|
1 // Copyright (c) 2006, Google Inc. |
|
2 // All rights reserved. |
|
3 // |
|
4 // Redistribution and use in source and binary forms, with or without |
|
5 // modification, are permitted provided that the following conditions are |
|
6 // met: |
|
7 // |
|
8 // * Redistributions of source code must retain the above copyright |
|
9 // notice, this list of conditions and the following disclaimer. |
|
10 // * Redistributions in binary form must reproduce the above |
|
11 // copyright notice, this list of conditions and the following disclaimer |
|
12 // in the documentation and/or other materials provided with the |
|
13 // distribution. |
|
14 // * Neither the name of Google Inc. nor the names of its |
|
15 // contributors may be used to endorse or promote products derived from |
|
16 // this software without specific prior written permission. |
|
17 // |
|
18 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
|
19 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
|
20 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
|
21 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
|
22 // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
|
23 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
|
24 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
|
25 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
|
26 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
|
27 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
|
28 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
|
29 |
|
30 // macho_id.cc: Functions to gather identifying information from a macho file |
|
31 // |
|
32 // See macho_id.h for documentation |
|
33 // |
|
34 // Author: Dan Waylonis |
|
35 |
|
36 extern "C" { // necessary for Leopard |
|
37 #include <fcntl.h> |
|
38 #include <mach-o/loader.h> |
|
39 #include <mach-o/swap.h> |
|
40 #include <stdio.h> |
|
41 #include <stdlib.h> |
|
42 #include <string.h> |
|
43 #include <sys/time.h> |
|
44 #include <sys/types.h> |
|
45 #include <unistd.h> |
|
46 } |
|
47 |
|
48 #include "common/mac/macho_id.h" |
|
49 #include "common/mac/macho_walker.h" |
|
50 #include "common/mac/macho_utilities.h" |
|
51 |
|
52 namespace MacFileUtilities { |
|
53 |
|
54 using google_breakpad::MD5Init; |
|
55 using google_breakpad::MD5Update; |
|
56 using google_breakpad::MD5Final; |
|
57 |
|
58 MachoID::MachoID(const char *path) |
|
59 : memory_(0), |
|
60 memory_size_(0), |
|
61 crc_(0), |
|
62 md5_context_(), |
|
63 update_function_(NULL) { |
|
64 strlcpy(path_, path, sizeof(path_)); |
|
65 } |
|
66 |
|
67 MachoID::MachoID(const char *path, void *memory, size_t size) |
|
68 : memory_(memory), |
|
69 memory_size_(size), |
|
70 crc_(0), |
|
71 md5_context_(), |
|
72 update_function_(NULL) { |
|
73 strlcpy(path_, path, sizeof(path_)); |
|
74 } |
|
75 |
|
76 MachoID::~MachoID() { |
|
77 } |
|
78 |
|
79 // The CRC info is from http://en.wikipedia.org/wiki/Adler-32 |
|
80 // With optimizations from http://www.zlib.net/ |
|
81 |
|
82 // The largest prime smaller than 65536 |
|
83 #define MOD_ADLER 65521 |
|
84 // MAX_BLOCK is the largest n such that 255n(n+1)/2 + (n+1)(MAX_BLOCK-1) <= 2^32-1 |
|
85 #define MAX_BLOCK 5552 |
|
86 |
|
87 void MachoID::UpdateCRC(unsigned char *bytes, size_t size) { |
|
88 // Unrolled loops for summing |
|
89 #define DO1(buf,i) {sum1 += (buf)[i]; sum2 += sum1;} |
|
90 #define DO2(buf,i) DO1(buf,i); DO1(buf,i+1); |
|
91 #define DO4(buf,i) DO2(buf,i); DO2(buf,i+2); |
|
