blob: 18114b10d5686ad9545458f55cda8d0ab3d5b0fa [file] [log] [blame]
//===- AsmParser.cpp - Parser for Assembly Files --------------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This class implements the parser for assembly files.
//
//===----------------------------------------------------------------------===//
#include "llvm/ADT/APFloat.h"
#include "llvm/ADT/APInt.h"
#include "llvm/ADT/ArrayRef.h"
#include "llvm/ADT/None.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/StringMap.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/StringSwitch.h"
#include "llvm/ADT/Twine.h"
#include "llvm/BinaryFormat/Dwarf.h"
#include "llvm/DebugInfo/CodeView/SymbolRecord.h"
#include "llvm/MC/MCAsmInfo.h"
#include "llvm/MC/MCCodeView.h"
#include "llvm/MC/MCContext.h"
#include "llvm/MC/MCDirectives.h"
#include "llvm/MC/MCDwarf.h"
#include "llvm/MC/MCExpr.h"
#include "llvm/MC/MCInstPrinter.h"
#include "llvm/MC/MCInstrDesc.h"
#include "llvm/MC/MCInstrInfo.h"
#include "llvm/MC/MCObjectFileInfo.h"
#include "llvm/MC/MCParser/AsmCond.h"
#include "llvm/MC/MCParser/AsmLexer.h"
#include "llvm/MC/MCParser/MCAsmLexer.h"
#include "llvm/MC/MCParser/MCAsmParser.h"
#include "llvm/MC/MCParser/MCAsmParserExtension.h"
#include "llvm/MC/MCParser/MCAsmParserUtils.h"
#include "llvm/MC/MCParser/MCParsedAsmOperand.h"
#include "llvm/MC/MCParser/MCTargetAsmParser.h"
#include "llvm/MC/MCRegisterInfo.h"
#include "llvm/MC/MCSection.h"
#include "llvm/MC/MCStreamer.h"
#include "llvm/MC/MCSymbol.h"
#include "llvm/MC/MCTargetOptions.h"
#include "llvm/MC/MCValue.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/Format.h"
#include "llvm/Support/MD5.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/SMLoc.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <cassert>
#include <cctype>
#include <climits>
#include <cstddef>
#include <cstdint>
#include <deque>
#include <memory>
#include <sstream>
#include <string>
#include <tuple>
#include <utility>
#include <vector>
using namespace llvm;
extern cl::opt<unsigned> AsmMacroMaxNestingDepth;
namespace {
/// Helper types for tracking macro definitions.
typedef std::vector<AsmToken> MCAsmMacroArgument;
typedef std::vector<MCAsmMacroArgument> MCAsmMacroArguments;
/// Helper class for storing information about an active macro instantiation.
struct MacroInstantiation {
/// The location of the instantiation.
SMLoc InstantiationLoc;
/// The buffer where parsing should resume upon instantiation completion.
unsigned ExitBuffer;
/// The location where parsing should resume upon instantiation completion.
SMLoc ExitLoc;
/// The depth of TheCondStack at the start of the instantiation.
size_t CondStackDepth;
};
struct ParseStatementInfo {
/// The parsed operands from the last parsed statement.
SmallVector<std::unique_ptr<MCParsedAsmOperand>, 8> ParsedOperands;
/// The opcode from the last parsed instruction.
unsigned Opcode = ~0U;
/// Was there an error parsing the inline assembly?
bool ParseError = false;
/// The value associated with a macro exit.
Optional<std::string> ExitValue;
SmallVectorImpl<AsmRewrite> *AsmRewrites = nullptr;
ParseStatementInfo() = delete;
ParseStatementInfo(SmallVectorImpl<AsmRewrite> *rewrites)
: AsmRewrites(rewrites) {}
};
enum FieldType {
FT_INTEGRAL, // Initializer: integer expression, stored as an MCExpr.
FT_REAL, // Initializer: real number, stored as an APInt.
FT_STRUCT // Initializer: struct initializer, stored recursively.
};
struct FieldInfo;
struct StructInfo {
StringRef Name;
bool IsUnion = false;
unsigned Alignment = 0;
unsigned Size = 0;
unsigned AlignmentSize = 0;
std::vector<FieldInfo> Fields;
StringMap<size_t> FieldsByName;
FieldInfo &addField(StringRef FieldName, FieldType FT,
unsigned FieldAlignmentSize);
StructInfo() = default;
StructInfo(StringRef StructName, bool Union, unsigned AlignmentValue)
: Name(StructName), IsUnion(Union), Alignment(AlignmentValue) {}
};
// FIXME: This should probably use a class hierarchy, raw pointers between the
// objects, and dynamic type resolution instead of a union. On the other hand,
// ownership then becomes much more complicated; the obvious thing would be to
// use BumpPtrAllocator, but the lack of a destructor makes that messy.
struct StructInitializer;
struct IntFieldInfo {
SmallVector<const MCExpr *, 1> Values;
IntFieldInfo() = default;
IntFieldInfo(const SmallVector<const MCExpr *, 1> &V) { Values = V; }
IntFieldInfo(SmallVector<const MCExpr *, 1> &&V) { Values = V; }
};
struct RealFieldInfo {
SmallVector<APInt, 1> AsIntValues;
RealFieldInfo() = default;
RealFieldInfo(const SmallVector<APInt, 1> &V) { AsIntValues = V; }
RealFieldInfo(SmallVector<APInt, 1> &&V) { AsIntValues = V; }
};
struct StructFieldInfo {
std::vector<StructInitializer> Initializers;
StructInfo Structure;
StructFieldInfo() = default;
StructFieldInfo(const std::vector<StructInitializer> &V, StructInfo S) {
Initializers = V;
Structure = S;
}
StructFieldInfo(std::vector<StructInitializer> &&V, StructInfo S) {
Initializers = V;
Structure = S;
}
};
class FieldInitializer {
public:
FieldType FT;
union {
IntFieldInfo IntInfo;
RealFieldInfo RealInfo;
StructFieldInfo StructInfo;
};
~FieldInitializer() {
switch (FT) {
case FT_INTEGRAL:
IntInfo.~IntFieldInfo();
break;
case FT_REAL:
RealInfo.~RealFieldInfo();
break;
case FT_STRUCT:
StructInfo.~StructFieldInfo();
break;
}
}
FieldInitializer(FieldType FT) : FT(FT) {
switch (FT) {
case FT_INTEGRAL:
new (&IntInfo) IntFieldInfo();
break;
case FT_REAL:
new (&RealInfo) RealFieldInfo();
break;
case FT_STRUCT:
new (&StructInfo) StructFieldInfo();
break;
}
}
FieldInitializer(SmallVector<const MCExpr *, 1> &&Values) : FT(FT_INTEGRAL) {
new (&IntInfo) IntFieldInfo(Values);
}
FieldInitializer(SmallVector<APInt, 1> &&AsIntValues) : FT(FT_REAL) {
new (&RealInfo) RealFieldInfo(AsIntValues);
}
FieldInitializer(std::vector<StructInitializer> &&Initializers,
struct StructInfo Structure)
: FT(FT_STRUCT) {
new (&StructInfo) StructFieldInfo(Initializers, Structure);
}
FieldInitializer(const FieldInitializer &Initializer) : FT(Initializer.FT) {
switch (FT) {
case FT_INTEGRAL:
new (&IntInfo) IntFieldInfo(Initializer.IntInfo);
break;
case FT_REAL:
new (&RealInfo) RealFieldInfo(Initializer.RealInfo);
break;
case FT_STRUCT:
new (&StructInfo) StructFieldInfo(Initializer.StructInfo);
break;
}
}
FieldInitializer(FieldInitializer &&Initializer) : FT(Initializer.FT) {
switch (FT) {
case FT_INTEGRAL:
new (&IntInfo) IntFieldInfo(Initializer.IntInfo);
break;
case FT_REAL:
new (&RealInfo) RealFieldInfo(Initializer.RealInfo);
break;
case FT_STRUCT:
new (&StructInfo) StructFieldInfo(Initializer.StructInfo);
break;
}
}
FieldInitializer &operator=(const FieldInitializer &Initializer) {
if (FT != Initializer.FT) {
switch (FT) {
case FT_INTEGRAL:
IntInfo.~IntFieldInfo();
break;
case FT_REAL:
RealInfo.~RealFieldInfo();
break;
case FT_STRUCT:
StructInfo.~StructFieldInfo();
break;
}
}
FT = Initializer.FT;
switch (FT) {
case FT_INTEGRAL:
IntInfo = Initializer.IntInfo;
break;
case FT_REAL:
RealInfo = Initializer.RealInfo;
break;
case FT_STRUCT:
StructInfo = Initializer.StructInfo;
break;
}
return *this;
}
FieldInitializer &operator=(FieldInitializer &&Initializer) {
if (FT != Initializer.FT) {
switch (FT) {
case FT_INTEGRAL:
IntInfo.~IntFieldInfo();
break;
case FT_REAL:
RealInfo.~RealFieldInfo();
break;
case FT_STRUCT:
StructInfo.~StructFieldInfo();
break;
}
}
FT = Initializer.FT;
switch (FT) {
case FT_INTEGRAL:
IntInfo = Initializer.IntInfo;
break;
case FT_REAL:
RealInfo = Initializer.RealInfo;
break;
case FT_STRUCT:
StructInfo = Initializer.StructInfo;
break;
}
return *this;
}
};
struct StructInitializer {
std::vector<FieldInitializer> FieldInitializers;
};
struct FieldInfo {
// Offset of the field within the containing STRUCT.
size_t Offset = 0;
// Total size of the field (= LengthOf * Type).
unsigned SizeOf = 0;
// Number of elements in the field (1 if scalar, >1 if an array).
unsigned LengthOf = 0;
// Size of a single entry in this field, in bytes ("type" in MASM standards).
unsigned Type = 0;
FieldInitializer Contents;
FieldInfo(FieldType FT) : Contents(FT) {}
};
FieldInfo &StructInfo::addField(StringRef FieldName, FieldType FT,
unsigned FieldAlignmentSize) {
if (!FieldName.empty())
FieldsByName[FieldName.lower()] = Fields.size();
Fields.emplace_back(FT);
FieldInfo &Field = Fields.back();
if (IsUnion) {
Field.Offset = 0;
} else {
Size = llvm::alignTo(Size, std::min(Alignment, FieldAlignmentSize));
Field.Offset = Size;
}
AlignmentSize = std::max(AlignmentSize, FieldAlignmentSize);
return Field;
}
/// The concrete assembly parser instance.
// Note that this is a full MCAsmParser, not an MCAsmParserExtension!
// It's a peer of AsmParser, not of COFFAsmParser, WasmAsmParser, etc.
class MasmParser : public MCAsmParser {
private:
AsmLexer Lexer;
MCContext &Ctx;
MCStreamer &Out;
const MCAsmInfo &MAI;
SourceMgr &SrcMgr;
SourceMgr::DiagHandlerTy SavedDiagHandler;
void *SavedDiagContext;
std::unique_ptr<MCAsmParserExtension> PlatformParser;
/// This is the current buffer index we're lexing from as managed by the
/// SourceMgr object.
unsigned CurBuffer;
std::vector<bool> EndStatementAtEOFStack;
AsmCond TheCondState;
std::vector<AsmCond> TheCondStack;
/// maps directive names to handler methods in parser
/// extensions. Extensions register themselves in this map by calling
/// addDirectiveHandler.
StringMap<ExtensionDirectiveHandler> ExtensionDirectiveMap;
/// maps assembly-time variable names to variables.
struct Variable {
StringRef Name;
bool Redefinable = true;
bool IsText = false;
int64_t NumericValue = 0;
std::string TextValue;
};
StringMap<Variable> Variables;
/// Stack of active struct definitions.
