symRef)); // disp + LEA.addOperand(MCOperand::createReg(0)); // seg + } else { + LEA.setOpcode(X86::LEA32r); + LEA.addOperand(MCOperand::createReg(X86::EAX)); // dest + LEA.addOperand(MCOperand::createReg(0)); // base + LEA.addOperand(MCOperand::createImm(1)); // scale + LEA.addOperand(MCOperand::createReg(X86::EBX)); // index + LEA.addOperand(MCOperand::createExpr(symRef)); // disp + LEA.addOperand(MCOperand::createReg(0)); // seg + } + EmitAndCountInstruction(LEA); + + if (needsPadding) { + EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX)); + EmitAndCountInstruction(MCInstBuilder(X86::DATA16_PREFIX)); + EmitAndCountInstruction(MCInstBuilder(X86::REX64_PREFIX)); + } + + StringRef name = is64Bits ? "__tls_get_addr" : "___tls_get_addr"; + MCSymbol *tlsGetAddr = context.getOrCreateSymbol(name); + const MCSymbolRefExpr *tlsRef = + MCSymbolRefExpr::create(tlsGetAddr, + MCSymbolRefExpr::VK_PLT, + context); + + EmitAndCountInstruction(MCInstBuilder(is64Bits ? X86::CALL64pcrel32 + : X86::CALLpcrel32) + .addExpr(tlsRef)); +} + +/// \brief Emit the optimal amount of multi-byte nops on X86. +static void EmitNops(MCStreamer &OS, unsigned NumBytes, bool Is64Bit, const MCSubtargetInfo &STI) { + // This works only for 64bit. For 32bit we have to do additional checking if + // the CPU supports multi-byte nops. + assert(Is64Bit && "EmitNops only supports X86-64"); + while (NumBytes) { + unsigned Opc, BaseReg, ScaleVal, IndexReg, Displacement, SegmentReg; + Opc = IndexReg = Displacement = SegmentReg = 0; + BaseReg = X86::RAX; ScaleVal = 1; + switch (NumBytes) { + case 0: llvm_unreachable("Zero nops?"); break; + case 1: NumBytes -= 1; Opc = X86::NOOP; break; + case 2: NumBytes -= 2; Opc = X86::XCHG16ar; break; + case 3: NumBytes -= 3; Opc = X86::NOOPL; break; + case 4: NumBytes -= 4; Opc = X86::NOOPL; Displacement = 8; break; + case 5: NumBytes -= 5; Opc = X86::NOOPL; Displacement = 8; + IndexReg = X86::RAX; break; + case 6: NumBytes -= 6; Opc = X86::NOOPW; Displacement = 8; + IndexReg = X86::RAX; break; + case 7: NumBytes -= 7; Opc = X86::NOOPL; Displacement = 512; break; + case 8: NumBytes -= 8; Opc = X86::NOOPL; Displacement = 512; + IndexReg = X86::RAX; break; + case 9: NumBytes -= 9; Opc = X86::NOOPW; Displacement = 512; + IndexReg = X86::RAX; break; + default: NumBytes -= 10; Opc = X86::NOOPW; Displacement = 512; + IndexReg = X86::RAX; SegmentReg = X86::CS; break; + } + + unsigned NumPrefixes = std::min(NumBytes, 5U); + NumBytes -= NumPrefixes; + for (unsigned i = 0; i != NumPrefixes; ++i) + OS.EmitBytes("\x66"); + + switch (Opc) { + default: llvm_unreachable("Unexpected opcode"); break; + case X86::NOOP: + OS.EmitInstruction(MCInstBuilder(Opc), STI); + break; + case X86::XCHG16ar: + OS.EmitInstruction(MCInstBuilder(Opc).addReg(X86::AX), STI); + break; + case X86::NOOPL: + case X86::NOOPW: + OS.EmitInstruction(MCInstBuilder(Opc).addReg(BaseReg) + .addImm(ScaleVal).addReg(IndexReg) + .addImm(Displacement).addReg(SegmentReg), STI); + break; + } + } // while (NumBytes) +} + +void X86AsmPrinter::LowerSTATEPOINT(const MachineInstr &MI, + X86MCInstLower &MCIL) { + assert(Subtarget->is64Bit() && "Statepoint currently only supports X86-64"); + + StatepointOpers SOpers(&MI); + if (unsigned PatchBytes = SOpers.getNumPatchBytes()) { + EmitNops(*OutStreamer, PatchBytes, Subtarget->is64Bit(), + getSubtargetInfo()); + } else { + // Lower call target and choose correct opcode + const MachineOperand &CallTarget = SOpers.getCallTarget(); + MCOperand