X86Disassembler.h revision 98c5ddabca1debf935a07d14d0cbc9732374bdb8
18ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//===- X86Disassembler.h - Disassembler for x86 and x86_64 ------*- C++ -*-===//
28ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
38ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//                     The LLVM Compiler Infrastructure
48ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
58ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// This file is distributed under the University of Illinois Open Source
68ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// License. See LICENSE.TXT for details.
78ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
88ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//===----------------------------------------------------------------------===//
98ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
108ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// The X86 disassembler is a table-driven disassembler for the 16-, 32-, and
118ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 64-bit X86 instruction sets.  The main decode sequence for an assembly
128ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// instruction in this disassembler is:
138ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
148ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 1. Read the prefix bytes and determine the attributes of the instruction.
158ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    These attributes, recorded in enum attributeBits
168ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    (X86DisassemblerDecoderCommon.h), form a bitmask.  The table CONTEXTS_SYM
178ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    provides a mapping from bitmasks to contexts, which are represented by
188ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    enum InstructionContext (ibid.).
198ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
208ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 2. Read the opcode, and determine what kind of opcode it is.  The
218ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    disassembler distinguishes four kinds of opcodes, which are enumerated in
228ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    OpcodeType (X86DisassemblerDecoderCommon.h): one-byte (0xnn), two-byte
238ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    (0x0f 0xnn), three-byte-38 (0x0f 0x38 0xnn), or three-byte-3a
248ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    (0x0f 0x3a 0xnn).  Mandatory prefixes are treated as part of the context.
258ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
268ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 3. Depending on the opcode type, look in one of four ClassDecision structures
278ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    (X86DisassemblerDecoderCommon.h).  Use the opcode class to determine which
288ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    OpcodeDecision (ibid.) to look the opcode in.  Look up the opcode, to get
298ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    a ModRMDecision (ibid.).
308ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
318ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 4. Some instructions, such as escape opcodes or extended opcodes, or even
328ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    instructions that have ModRM*Reg / ModRM*Mem forms in LLVM, need the
338ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    ModR/M byte to complete decode.  The ModRMDecision's type is an entry from
348ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    ModRMDecisionType (X86DisassemblerDecoderCommon.h) that indicates if the
358ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    ModR/M byte is required and how to interpret it.
368ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
378ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 5. After resolving the ModRMDecision, the disassembler has a unique ID
388ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    of type InstrUID (X86DisassemblerDecoderCommon.h).  Looking this ID up in
398ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    INSTRUCTIONS_SYM yields the name of the instruction and the encodings and
408ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    meanings of its operands.
418ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
428ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 6. For each operand, its encoding is an entry from OperandEncoding
438ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    (X86DisassemblerDecoderCommon.h) and its type is an entry from
448ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    OperandType (ibid.).  The encoding indicates how to read it from the
458ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    instruction; the type indicates how to interpret the value once it has
468ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    been read.  For example, a register operand could be stored in the R/M
478ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    field of the ModR/M byte, the REG field of the ModR/M byte, or added to
488ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    the main opcode.  This is orthogonal from its meaning (an GPR or an XMM
498ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    register, for instance).  Given this information, the operands can be
508ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    extracted and interpreted.
518ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
528ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// 7. As the last step, the disassembler translates the instruction information
538ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    and operands into a format understandable by the client - in this case, an
548ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//    MCInst for use by the MC infrastructure.
558ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
568ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// The disassembler is broken broadly into two parts: the table emitter that
578ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// emits the instruction decode tables discussed above during compilation, and
588ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// the disassembler itself.  The table emitter is documented in more detail in
598ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// utils/TableGen/X86DisassemblerEmitter.h.
608ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
618ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// X86Disassembler.h contains the public interface for the disassembler,
628ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//   adhering to the MCDisassembler interface.
638ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// X86Disassembler.cpp contains the code responsible for step 7, and for
648ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//   invoking the decoder to execute steps 1-6.
658ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// X86DisassemblerDecoderCommon.h contains the definitions needed by both the
668ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//   table emitter and the disassembler.
678ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// X86DisassemblerDecoder.h contains the public interface of the decoder,
688ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//   factored out into C for possible use by other projects.
698ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan// X86DisassemblerDecoder.c contains the source code of the decoder, which is
708ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//   responsible for steps 1-6.
718ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//
728ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan//===----------------------------------------------------------------------===//
738ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
748ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#ifndef X86DISASSEMBLER_H
758ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#define X86DISASSEMBLER_H
768ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
778ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#define INSTRUCTION_SPECIFIER_FIELDS  \
788ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  const char*             name;
798ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
808ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#define INSTRUCTION_IDS               \
814d1dca92bd6d4aad7121e28c7ffc93c0a6a187d7Benjamin Kramer  const InstrUID *instructionIDs;
828ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
838ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#include "X86DisassemblerDecoderCommon.h"
848ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
858ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#undef INSTRUCTION_SPECIFIER_FIELDS
868ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#undef INSTRUCTION_IDS
878ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
888ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#include "llvm/MC/MCDisassembler.h"
898ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
908ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananstruct InternalInstruction;
918ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
928ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanannamespace llvm {
938ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
948ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass MCInst;
95b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloyclass MCSubtargetInfo;
968ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass MemoryObject;
978ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass raw_ostream;
989899f70a7406d632c82849978bf6981f1ee4ccb5Sean Callanan
999899f70a7406d632c82849978bf6981f1ee4ccb5Sean Callananstruct EDInstInfo;
1008ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1018ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanannamespace X86Disassembler {
1028ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1038ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan/// X86GenericDisassembler - Generic disassembler for all X86 platforms.
1048ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan///   All each platform class should have to do is subclass the constructor, and
1058ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan///   provide a different disassemblerMode value.
1068ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass X86GenericDisassembler : public MCDisassembler {
1078ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananprotected:
1088ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  /// Constructor     - Initializes the disassembler.
1098ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  ///
1108ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  /// @param mode     - The X86 architecture mode to decode for.
111b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy  X86GenericDisassembler(const MCSubtargetInfo &STI, DisassemblerMode mode);
1128ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananpublic:
1138ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  ~X86GenericDisassembler();
1148ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1158ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  /// getInstruction - See MCDisassembler.
11683e3f67fb68d497b600da83a62f000fcce7868a9Owen Anderson  DecodeStatus getInstruction(MCInst &instr,
11783e3f67fb68d497b600da83a62f000fcce7868a9Owen Anderson                              uint64_t &size,
11883e3f67fb68d497b600da83a62f000fcce7868a9Owen Anderson                              const MemoryObject &region,
11983e3f67fb68d497b600da83a62f000fcce7868a9Owen Anderson                              uint64_t address,
12098c5ddabca1debf935a07d14d0cbc9732374bdb8Owen Anderson                              raw_ostream &vStream,
12198c5ddabca1debf935a07d14d0cbc9732374bdb8Owen Anderson                              raw_ostream &cStream) const;
1229899f70a7406d632c82849978bf6981f1ee4ccb5Sean Callanan
1239899f70a7406d632c82849978bf6981f1ee4ccb5Sean Callanan  /// getEDInfo - See MCDisassembler.
1249899f70a7406d632c82849978bf6981f1ee4ccb5Sean Callanan  EDInstInfo *getEDInfo() const;
1258ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananprivate:
1268ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  DisassemblerMode              fMode;
1278ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan};
1288ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1298ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan/// X86_16Disassembler - 16-bit X86 disassembler.
1308ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass X86_16Disassembler : public X86GenericDisassembler {
1318ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananpublic:
132b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy  X86_16Disassembler(const MCSubtargetInfo &STI) :
133b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy    X86GenericDisassembler(STI, MODE_16BIT) {
1348ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  }
1358ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan};
1368ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1378ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan/// X86_16Disassembler - 32-bit X86 disassembler.
1388ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass X86_32Disassembler : public X86GenericDisassembler {
1398ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananpublic:
140b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy  X86_32Disassembler(const MCSubtargetInfo &STI) :
141b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy    X86GenericDisassembler(STI, MODE_32BIT) {
1428ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  }
1438ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan};
1448ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1458ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan/// X86_16Disassembler - 64-bit X86 disassembler.
1468ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananclass X86_64Disassembler : public X86GenericDisassembler {
1478ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callananpublic:
148b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy  X86_64Disassembler(const MCSubtargetInfo &STI) :
149b950585cc5a0d665e9accfe5ce490cd269756f2eJames Molloy    X86GenericDisassembler(STI, MODE_64BIT) {
1508ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan  }
1518ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan};
1528ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1538ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan} // namespace X86Disassembler
1548ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1558ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan} // namespace llvm
1568ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan
1578ed9f51663bc5533f36ca62e5668ae08e9a1313fSean Callanan#endif
158