SmallVector<StructInfo, 1> StructInProgress;
/// Maps struct tags to struct definitions.
StringMap<StructInfo> Structs;
/// Maps data location names to types.
StringMap<AsmTypeInfo> KnownType;
/// Stack of active macro instantiations.
std::vector<MacroInstantiation*> ActiveMacros;
/// List of bodies of anonymous macros.
std::deque<MCAsmMacro> MacroLikeBodies;
/// Keeps track of how many .macro's have been instantiated.
unsigned NumOfMacroInstantiations;
/// The values from the last parsed cpp hash file line comment if any.
struct CppHashInfoTy {
StringRef Filename;
int64_t LineNumber;
SMLoc Loc;
unsigned Buf;
CppHashInfoTy() : Filename(), LineNumber(0), Loc(), Buf(0) {}
};
CppHashInfoTy CppHashInfo;
/// The filename from the first cpp hash file line comment, if any.
StringRef FirstCppHashFilename;
/// List of forward directional labels for diagnosis at the end.
SmallVector<std::tuple<SMLoc, CppHashInfoTy, MCSymbol *>, 4> DirLabels;
/// AssemblerDialect. ~OU means unset value and use value provided by MAI.
/// Defaults to 1U, meaning Intel.
unsigned AssemblerDialect = 1U;
/// is Darwin compatibility enabled?
bool IsDarwin = false;
/// Are we parsing ms-style inline assembly?
bool ParsingMSInlineAsm = false;
/// Did we already inform the user about inconsistent MD5 usage?
bool ReportedInconsistentMD5 = false;
// Current <...> expression depth.
unsigned AngleBracketDepth = 0U;
// Number of locals defined.
uint16_t LocalCounter = 0;
public:
MasmParser(SourceMgr &SM, MCContext &Ctx, MCStreamer &Out,
const MCAsmInfo &MAI, unsigned CB);
MasmParser(const MasmParser &) = delete;
MasmParser &operator=(const MasmParser &) = delete;
~MasmParser() override;
bool Run(bool NoInitialTextSection, bool NoFinalize = false) override;
void addDirectiveHandler(StringRef Directive,
ExtensionDirectiveHandler Handler) override {
ExtensionDirectiveMap[Directive] = Handler;
if (DirectiveKindMap.find(Directive) == DirectiveKindMap.end()) {
DirectiveKindMap[Directive] = DK_HANDLER_DIRECTIVE;
}
}
void addAliasForDirective(StringRef Directive, StringRef Alias) override {
DirectiveKindMap[Directive] = DirectiveKindMap[Alias];
}
/// @name MCAsmParser Interface
/// {
SourceMgr &getSourceManager() override { return SrcMgr; }
MCAsmLexer &getLexer() override { return Lexer; }
MCContext &getContext() override { return Ctx; }
MCStreamer &getStreamer() override { return Out; }
CodeViewContext &getCVContext() { return Ctx.getCVContext(); }
unsigned getAssemblerDialect() override {
if (AssemblerDialect == ~0U)
return MAI.getAssemblerDialect();
else
return AssemblerDialect;
}
void setAssemblerDialect(unsigned i) override {
AssemblerDialect = i;
}
void Note(SMLoc L, const Twine &Msg, SMRange Range = None) override;
bool Warning(SMLoc L, const Twine &Msg, SMRange Range = None) override;
bool printError(SMLoc L, const Twine &Msg, SMRange Range = None) override;
const AsmToken &Lex() override;
void setParsingMSInlineAsm(bool V) override {
ParsingMSInlineAsm = V;
// When parsing MS inline asm, we must lex 0b1101 and 0ABCH as binary and
// hex integer literals.
Lexer.setLexMasmIntegers(V);
}
bool isParsingMSInlineAsm() override { return ParsingMSInlineAsm; }
bool isParsingMasm() const override { return true; }
bool defineMacro(StringRef Name, StringRef Value) override;
bool lookUpField(StringRef Name, AsmFieldInfo &Info) const override;
bool lookUpField(StringRef Base, StringRef Member,
AsmFieldInfo &Info) const override;
bool lookUpType(StringRef Name, AsmTypeInfo &Info) const override;
bool parseMSInlineAsm(void *AsmLoc, std::string &AsmString,
unsigned &NumOutputs, unsigned &NumInputs,
SmallVectorImpl<std::pair<void *,bool>> &OpDecls,
SmallVectorImpl<std::string> &Constraints,
SmallVectorImpl<std::string> &Clobbers,
const MCInstrInfo *MII, const MCInstPrinter *IP,
MCAsmParserSemaCallback &SI) override;
bool parseExpression(const MCExpr *&Res);
bool parseExpression(const MCExpr *&Res, SMLoc &EndLoc) override;
bool parsePrimaryExpr(const MCExpr *&Res, SMLoc &EndLoc,
AsmTypeInfo *TypeInfo) override;
bool parseParenExpression(const MCExpr *&Res, SMLoc &EndLoc) override;
bool parseParenExprOfDepth(unsigned ParenDepth, const MCExpr *&Res,
SMLoc &EndLoc) override;
bool parseAbsoluteExpression(int64_t &Res) override;
/// Parse a floating point expression using the float \p Semantics
/// and set \p Res to the value.
bool parseRealValue(const fltSemantics &Semantics, APInt &Res);
/// Parse an identifier or string (as a quoted identifier)
/// and set \p Res to the identifier contents.
bool parseIdentifier(StringRef &Res) override;
void eatToEndOfStatement() override;
bool checkForValidSection() override;
/// }
private:
bool parseStatement(ParseStatementInfo &Info,
MCAsmParserSemaCallback *SI);
bool parseCurlyBlockScope(SmallVectorImpl<AsmRewrite>& AsmStrRewrites);
bool parseCppHashLineFilenameComment(SMLoc L);
bool expandMacro(raw_svector_ostream &OS, StringRef Body,
ArrayRef<MCAsmMacroParameter> Parameters,
ArrayRef<MCAsmMacroArgument> A,
const std::vector<std::string> &Locals, SMLoc L);
/// Are we inside a macro instantiation?
bool isInsideMacroInstantiation() {return !ActiveMacros.empty();}
/// Handle entry to macro instantiation.
///
/// \param M The macro.
/// \param NameLoc Instantiation location.
bool handleMacroEntry(
const MCAsmMacro *M, SMLoc NameLoc,
AsmToken::TokenKind ArgumentEndTok = AsmToken::EndOfStatement);
/// Handle invocation of macro function.
///
/// \param M The macro.
/// \param NameLoc Invocation location.
bool handleMacroInvocation(const MCAsmMacro *M, SMLoc NameLoc);
/// Handle exit from macro instantiation.
void handleMacroExit();
/// Extract AsmTokens for a macro argument.
bool
parseMacroArgument(const MCAsmMacroParameter *MP, MCAsmMacroArgument &MA,
AsmToken::TokenKind EndTok = AsmToken::EndOfStatement);
/// Parse all macro arguments for a given macro.
bool
parseMacroArguments(const MCAsmMacro *M, MCAsmMacroArguments &A,
AsmToken::TokenKind EndTok = AsmToken::EndOfStatement);
void printMacroInstantiations();
bool expandStatement(SMLoc Loc);
void printMessage(SMLoc Loc, SourceMgr::DiagKind Kind, const Twine &Msg,
SMRange Range = None) const {
ArrayRef<SMRange> Ranges(Range);
SrcMgr.PrintMessage(Loc, Kind, Msg, Ranges);
}
static void DiagHandler(const SMDiagnostic &Diag, void *Context);
bool lookUpField(const StructInfo &Structure, StringRef Member,
AsmFieldInfo &Info) const;
/// Should we emit DWARF describing this assembler source? (Returns false if
/// the source has .file directives, which means we don't want to generate
/// info describing the assembler source itself.)
bool enabledGenDwarfForAssembly();
/// Enter the specified file. This returns true on failure.
bool enterIncludeFile(const std::string &Filename);
/// Reset the current lexer position to that given by \p Loc. The
/// current token is not set; clients should ensure Lex() is called
/// subsequently.
///
/// \param InBuffer If not 0, should be the known buffer id that contains the
/// location.
void jumpToLoc(SMLoc Loc, unsigned InBuffer = 0,
bool EndStatementAtEOF = true);
/// Parse up to a token of kind \p EndTok and return the contents from the
/// current token up to (but not including) this token; the current token on
/// exit will be either this kind or EOF. Reads through instantiated macro
/// functions and text macros.
SmallVector<StringRef, 1> parseStringRefsTo(AsmToken::TokenKind EndTok);
std::string parseStringTo(AsmToken::TokenKind EndTok);
/// Parse up to the end of statement and return the contents from the current
/// token until the end of the statement; the current token on exit will be
/// either the EndOfStatement or EOF.
StringRef parseStringToEndOfStatement() override;
bool parseTextItem(std::string &Data);
unsigned getBinOpPrecedence(AsmToken::TokenKind K,
MCBinaryExpr::Opcode &Kind);
bool parseBinOpRHS(unsigned Precedence, const MCExpr *&Res, SMLoc &EndLoc);
bool parseParenExpr(const MCExpr *&Res, SMLoc &EndLoc);
bool parseBracketExpr(const MCExpr *&Res, SMLoc &EndLoc);
bool parseRegisterOrRegisterNumber(int64_t &Register, SMLoc DirectiveLoc);
bool parseCVFunctionId(int64_t &FunctionId, StringRef DirectiveName);
bool parseCVFileId(int64_t &FileId, StringRef DirectiveName);
// Generic (target and platform independent) directive parsing.
enum DirectiveKind {
DK_NO_DIRECTIVE, // Placeholder
DK_HANDLER_DIRECTIVE,
DK_ASSIGN,
DK_EQU,
DK_TEXTEQU,
DK_ASCII,
DK_ASCIZ,
DK_STRING,
DK_BYTE,
DK_SBYTE,
DK_WORD,
DK_SWORD,
DK_DWORD,
DK_SDWORD,
DK_FWORD,
DK_QWORD,
DK_SQWORD,
DK_DB,
DK_DD,
DK_DF,
DK_DQ,
DK_DW,
DK_REAL4,
DK_REAL8,
DK_REAL10,
DK_ALIGN,
DK_ORG,
DK_ENDR,
DK_EXTERN,
DK_PUBLIC,
DK_COMM,
DK_COMMENT,
DK_INCLUDE,
DK_REPEAT,
DK_WHILE,
DK_FOR,
DK_FORC,
DK_IF,
DK_IFE,
DK_IFB,
DK_IFNB,
DK_IFDEF,
DK_IFNDEF,
DK_IFDIF,
DK_IFDIFI,
DK_IFIDN,
DK_IFIDNI,
DK_ELSEIF,
DK_ELSEIFE,
DK_ELSEIFB,
DK_ELSEIFNB,
DK_ELSEIFDEF,
DK_ELSEIFNDEF,
DK_ELSEIFDIF,
DK_ELSEIFDIFI,
DK_ELSEIFIDN,
DK_ELSEIFIDNI,
DK_ELSE,
DK_ENDIF,
DK_FILE,
DK_LINE,
DK_LOC,
DK_STABS,
DK_CV_FILE,
DK_CV_FUNC_ID,
DK_CV_INLINE_SITE_ID,
DK_CV_LOC,
DK_CV_LINETABLE,
DK_CV_INLINE_LINETABLE,
DK_CV_DEF_RANGE,
DK_CV_STRINGTABLE,
DK_CV_STRING,
DK_CV_FILECHECKSUMS,
DK_CV_FILECHECKSUM_OFFSET,
DK_CV_FPO_DATA,
DK_CFI_SECTIONS,
DK_CFI_STARTPROC,
DK_CFI_ENDPROC,
DK_CFI_DEF_CFA,
DK_CFI_DEF_CFA_OFFSET,
DK_CFI_ADJUST_CFA_OFFSET,
DK_CFI_DEF_CFA_REGISTER,
DK_CFI_OFFSET,
DK_CFI_REL_OFFSET,
DK_CFI_PERSONALITY,
DK_CFI_LSDA,
DK_CFI_REMEMBER_STATE,
DK_CFI_RESTORE_STATE,
DK_CFI_SAME_VALUE,
DK_CFI_RESTORE,
DK_CFI_ESCAPE,
DK_CFI_RETURN_COLUMN,
DK_CFI_SIGNAL_FRAME,
DK_CFI_UNDEFINED,
DK_CFI_REGISTER,
DK_CFI_WINDOW_SAVE,
DK_CFI_B_KEY_FRAME,
DK_MACRO,
DK_EXITM,
DK_ENDM,
DK_PURGE,
DK_ERR,
DK_ERRB,
DK_ERRNB,
DK_ERRDEF,
DK_ERRNDEF,
DK_ERRDIF,
DK_ERRDIFI,
DK_ERRIDN,
DK_ERRIDNI,
DK_ERRE,
DK_ERRNZ,
DK_ECHO,
DK_STRUCT,
DK_UNION,
DK_ENDS,
DK_END,
DK_PUSHFRAME,
DK_PUSHREG,
DK_SAVEREG,
DK_SAVEXMM128,
DK_SETFRAME,
DK_RADIX,
};
/// Maps directive name --> DirectiveKind enum, for directives parsed by this
/// class.