CallTargetMCOp; + unsigned CallOpcode; + switch (CallTarget.getType()) { + case MachineOperand::MO_GlobalAddress: + case MachineOperand::MO_ExternalSymbol: + CallTargetMCOp = MCIL.LowerSymbolOperand( + CallTarget, MCIL.GetSymbolFromOperand(CallTarget)); + CallOpcode = X86::CALL64pcrel32; + // Currently, we only support relative addressing with statepoints. + // Otherwise, we'll need a scratch register to hold the target + // address. You'll fail asserts during load & relocation if this + // symbol is to far away. (TODO: support non-relative addressing) + break; + case MachineOperand::MO_Immediate: + CallTargetMCOp = MCOperand::createImm(CallTarget.getImm()); + CallOpcode = X86::CALL64pcrel32; + // Currently, we only support relative addressing with statepoints. + // Otherwise, we'll need a scratch register to hold the target + // immediate. You'll fail asserts during load & relocation if this + // address is to far away. (TODO: support non-relative addressing) + break; + case MachineOperand::MO_Register: + CallTargetMCOp = MCOperand::createReg(CallTarget.getReg()); + CallOpcode = X86::CALL64r; + break; + default: + llvm_unreachable("Unsupported operand type in statepoint call target"); + break; + } + + // Emit call + MCInst CallInst; + CallInst.setOpcode(CallOpcode); + CallInst.addOperand(CallTargetMCOp); + OutStreamer->EmitInstruction(CallInst, getSubtargetInfo()); } + + // Record our statepoint node in the same section used by STACKMAP + // and PATCHPOINT + SM.recordStatepoint(MI); +} + +void X86AsmPrinter::LowerFAULTING_LOAD_OP(const MachineInstr &MI, + X86MCInstLower &MCIL) { + // FAULTING_LOAD_OP , , , + + unsigned LoadDefRegister = MI.getOperand(0).getReg(); + MCSymbol *HandlerLabel = MI.getOperand(1).getMCSymbol(); + unsigned LoadOpcode = MI.getOperand(2).getImm(); + unsigned LoadOperandsBeginIdx = 3; + + FM.recordFaultingOp(FaultMaps::FaultingLoad, HandlerLabel); + + MCInst LoadMI; + LoadMI.setOpcode(LoadOpcode); + + if (LoadDefRegister != X86::NoRegister) + LoadMI.addOperand(MCOperand::createReg(LoadDefRegister)); + + for (auto I = MI.operands_begin() + LoadOperandsBeginIdx, + E = MI.operands_end(); + I != E; ++I) + if (auto MaybeOperand = MCIL.LowerMachineOperand(&MI, *I)) + LoadMI.addOperand(MaybeOperand.getValue()); + + OutStreamer->EmitInstruction(LoadMI, getSubtargetInfo()); } +// Lower a stackmap of the form: +// , , ... +void X86AsmPrinter::LowerSTACKMAP(const MachineInstr &MI) { + SMShadowTracker.emitShadowPadding(*OutStreamer, getSubtargetInfo()); + SM.recordStackMap(MI); + unsigned NumShadowBytes = MI.getOperand(1).getImm(); + SMShadowTracker.reset(NumShadowBytes); +} + +// Lower a patchpoint of the form: +// [], , , , , , ... +void X86AsmPrinter::LowerPATCHPOINT(const MachineInstr &MI, + X86MCInstLower &MCIL) { + assert(Subtarget->is64Bit() && "Patchpoint currently only supports X86-64"); + + SMShadowTracker.emitShadowPadding(*OutStreamer, getSubtargetInfo()); + + SM.recordPatchPoint(MI); + + PatchPointOpers opers(&MI); + unsigned ScratchIdx = opers.getNextScratchIdx(); + unsigned EncodedBytes = 0; + const MachineOperand &CalleeMO = + opers.getMetaOper(PatchPointOpers::TargetPos); + + // Check for null target. If target is non-null (i.e. is non-zero or is + // symbolic) then emit a call. + if (!