-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathcode_attribute.hh
163 lines (144 loc) · 6.5 KB
/
code_attribute.hh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
/**
* Copyright 2019 Anthony Calandra
*
* Permission is hereby granted, free of charge, to any person obtaining a copy of this software
* and associated documentation files (the "Software"), to deal in the Software without
* restriction, including without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all copies or
* substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING
* BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
* DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/
#pragma once
#include <array>
#include <functional>
#include <iostream>
#include <memory>
#include <vector>
#include "attribute_info.hh"
#include "bytecode.hh"
#include "constant_pool.hh"
#include "util.hh"
struct exception_table_entry
{
uint16_t start_pc;
uint16_t end_pc;
uint16_t handler_pc;
uint16_t catch_pc;
explicit exception_table_entry(uint16_t start_pc, uint16_t end_pc, uint16_t handler_pc,
uint16_t catch_pc) :
start_pc{start_pc},
end_pc{end_pc},
handler_pc{handler_pc},
catch_pc{catch_pc}
{}
};
class code_attribute: public attribute_info
{
const constant_pool& cp;
[[maybe_unused]] uint16_t max_stack;
[[maybe_unused]] uint16_t max_locals;
// Despite the program counter being a 2-byte short, the code length is a 4-byte int because of
// the case where an instruction at address 0xFFFF is bigger than a byte (e.g. `lookupswitch` is a
// variable-sized instruction thus exceeding the max size of the pc).
uint32_t code_length;
std::unique_ptr<uint8_t[]> bytecode;
std::vector<exception_table_entry> exception_table;
entry_attributes code_attributes;
public:
explicit code_attribute(const constant_pool& cp, uint16_t max_stack, uint16_t max_locals,
uint32_t code_length, std::unique_ptr<uint8_t[]> bytecode,
std::vector<exception_table_entry> exception_table, entry_attributes code_attributes) :
cp{cp},
max_stack{max_stack},
max_locals{max_locals},
code_length{code_length},
bytecode{std::move(bytecode)},
exception_table{std::move(exception_table)},
code_attributes{std::move(code_attributes)}
{}
void print_code(std::ostream& out) const;
template <bytecode_tag instr>
using find_instructions_cb =
// Since each instruction tag is one byte, the instruction size minus one is the number of
// instruction operands.
typename find_instructions_cb_fn_type<get_instruction_size(instr) - 1>::type;
template <bytecode_tag instr>
void find_instruction(find_instructions_cb<instr> cb) const
{
constexpr size_t target_instr_operands_size = get_instruction_size(instr) - 1;
for (uint32_t pc = 0; pc < code_length;)
{
const bytecode_tag curr_instr = static_cast<bytecode_tag>(bytecode[pc]);
const size_t curr_instr_size = get_instruction_size(curr_instr);
// These are variable-sized instructions and require special parsing.
if (curr_instr == bytecode_tag::LOOKUPSWITCH ||
curr_instr == bytecode_tag::TABLESWITCH)
{
// Skip past padding bytes until we get to the first address that is a multiple of 4.
if (pc % 4 != 0)
{
for (; pc % 4 != 0; pc++);
}
// Skip `default` bytes.
pc += 4;
if (curr_instr == bytecode_tag::LOOKUPSWITCH)
{
// `npairs` is a signed 4-byte, big-endian integer. This should probably never
// be negative.
uint32_t* npairs_bytes = reinterpret_cast<uint32_t*>(&bytecode[pc]);
uint32_t npairs = to_little_endian_int(*npairs_bytes);
// Each pair consists of two 4-byte ints.
pc += 8 * npairs;
}
else if (curr_instr == bytecode_tag::TABLESWITCH)
{
// Read both `low` and `high` signed ints. These should probably never be
// negative.
uint32_t* low_bytes = reinterpret_cast<uint32_t*>(&bytecode[pc]);
uint32_t low = to_little_endian_int(*low_bytes);
pc += 4;
uint32_t* high_bytes = reinterpret_cast<uint32_t*>(&bytecode[pc]);
uint32_t high = to_little_endian_int(*high_bytes);
// There are `high - low + 1` signed integer offsets that must be skipped.
pc += 4 * (high - low + 1 + 1);
}
continue;
}
if (curr_instr == instr)
{
// The type of `operands` is a `std::tuple` with a variable number of types
// (depends on the cb function given).
typename function_args_tuple<decltype(cb)>::type operands;
std::get<0>(operands) = pc & 0xFFFF;
// A `constexpr-for` construct would be perfect here...
if constexpr (target_instr_operands_size >= 1)
std::get<1>(operands) = bytecode[pc + 1];
if constexpr (target_instr_operands_size >= 2)
std::get<2>(operands) = bytecode[pc + 2];
if constexpr (target_instr_operands_size >= 3)
std::get<3>(operands) = bytecode[pc + 3];
if constexpr (target_instr_operands_size >= 4)
std::get<4>(operands) = bytecode[pc + 4];
std::apply(cb, operands);
}
pc += curr_instr_size;
}
}
const entry_attributes& get_code_attributes() const
{
return code_attributes;
}
virtual attribute_info_type get_type() const
{
return attribute_info_type::code;
}
};
std::unique_ptr<attribute_info> parse_code_attribute(std::ifstream& file, const constant_pool& cp);