Add files via upload
This commit is contained in:
GitHub
parent
cb03d9a148
commit
95ba10eacc
560
objdump_readelf/kvs_.py
Normal file
560
objdump_readelf/kvs_.py
Normal file
|
|
@ -0,0 +1,560 @@
|
|||
#!/usr/bin/env python3
|
||||
# -*- coding: utf-8 -*-
|
||||
|
||||
"""
|
||||
КВС - язык программирования
|
||||
версия minimal.py (обновлённая, с полной поддержкой ELF)
|
||||
без использования re
|
||||
"""
|
||||
|
||||
import sys
|
||||
import struct
|
||||
|
||||
# Глобальная переменная для имени исходного файла
|
||||
source_filename = ""
|
||||
|
||||
def safe_close(f):
|
||||
if f is not None:
|
||||
f.close()
|
||||
|
||||
# === Глобальные данные ===
|
||||
INSTRUCTIONS = {
|
||||
"переместить_имм": {"code": None, "type": "reg_imm"},
|
||||
"сравнить_с": {"code": None, "type": "cmp_reg_imm"},
|
||||
"переход_если_неравно": {"code": b"\x0F\x85", "type": "jcc"},
|
||||
"вызов_системы": {"code": b"\x0F\x05", "type": "none"},
|
||||
"переход": {"code": b"\xE9", "type": "jmp"},
|
||||
}
|
||||
|
||||
REGISTERS = {
|
||||
"раикс": 0, # RAX
|
||||
"рдикс": 2, # RDX
|
||||
"рсиай": 6, # RSI
|
||||
"рдиай": 7, # RDI
|
||||
}
|
||||
|
||||
# Глобальное состояние ассемблера
|
||||
labels = {}
|
||||
label_sections = {}
|
||||
symbols = {}
|
||||
sections = {".text": bytearray(), ".data": bytearray()}
|
||||
current_section = ".text"
|
||||
entry_point = "_start"
|
||||
position = {".text": 0, ".data": 0}
|
||||
pass_num = 0
|
||||
|
||||
# === ELF layout (глобальные переменные для второго прохода и записи) ===
|
||||
PAGE_SIZE = 0x1000
|
||||
elf_base_vaddr = 0x400000 # первая страница — заголовки
|
||||
text_vaddr_base = 0x401000 # вторая страница — .text
|
||||
data_vaddr_base = 0x402000 # третья страница — .data
|
||||
|
||||
# Смещения и адреса (вычисляются после первого прохода)
|
||||
text_size = 0
|
||||
data_size = 0
|
||||
offset_text = 0
|
||||
offset_data = 0
|
||||
shstrtab_offset = 0
|
||||
shdr_offset = 0
|
||||
vaddr_text = 0
|
||||
vaddr_data = 0
|
||||
|
||||
def align_up(x, align):
|
||||
return (x + align - 1) & ~(align - 1)
|
||||
|
||||
def get_reg_info(reg_name):
|
||||
return {"size": 64, "index": REGISTERS[reg_name]}
|
||||
|
||||
# === Функции формирования ошибок ===
|
||||
|
||||
def make_error_msg(line_num, line_text, detail):
|
||||
return f"Ошибка в файле {source_filename}, строка {line_num}:\n {line_text}\n{detail}"
|
||||
|
||||
def error_invalid_operand_count(mnemonic, expected, got):
|
||||
return f"Инструкция '{mnemonic}':\n ожидается {expected} операнд(а/ов), получено: {got}"
|
||||
|
||||
def error_unknown_mnemonic(mnemonic):
|
||||
return f"Неизвестная инструкция: '{mnemonic}'"
|
||||
|
||||
def error_invalid_register(op_name, reg_name, mnemonic):
|
||||
return f"Инструкция '{mnemonic}':\n операнд '{op_name}' = '{reg_name}' не является допустимым регистром.\n Допустимые регистры: {', '.join(list(REGISTERS.keys()))}"
|
||||
|
||||
def error_invalid_number_format(s):
|
||||
return f"Недопустимый формат числа: '{s}'"
|
||||
|
||||