StringMap<DirectiveKind> DirectiveKindMap;
bool isMacroLikeDirective();
// Codeview def_range type parsing.
enum CVDefRangeType {
CVDR_DEFRANGE = 0, // Placeholder
CVDR_DEFRANGE_REGISTER,
CVDR_DEFRANGE_FRAMEPOINTER_REL,
CVDR_DEFRANGE_SUBFIELD_REGISTER,
CVDR_DEFRANGE_REGISTER_REL
};
/// Maps Codeview def_range types --> CVDefRangeType enum, for Codeview
/// def_range types parsed by this class.
StringMap<CVDefRangeType> CVDefRangeTypeMap;
// ".ascii", ".asciz", ".string"
bool parseDirectiveAscii(StringRef IDVal, bool ZeroTerminated);
// "byte", "word", ...
bool emitIntValue(const MCExpr *Value, unsigned Size);
bool parseScalarInitializer(unsigned Size,
SmallVectorImpl<const MCExpr *> &Values,
unsigned StringPadLength = 0);
bool parseScalarInstList(
unsigned Size, SmallVectorImpl<const MCExpr *> &Values,
const AsmToken::TokenKind EndToken = AsmToken::EndOfStatement);
bool emitIntegralValues(unsigned Size, unsigned *Count = nullptr);
bool addIntegralField(StringRef Name, unsigned Size);
bool parseDirectiveValue(StringRef IDVal, unsigned Size);
bool parseDirectiveNamedValue(StringRef TypeName, unsigned Size,
StringRef Name, SMLoc NameLoc);
// "real4", "real8", "real10"
bool emitRealValues(const fltSemantics &Semantics, unsigned *Count = nullptr);
bool addRealField(StringRef Name, const fltSemantics &Semantics, size_t Size);
bool parseDirectiveRealValue(StringRef IDVal, const fltSemantics &Semantics,
size_t Size);
bool parseRealInstList(
const fltSemantics &Semantics, SmallVectorImpl<APInt> &Values,
const AsmToken::TokenKind EndToken = AsmToken::EndOfStatement);
bool parseDirectiveNamedRealValue(StringRef TypeName,
const fltSemantics &Semantics,
unsigned Size, StringRef Name,
SMLoc NameLoc);
bool parseOptionalAngleBracketOpen();
bool parseAngleBracketClose(const Twine &Msg = "expected '>'");
bool parseFieldInitializer(const FieldInfo &Field,
FieldInitializer &Initializer);
bool parseFieldInitializer(const FieldInfo &Field,
const IntFieldInfo &Contents,
FieldInitializer &Initializer);
bool parseFieldInitializer(const FieldInfo &Field,
const RealFieldInfo &Contents,
FieldInitializer &Initializer);
bool parseFieldInitializer(const FieldInfo &Field,
const StructFieldInfo &Contents,
FieldInitializer &Initializer);
bool parseStructInitializer(const StructInfo &Structure,
StructInitializer &Initializer);
bool parseStructInstList(
const StructInfo &Structure, std::vector<StructInitializer> &Initializers,
const AsmToken::TokenKind EndToken = AsmToken::EndOfStatement);
bool emitFieldValue(const FieldInfo &Field);
bool emitFieldValue(const FieldInfo &Field, const IntFieldInfo &Contents);
bool emitFieldValue(const FieldInfo &Field, const RealFieldInfo &Contents);
bool emitFieldValue(const FieldInfo &Field, const StructFieldInfo &Contents);
bool emitFieldInitializer(const FieldInfo &Field,
const FieldInitializer &Initializer);
bool emitFieldInitializer(const FieldInfo &Field,
const IntFieldInfo &Contents,
const IntFieldInfo &Initializer);
bool emitFieldInitializer(const FieldInfo &Field,
const RealFieldInfo &Contents,
const RealFieldInfo &Initializer);
bool emitFieldInitializer(const FieldInfo &Field,
const StructFieldInfo &Contents,
const StructFieldInfo &Initializer);
bool emitStructInitializer(const StructInfo &Structure,
const StructInitializer &Initializer);
// User-defined types (structs, unions):
bool emitStructValues(const StructInfo &Structure, unsigned *Count = nullptr);
bool addStructField(StringRef Name, const StructInfo &Structure);
bool parseDirectiveStructValue(const StructInfo &Structure,
StringRef Directive, SMLoc DirLoc);
bool parseDirectiveNamedStructValue(const StructInfo &Structure,
StringRef Directive, SMLoc DirLoc,
StringRef Name);
// "=", "equ", "textequ"
bool parseDirectiveEquate(StringRef IDVal, StringRef Name,
DirectiveKind DirKind);
bool parseDirectiveOrg(); // ".org"
bool parseDirectiveAlign(); // "align"
// ".file", ".line", ".loc", ".stabs"
bool parseDirectiveFile(SMLoc DirectiveLoc);
bool parseDirectiveLine();
bool parseDirectiveLoc();
bool parseDirectiveStabs();
// ".cv_file", ".cv_func_id", ".cv_inline_site_id", ".cv_loc", ".cv_linetable",
// ".cv_inline_linetable", ".cv_def_range", ".cv_string"
bool parseDirectiveCVFile();
bool parseDirectiveCVFuncId();
bool parseDirectiveCVInlineSiteId();
bool parseDirectiveCVLoc();
bool parseDirectiveCVLinetable();
bool parseDirectiveCVInlineLinetable();
bool parseDirectiveCVDefRange();
bool parseDirectiveCVString();
bool parseDirectiveCVStringTable();
bool parseDirectiveCVFileChecksums();
bool parseDirectiveCVFileChecksumOffset();
bool parseDirectiveCVFPOData();
// .cfi directives
bool parseDirectiveCFIRegister(SMLoc DirectiveLoc);
bool parseDirectiveCFIWindowSave();
bool parseDirectiveCFISections();
bool parseDirectiveCFIStartProc();
bool parseDirectiveCFIEndProc();
bool parseDirectiveCFIDefCfaOffset();
bool parseDirectiveCFIDefCfa(SMLoc DirectiveLoc);
bool parseDirectiveCFIAdjustCfaOffset();
bool parseDirectiveCFIDefCfaRegister(SMLoc DirectiveLoc);
bool parseDirectiveCFIOffset(SMLoc DirectiveLoc);
bool parseDirectiveCFIRelOffset(SMLoc DirectiveLoc);
bool parseDirectiveCFIPersonalityOrLsda(bool IsPersonality);
bool parseDirectiveCFIRememberState();
bool parseDirectiveCFIRestoreState();
bool parseDirectiveCFISameValue(SMLoc DirectiveLoc);
bool parseDirectiveCFIRestore(SMLoc DirectiveLoc);
bool parseDirectiveCFIEscape();
bool parseDirectiveCFIReturnColumn(SMLoc DirectiveLoc);
bool parseDirectiveCFISignalFrame();
bool parseDirectiveCFIUndefined(SMLoc DirectiveLoc);
// macro directives
bool parseDirectivePurgeMacro(SMLoc DirectiveLoc);
bool parseDirectiveExitMacro(SMLoc DirectiveLoc, StringRef Directive,
std::string &Value);
bool parseDirectiveEndMacro(StringRef Directive);
bool parseDirectiveMacro(StringRef Name, SMLoc NameLoc);
bool parseDirectiveStruct(StringRef Directive, DirectiveKind DirKind,
StringRef Name, SMLoc NameLoc);
bool parseDirectiveNestedStruct(StringRef Directive, DirectiveKind DirKind);
bool parseDirectiveEnds(StringRef Name, SMLoc NameLoc);
bool parseDirectiveNestedEnds();
/// Parse a directive like ".globl" which accepts a single symbol (which
/// should be a label or an external).
bool parseDirectiveSymbolAttribute(MCSymbolAttr Attr);
bool parseDirectiveComm(bool IsLocal); // ".comm" and ".lcomm"
bool parseDirectiveComment(SMLoc DirectiveLoc); // "comment"
bool parseDirectiveInclude(); // "include"
// "if" or "ife"
bool parseDirectiveIf(SMLoc DirectiveLoc, DirectiveKind DirKind);
// "ifb" or "ifnb", depending on ExpectBlank.
bool parseDirectiveIfb(SMLoc DirectiveLoc, bool ExpectBlank);
// "ifidn", "ifdif", "ifidni", or "ifdifi", depending on ExpectEqual and
// CaseInsensitive.
bool parseDirectiveIfidn(SMLoc DirectiveLoc, bool ExpectEqual,
bool CaseInsensitive);
// "ifdef" or "ifndef", depending on expect_defined
bool parseDirectiveIfdef(SMLoc DirectiveLoc, bool expect_defined);
// "elseif" or "elseife"
bool parseDirectiveElseIf(SMLoc DirectiveLoc, DirectiveKind DirKind);
// "elseifb" or "elseifnb", depending on ExpectBlank.
bool parseDirectiveElseIfb(SMLoc DirectiveLoc, bool ExpectBlank);
// ".elseifdef" or ".elseifndef", depending on expect_defined
bool parseDirectiveElseIfdef(SMLoc DirectiveLoc, bool expect_defined);
// "elseifidn", "elseifdif", "elseifidni", or "elseifdifi", depending on
// ExpectEqual and CaseInsensitive.
