1//===- llvm/MC/MCDisassembler.h - Disassembler interface --------*- C++ -*-===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8
9#ifndef LLVM_MC_MCDISASSEMBLER_MCDISASSEMBLER_H
10#define LLVM_MC_MCDISASSEMBLER_MCDISASSEMBLER_H
11
12#include "llvm/MC/MCDisassembler/MCSymbolizer.h"
13#include <cstdint>
14#include <memory>
15
16namespace llvm {
17
18template <typename T> class ArrayRef;
19class StringRef;
20class MCContext;
21class MCInst;
22class MCSubtargetInfo;
23class raw_ostream;
24
25/// Superclass for all disassemblers. Consumes a memory region and provides an
26/// array of assembly instructions.
27class MCDisassembler {
28public:
29  /// Ternary decode status. Most backends will just use Fail and
30  /// Success, however some have a concept of an instruction with
31  /// understandable semantics but which is architecturally
32  /// incorrect. An example of this is ARM UNPREDICTABLE instructions
33  /// which are disassemblable but cause undefined behaviour.
34  ///
35  /// Because it makes sense to disassemble these instructions, there
36  /// is a "soft fail" failure mode that indicates the MCInst& is
37  /// valid but architecturally incorrect.
38  ///
39  /// The enum numbers are deliberately chosen such that reduction
40  /// from Success->SoftFail ->Fail can be done with a simple
41  /// bitwise-AND:
42  ///
43  ///   LEFT & TOP =  | Success       Unpredictable   Fail
44  ///   --------------+-----------------------------------
45  ///   Success       | Success       Unpredictable   Fail
46  ///   Unpredictable | Unpredictable Unpredictable   Fail
47  ///   Fail          | Fail          Fail            Fail
48  ///
49  /// An easy way of encoding this is as 0b11, 0b01, 0b00 for
50  /// Success, SoftFail, Fail respectively.
51  enum DecodeStatus {
52    Fail = 0,
53    SoftFail = 1,
54    Success = 3
55  };
56
57  MCDisassembler(const MCSubtargetInfo &STI, MCContext &Ctx)
58    : Ctx(Ctx), STI(STI) {}
59
60  virtual ~MCDisassembler();
61
62  /// Returns the disassembly of a single instruction.
63  ///
64  /// \param Instr    - An MCInst to populate with the contents of the
65  ///                   instruction.
66  /// \param Size     - A value to populate with the size of the instruction, or
67  ///                   the number of bytes consumed while attempting to decode
68  ///                   an invalid instruction.
69  /// \param Address  - The address, in the memory space of region, of the first
70  ///                   byte of the instruction.
71  /// \param Bytes    - A reference to the actual bytes of the instruction.
72  /// \param CStream  - The stream to print comments and annotations on.
73  /// \return         - MCDisassembler::Success if the instruction is valid,
74  ///                   MCDisassembler::SoftFail if the instruction was
75  ///                                            disassemblable but invalid,
76  ///                   MCDisassembler::Fail if the instruction was invalid.
77  virtual DecodeStatus getInstruction(MCInst &Instr, uint64_t &Size,
78                                      ArrayRef<uint8_t> Bytes, uint64_t Address,
79                                      raw_ostream &CStream) const = 0;
80
81  /// May parse any prelude that precedes instructions after the start of a
82  /// symbol. Needed for some targets, e.g. WebAssembly.
83  ///
84  /// \param Name     - The name of the symbol.
85  /// \param Size     - The number of bytes consumed.
86  /// \param Address  - The address, in the memory space of region, of the first
87  ///                   byte of the symbol.
88  /// \param Bytes    - A reference to the actual bytes at the symbol location.
89  /// \param CStream  - The stream to print comments and annotations on.
90  /// \return         - MCDisassembler::Success if the bytes are valid,
91  ///                   MCDisassembler::Fail if the bytes were invalid.
92  virtual DecodeStatus onSymbolStart(StringRef Name, uint64_t &Size,
93                                     ArrayRef<uint8_t> Bytes, uint64_t Address,
94                                     raw_ostream &CStream) const;
95
96private:
97  MCContext &Ctx;
98
99protected:
100  // Subtarget information, for instruction decoding predicates if required.
101  const MCSubtargetInfo &STI;
102  std::unique_ptr<MCSymbolizer> Symbolizer;
103
104public:
105  // Helpers around MCSymbolizer
106  bool tryAddingSymbolicOperand(MCInst &Inst,
107                                int64_t Value,
108                                uint64_t Address, bool IsBranch,
109                                uint64_t Offset, uint64_t InstSize) const;
110
111  void tryAddingPcLoadReferenceComment(int64_t Value, uint64_t Address) const;
112
113  /// Set \p Symzer as the current symbolizer.
114  /// This takes ownership of \p Symzer, and deletes the previously set one.
115  void setSymbolizer(std::unique_ptr<MCSymbolizer> Symzer);
116
117  MCContext& getContext() const { return Ctx; }
118
119  const MCSubtargetInfo& getSubtargetInfo() const { return STI; }
120
121  // Marked mutable because we cache it inside the disassembler, rather than
122  // having to pass it around as an argument through all the autogenerated code.
123  mutable raw_ostream *CommentStream = nullptr;
124};
125
126} // end namespace llvm
127
128#endif // LLVM_MC_MCDISASSEMBLER_MCDISASSEMBLER_H
129