def error_unknown_directive(word):
|
||||
return f"Неизвестная директива: '{word}'"
|
||||
|
||||
def error_missing_label(label):
|
||||
return f"Метка не найдена: '{label}'"
|
||||
|
||||
def error_unexpected_string_in_byte():
|
||||
return ".байт требует числовые значения, не строки"
|
||||
|
||||
def error_unterminated_string():
|
||||
return "Незакрытая кавычка в строке"
|
||||
|
||||
# === Лексер ===
|
||||
|
||||
def tokenize_line(line):
|
||||
semi = line.find(';')
|
||||
if semi != -1:
|
||||
line = line[:semi]
|
||||
line = line.rstrip()
|
||||
if not line:
|
||||
return []
|
||||
tokens = []
|
||||
i = 0
|
||||
n = len(line)
|
||||
while i < n:
|
||||
ch = line[i]
|
||||
if ch.isspace():
|
||||
i += 1
|
||||
continue
|
||||
if ch == '"':
|
||||
i += 1
|
||||
s = ''
|
||||
while i < n and line[i] != '"':
|
||||
if line[i] == '\\' and i + 1 < n:
|
||||
i += 1
|
||||
esc = line[i]
|
||||
if esc == 'n':
|
||||
s += '\n'
|
||||
elif esc == 't':
|
||||
s += '\t'
|
||||
elif esc == '"':
|
||||
s += '"'
|
||||
elif esc == '\\':
|
||||
s += '\\'
|
||||
else:
|
||||
s += '\\' + esc
|
||||
i += 1
|
||||
else:
|
||||
s += line[i]
|
||||
i += 1
|
||||
if i >= n:
|
||||
raise ValueError(error_unterminated_string())
|
||||
i += 1
|
||||
tokens.append(('string', s))
|
||||
continue
|
||||
if ch == ',':
|
||||
tokens.append(('comma', ','))
|
||||
i += 1
|
||||
continue
|
||||
if ch == ':':
|
||||
tokens.append(('colon', ':'))
|
||||
i += 1
|
||||
continue
|
||||
j = i
|
||||
while j < n and not (line[j].isspace() or line[j] in ',:;'):
|
||||
j += 1
|
||||
word = line[i:j]
|
||||
if word:
|
||||
tokens.append(('word', word))
|
||||
i = j
|
||||
return tokens
|
||||
|
||||
# === Парсинг ===
|
||||
|
||||
def parse_operand(operand):
|
||||
if operand.isdigit():
|
||||
return int(operand)
|
||||
if operand.startswith("0x"):
|
||||
return int(operand, 16)
|
||||
if operand in labels:
|
||||
section = label_sections[operand]
|
||||
if section == ".text":
|
||||
return labels[operand] + vaddr_text
|
||||
elif section == ".data":
|
||||
return labels[operand] + vaddr_data
|
||||
else:
|
||||
raise ValueError(f"Метка в неизвестной секции: {section}")
|
||||
if operand in symbols:
|
||||
return symbols[operand]
|
||||
raise ValueError(f"Неизвестный операнд: {operand}")
|
||||
|
||||
def parse_number_token(token):
|
||||
if token[0] != 'word':
|
||||
raise ValueError(f"Ожидалось число, получено: {token}")
|
||||
s = token[1]
|
||||
if s.isdigit():
|
||||
return int(s)
|
||||
if s.startswith("0x"):
|
||||
return int(s, 16)
|
||||
raise ValueError(error_invalid_number_format(s))
|
||||
|
||||
def encode_instruction(mnemonic, operands, line_num, line_text):
|
||||
instr = INSTRUCTIONS[mnemonic]
|
||||
code = bytearray()
|
||||
itype = instr["type"]
|
||||
|
||||
if itype == "none":
|
||||
code.extend(instr["code"])
|
||||
|
||||
elif itype == "reg_imm":
|
||||
if len(operands) != 2:
|
||||
raise ValueError(error_invalid_operand_count(mnemonic, 2, len(operands)))
|
||||