bool parseDirectiveElseIfidn(SMLoc DirectiveLoc, bool ExpectEqual,
bool CaseInsensitive);
bool parseDirectiveElse(SMLoc DirectiveLoc); // "else"
bool parseDirectiveEndIf(SMLoc DirectiveLoc); // "endif"
bool parseEscapedString(std::string &Data) override;
bool parseAngleBracketString(std::string &Data) override;
// Macro-like directives
MCAsmMacro *parseMacroLikeBody(SMLoc DirectiveLoc);
void instantiateMacroLikeBody(MCAsmMacro *M, SMLoc DirectiveLoc,
raw_svector_ostream &OS);
void instantiateMacroLikeBody(MCAsmMacro *M, SMLoc DirectiveLoc,
SMLoc ExitLoc, raw_svector_ostream &OS);
bool parseDirectiveRepeat(SMLoc DirectiveLoc, StringRef Directive);
bool parseDirectiveFor(SMLoc DirectiveLoc, StringRef Directive);
bool parseDirectiveForc(SMLoc DirectiveLoc, StringRef Directive);
bool parseDirectiveWhile(SMLoc DirectiveLoc);
// "_emit" or "__emit"
bool parseDirectiveMSEmit(SMLoc DirectiveLoc, ParseStatementInfo &Info,
size_t Len);
// "align"
bool parseDirectiveMSAlign(SMLoc DirectiveLoc, ParseStatementInfo &Info);
// "end"
bool parseDirectiveEnd(SMLoc DirectiveLoc);
// ".err"
bool parseDirectiveError(SMLoc DirectiveLoc);
// ".errb" or ".errnb", depending on ExpectBlank.
bool parseDirectiveErrorIfb(SMLoc DirectiveLoc, bool ExpectBlank);
// ".errdef" or ".errndef", depending on ExpectBlank.
bool parseDirectiveErrorIfdef(SMLoc DirectiveLoc, bool ExpectDefined);
// ".erridn", ".errdif", ".erridni", or ".errdifi", depending on ExpectEqual
// and CaseInsensitive.
bool parseDirectiveErrorIfidn(SMLoc DirectiveLoc, bool ExpectEqual,
bool CaseInsensitive);
// ".erre" or ".errnz", depending on ExpectZero.
bool parseDirectiveErrorIfe(SMLoc DirectiveLoc, bool ExpectZero);
// ".radix"
bool parseDirectiveRadix(SMLoc DirectiveLoc);
// "echo"
bool parseDirectiveEcho();
void initializeDirectiveKindMap();
void initializeCVDefRangeTypeMap();
};
} // end anonymous namespace
namespace llvm {
extern MCAsmParserExtension *createCOFFMasmParser();
} // end namespace llvm
enum { DEFAULT_ADDRSPACE = 0 };
MasmParser::MasmParser(SourceMgr &SM, MCContext &Ctx, MCStreamer &Out,
const MCAsmInfo &MAI, unsigned CB = 0)
: Lexer(MAI), Ctx(Ctx), Out(Out), MAI(MAI), SrcMgr(SM),
CurBuffer(CB ? CB : SM.getMainFileID()) {
HadError = false;
// Save the old handler.
SavedDiagHandler = SrcMgr.getDiagHandler();
SavedDiagContext = SrcMgr.getDiagContext();
// Set our own handler which calls the saved handler.
SrcMgr.setDiagHandler(DiagHandler, this);
Lexer.setBuffer(SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer());
EndStatementAtEOFStack.push_back(true);
// Initialize the platform / file format parser.
switch (Ctx.getObjectFileInfo()->getObjectFileType()) {
case MCObjectFileInfo::IsCOFF:
PlatformParser.reset(createCOFFMasmParser());
break;
default:
report_fatal_error("llvm-ml currently supports only COFF output.");
break;
}
initializeDirectiveKindMap();
PlatformParser->Initialize(*this);
initializeCVDefRangeTypeMap();
NumOfMacroInstantiations = 0;
}
MasmParser::~MasmParser() {
assert((HadError || ActiveMacros.empty()) &&
"Unexpected active macro instantiation!");
// Restore the saved diagnostics handler and context for use during
// finalization.
SrcMgr.setDiagHandler(SavedDiagHandler, SavedDiagContext);
}
void MasmParser::printMacroInstantiations() {
// Print the active macro instantiation stack.
for (std::vector<MacroInstantiation *>::const_reverse_iterator
it = ActiveMacros.rbegin(),
ie = ActiveMacros.rend();
it != ie; ++it)
printMessage((*it)->InstantiationLoc, SourceMgr::DK_Note,
"while in macro instantiation");
}
void MasmParser::Note(SMLoc L, const Twine &Msg, SMRange Range) {
printPendingErrors();
printMessage(L, SourceMgr::DK_Note, Msg, Range);
printMacroInstantiations();
}
bool MasmParser::Warning(SMLoc L, const Twine &Msg, SMRange Range) {
if (getTargetParser().getTargetOptions().MCNoWarn)
return false;
if (getTargetParser().getTargetOptions().MCFatalWarnings)
return Error(L, Msg, Range);
printMessage(L, SourceMgr::DK_Warning, Msg, Range);
printMacroInstantiations();
return false;
}
bool MasmParser::printError(SMLoc L, const Twine &Msg, SMRange Range) {
HadError = true;
printMessage(L, SourceMgr::DK_Error, Msg, Range);
printMacroInstantiations();
return true;
}
bool MasmParser::enterIncludeFile(const std::string &Filename) {
std::string IncludedFile;
unsigned NewBuf =
SrcMgr.AddIncludeFile(Filename, Lexer.getLoc(), IncludedFile);
if (!NewBuf)
return true;
CurBuffer = NewBuf;
Lexer.setBuffer(SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer());
EndStatementAtEOFStack.push_back(true);
return false;
}
void MasmParser::jumpToLoc(SMLoc Loc, unsigned InBuffer,
bool EndStatementAtEOF) {
CurBuffer = InBuffer ? InBuffer : SrcMgr.FindBufferContainingLoc(Loc);
Lexer.setBuffer(SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer(),
Loc.getPointer(), EndStatementAtEOF);
}
const AsmToken &MasmParser::Lex() {
if (Lexer.getTok().is(AsmToken::Error))
Error(Lexer.getErrLoc(), Lexer.getErr());
// if it's a end of statement with a comment in it
if (getTok().is(AsmToken::EndOfStatement)) {
// if this is a line comment output it.
if (!getTok().getString().empty() && getTok().getString().front() != '\n' &&
getTok().getString().front() != '\r' && MAI.preserveAsmComments())
Out.addExplicitComment(Twine(getTok().getString()));
}
const AsmToken *tok = &Lexer.Lex();
while (tok->is(AsmToken::Identifier)) {
auto it = Variables.find(tok->getIdentifier().lower());
const llvm::MCAsmMacro *M =
getContext().lookupMacro(tok->getIdentifier().lower());
if (it != Variables.end() && it->second.IsText) {
// This is a textmacro; expand it in place.
std::unique_ptr<MemoryBuffer> Instantiation =
MemoryBuffer::getMemBufferCopy(it->second.TextValue,
"<instantiation>");
// Jump to the macro instantiation and prime the lexer.
CurBuffer = SrcMgr.AddNewSourceBuffer(std::move(Instantiation),
getTok().getEndLoc());
Lexer.setBuffer(SrcMgr.getMemoryBuffer(CurBuffer)->getBuffer(), nullptr,
/*EndStatementAtEOF=*/false);
EndStatementAtEOFStack.push_back(false);
tok = &Lexer.Lex();
} else if (M && M->IsFunction && Lexer.peekTok().is(AsmToken::LParen)) {
// This is a macro function invocation; expand it in place.
const AsmToken MacroTok = *tok;
tok = &Lexer.Lex();
if (handleMacroInvocation(M, MacroTok.getLoc())) {
Lexer.UnLex(AsmToken(AsmToken::Error, MacroTok.getIdentifier()));
tok = &Lexer.Lex();
}
continue;
} else {
break;
}
}
// Parse comments here to be deferred until end of next statement.
while (tok->is(AsmToken::Comment)) {
if (MAI.preserveAsmComments())
Out.addExplicitComment(Twine(tok->getString()));
tok = &Lexer.Lex();
}
// Recognize and bypass line continuations.
while (tok->is(AsmToken::BackSlash) &&
Lexer.peekTok().is(AsmToken::EndOfStatement)) {
// Eat both the backslash and the end of statement.
Lexer.Lex();
tok = &Lexer.Lex();
}
if (tok->is(AsmToken::Eof)) {
// If this is the end of an included file, pop the parent file off the
// include stack.
SMLoc ParentIncludeLoc = SrcMgr.getParentIncludeLoc(CurBuffer);
if (ParentIncludeLoc != SMLoc()) {
EndStatementAtEOFStack.pop_back();
jumpToLoc(ParentIncludeLoc, 0, EndStatementAtEOFStack.back());
return Lex();
}
EndStatementAtEOFStack.pop_back();
assert(EndStatementAtEOFStack.empty());
}
return *tok;
}
bool MasmParser::enabledGenDwarfForAssembly() {
// Check whether the user specified -g.
if (!getContext().getGenDwarfForAssembly())
return false;
// If we haven't encountered any .file directives (which would imply that
// the assembler source was produced with debug info already) then emit one
// describing the assembler source file itself.
if (getContext().getGenDwarfFileNumber() == 0) {
// Use the first #line directive for this, if any. It's preprocessed, so
// there is no checksum, and of course no source directive.
if (!FirstCppHashFilename.empty())
getContext().setMCLineTableRootFile(/*CUID=*/0,
getContext().getCompilationDir(),
FirstCppHashFilename,
/*Cksum=*/None, /*Source=*/None);
const MCDwarfFile &RootFile =
getContext().getMCDwarfLineTable(/*CUID=*/0).getRootFile();
getContext().setGenDwarfFileNumber(getStreamer().emitDwarfFileDirective(
/*CUID=*/0, getContext().getCompilationDir(), RootFile.Name,
RootFile.Checksum, RootFile.Source));
}
return true;
}
bool MasmParser::Run(bool NoInitialTextSection, bool NoFinalize) {
// Create the initial section, if requested.
if (!NoInitialTextSection)
Out.InitSections(false);
// Prime the lexer.
Lex();
HadError = false;
AsmCond StartingCondState = TheCondState;
SmallVector<AsmRewrite, 4> AsmStrRewrites;
// If we are generating dwarf for assembly source files save the initial text
// section. (Don't use enabledGenDwarfForAssembly() here, as we aren't
// emitting any actual debug info yet and haven't had a chance to parse any
// embedded .file directives.)
if (getContext().getGenDwarfForAssembly()) {
MCSection *Sec = getStreamer().getCurrentSectionOnly();
if (!Sec->getBeginSymbol()) {
MCSymbol *SectionStartSym = getContext().createTempSymbol();
getStreamer().emitLabel(SectionStartSym);
Sec->setBeginSymbol(SectionStartSym);
}
bool InsertResult = getContext().addGenDwarfSection(Sec);
assert(InsertResult && ".text section should not have debug info yet");
(void)InsertResult;
}
getTargetParser().onBeginOfFile();
// While we have input, parse each statement.
while (Lexer.isNot(AsmToken::Eof) ||
SrcMgr.getParentIncludeLoc(CurBuffer) != SMLoc()) {
// Skip through the EOF at the end of an inclusion.
if (Lexer.is(AsmToken::Eof))
Lex();
ParseStatementInfo Info(&AsmStrRewrites);
bool Parsed = parseStatement(Info, nullptr);
// If we have a Lexer Error we are on an Error Token. Load in Lexer Error
// for printing ErrMsg via Lex() only if no (presumably better) parser error
// exists.
if (Parsed && !hasPendingError() && Lexer.getTok().is(AsmToken::Error)) {
Lex();
}
// parseStatement returned true so may need to emit an error.
printPendingErrors();
// Skipping to the next line if needed.
if (Parsed && !getLexer().isAtStartOfStatement())
eatToEndOfStatement();
}
getTargetParser().onEndOfFile();
printPendingErrors();
// All errors should have been emitted.
assert(!hasPendingError() && "unexpected error from parseStatement");
getTargetParser().flushPendingInstructions(getStreamer());
if (TheCondState.TheCond != StartingCondState.TheCond ||
TheCondState.Ignore != StartingCondState.Ignore)
printError(getTok().getLoc(), "unmatched .ifs or .elses");
// Check to see there are no empty DwarfFile slots.
const auto &LineTables = getContext().getMCDwarfLineTables();
if (!LineTables.empty()) {
unsigned Index = 0;
for (const auto &File : LineTables.begin()->second.getMCDwarfFiles()) {
if (File.Name.empty() && Index != 0)
printError(getTok().getLoc(), "unassigned file number: " +
Twine(Index) +
" for .file directives");
++Index;
}
}
// Check to see that all assembler local symbols were actually defined.