(CalleeMO.isImm() && !CalleeMO.getImm())) { + MCOperand CalleeMCOp; + switch (CalleeMO.getType()) { + default: + /// FIXME: Add a verifier check for bad callee types. + llvm_unreachable("Unrecognized callee operand type."); + case MachineOperand::MO_Immediate: + if (CalleeMO.getImm()) + CalleeMCOp = MCOperand::createImm(CalleeMO.getImm()); + break; + case MachineOperand::MO_ExternalSymbol: + case MachineOperand::MO_GlobalAddress: + CalleeMCOp = + MCIL.LowerSymbolOperand(CalleeMO, + MCIL.GetSymbolFromOperand(CalleeMO)); + break; + } + + // Emit MOV to materialize the target address and the CALL to target. + // This is encoded with 12-13 bytes, depending on which register is used. + unsigned ScratchReg = MI.getOperand(ScratchIdx).getReg(); + if (X86II::isX86_64ExtendedReg(ScratchReg)) + EncodedBytes = 13; + else + EncodedBytes = 12; + + EmitAndCountInstruction( + MCInstBuilder(X86::MOV64ri).addReg(ScratchReg).addOperand(CalleeMCOp)); + EmitAndCountInstruction(MCInstBuilder(X86::CALL64r).addReg(ScratchReg)); + } + + // Emit padding. + unsigned NumBytes = opers.getMetaOper(PatchPointOpers::NBytesPos).getImm(); + assert(NumBytes >= EncodedBytes && + "Patchpoint can't request size less than the length of a call."); + + EmitNops(*OutStreamer, NumBytes - EncodedBytes, Subtarget->is64Bit(), + getSubtargetInfo()); +} + +// Returns instruction preceding MBBI in MachineFunction. +// If MBBI is the first instruction of the first basic block, returns null. +static MachineBasicBlock::const_iterator +PrevCrossBBInst(MachineBasicBlock::const_iterator MBBI) { + const MachineBasicBlock *MBB = MBBI->getParent(); + while (MBBI == MBB->begin()) { + if (MBB == MBB->getParent()->begin()) + return nullptr; + MBB = MBB->getPrevNode(); + MBBI = MBB->end(); + } + return --MBBI; +} + +static const Constant *getConstantFromPool(const MachineInstr &MI, + const MachineOperand &Op) { + if (!Op.isCPI()) + return nullptr; + + ArrayRef Constants = + MI.getParent()->getParent()->getConstantPool()->getConstants(); + const MachineConstantPoolEntry &ConstantEntry = + Constants[Op.getIndex()]; + + // Bail if this is a machine constant pool entry, we won't be able to dig out + // anything useful. + if (ConstantEntry.isMachineConstantPoolEntry()) + return nullptr; + + auto *C = dyn_cast(ConstantEntry.Val.ConstVal); + assert((!C || ConstantEntry.getType() == C->getType()) && + "Expected a constant of the same type!"); + return C; +} + +static std::string getShuffleComment(const MachineOperand &DstOp, + const MachineOperand &SrcOp, + ArrayRef Mask) { + std::string Comment; + + // Compute the name for a register. This is really goofy because we have + // multiple instruction printers that could (in theory) use different + // names. Fortunately most people use the ATT style (outside of Windows) + // and they actually agree on register naming here. Ultimately, this is + // a comment, and so its OK if it isn't perfect. + auto GetRegisterName = [](unsigned RegNum) -> StringRef { + return X86ATTInstPrinter::getRegisterName(RegNum); + }; + + StringRef DstName = DstOp.isReg() ? GetRegisterName(DstOp.getReg()) : "mem"; + StringRef SrcName = SrcOp.isReg() ? GetRegisterName(SrcOp.getReg()) : "mem"; + + raw_string_ostream CS(Comment); + CS << DstName << " = "; + bool NeedComma = false; + bool InSrc = false; + for (int M : Mask) { + // Wrap up any prior entry... + if (M == SM_SentinelZero && InSrc) { + InSrc = false; + CS << "]"; + } + if (NeedComma) + CS << ","; + else + NeedComma = true; + + // Print this shuffle... + if (M == SM_SentinelZero) { + CS << "zero"; + } else { + if (!InSrc) { + InSrc = true; + CS << SrcName << "["; + } + if (M == SM_SentinelUndef) + CS << "u"; + else + CS << M; + } + } + if (InSrc) + CS << "]"; + CS.flush(); + + return Comment; +} void X86AsmPrinter::EmitInstruction(const MachineInstr *MI) { - X86MCInstLower MCInstLowering(Mang, *MF, *this); + X86MCInstLower MCInstLowering(*MF, *this); + const X86RegisterInfo *RI = MF->getSubtarget().getRegisterInfo(); + switch (MI->getOpcode()) { case TargetOpcode::DBG_VALUE: - if (isVerbose() && OutStreamer.hasRawTextSupport()) { - std::string TmpStr; - raw_string_ostream OS(TmpStr); - PrintDebugValueComment(MI, OS); - OutStreamer.EmitRawText(StringRef(OS.str())); - } - return; + llvm_unreachable("Should be handled target independently"); // Emit nothing here but a comment if we can. case X86::Int_MemBarrier: - if (OutStreamer.hasRawTextSupport()) - OutStreamer.EmitRawText(StringRef("\t#MEMBARRIER")); + OutStreamer->emitRawComment("MEMBARRIER"); return; - + case X86::EH_RETURN: case X86::EH_RETURN64: { // Lower these as normal, but add some comments. unsigned Reg = MI->getOperand(0).getReg(); - OutStreamer.AddComment(StringRef("eh_return, addr: %") + - X86ATTInstPrinter::getRegisterName(Reg)); + OutStreamer->AddComment(StringRef("eh_return, addr: %") + + X86ATTInstPrinter::getRegisterName(Reg)); break; } case X86::TAILJMPr: + case X86::TAILJMPm: case X86::TAILJMPd: + case X86::TAILJMPr64: + case X86::TAILJMPm64: case X86::TAILJMPd64: + case X86::TAILJMPr64_REX: + case X86::TAILJMPm64_REX: + case X86::TAILJMPd64_REX: // Lower these as normal, but add some comments. - OutStreamer.AddComment("TAILCALL"); + OutStreamer->AddComment("TAILCALL"); break; - + + case X86::TLS_addr32: + case X86::TLS_addr64: + case X86::TLS_base_addr32: + case X86::TLS_base_addr64: + return LowerTlsAddr(MCInstLowering, *MI); + case X86::MOVPC32r: { - MCInst TmpInst; // This is a pseudo op for a two instruction sequence with a label, which // looks like: // call "L1$pb" // "L1$pb": // popl %esi - + // Emit the call. - MCSymbol *PICBase = MCInstLowering.GetPICBaseSymbol(); - TmpInst.setOpcode(X86::CALLpcrel32); + MCSymbol *PICBase = MF->getPICBaseSymbol(); // FIXME: We would like an efficient form for this, so we don't have to do a // lot of extra uniquing. - TmpInst.addOperand(MCOperand::CreateExpr(MCSymbolRefExpr::Create(PICBase, - OutContext))); - OutStreamer.EmitInstruction(TmpInst); - + EmitAndCountInstruction(MCInstBuilder(X86::CALLpcrel32) + .addExpr(MCSymbolRefExpr::create(PICBase, OutContext))); + // Emit the label. - OutStreamer.EmitLabel(PICBase); - + OutStreamer->EmitLabel(PICBase); + // popl $reg - TmpInst.setOpcode(X86::POP32r); - TmpInst.getOperand(0) = MCOperand::CreateReg(MI->getOperand(0).getReg()); - OutStreamer.EmitInstruction(TmpInst); + EmitAndCountInstruction(MCInstBuilder(X86::POP32r) + .addReg(MI->getOperand(0).getReg())); return; } - + case X86::ADD32ri: { // Lower the MO_GOT_ABSOLUTE_ADDRESS form of ADD32ri. if (MI->getOperand(2).getTargetFlags() != X86II::MO_GOT_ABSOLUTE_ADDRESS) break; - + // Okay, we have something like: // EAX = ADD32ri EAX, MO_GOT_ABSOLUTE_ADDRESS(@MYGLOBAL) - + // For this, we want to print something like: // MYGLOBAL + (. - PICBASE) // However, we can't generate a ".", so just emit a new label here and refer // to it. - MCSymbol *DotSym = OutContext.CreateTempSymbol(); - OutStreamer.EmitLabel(DotSym); - + MCSymbol *DotSym = OutContext.createTempSymbol(); + OutStreamer->EmitLabel(DotSym); + // Now that we have emitted the label, lower the complex operand expression. MCSymbol *OpSym = MCInstLowering.GetSymbolFromOperand(MI->getOperand(2)); - - const MCExpr *DotExpr = MCSymbolRefExpr::Create(DotSym, OutContext); + + const MCExpr *DotExpr = MCSymbolRefExpr::create(DotSym, OutContext); const MCExpr *PICBase = - MCSymbolRefExpr::Create(MCInstLowering.GetPICBaseSymbol(), OutContext); - DotExpr = MCBinaryExpr::CreateSub(DotExpr, PICBase, OutContext); - - DotExpr = MCBinaryExpr::CreateAdd(MCSymbolRefExpr::Create(OpSym,OutContext), + MCSymbolRefExpr::create(MF->getPICBaseSymbol(), OutContext); + DotExpr = MCBinaryExpr::createSub(DotExpr, PICBase, OutContext); + + DotExpr = MCBinaryExpr::createAdd(MCSymbolRefExpr::create(OpSym,OutContext), DotExpr, OutContext); - - MCInst TmpInst; - TmpInst.setOpcode(X86::ADD32ri); - TmpInst.addOperand(MCOperand::CreateReg(MI->getOperand(0).getReg())); - TmpInst.addOperand(MCOperand::CreateReg(MI->getOperand(1).getReg())); - TmpInst.addOperand(MCOperand::CreateExpr(DotExpr)); - OutStreamer.EmitInstruction(TmpInst); + + EmitAndCountInstruction(MCInstBuilder(X86::ADD32ri) + .addReg(MI->getOperand(0).getReg()) + .addReg(MI->getOperand(1).getReg()) + .addExpr(DotExpr)); return; } + case TargetOpcode::STATEPOINT: + return LowerSTATEPOINT(*MI, MCInstLowering); + + case TargetOpcode::FAULTING_LOAD_OP: + return LowerFAULTING_LOAD_OP(*MI, MCInstLowering); + + case TargetOpcode::STACKMAP: + return LowerSTACKMAP(*MI); + + case TargetOpcode::PATCHPOINT: + return LowerPATCHPOINT(*MI, MCInstLowering); + + case X86::MORESTACK_RET: + EmitAndCountInstruction(MCInstBuilder(getRetOpcode(*Subtarget))); + return; + + case X86::MORESTACK_RET_RESTORE_R10: + // Return, then restore R10. + EmitAndCountInstruction(MCInstBuilder(getRetOpcode(*Subtarget))); + EmitAndCountInstruction(MCInstBuilder(X86::MOV64rr) + .addReg(X86::R10) + .addReg(X86::RAX)); + return; + + case X86::SEH_PushReg: + OutStreamer->EmitWinCFIPushReg(RI->getSEHRegNum(MI->getOperand(0).getImm())); + return; + + case X86::SEH_SaveReg: + OutStreamer->EmitWinCFISaveReg(RI->getSEHRegNum(MI->getOperand(0).getImm()), + MI->getOperand(1).getImm()); + return; + + case X86::SEH_SaveXMM: + OutStreamer->EmitWinCFISaveXMM(RI->getSEHRegNum(MI->getOperand(0).getImm()), + MI->getOperand(1).getImm()); + return; + + case X86::SEH_StackAlloc: + OutStreamer->EmitWinCFIAllocStack(MI->getOperand(0).getImm()); + return; + + case X86::SEH_SetFrame: + OutStreamer->EmitWinCFISetFrame(RI->getSEHRegNum(MI->getOperand(0).getImm()), + MI->getOperand(1).getImm()); + return; + + case X86::SEH_PushFrame: + OutStreamer->EmitWinCFIPushFrame(MI->getOperand(0).getImm()); + return; + + case X86::SEH_EndPrologue: + OutStreamer->EmitWinCFIEndProlog(); + return; + + case X86::SEH_Epilogue: { + MachineBasicBlock::const_iterator MBBI(MI); + // Check if preceded by a call and emit nop if so. + for (MBBI = PrevCrossBBInst(MBBI); MBBI; MBBI = PrevCrossBBInst(MBBI)) { + // Conservatively assume that pseudo instructions don't emit code and keep + // looking for a call. We may emit an unnecessary nop in some cases. + if (!MBBI->isPseudo()) { + if (MBBI->isCall()) + EmitAndCountInstruction(MCInstBuilder(X86::NOOP)); + break; + } + } + return; + } + + // Lower PSHUFB and VPERMILP normally but add a comment if we can find + // a constant shuffle mask. We won't be able to do this at the MC layer + // because the mask isn't an immediate. + case X86::PSHUFBrm: + case X86::VPSHUFBrm: + case X86::VPSHUFBYrm: { + if (!OutStreamer->isVerboseAsm()) + break; + assert(MI->getNumOperands() > 5 && + "We should always have at least 5 operands!"); + const MachineOperand &DstOp = MI->getOperand(0); + const