reg_info = get_reg_info(operands[0])
|
||||
if reg_info is None:
|
||||
raise ValueError(error_invalid_register("назначения", operands[0], mnemonic))
|
||||
reg = reg_info["index"]
|
||||
imm = parse_operand(operands[1])
|
||||
|
||||
if 0 <= reg <= 7:
|
||||
code.extend(b'\x48')
|
||||
code.append(0xB8 + reg)
|
||||
code.extend(struct.pack('<Q', imm))
|
||||
elif 8 <= reg <= 15:
|
||||
code.extend(b'\x49')
|
||||
code.append(0xB8 + (reg - 8))
|
||||
code.extend(struct.pack('<Q', imm))
|
||||
else:
|
||||
raise ValueError(f"Недопустимый регистр в '{mnemonic}'")
|
||||
|
||||
elif itype == "cmp_reg_imm":
|
||||
if len(operands) != 2:
|
||||
raise ValueError(error_invalid_operand_count(mnemonic, 2, len(operands)))
|
||||
reg_info = get_reg_info(operands[0])
|
||||
if reg_info is None:
|
||||
raise ValueError(error_invalid_register("регистра", operands[0], mnemonic))
|
||||
reg = reg_info["index"]
|
||||
imm = parse_operand(operands[1])
|
||||
|
||||
if -0x80000000 <= imm <= 0x7FFFFFFF:
|
||||
rex = 0x48
|
||||
if reg >= 8:
|
||||
rex |= 0x01
|
||||
code.append(rex)
|
||||
code.append(0x81)
|
||||
modrm = 0xF8 | (reg & 7)
|
||||
code.append(modrm)
|
||||
code.extend(struct.pack('<i', imm))
|
||||
else:
|
||||
raise ValueError(f"Непосредственное значение слишком велико для сравнения: {imm}")
|
||||
|
||||
elif itype == "jmp":
|
||||
if len(operands) != 1:
|
||||
raise ValueError(error_invalid_operand_count(mnemonic, 1, len(operands)))
|
||||
code.extend(instr["code"])
|
||||
target = parse_operand(operands[0])
|
||||
current_addr = vaddr_text + position[".text"]
|
||||
offset = target - (current_addr + 5)
|
||||
if offset < -0x80000000 or offset > 0x7FFFFFFF:
|
||||
raise ValueError(f"Цель слишком далеко для 32-битного смещения в инструкции '{mnemonic}': {offset} (должно быть в [-2^31, 2^31-1])")
|
||||
code.extend(struct.pack('<i', offset))
|
||||
|
||||
elif itype == "jcc":
|
||||
if len(operands) != 1:
|
||||
raise ValueError(error_invalid_operand_count(mnemonic, 1, len(operands)))
|
||||
code.extend(instr["code"])
|
||||
target = parse_operand(operands[0])
|
||||
current_addr = vaddr_text + position[".text"]
|
||||
offset = target - (current_addr + 6)
|
||||
if offset < -0x80000000 or offset > 0x7FFFFFFF:
|
||||
raise ValueError(f"Цель условного перехода слишком далеко: {offset}")
|
||||
code.extend(struct.pack('<i', offset))
|
||||
|
||||
else:
|
||||
raise ValueError(f"Неизвестный тип инструкции: {itype}")
|
||||
|
||||
return bytes(code)
|
||||
|
||||
def parse_instruction_or_directive(tokens, line_num, line_text):
|
||||
global current_section, entry_point
|
||||
first = tokens[0]
|
||||
if first[0] != 'word':
|
||||
raise ValueError(f"Ожидалось слово, получено: {first}")
|
||||
word = first[1]
|
||||
if word.startswith('.'):
|
||||
if word == '.текст':
|
||||
current_section = ".text"
|
||||
elif word == '.данные':
|
||||
current_section = ".data"
|
||||
elif word == '.глобал':
|
||||
if len(tokens) < 2 or tokens[1][0] != 'word':
|
||||
raise ValueError(".глобал требует имя метки")