// Targets that don't do subsections via symbols may not want this, though,
// so conservatively exclude them. Only do this if we're finalizing, though,
// as otherwise we won't necessarilly have seen everything yet.
if (!NoFinalize) {
if (MAI.hasSubsectionsViaSymbols()) {
for (const auto &TableEntry : getContext().getSymbols()) {
MCSymbol *Sym = TableEntry.getValue();
// Variable symbols may not be marked as defined, so check those
// explicitly. If we know it's a variable, we have a definition for
// the purposes of this check.
if (Sym->isTemporary() && !Sym->isVariable() && !Sym->isDefined())
// FIXME: We would really like to refer back to where the symbol was
// first referenced for a source location. We need to add something
// to track that. Currently, we just point to the end of the file.
printError(getTok().getLoc(), "assembler local symbol '" +
Sym->getName() + "' not defined");
}
}
// Temporary symbols like the ones for directional jumps don't go in the
// symbol table. They also need to be diagnosed in all (final) cases.
for (std::tuple<SMLoc, CppHashInfoTy, MCSymbol *> &LocSym : DirLabels) {
if (std::get<2>(LocSym)->isUndefined()) {
// Reset the state of any "# line file" directives we've seen to the
// context as it was at the diagnostic site.
CppHashInfo = std::get<1>(LocSym);
printError(std::get<0>(LocSym), "directional label undefined");
}
}
}
// Finalize the output stream if there are no errors and if the client wants
// us to.
if (!HadError && !NoFinalize)
Out.Finish(Lexer.getLoc());
return HadError || getContext().hadError();
}
bool MasmParser::checkForValidSection() {
if (!ParsingMSInlineAsm && !getStreamer().getCurrentSectionOnly()) {
Out.InitSections(false);
return Error(getTok().getLoc(),
"expected section directive before assembly directive");
}
return false;
}
/// Throw away the rest of the line for testing purposes.
void MasmParser::eatToEndOfStatement() {
while (Lexer.isNot(AsmToken::EndOfStatement)) {
if (Lexer.is(AsmToken::Eof)) {
SMLoc ParentIncludeLoc = SrcMgr.getParentIncludeLoc(CurBuffer);
if (ParentIncludeLoc == SMLoc()) {
break;
}
EndStatementAtEOFStack.pop_back();
jumpToLoc(ParentIncludeLoc, 0, EndStatementAtEOFStack.back());
}
Lexer.Lex();
}
// Eat EOL.
if (Lexer.is(AsmToken::EndOfStatement))
Lexer.Lex();
}
SmallVector<StringRef, 1>
MasmParser::parseStringRefsTo(AsmToken::TokenKind EndTok) {
SmallVector<StringRef, 1> Refs;
const char *Start = getTok().getLoc().getPointer();
while (Lexer.isNot(EndTok)) {
if (Lexer.is(AsmToken::Eof)) {
SMLoc ParentIncludeLoc = SrcMgr.getParentIncludeLoc(CurBuffer);
if (ParentIncludeLoc == SMLoc()) {
break;
}
Refs.emplace_back(Start, getTok().getLoc().getPointer() - Start);
EndStatementAtEOFStack.pop_back();
jumpToLoc(ParentIncludeLoc, 0, EndStatementAtEOFStack.back());
Lexer.Lex();
Start = getTok().getLoc().getPointer();
} else {
Lexer.Lex();
}
}
Refs.emplace_back(Start, getTok().getLoc().getPointer() - Start);
return Refs;
}
std::string MasmParser::parseStringTo(AsmToken::TokenKind EndTok) {
SmallVector<StringRef, 1> Refs = parseStringRefsTo(EndTok);
std::string Str;
for (StringRef S : Refs) {
Str.append(S.str());
}
return Str;
}
StringRef MasmParser::parseStringToEndOfStatement() {
const char *Start = getTok().getLoc().getPointer();
while (Lexer.isNot(AsmToken::EndOfStatement) && Lexer.isNot(AsmToken::Eof))
Lexer.Lex();
const char *End = getTok().getLoc().getPointer();
return StringRef(Start, End - Start);
}
/// Parse a paren expression and return it.
/// NOTE: This assumes the leading '(' has already been consumed.
///
/// parenexpr ::= expr)
///
bool MasmParser::parseParenExpr(const MCExpr *&Res, SMLoc &EndLoc) {
if (parseExpression(Res))
return true;
if (Lexer.isNot(AsmToken::RParen))
return TokError("expected ')' in parentheses expression");
EndLoc = Lexer.getTok().getEndLoc();
Lex();
return false;
}
/// Parse a bracket expression and return it.
/// NOTE: This assumes the leading '[' has already been consumed.
///
/// bracketexpr ::= expr]
///
bool MasmParser::parseBracketExpr(const MCExpr *&Res, SMLoc &EndLoc) {
if (parseExpression(Res))
return true;
EndLoc = getTok().getEndLoc();
if (parseToken(AsmToken::RBrac, "expected ']' in brackets expression"))
return true;
return false;
}
/// Parse a primary expression and return it.
/// primaryexpr ::= (parenexpr
/// primaryexpr ::= symbol
/// primaryexpr ::= number
/// primaryexpr ::= '.'
/// primaryexpr ::= ~,+,-,'not' primaryexpr
/// primaryexpr ::= string
/// (a string is interpreted as a 64-bit number in big-endian base-256)
bool MasmParser::parsePrimaryExpr(const MCExpr *&Res, SMLoc &EndLoc,
AsmTypeInfo *TypeInfo) {
SMLoc FirstTokenLoc = getLexer().getLoc();
AsmToken::TokenKind FirstTokenKind = Lexer.getKind();
switch (FirstTokenKind) {
default:
return TokError("unknown token in expression");
// If we have an error assume that we've already handled it.
case AsmToken::Error:
return true;
case AsmToken::Exclaim:
Lex(); // Eat the operator.
if (parsePrimaryExpr(Res, EndLoc, nullptr))
return true;
Res = MCUnaryExpr::createLNot(Res, getContext(), FirstTokenLoc);
return false;
case AsmToken::Dollar:
case AsmToken::At:
case AsmToken::Identifier: {
StringRef Identifier;
if (parseIdentifier(Identifier)) {
// We may have failed but $ may be a valid token.
if (getTok().is(AsmToken::Dollar)) {
if (Lexer.getMAI().getDollarIsPC()) {
Lex();
// This is a '$' reference, which references the current PC. Emit a
// temporary label to the streamer and refer to it.
MCSymbol *Sym = Ctx.createTempSymbol();
Out.emitLabel(Sym);
Res = MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None,
getContext());
EndLoc = FirstTokenLoc;
return false;
}
return Error(FirstTokenLoc, "invalid token in expression");
}
}
// Parse named bitwise negation.
if (Identifier.equals_lower("not")) {
if (parsePrimaryExpr(Res, EndLoc, nullptr))
return true;
Res = MCUnaryExpr::createNot(Res, getContext(), FirstTokenLoc);
return false;
}
// Parse symbol variant.
std::pair<StringRef, StringRef> Split;
if (!MAI.useParensForSymbolVariant()) {
if (FirstTokenKind == AsmToken::String) {
if (Lexer.is(AsmToken::At)) {
Lex(); // eat @
SMLoc AtLoc = getLexer().getLoc();
StringRef VName;
if (parseIdentifier(VName))
return Error(AtLoc, "expected symbol variant after '@'");
Split = std::make_pair(Identifier, VName);
}
} else {
Split = Identifier.split('@');
}
} else if (Lexer.is(AsmToken::LParen)) {
Lex(); // eat '('.
StringRef VName;
parseIdentifier(VName);
// eat ')'.
if (parseToken(AsmToken::RParen,
"unexpected token in variant, expected ')'"))
return true;
Split = std::make_pair(Identifier, VName);
}
EndLoc = SMLoc::getFromPointer(Identifier.end());
// This is a symbol reference.
StringRef SymbolName = Identifier;
if (SymbolName.empty())
return Error(getLexer().getLoc(), "expected a symbol reference");
MCSymbolRefExpr::VariantKind Variant = MCSymbolRefExpr::VK_None;
// Look up the symbol variant if used.
if (!Split.second.empty()) {
Variant = MCSymbolRefExpr::getVariantKindForName(Split.second);
if (Variant != MCSymbolRefExpr::VK_Invalid) {
SymbolName = Split.first;
} else if (MAI.doesAllowAtInName() && !MAI.useParensForSymbolVariant()) {
Variant = MCSymbolRefExpr::VK_None;
} else {
return Error(SMLoc::getFromPointer(Split.second.begin()),
"invalid variant '" + Split.second + "'");
}
}
// Find the field offset if used.
AsmFieldInfo Info;
Split = SymbolName.split('.');
if (Split.second.empty()) {
} else {
SymbolName = Split.first;
if (lookUpField(SymbolName, Split.second, Info)) {
std::pair<StringRef, StringRef> BaseMember = Split.second.split('.');
StringRef Base = BaseMember.first, Member = BaseMember.second;
lookUpField(Base, Member, Info);
} else if (Structs.count(SymbolName.lower())) {
// This is actually a reference to a field offset.
Res = MCConstantExpr::create(Info.Offset, getContext());
return false;
}
}
MCSymbol *Sym = getContext().getInlineAsmLabel(SymbolName);
if (!Sym)
Sym = getContext().getOrCreateSymbol(SymbolName);
// If this is an absolute variable reference, substitute it now to preserve
// semantics in the face of reassignment.
if (Sym->isVariable()) {
auto V = Sym->getVariableValue(/*SetUsed*/ false);
bool DoInline = isa<MCConstantExpr>(V) && !Variant;
if (auto TV = dyn_cast<MCTargetExpr>(V))
DoInline = TV->inlineAssignedExpr();
if (DoInline) {
if (Variant)
return Error(EndLoc, "unexpected modifier on variable reference");
Res = Sym->getVariableValue(/*SetUsed*/ false);
return false;
}
}
// Otherwise create a symbol ref.
const MCExpr *SymRef =
MCSymbolRefExpr::create(Sym, Variant, getContext(), FirstTokenLoc);
if (Info.Offset) {
Res = MCBinaryExpr::create(
MCBinaryExpr::Add, SymRef,
MCConstantExpr::create(Info.Offset, getContext()), getContext());
} else {
Res = SymRef;
}
if (TypeInfo) {
if (Info.Type.Name.empty()) {
auto TypeIt = KnownType.find(Identifier.lower());
if (TypeIt != KnownType.end()) {
Info.Type = TypeIt->second;
}
}
*TypeInfo = Info.Type;
}
return false;
}
case AsmToken::BigNum:
return TokError("literal value out of range for directive");
case AsmToken::Integer: {
SMLoc Loc = getTok().getLoc();
int64_t IntVal = getTok().getIntVal();
Res = MCConstantExpr::create(IntVal, getContext());
EndLoc = Lexer.getTok().getEndLoc();
Lex(); // Eat token.
