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
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
|
#!/usr/bin/env python3
# SPDX-License-Identifier: GPL-2.0-or-later
# Copyright (C) 2026 Richard Qin
import argparse
import struct
import sys
MAGIC = 0x4C554D45434F4D50 # "LUMECOMP"
COMP_NONE = 0
COMP_LZ4 = 1
HDR_FMT = "<QIHHIIQQQQQQII"
HDR_SIZE = struct.calcsize(HDR_FMT)
HDR_OFFSET = 64
def crc32_update(crc: int, data: bytes) -> int:
for b in data:
crc ^= b
for _ in range(8):
mask = -(crc & 1) & 0xFFFFFFFF
crc = ((crc >> 1) ^ (0xEDB88320 & mask)) & 0xFFFFFFFF
return crc
def crc32_ieee(data: bytes) -> int:
crc = 0xFFFFFFFF
crc = crc32_update(crc, data)
return (crc ^ 0xFFFFFFFF) & 0xFFFFFFFF
def lz4_compress(data: bytes) -> bytes:
n = len(data)
if n == 0:
return b""
min_match = 4
hash_log = 16
hash_size = 1 << hash_log
max_offset = 65535
def hash4(b0, b1, b2, b3):
v = b0 | (b1 << 8) | (b2 << 16) | (b3 << 24)
return (v * 2654435761) & (hash_size - 1)
htab = [-1] * hash_size
out = bytearray()
anchor = 0
i = 0
while i + min_match <= n:
h = hash4(data[i], data[i + 1], data[i + 2], data[i + 3])
ref = htab[h]
htab[h] = i
if ref >= 0 and i - ref <= max_offset and data[ref:ref + min_match] == data[i:i + min_match]:
match_len = min_match
while i + match_len < n and data[ref + match_len] == data[i + match_len]:
match_len += 1
lit_len = i - anchor
token = (15 if lit_len >= 15 else lit_len) << 4
match_field = match_len - min_match
token |= 15 if match_field >= 15 else match_field
out.append(token)
if lit_len >= 15:
l = lit_len - 15
while l >= 255:
out.append(255)
l -= 255
out.append(l)
out.extend(data[anchor:i])
offset = i - ref
out.append(offset & 0xFF)
out.append((offset >> 8) & 0xFF)
if match_field >= 15:
l = match_field - 15
while l >= 255:
out.append(255)
l -= 255
out.append(l)
i += match_len
anchor = i
continue
i += 1
lit_len = n - anchor
token = (15 if lit_len >= 15 else lit_len) << 4
out.append(token)
if lit_len >= 15:
l = lit_len - 15
while l >= 255:
out.append(255)
l -= 255
out.append(l)
out.extend(data[anchor:])
return bytes(out)
def get_header_offset(data: bytes) -> int:
if len(data) < HDR_OFFSET + HDR_SIZE:
return -1
if data[HDR_OFFSET:HDR_OFFSET + 8] != struct.pack("<Q", MAGIC):
return -1
return HDR_OFFSET
def read_struct(fmt, data, offset):
size = struct.calcsize(fmt)
return struct.unpack_from(fmt, data, offset), size
def parse_elf_symbols(path: str):
data = open(path, "rb").read()
if data[:4] != b"\x7fELF":
return {}
if data[4] != 2 or data[5] != 1:
return {}
(e_ident, e_type, e_machine, e_version, e_entry, e_phoff, e_shoff,
e_flags, e_ehsize, e_phentsize, e_phnum, e_shentsize, e_shnum,
e_shstrndx), _ = read_struct("<16sHHIQQQIHHHHHH", data, 0)
symbols = {}
for i in range(e_shnum):
off = e_shoff + i * e_shentsize
(sh_name, sh_type, sh_flags, sh_addr, sh_offset, sh_size,
sh_link, sh_info, sh_addralign, sh_entsize), _ = read_struct(
"<IIQQQQIIQQ", data, off)
if sh_type != 2:
continue
if sh_link >= e_shnum:
continue
str_off = e_shoff + sh_link * e_shentsize
(sname, stype, sflags, saddr, soff, ssize, slink, sinfo,
salign, sentsz), _ = read_struct("<IIQQQQIIQQ", data, str_off)
strtab = data[soff:soff + ssize]
entsize = sh_entsize or 24
count = sh_size // entsize
for j in range(count):
ent_off = sh_offset + j * entsize
(st_name, st_info, st_other, st_shndx, st_value, st_size), _ = read_struct(
"<IBBHQQ", data, ent_off)
if st_name == 0:
continue
end = strtab.find(b"\x00", st_name)
if end < 0:
continue
name = strtab[st_name:end].decode("ascii", errors="ignore")
if name:
symbols[name] = st_value
break
return symbols
def main():
ap = argparse.ArgumentParser()
ap.add_argument("--elf")
ap.add_argument("--input", required=True)
ap.add_argument("--output", required=True)
ap.add_argument("--compress", choices=["none", "lz4"], default="none")
args = ap.parse_args()
raw = open(args.input, "rb").read()
hdr_off = get_header_offset(raw)
if hdr_off < 0:
raise ValueError("compressed header magic not found")
header = list(struct.unpack_from(HDR_FMT, raw, hdr_off))
if header[0] != MAGIC:
raise ValueError("bad header magic")
if header[2] < HDR_SIZE:
raise ValueError("header_size too small")
if args.compress == "lz4":
payload = lz4_compress(raw)
comp = COMP_LZ4
else:
payload = b""
comp = COMP_NONE
uncompressed_size = len(raw)
compressed_size = len(payload)
payload_offset = (len(raw) + 7) & ~7 if payload else 0
if payload and args.elf:
syms = parse_elf_symbols(args.elf)
link_base = syms.get("_start")
bss_end = syms.get("_ebss")
if link_base is not None and bss_end is not None and bss_end >= link_base:
payload_offset = (bss_end - link_base + 7) & ~7
header[4] = comp
header[6] = uncompressed_size
header[7] = compressed_size
header[8] = 0 # load_phys_base auto
header[11] = payload_offset
header[12] = 0
header[13] = crc32_ieee(payload) if payload else 0
header_bytes = struct.pack(HDR_FMT, *header)
header[12] = crc32_ieee(header_bytes)
out = bytearray(raw)
struct.pack_into(HDR_FMT, out, hdr_off, *header)
if payload:
if payload_offset > len(out):
out.extend(b"\x00" * (payload_offset - len(out)))
out.extend(payload)
with open(args.output, "wb") as f:
f.write(out)
return 0
if __name__ == "__main__":
try:
sys.exit(main())
except Exception as e:
print("pack_image: " + str(e), file=sys.stderr)
sys.exit(1)
|