|
||||
entry_point = tokens[1][1]
|
||||
elif word == '.строка_нуль' or word == '.строка':
|
||||
if len(tokens) < 2 or tokens[1][0] != 'string':
|
||||
raise ValueError(f"{word} требует строку в кавычках")
|
||||
if current_section != ".data":
|
||||
raise ValueError("Строки разрешены только в секции .data")
|
||||
s = tokens[1][1]
|
||||
bstring = s.encode('utf-8')
|
||||
add_null = (word == '.строка_нуль')
|
||||
size = len(bstring) + (1 if add_null else 0)
|
||||
if pass_num == 1:
|
||||
position[".data"] += size
|
||||
elif pass_num == 2:
|
||||
sections[".data"] += bstring
|
||||
if add_null:
|
||||
sections[".data"] += b'\x00'
|
||||
position[".data"] += size
|
||||
else:
|
||||
raise ValueError(error_unknown_directive(word))
|
||||
return
|
||||
|
||||
mnemonic = word
|
||||
if mnemonic not in INSTRUCTIONS:
|
||||
raise ValueError(error_unknown_mnemonic(mnemonic))
|
||||
|
||||
operands = []
|
||||
for tok in tokens[1:]:
|
||||
if tok[0] == 'word':
|
||||
operands.append(tok[1])
|
||||
elif tok[0] == 'comma':
|
||||
continue
|
||||
else:
|
||||
raise ValueError(f"Недопустимый токен в операндах: {tok}")
|
||||
|
||||
if pass_num == 2:
|
||||
code = encode_instruction(mnemonic, operands, line_num, line_text)
|
||||
sections[current_section] += code
|
||||
position[current_section] += len(code)
|
||||
else:
|
||||
instr_info = INSTRUCTIONS[mnemonic]
|
||||
itype = instr_info["type"]
|
||||
size_map = {
|
||||
"reg_imm": 10,
|
||||
"cmp_reg_imm": 7,
|
||||
"jmp": 5,
|
||||
"jcc": 6,
|
||||
"none": lambda: len(instr_info["code"]),
|
||||
}
|
||||
if itype in size_map:
|
||||
sz = size_map[itype]
|
||||
position[current_section] += sz() if callable(sz) else sz
|
||||
else:
|
||||
raise ValueError(f"Неподдерживаемый тип инструкции при подсчёте размера: {itype}")
|
||||
|
||||
def parse_tokens(tokens, line_num, line_text):
|
||||
if not tokens:
|
||||
return
|
||||
if len(tokens) >= 2 and tokens[0][0] == 'word' and tokens[1][0] == 'colon':
|
||||
label = tokens[0][1]
|
||||
labels[label] = position[current_section]
|
||||
label_sections[label] = current_section
|
||||
rest = tokens[2:]
|
||||
if rest:
|
||||
parse_instruction_or_directive(rest, line_num, line_text)
|
||||
return
|
||||
parse_instruction_or_directive(tokens, line_num, line_text)
|
||||
|
||||
def parse(source):
|
||||
global pass_num, text_size, data_size
|
||||
global offset_text, offset_data, shstrtab_offset, shdr_offset
|
||||
global vaddr_text, vaddr_data
|
||||
global sections, position
|
||||
|
||||
lines = source.split('\n')
|
||||
# === ПРОХОД 1: анализ ===
|
||||
pass_num = 1
|
||||
position[".text"] = 0
|
||||
position[".data"] = 0
|
||||
sections[".text"] = bytearray()
|
||||
sections[".data"] = bytearray()
|
||||
labels.clear()
|
||||
label_sections.clear()
|
||||
|
||||
for line_num, line in enumerate(lines, start=1):
|
||||
original_line = line.rstrip()
|
||||
try:
|
||||
tokens = tokenize_line(line)
|
||||
parse_tokens(tokens, line_num, original_line)
|
||||