// Look for 'b' or 'f' following an Integer as a directional label.
if (Lexer.getKind() == AsmToken::Identifier) {
StringRef IDVal = getTok().getString();
// Look up the symbol variant if used.
std::pair<StringRef, StringRef> Split = IDVal.split('@');
MCSymbolRefExpr::VariantKind Variant = MCSymbolRefExpr::VK_None;
if (Split.first.size() != IDVal.size()) {
Variant = MCSymbolRefExpr::getVariantKindForName(Split.second);
if (Variant == MCSymbolRefExpr::VK_Invalid)
return TokError("invalid variant '" + Split.second + "'");
IDVal = Split.first;
}
if (IDVal == "f" || IDVal == "b") {
MCSymbol *Sym =
Ctx.getDirectionalLocalSymbol(IntVal, IDVal == "b");
Res = MCSymbolRefExpr::create(Sym, Variant, getContext());
if (IDVal == "b" && Sym->isUndefined())
return Error(Loc, "directional label undefined");
DirLabels.push_back(std::make_tuple(Loc, CppHashInfo, Sym));
EndLoc = Lexer.getTok().getEndLoc();
Lex(); // Eat identifier.
}
}
return false;
}
case AsmToken::String: {
// MASM strings (used as constants) are interpreted as big-endian base-256.
SMLoc ValueLoc = getTok().getLoc();
std::string Value;
if (parseEscapedString(Value))
return true;
if (Value.size() > 8)
return Error(ValueLoc, "literal value out of range");
uint64_t IntValue = 0;
for (const unsigned char CharVal : Value)
IntValue = (IntValue << 8) | CharVal;
Res = MCConstantExpr::create(IntValue, getContext());
return false;
}
case AsmToken::Real: {
APFloat RealVal(APFloat::IEEEdouble(), getTok().getString());
uint64_t IntVal = RealVal.bitcastToAPInt().getZExtValue();
Res = MCConstantExpr::create(IntVal, getContext());
EndLoc = Lexer.getTok().getEndLoc();
Lex(); // Eat token.
return false;
}
case AsmToken::Dot: {
// This is a '.' reference, which references the current PC. Emit a
// temporary label to the streamer and refer to it.
MCSymbol *Sym = Ctx.createTempSymbol();
Out.emitLabel(Sym);
Res = MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
EndLoc = Lexer.getTok().getEndLoc();
Lex(); // Eat identifier.
return false;
}
case AsmToken::LParen:
Lex(); // Eat the '('.
return parseParenExpr(Res, EndLoc);
case AsmToken::LBrac:
if (!PlatformParser->HasBracketExpressions())
return TokError("brackets expression not supported on this target");
Lex(); // Eat the '['.
return parseBracketExpr(Res, EndLoc);
case AsmToken::Minus:
Lex(); // Eat the operator.
if (parsePrimaryExpr(Res, EndLoc, nullptr))
return true;
Res = MCUnaryExpr::createMinus(Res, getContext(), FirstTokenLoc);
return false;
case AsmToken::Plus:
Lex(); // Eat the operator.
if (parsePrimaryExpr(Res, EndLoc, nullptr))
return true;
Res = MCUnaryExpr::createPlus(Res, getContext(), FirstTokenLoc);
return false;
case AsmToken::Tilde:
Lex(); // Eat the operator.
if (parsePrimaryExpr(Res, EndLoc, nullptr))
return true;
Res = MCUnaryExpr::createNot(Res, getContext(), FirstTokenLoc);
return false;
// MIPS unary expression operators. The lexer won't generate these tokens if
// MCAsmInfo::HasMipsExpressions is false for the target.
case AsmToken::PercentCall16:
case AsmToken::PercentCall_Hi:
case AsmToken::PercentCall_Lo:
case AsmToken::PercentDtprel_Hi:
case AsmToken::PercentDtprel_Lo:
case AsmToken::PercentGot:
case AsmToken::PercentGot_Disp:
case AsmToken::PercentGot_Hi:
case AsmToken::PercentGot_Lo:
case AsmToken::PercentGot_Ofst:
case AsmToken::PercentGot_Page:
case AsmToken::PercentGottprel:
case AsmToken::PercentGp_Rel:
case AsmToken::PercentHi:
case AsmToken::PercentHigher:
case AsmToken::PercentHighest:
case AsmToken::PercentLo:
case AsmToken::PercentNeg:
case AsmToken::PercentPcrel_Hi:
case AsmToken::PercentPcrel_Lo:
case AsmToken::PercentTlsgd:
case AsmToken::PercentTlsldm:
case AsmToken::PercentTprel_Hi:
case AsmToken::PercentTprel_Lo:
Lex(); // Eat the operator.
if (Lexer.isNot(AsmToken::LParen))
return TokError("expected '(' after operator");
Lex(); // Eat the operator.
if (parseExpression(Res, EndLoc))
return true;
if (Lexer.isNot(AsmToken::RParen))
return TokError("expected ')'");
Lex(); // Eat the operator.
Res = getTargetParser().createTargetUnaryExpr(Res, FirstTokenKind, Ctx);
return !Res;
}
}
bool MasmParser::parseExpression(const MCExpr *&Res) {
SMLoc EndLoc;
return parseExpression(Res, EndLoc);
}
/// This function checks if the next token is <string> type or arithmetic.
/// string that begin with character '<' must end with character '>'.
/// otherwise it is arithmetics.
/// If the function returns a 'true' value,
/// the End argument will be filled with the last location pointed to the '>'
/// character.
static bool isAngleBracketString(SMLoc &StrLoc, SMLoc &EndLoc) {
assert((StrLoc.getPointer() != nullptr) &&
"Argument to the function cannot be a NULL value");
const char *CharPtr = StrLoc.getPointer();
while ((*CharPtr != '>') && (*CharPtr != '\n') && (*CharPtr != '\r') &&
(*CharPtr != '\0')) {
if (*CharPtr == '!')
CharPtr++;
CharPtr++;
}
if (*CharPtr == '>') {
EndLoc = StrLoc.getFromPointer(CharPtr + 1);
return true;
}
return false;
}
/// creating a string without the escape characters '!'.
static std::string angleBracketString(StringRef BracketContents) {
std::string Res;
for (size_t Pos = 0; Pos < BracketContents.size(); Pos++) {
if (BracketContents[Pos] == '!')
Pos++;
Res += BracketContents[Pos];
}
return Res;
}
/// Parse an expression and return it.
///
/// expr ::= expr &&,|| expr -> lowest.
/// expr ::= expr |,^,&,! expr
/// expr ::= expr ==,!=,<>,<,<=,>,>= expr
/// expr ::= expr <<,>> expr
/// expr ::= expr +,- expr
/// expr ::= expr *,/,% expr -> highest.
/// expr ::= primaryexpr
///
bool MasmParser::parseExpression(const MCExpr *&Res, SMLoc &EndLoc) {
// Parse the expression.
Res = nullptr;
if (getTargetParser().parsePrimaryExpr(Res, EndLoc) ||
parseBinOpRHS(1, Res, EndLoc))
return true;
// Try to constant fold it up front, if possible. Do not exploit
// assembler here.
int64_t Value;
if (Res->evaluateAsAbsolute(Value))
Res = MCConstantExpr::create(Value, getContext());
return false;
}
bool MasmParser::parseParenExpression(const MCExpr *&Res, SMLoc &EndLoc) {
Res = nullptr;
return parseParenExpr(Res, EndLoc) || parseBinOpRHS(1, Res, EndLoc);
}
bool MasmParser::parseParenExprOfDepth(unsigned ParenDepth, const MCExpr *&Res,
SMLoc &EndLoc) {
if (parseParenExpr(Res, EndLoc))
return true;
for (; ParenDepth > 0; --ParenDepth) {
if (parseBinOpRHS(1, Res, EndLoc))
return true;
// We don't Lex() the last RParen.
// This is the same behavior as parseParenExpression().
if (ParenDepth - 1 > 0) {
EndLoc = getTok().getEndLoc();
if (parseToken(AsmToken::RParen,
"expected ')' in parentheses expression"))
return true;
}
}
return false;
}
bool MasmParser::parseAbsoluteExpression(int64_t &Res) {
const MCExpr *Expr;
SMLoc StartLoc = Lexer.getLoc();
if (parseExpression(Expr))
return true;
if (!Expr->evaluateAsAbsolute(Res, getStreamer().getAssemblerPtr()))
return Error(StartLoc, "expected absolute expression");
return false;
}
static unsigned getGNUBinOpPrecedence(AsmToken::TokenKind K,
MCBinaryExpr::Opcode &Kind,
bool ShouldUseLogicalShr,
bool EndExpressionAtGreater) {
switch (K) {
default:
return 0; // not a binop.
// Lowest Precedence: &&, ||
case AsmToken::AmpAmp:
Kind = MCBinaryExpr::LAnd;
return 2;
case AsmToken::PipePipe:
Kind = MCBinaryExpr::LOr;
return 1;
// Low Precedence: ==, !=, <>, <, <=, >, >=
case AsmToken::EqualEqual:
Kind = MCBinaryExpr::EQ;
return 3;
case AsmToken::ExclaimEqual:
case AsmToken::LessGreater:
Kind = MCBinaryExpr::NE;
return 3;
case AsmToken::Less:
Kind = MCBinaryExpr::LT;
return 3;
case AsmToken::LessEqual:
Kind = MCBinaryExpr::LTE;
return 3;
case AsmToken::Greater:
if (EndExpressionAtGreater)
return 0;
Kind = MCBinaryExpr::GT;
return 3;
case AsmToken::GreaterEqual:
Kind = MCBinaryExpr::GTE;
return 3;
// Low Intermediate Precedence: +, -
case AsmToken::Plus:
Kind = MCBinaryExpr::Add;
return 4;
case AsmToken::Minus:
Kind = MCBinaryExpr::Sub;
return 4;
// High Intermediate Precedence: |, &, ^
case AsmToken::Pipe:
Kind = MCBinaryExpr::Or;
return 5;
case AsmToken::Caret:
Kind = MCBinaryExpr::Xor;
return 5;
case AsmToken::Amp:
Kind = MCBinaryExpr::And;
return 5;
// Highest Precedence: *, /, %, <<, >>
case AsmToken::Star:
Kind = MCBinaryExpr::Mul;
return 6;
case AsmToken::Slash:
Kind = MCBinaryExpr::Div;
return 6;
case AsmToken::Percent:
Kind = MCBinaryExpr::Mod;
return 6;
case AsmToken::LessLess:
Kind = MCBinaryExpr::Shl;
return 6;
case AsmToken::GreaterGreater:
if (EndExpressionAtGreater)
return 0;
Kind = ShouldUseLogicalShr ? MCBinaryExpr::LShr : MCBinaryExpr::AShr;
return 6;
}
}
unsigned MasmParser::getBinOpPrecedence(AsmToken::TokenKind K,
MCBinaryExpr::Opcode &Kind) {
bool ShouldUseLogicalShr = MAI.shouldUseLogicalShr();
return getGNUBinOpPrecedence(K, Kind, ShouldUseLogicalShr,
AngleBracketDepth > 0);
}
/// Parse all binary operators with precedence >= 'Precedence'.