MachineOperand &SrcOp = MI->getOperand(1); + const MachineOperand &MaskOp = MI->getOperand(5); + + if (auto *C = getConstantFromPool(*MI, MaskOp)) { + SmallVector Mask; + DecodePSHUFBMask(C, Mask); + if (!Mask.empty()) + OutStreamer->AddComment(getShuffleComment(DstOp, SrcOp, Mask)); + } + break; } - + case X86::VPERMILPSrm: + case X86::VPERMILPDrm: + case X86::VPERMILPSYrm: + case X86::VPERMILPDYrm: { + if (!OutStreamer->isVerboseAsm()) + break; + assert(MI->getNumOperands() > 5 && + "We should always have at least 5 operands!"); + const MachineOperand &DstOp = MI->getOperand(0); + const MachineOperand &SrcOp = MI->getOperand(1); + const MachineOperand &MaskOp = MI->getOperand(5); + + if (auto *C = getConstantFromPool(*MI, MaskOp)) { + SmallVector Mask; + DecodeVPERMILPMask(C, Mask); + if (!Mask.empty()) + OutStreamer->AddComment(getShuffleComment(DstOp, SrcOp, Mask)); + } + break; + } + + // For loads from a constant pool to a vector register, print the constant + // loaded. + case X86::MOVAPDrm: + case X86::VMOVAPDrm: + case X86::VMOVAPDYrm: + case X86::MOVUPDrm: + case X86::VMOVUPDrm: + case X86::VMOVUPDYrm: + case X86::MOVAPSrm: + case X86::VMOVAPSrm: + case X86::VMOVAPSYrm: + case X86::MOVUPSrm: + case X86::VMOVUPSrm: + case X86::VMOVUPSYrm: + case X86::MOVDQArm: + case X86::VMOVDQArm: + case X86::VMOVDQAYrm: + case X86::MOVDQUrm: + case X86::VMOVDQUrm: + case X86::VMOVDQUYrm: + if (!OutStreamer->isVerboseAsm()) + break; + if (MI->getNumOperands() > 4) + if (auto *C = getConstantFromPool(*MI, MI->getOperand(4))) { + std::string Comment; + raw_string_ostream CS(Comment); + const MachineOperand &DstOp = MI->getOperand(0); + CS << X86ATTInstPrinter::getRegisterName(DstOp.getReg()) << " = "; + if (auto *CDS = dyn_cast(C)) { + CS << "["; + for (int i = 0, NumElements = CDS->getNumElements(); i < NumElements; ++i) { + if (i != 0) + CS << ","; + if (CDS->getElementType()->isIntegerTy()) + CS << CDS->getElementAsInteger(i); + else if (CDS->getElementType()->isFloatTy()) + CS << CDS->getElementAsFloat(i); + else if (CDS->getElementType()->isDoubleTy()) + CS << CDS->getElementAsDouble(i); + else + CS << "?"; + } + CS << "]"; + OutStreamer->AddComment(CS.str()); + } else if (auto *CV = dyn_cast(C)) { + CS << "<"; + for (int i = 0, NumOperands = CV->getNumOperands(); i < NumOperands; ++i) { + if (i != 0) + CS << ","; + Constant *COp = CV->getOperand(i); + if (isa(COp)) { + CS << "u"; + } else if (auto *CI = dyn_cast(COp)) { + CS << CI->getZExtValue(); + } else if (auto *CF = dyn_cast(COp)) { + SmallString<32> Str; + CF->getValueAPF().toString(Str); + CS << Str; + } else { + CS << "?"; + } + } + CS << ">"; + OutStreamer->AddComment(CS.str()); + } + } + break; + } + MCInst TmpInst; MCInstLowering.Lower(MI, TmpInst); - OutStreamer.EmitInstruction(TmpInst); -} + // Stackmap shadows cannot include branch targets, so we can count the bytes + // in a call towards the shadow, but must ensure that the no thread returns + // in to the stackmap shadow. The only way to achieve this is if the call + // is at the end of the shadow. + if (MI->isCall()) { + // Count then size of the call towards the shadow + SMShadowTracker.count(TmpInst, getSubtargetInfo()); + // Then flush the shadow so that we fill with nops before the call, not + // after it. + SMShadowTracker.emitShadowPadding(*OutStreamer, getSubtargetInfo()); + // Then emit the call + OutStreamer->EmitInstruction(TmpInst, getSubtargetInfo()); + return; + } + + EmitAndCountInstruction(TmpInst); +}