except Exception as e:
|
||||
error_full = make_error_msg(line_num, original_line, str(e))
|
||||
print(error_full, file=sys.stderr)
|
||||
sys.exit(1)
|
||||
|
||||
text_size = position[".text"]
|
||||
data_size = position[".data"]
|
||||
|
||||
# === ВЫЧИСЛЕНИЕ РАЗМЕЩЕНИЯ ===
|
||||
elf_header_size = 64
|
||||
ph_size = 56
|
||||
ph_num = 3
|
||||
ph_table_size = ph_num * ph_size
|
||||
|
||||
offset_text = align_up(elf_header_size + ph_table_size, PAGE_SIZE)
|
||||
offset_data = align_up(offset_text + text_size, PAGE_SIZE)
|
||||
|
||||
vaddr_text = text_vaddr_base
|
||||
vaddr_data = data_vaddr_base
|
||||
|
||||
# Section headers
|
||||
shstrtab_content = b"\x00.text\x00.data\x00.shstrtab\x00"
|
||||
shstrtab_size = len(shstrtab_content)
|
||||
shstrtab_offset = align_up(offset_data + data_size, 8)
|
||||
shdr_size = 64
|
||||
shdr_num = 4
|
||||
shdr_offset = align_up(shstrtab_offset + shstrtab_size, 16)
|
||||
|
||||
# === ПРОХОД 2: генерация ===
|
||||
pass_num = 2
|
||||
position[".text"] = 0
|
||||
position[".data"] = 0
|
||||
sections[".text"] = bytearray()
|
||||
sections[".data"] = bytearray()
|
||||
for line_num, line in enumerate(lines, start=1):
|
||||
original_line = line.rstrip()
|
||||
try:
|
||||
tokens = tokenize_line(line)
|
||||
parse_tokens(tokens, line_num, original_line)
|
||||
except Exception as e:
|
||||
error_full = make_error_msg(line_num, original_line, str(e))
|
||||
print(error_full, file=sys.stderr)
|
||||
sys.exit(1)
|
||||
|
||||
def create_elf(filename):
|
||||
global sections, entry_point, labels, vaddr_text, vaddr_data
|
||||
global offset_text, offset_data, shstrtab_offset, shdr_offset
|
||||
|
||||
text = sections[".text"]
|
||||
data = sections[".data"]
|
||||
|
||||
# Точка входа
|
||||
entry_addr = vaddr_text + labels.get(entry_point, 0)
|
||||
|
||||
# .shstrtab
|
||||
shstrtab_content = b"\x00.text\x00.data\x00.shstrtab\x00"
|
||||
shstrtab_size = len(shstrtab_content)
|
||||
|
||||
# Размеры
|
||||
elf_header_size = 64
|
||||
ph_size = 56
|
||||
ph_num = 3
|
||||
shdr_size = 64
|
||||
shdr_num = 4
|
||||
|
||||
file_size = shdr_offset + shdr_num * shdr_size
|
||||
elf_data = bytearray(file_size)
|
||||
|
||||
# === 1. ELF-заголовок ===
|
||||
e_ident = b"\x7fELF\x02\x01\x01\x00" + b"\x00" * 8
|
||||
elf_header = struct.pack(
|
||||
'<16sHHIQQQIHHHHHH',
|
||||
e_ident,
|
||||
2, # ET_EXEC
|
||||
0x3e, # EM_X86_64
|
||||
1, # version
|
||||
entry_addr, # entry
|
||||
elf_header_size,# phoff
|
||||
shdr_offset, # shoff
|
||||
0, # flags
|
||||
elf_header_size,
|
||||
ph_size,
|
||||
ph_num,
|
||||
shdr_size,
|
||||
shdr_num,
|
||||
3 # .shstrtab index
|
||||
)
|
||||
elf_data[0:64] = elf_header
|
||||
|
||||
# === 2. Program Headers ===
|
||||
def phdr(p_type, p_flags, p_offset, p_vaddr, p_filesz, p_memsz):
|
||||
return struct.pack('<IIQQQQQQ',
|
||||
p_type, p_flags,
|
||||
p_offset, p_vaddr, p_vaddr,
|
||||
p_filesz, p_memsz,
|
||||
PAGE_SIZE
|
||||
)
|
||||
|
||||
ph0 = phdr(
|
||||
p_type=1,
|
||||
p_flags=4,
|
||||