/// Res contains the LHS of the expression on input.
bool MasmParser::parseBinOpRHS(unsigned Precedence, const MCExpr *&Res,
SMLoc &EndLoc) {
SMLoc StartLoc = Lexer.getLoc();
while (true) {
AsmToken::TokenKind TokKind = Lexer.getKind();
if (Lexer.getKind() == AsmToken::Identifier) {
TokKind = StringSwitch<AsmToken::TokenKind>(Lexer.getTok().getString())
.CaseLower("and", AsmToken::Amp)
.CaseLower("not", AsmToken::Exclaim)
.CaseLower("or", AsmToken::Pipe)
.CaseLower("eq", AsmToken::EqualEqual)
.CaseLower("ne", AsmToken::ExclaimEqual)
.CaseLower("lt", AsmToken::Less)
.CaseLower("le", AsmToken::LessEqual)
.CaseLower("gt", AsmToken::Greater)
.CaseLower("ge", AsmToken::GreaterEqual)
.Default(TokKind);
}
MCBinaryExpr::Opcode Kind = MCBinaryExpr::Add;
unsigned TokPrec = getBinOpPrecedence(TokKind, Kind);
// If the next token is lower precedence than we are allowed to eat, return
// successfully with what we ate already.
if (TokPrec < Precedence)
return false;
Lex();
// Eat the next primary expression.
const MCExpr *RHS;
if (getTargetParser().parsePrimaryExpr(RHS, EndLoc))
return true;
// If BinOp binds less tightly with RHS than the operator after RHS, let
// the pending operator take RHS as its LHS.
MCBinaryExpr::Opcode Dummy;
unsigned NextTokPrec = getBinOpPrecedence(Lexer.getKind(), Dummy);
if (TokPrec < NextTokPrec && parseBinOpRHS(TokPrec + 1, RHS, EndLoc))
return true;
// Merge LHS and RHS according to operator.
Res = MCBinaryExpr::create(Kind, Res, RHS, getContext(), StartLoc);
}
}
/// ParseStatement:
/// ::= % statement
/// ::= EndOfStatement
/// ::= Label* Directive ...Operands... EndOfStatement
/// ::= Label* Identifier OperandList* EndOfStatement
bool MasmParser::parseStatement(ParseStatementInfo &Info,
MCAsmParserSemaCallback *SI) {
assert(!hasPendingError() && "parseStatement started with pending error");
// Eat initial spaces and comments.
while (Lexer.is(AsmToken::Space))
Lex();
if (Lexer.is(AsmToken::EndOfStatement)) {
// If this is a line comment we can drop it safely.
if (getTok().getString().empty() || getTok().getString().front() == '\r' ||
getTok().getString().front() == '\n')
Out.AddBlankLine();
Lex();
return false;
}
// If preceded by an expansion operator, first expand all text macros and
// macro functions.
if (getTok().is(AsmToken::Percent)) {
SMLoc ExpansionLoc = getTok().getLoc();
if (parseToken(AsmToken::Percent) || expandStatement(ExpansionLoc))
return true;
}
// Statements always start with an identifier, unless we're dealing with a
// processor directive (.386, .686, etc.) that lexes as a real.
AsmToken ID = getTok();
SMLoc IDLoc = ID.getLoc();
StringRef IDVal;
int64_t LocalLabelVal = -1;
if (Lexer.is(AsmToken::HashDirective))
return parseCppHashLineFilenameComment(IDLoc);
// Allow an integer followed by a ':' as a directional local label.
if (Lexer.is(AsmToken::Integer)) {
LocalLabelVal = getTok().getIntVal();
if (LocalLabelVal < 0) {
if (!TheCondState.Ignore) {
Lex(); // always eat a token
return Error(IDLoc, "unexpected token at start of statement");
}
IDVal = "";
} else {
IDVal = getTok().getString();
Lex(); // Consume the integer token to be used as an identifier token.
if (Lexer.getKind() != AsmToken::Colon) {
if (!TheCondState.Ignore) {
Lex(); // always eat a token
return Error(IDLoc, "unexpected token at start of statement");
}
}
}
} else if (Lexer.is(AsmToken::Dot)) {
// Treat '.' as a valid identifier in this context.
Lex();
IDVal = ".";
} else if (Lexer.is(AsmToken::LCurly)) {
// Treat '{' as a valid identifier in this context.
Lex();
IDVal = "{";
} else if (Lexer.is(AsmToken::RCurly)) {
// Treat '}' as a valid identifier in this context.
Lex();
IDVal = "}";
} else if (Lexer.is(AsmToken::Star) &&
getTargetParser().starIsStartOfStatement()) {
// Accept '*' as a valid start of statement.
Lex();
IDVal = "*";
} else if (Lexer.is(AsmToken::Real)) {
// Treat ".<number>" as a valid identifier in this context.
IDVal = getTok().getString();
Lex(); // always eat a token
if (!IDVal.startswith("."))
return Error(IDLoc, "unexpected token at start of statement");
} else if (Lexer.is(AsmToken::Identifier) &&
getTok().getString().equals_lower("echo")) {
// Intercept echo early to avoid lexical substitution in its message, and
// delegate all handling to the appropriate function.
return parseDirectiveEcho();
} else if (parseIdentifier(IDVal)) {
if (!TheCondState.Ignore) {
Lex(); // always eat a token
return Error(IDLoc, "unexpected token at start of statement");
}
IDVal = "";
}
// Handle conditional assembly here before checking for skipping. We
// have to do this so that .endif isn't skipped in a ".if 0" block for
// example.
StringMap<DirectiveKind>::const_iterator DirKindIt =
DirectiveKindMap.find(IDVal.lower());
DirectiveKind DirKind = (DirKindIt == DirectiveKindMap.end())
? DK_NO_DIRECTIVE
: DirKindIt->getValue();
switch (DirKind) {
default:
break;
case DK_IF:
case DK_IFE:
return parseDirectiveIf(IDLoc, DirKind);
case DK_IFB:
return parseDirectiveIfb(IDLoc, true);
case DK_IFNB:
return parseDirectiveIfb(IDLoc, false);
case DK_IFDEF:
return parseDirectiveIfdef(IDLoc, true);
case DK_IFNDEF:
return parseDirectiveIfdef(IDLoc, false);
case DK_IFDIF:
return parseDirectiveIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/false);
case DK_IFDIFI:
return parseDirectiveIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/true);
case DK_IFIDN:
return parseDirectiveIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/false);
case DK_IFIDNI:
return parseDirectiveIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/true);
case DK_ELSEIF:
case DK_ELSEIFE:
return parseDirectiveElseIf(IDLoc, DirKind);
case DK_ELSEIFB:
return parseDirectiveElseIfb(IDLoc, true);
case DK_ELSEIFNB:
return parseDirectiveElseIfb(IDLoc, false);
case DK_ELSEIFDEF:
return parseDirectiveElseIfdef(IDLoc, true);
case DK_ELSEIFNDEF:
return parseDirectiveElseIfdef(IDLoc, false);
case DK_ELSEIFDIF:
return parseDirectiveElseIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/false);
case DK_ELSEIFDIFI:
return parseDirectiveElseIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/true);
case DK_ELSEIFIDN:
return parseDirectiveElseIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/false);
case DK_ELSEIFIDNI:
return parseDirectiveElseIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/true);
case DK_ELSE:
return parseDirectiveElse(IDLoc);
case DK_ENDIF:
return parseDirectiveEndIf(IDLoc);
}
// Ignore the statement if in the middle of inactive conditional
// (e.g. ".if 0").
if (TheCondState.Ignore) {
eatToEndOfStatement();
return false;
}
// FIXME: Recurse on local labels?
// See what kind of statement we have.
switch (Lexer.getKind()) {
case AsmToken::Colon: {
if (!getTargetParser().isLabel(ID))
break;
if (checkForValidSection())
return true;
// identifier ':' -> Label.
Lex();
// Diagnose attempt to use '.' as a label.
if (IDVal == ".")
return Error(IDLoc, "invalid use of pseudo-symbol '.' as a label");
// Diagnose attempt to use a variable as a label.
//
// FIXME: Diagnostics. Note the location of the definition as a label.
// FIXME: This doesn't diagnose assignment to a symbol which has been
// implicitly marked as external.
MCSymbol *Sym;
if (LocalLabelVal == -1) {
if (ParsingMSInlineAsm && SI) {
StringRef RewrittenLabel =
SI->LookupInlineAsmLabel(IDVal, getSourceManager(), IDLoc, true);
assert(!RewrittenLabel.empty() &&
"We should have an internal name here.");
Info.AsmRewrites->emplace_back(AOK_Label, IDLoc, IDVal.size(),
RewrittenLabel);
IDVal = RewrittenLabel;
}
Sym = getContext().getOrCreateSymbol(IDVal);
} else
Sym = Ctx.createDirectionalLocalSymbol(LocalLabelVal);
// End of Labels should be treated as end of line for lexing
// purposes but that information is not available to the Lexer who
// does not understand Labels. This may cause us to see a Hash
// here instead of a preprocessor line comment.
if (getTok().is(AsmToken::Hash)) {
std::string CommentStr = parseStringTo(AsmToken::EndOfStatement);
Lexer.Lex();
Lexer.UnLex(AsmToken(AsmToken::EndOfStatement, CommentStr));
}
// Consume any end of statement token, if present, to avoid spurious
// AddBlankLine calls().
if (getTok().is(AsmToken::EndOfStatement)) {
Lex();
}
getTargetParser().doBeforeLabelEmit(Sym);
// Emit the label.
if (!getTargetParser().isParsingMSInlineAsm())
Out.emitLabel(Sym, IDLoc);
// If we are generating dwarf for assembly source files then gather the
// info to make a dwarf label entry for this label if needed.
if (enabledGenDwarfForAssembly())
MCGenDwarfLabelEntry::Make(Sym, &getStreamer(), getSourceManager(),
IDLoc);
getTargetParser().onLabelParsed(Sym);
return false;
}
default: // Normal instruction or directive.
break;
}
// If macros are enabled, check to see if this is a macro instantiation.
if (const MCAsmMacro *M = getContext().lookupMacro(IDVal.lower())) {
return handleMacroEntry(M, IDLoc);
}
// Otherwise, we have a normal instruction or directive.
if (DirKind != DK_NO_DIRECTIVE) {
// There are several entities interested in parsing directives:
//
// 1. Asm parser extensions. For example, platform-specific parsers
// (like the ELF parser) register themselves as extensions.
// 2. The target-specific assembly parser. Some directives are target
// specific or may potentially behave differently on certain targets.
// 3. The generic directive parser implemented by this class. These are
// all the directives that behave in a target and platform independent
// manner, or at least have a default behavior that's shared between
// all targets and platforms.
getTargetParser().flushPendingInstructions(getStreamer());
// Special-case handling of structure-end directives at higher priority,
// since ENDS is overloaded as a segment-end directive.
if (IDVal.equals_lower("ends") && StructInProgress.size() > 1 &&
getTok().is(AsmToken::EndOfStatement)) {
return parseDirectiveNestedEnds();
}
// First, check the extension directive map to see if any extension has
// registered itself to parse this directive.
std::pair<MCAsmParserExtension *, DirectiveHandler> Handler =
ExtensionDirectiveMap.lookup(IDVal.lower());
if (Handler.first)
return (*Handler.second)(Handler.first, IDVal, IDLoc);
// Next, let the target-specific assembly parser try.