p_offset=0,
|
||||
p_vaddr=elf_base_vaddr,
|
||||
p_filesz=elf_header_size + ph_num * ph_size,
|
||||
p_memsz=PAGE_SIZE
|
||||
)
|
||||
|
||||
ph1 = phdr(
|
||||
p_type=1,
|
||||
p_flags=5,
|
||||
p_offset=offset_text,
|
||||
p_vaddr=vaddr_text,
|
||||
p_filesz=len(text),
|
||||
p_memsz=len(text)
|
||||
)
|
||||
|
||||
ph2 = phdr(
|
||||
p_type=1,
|
||||
p_flags=6,
|
||||
p_offset=offset_data,
|
||||
p_vaddr=vaddr_data,
|
||||
p_filesz=len(data),
|
||||
p_memsz=len(data)
|
||||
)
|
||||
|
||||
phdrs = ph0 + ph1 + ph2
|
||||
elf_data[elf_header_size : elf_header_size + len(phdrs)] = phdrs
|
||||
|
||||
# === 3. Секции ===
|
||||
elf_data[offset_text : offset_text + len(text)] = text
|
||||
elf_data[offset_data : offset_data + len(data)] = data
|
||||
elf_data[shstrtab_offset : shstrtab_offset + shstrtab_size] = shstrtab_content
|
||||
|
||||
# === 4. Section Headers ===
|
||||
def shdr(name_idx, sh_type, flags, addr, offset, size, addralign=1):
|
||||
return struct.pack('<IIQQQQIIQQ',
|
||||
name_idx, sh_type, flags, addr, offset, size,
|
||||
0, 0, addralign, 0
|
||||
)
|
||||
|
||||
# Индексы: ""=0, ".text"=1, ".data"=7, ".shstrtab"=13
|
||||
sh0 = shdr(0, 0, 0, 0, 0, 0)
|
||||
sh1 = shdr(1, 1, 6, vaddr_text, offset_text, len(text), 16)
|
||||
sh2 = shdr(7, 1, 3, vaddr_data, offset_data, len(data), 8)
|
||||
sh3 = shdr(13, 3, 0, 0, shstrtab_offset, shstrtab_size, 1)
|
||||
|
||||
shdrs = sh0 + sh1 + sh2 + sh3
|
||||
elf_data[shdr_offset : shdr_offset + len(shdrs)] = shdrs
|
||||
|
||||
# === Запись ===
|
||||
with open(filename, "wb") as f:
|
||||
f.write(elf_data)
|
||||
|
||||
import os
|
||||
os.chmod(filename, 0o755)
|
||||
|
||||
# === Основной запуск ===
|
||||
|
||||
if len(sys.argv) != 2:
|
||||
print("Использование: python minimal.py <файл.квс>")
|
||||
sys.exit(1)
|
||||
|
||||
source_file = sys.argv[1]
|
||||
source_filename = source_file
|
||||
if not source_file.endswith('.квс'):
|
||||
print("Ошибка: файл должен иметь расширение .квс")
|
||||
sys.exit(1)
|
||||
|
||||
f_in = None
|
||||
try:
|
||||
f_in = open(source_file, "r", encoding="utf-8")
|
||||
source = f_in.read()
|
||||
f_in.close()
|
||||
except FileNotFoundError:
|
||||
safe_close(f_in)
|
||||
print(f"Ошибка: файл '{source_file}' не найден.")
|
||||
sys.exit(1)
|
||||
except UnicodeDecodeError as e:
|
||||
safe_close(f_in)
|
||||
print(f"Ошибка кодировки в файле '{source_file}': {e}")
|
||||
sys.exit(1)
|
||||
|
||||
try:
|
||||
parse(source)
|
||||
elf_file = source_file[:-4] + ".elf"
|
||||
create_elf(elf_file)
|
||||
print(f"ELF-файл создан: {elf_file}")
|
||||
print("Запустить: ./" + elf_file)
|
||||
print(f"Проверить: objdump -D {elf_file}")
|
||||
print(f"Проверить: eu-readelf -a {elf_file}")
|
||||
print(f"Проверить: readelf -l {elf_file}")
|
||||
print(f"Проверить: file {elf_file}")
|
||||
except SystemExit:
|
||||
raise
|
||||
except Exception as e:
|
||||
print(f"Внутренняя ошибка ассемблера: {e}", file=sys.stderr)
|
||||
sys.exit(1)
|
||||
Loading…
Reference in a new issue