SMLoc StartTokLoc = getTok().getLoc();
bool TPDirectiveReturn =
ID.is(AsmToken::Identifier) && getTargetParser().ParseDirective(ID);
if (hasPendingError())
return true;
// Currently the return value should be true if we are
// uninterested but as this is at odds with the standard parsing
// convention (return true = error) we have instances of a parsed
// directive that fails returning true as an error. Catch these
// cases as best as possible errors here.
if (TPDirectiveReturn && StartTokLoc != getTok().getLoc())
return true;
// Return if we did some parsing or believe we succeeded.
if (!TPDirectiveReturn || StartTokLoc != getTok().getLoc())
return false;
// Finally, if no one else is interested in this directive, it must be
// generic and familiar to this class.
switch (DirKind) {
default:
break;
case DK_ASCII:
return parseDirectiveAscii(IDVal, false);
case DK_ASCIZ:
case DK_STRING:
return parseDirectiveAscii(IDVal, true);
case DK_BYTE:
case DK_SBYTE:
case DK_DB:
return parseDirectiveValue(IDVal, 1);
case DK_WORD:
case DK_SWORD:
case DK_DW:
return parseDirectiveValue(IDVal, 2);
case DK_DWORD:
case DK_SDWORD:
case DK_DD:
return parseDirectiveValue(IDVal, 4);
case DK_FWORD:
case DK_DF:
return parseDirectiveValue(IDVal, 6);
case DK_QWORD:
case DK_SQWORD:
case DK_DQ:
return parseDirectiveValue(IDVal, 8);
case DK_REAL4:
return parseDirectiveRealValue(IDVal, APFloat::IEEEsingle(), 4);
case DK_REAL8:
return parseDirectiveRealValue(IDVal, APFloat::IEEEdouble(), 8);
case DK_REAL10:
return parseDirectiveRealValue(IDVal, APFloat::x87DoubleExtended(), 10);
case DK_STRUCT:
case DK_UNION:
return parseDirectiveNestedStruct(IDVal, DirKind);
case DK_ENDS:
return parseDirectiveNestedEnds();
case DK_ALIGN:
return parseDirectiveAlign();
case DK_ORG:
return parseDirectiveOrg();
case DK_EXTERN:
eatToEndOfStatement(); // .extern is the default, ignore it.
return false;
case DK_PUBLIC:
return parseDirectiveSymbolAttribute(MCSA_Global);
case DK_COMM:
return parseDirectiveComm(/*IsLocal=*/false);
case DK_COMMENT:
return parseDirectiveComment(IDLoc);
case DK_INCLUDE:
return parseDirectiveInclude();
case DK_REPEAT:
return parseDirectiveRepeat(IDLoc, IDVal);
case DK_WHILE:
return parseDirectiveWhile(IDLoc);
case DK_FOR:
return parseDirectiveFor(IDLoc, IDVal);
case DK_FORC:
return parseDirectiveForc(IDLoc, IDVal);
case DK_FILE:
return parseDirectiveFile(IDLoc);
case DK_LINE:
return parseDirectiveLine();
case DK_LOC:
return parseDirectiveLoc();
case DK_STABS:
return parseDirectiveStabs();
case DK_CV_FILE:
return parseDirectiveCVFile();
case DK_CV_FUNC_ID:
return parseDirectiveCVFuncId();
case DK_CV_INLINE_SITE_ID:
return parseDirectiveCVInlineSiteId();
case DK_CV_LOC:
return parseDirectiveCVLoc();
case DK_CV_LINETABLE:
return parseDirectiveCVLinetable();
case DK_CV_INLINE_LINETABLE:
return parseDirectiveCVInlineLinetable();
case DK_CV_DEF_RANGE:
return parseDirectiveCVDefRange();
case DK_CV_STRING:
return parseDirectiveCVString();
case DK_CV_STRINGTABLE:
return parseDirectiveCVStringTable();
case DK_CV_FILECHECKSUMS:
return parseDirectiveCVFileChecksums();
case DK_CV_FILECHECKSUM_OFFSET:
return parseDirectiveCVFileChecksumOffset();
case DK_CV_FPO_DATA:
return parseDirectiveCVFPOData();
case DK_CFI_SECTIONS:
return parseDirectiveCFISections();
case DK_CFI_STARTPROC:
return parseDirectiveCFIStartProc();
case DK_CFI_ENDPROC:
return parseDirectiveCFIEndProc();
case DK_CFI_DEF_CFA:
return parseDirectiveCFIDefCfa(IDLoc);
case DK_CFI_DEF_CFA_OFFSET:
return parseDirectiveCFIDefCfaOffset();
case DK_CFI_ADJUST_CFA_OFFSET:
return parseDirectiveCFIAdjustCfaOffset();
case DK_CFI_DEF_CFA_REGISTER:
return parseDirectiveCFIDefCfaRegister(IDLoc);
case DK_CFI_OFFSET:
return parseDirectiveCFIOffset(IDLoc);
case DK_CFI_REL_OFFSET:
return parseDirectiveCFIRelOffset(IDLoc);
case DK_CFI_PERSONALITY:
return parseDirectiveCFIPersonalityOrLsda(true);
case DK_CFI_LSDA:
return parseDirectiveCFIPersonalityOrLsda(false);
case DK_CFI_REMEMBER_STATE:
return parseDirectiveCFIRememberState();
case DK_CFI_RESTORE_STATE:
return parseDirectiveCFIRestoreState();
case DK_CFI_SAME_VALUE:
return parseDirectiveCFISameValue(IDLoc);
case DK_CFI_RESTORE:
return parseDirectiveCFIRestore(IDLoc);
case DK_CFI_ESCAPE:
return parseDirectiveCFIEscape();
case DK_CFI_RETURN_COLUMN:
return parseDirectiveCFIReturnColumn(IDLoc);
case DK_CFI_SIGNAL_FRAME:
return parseDirectiveCFISignalFrame();
case DK_CFI_UNDEFINED:
return parseDirectiveCFIUndefined(IDLoc);
case DK_CFI_REGISTER:
return parseDirectiveCFIRegister(IDLoc);
case DK_CFI_WINDOW_SAVE:
return parseDirectiveCFIWindowSave();
case DK_EXITM:
Info.ExitValue = "";
return parseDirectiveExitMacro(IDLoc, IDVal, *Info.ExitValue);
case DK_ENDM:
Info.ExitValue = "";
return parseDirectiveEndMacro(IDVal);
case DK_PURGE:
return parseDirectivePurgeMacro(IDLoc);
case DK_END:
return parseDirectiveEnd(IDLoc);
case DK_ERR:
return parseDirectiveError(IDLoc);
case DK_ERRB:
return parseDirectiveErrorIfb(IDLoc, true);
case DK_ERRNB:
return parseDirectiveErrorIfb(IDLoc, false);
case DK_ERRDEF:
return parseDirectiveErrorIfdef(IDLoc, true);
case DK_ERRNDEF:
return parseDirectiveErrorIfdef(IDLoc, false);
case DK_ERRDIF:
return parseDirectiveErrorIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/false);
case DK_ERRDIFI:
return parseDirectiveErrorIfidn(IDLoc, /*ExpectEqual=*/false,
/*CaseInsensitive=*/true);
case DK_ERRIDN:
return parseDirectiveErrorIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/false);
case DK_ERRIDNI:
return parseDirectiveErrorIfidn(IDLoc, /*ExpectEqual=*/true,
/*CaseInsensitive=*/true);
case DK_ERRE:
return parseDirectiveErrorIfe(IDLoc, true);
case DK_ERRNZ:
return parseDirectiveErrorIfe(IDLoc, false);
case DK_RADIX:
return parseDirectiveRadix(IDLoc);
}
return Error(IDLoc, "unknown directive");
}
// We also check if this is allocating memory with user-defined type.
auto IDIt = Structs.find(IDVal.lower());
if (IDIt != Structs.end())
return parseDirectiveStructValue(/*Structure=*/IDIt->getValue(), IDVal,
IDLoc);
// Non-conditional Microsoft directives sometimes follow their first argument.
const AsmToken nextTok = getTok();
const StringRef nextVal = nextTok.getString();
const SMLoc nextLoc = nextTok.getLoc();
// There are several entities interested in parsing infix directives:
//
// 1. Asm parser extensions. For example, platform-specific parsers
// (like the ELF parser) register themselves as extensions.
// 2. The generic directive parser implemented by this class. These are
// all the directives that behave in a target and platform independent
// manner, or at least have a default behavior that's shared between
// all targets and platforms.
getTargetParser().flushPendingInstructions(getStreamer());
// Special-case handling of structure-end directives at higher priority, since
// ENDS is overloaded as a segment-end directive.
if (nextVal.equals_lower("ends") && StructInProgress.size() == 1) {
Lex();
return parseDirectiveEnds(IDVal, IDLoc);
}
// First, check the extension directive map to see if any extension has
// registered itself to parse this directive.
std::pair<MCAsmParserExtension *, DirectiveHandler> Handler =
ExtensionDirectiveMap.lookup(nextVal.lower());
if (Handler.first) {
Lex();
Lexer.UnLex(ID);
return (*Handler.second)(Handler.first, nextVal, nextLoc);
}
// If no one else is interested in this directive, it must be
// generic and familiar to this class.
DirKindIt = DirectiveKindMap.find(nextVal.lower());
DirKind = (DirKindIt == DirectiveKindMap.end())
? DK_NO_DIRECTIVE
: DirKindIt->getValue();
switch (DirKind) {
default:
break;
case DK_ASSIGN:
case DK_EQU:
case DK_TEXTEQU:
Lex();
return parseDirectiveEquate(nextVal, IDVal, DirKind);
case DK_BYTE:
case DK_SBYTE:
case DK_DB:
Lex();
return parseDirectiveNamedValue(nextVal, 1, IDVal, IDLoc);
case DK_WORD:
case DK_SWORD:
case DK_DW:
Lex();
return parseDirectiveNamedValue(nextVal, 2, IDVal, IDLoc);
case DK_DWORD:
case DK_SDWORD:
case DK_DD:
Lex();
return parseDirectiveNamedValue(nextVal, 4, IDVal, IDLoc);
case DK_FWORD:
case DK_DF:
Lex();
return parseDirectiveNamedValue(nextVal, 6, IDVal, IDLoc);
case DK_QWORD:
case DK_SQWORD:
case DK_DQ:
Lex();
return parseDirectiveNamedValue(nextVal, 8, IDVal, IDLoc);
case DK_REAL4:
Lex();
return parseDirectiveNamedRealValue(nextVal, APFloat::IEEEsingle(), 4,
IDVal, IDLoc);
case DK_REAL8:
Lex();
return parseDirectiveNamedRealValue(nextVal, APFloat::IEEEdouble(), 8,
IDVal, IDLoc);
case DK_REAL10:
Lex();
return parseDirectiveNamedRealValue(nextVal, APFloat::x87DoubleExtended(),
10, IDVal, IDLoc);
case DK_STRUCT:
case DK_UNION:
Lex();
return parseDirectiveStruct(nextVal, DirKind, IDVal, IDLoc);
case DK_ENDS:
Lex();
return parseDirectiveEnds(IDVal, IDLoc);
case DK_MACRO:
Lex();
return parseDirectiveMacro(IDVal, IDLoc);
}
// Finally, we check if this is allocating a variable with user-defined type.
auto NextIt = Structs.find(nextVal.lower());
if (NextIt != Structs.end()) {
Lex();
return parseDirectiveNamedStructValue(/*Structure=*/NextIt->getValue(),
nextVal, nextLoc, IDVal);
}
// __asm _emit or __asm __emit
if (ParsingMSInlineAsm && (IDVal == "_emit" || IDVal == "__emit" ||
IDVal == "_EMIT" || IDVal == "__EMIT"))
return parseDirectiveMSEmit(IDLoc, Info, IDVal.size());
// __asm align
if (ParsingMSInlineAsm && (IDVal == "align" || IDVal == "ALIGN"))
return parseDirectiveMSAlign(IDLoc, Info);
if (ParsingMSInlineAsm && (IDVal == "even" || IDVal == "EVEN"))
Info.AsmRewrites->emplace_back(AOK_EVEN, IDLoc, 4);
if (checkForValidSection())
return true;
// Canonicalize the opcode to lower case.
std::string OpcodeStr = IDVal.lower();
ParseInstructionInfo IInfo(Info.AsmRewrites);
bool ParseHadError = getTargetParser().ParseInstruction(IInfo, OpcodeStr, ID,
Info.ParsedOperands);
Info.ParseError = ParseHadError;