92 #define DO8(buf,i) DO4(buf,i); DO4(buf,i+4); |
|
93 #define DO16(buf) DO8(buf,0); DO8(buf,8); |
|
94 // Split up the crc |
|
95 uint32_t sum1 = crc_ & 0xFFFF; |
|
96 uint32_t sum2 = (crc_ >> 16) & 0xFFFF; |
|
97 |
|
98 // Do large blocks |
|
99 while (size >= MAX_BLOCK) { |
|
100 size -= MAX_BLOCK; |
|
101 int block_count = MAX_BLOCK / 16; |
|
102 do { |
|
103 DO16(bytes); |
|
104 bytes += 16; |
|
105 } while (--block_count); |
|
106 sum1 %= MOD_ADLER; |
|
107 sum2 %= MOD_ADLER; |
|
108 } |
|
109 |
|
110 // Do remaining bytes |
|
111 if (size) { |
|
112 while (size >= 16) { |
|
113 size -= 16; |
|
114 DO16(bytes); |
|
115 bytes += 16; |
|
116 } |
|
117 while (size--) { |
|
118 sum1 += *bytes++; |
|
119 sum2 += sum1; |
|
120 } |
|
121 sum1 %= MOD_ADLER; |
|
122 sum2 %= MOD_ADLER; |
|
123 crc_ = (sum2 << 16) | sum1; |
|
124 } |
|
125 } |
|
126 |
|
127 void MachoID::UpdateMD5(unsigned char *bytes, size_t size) { |
|
128 MD5Update(&md5_context_, bytes, size); |
|
129 } |
|
130 |
|
131 void MachoID::Update(MachoWalker *walker, off_t offset, size_t size) { |
|
132 if (!update_function_ || !size) |
|
133 return; |
|
134 |
|
135 // Read up to 4k bytes at a time |
|
136 unsigned char buffer[4096]; |
|
137 size_t buffer_size; |
|
138 off_t file_offset = offset; |
|
139 while (size > 0) { |
|
140 if (size > sizeof(buffer)) { |
|
141 buffer_size = sizeof(buffer); |
|
142 size -= buffer_size; |
|
143 } else { |
|
144 buffer_size = size; |
|
145 size = 0; |
|
146 } |
|
147 |
|
148 if (!walker->ReadBytes(buffer, buffer_size, file_offset)) |
|
149 return; |
|
150 |
|
151 (this->*update_function_)(buffer, buffer_size); |
|
152 file_offset += buffer_size; |
|
153 } |
|
154 } |
|
155 |
|
156 bool MachoID::UUIDCommand(cpu_type_t cpu_type, |
|
157 cpu_subtype_t cpu_subtype, |
|
158 unsigned char bytes[16]) { |
|
159 struct breakpad_uuid_command uuid_cmd; |
|
160 uuid_cmd.cmd = 0; |
|
161 if (!WalkHeader(cpu_type, cpu_subtype, UUIDWalkerCB, &uuid_cmd)) |
|
162 return false; |
|
163 |
|
164 // If we found the command, we'll have initialized the uuid_command |
|
165 // structure |
|
166 if (uuid_cmd.cmd == LC_UUID) { |
|
167 memcpy(bytes, uuid_cmd.uuid, sizeof(uuid_cmd.uuid)); |
|
168 return true; |
|
169 } |
|
170 |
|
171 return false; |
|
172 } |
|
173 |
|
174 bool MachoID::IDCommand(cpu_type_t cpu_type, |
|
175 cpu_subtype_t cpu_subtype, |
|
176 unsigned char identifier[16]) { |
|
177 struct dylib_command dylib_cmd; |
|
178 dylib_cmd.cmd = 0; |
|
179 if (!WalkHeader(cpu_type, cpu_subtype, IDWalkerCB, &dylib_cmd)) |
|
180 return false; |
|
181 |
|
182 // If we found the command, we'll have initialized the dylib_command |
|
183 // structure |
|
184 if (dylib_cmd.cmd == LC_ID_DYLIB) { |
|
185 // Take the hashed filename, version, and compatability version bytes |
|
186 // to form the first 12 bytes, pad the rest with zeros |
|
187 |
|
188 // create a crude hash of the filename to generate the first 4 bytes |
|
189 identifier[0] = 0; |
|
190 identifier[1] = 0; |
|
191 identifier[2] = 0; |
|
192 identifier[3] = 0; |
|
193 |
|
194 for (int j = 0, i = (int)strlen(path_)-1; i>=0 && path_[i]!='/'; ++j, --i) { |
|
195 identifier[j%4] += path_[i]; |
|
196 } |
|
197 |
|
198 identifier[4] = (dylib_cmd.dylib.current_version >> 24) & 0xFF; |
|
199 identifier[5] = (dylib_cmd.dylib.current_version >> 16) & 0xFF; |
|
200 identifier[6] = (dylib_cmd.dylib.current_version >> 8) & 0xFF; |
|
201 identifier[7] = dylib_cmd.dylib.current_version & 0xFF; |
|
202 identifier[8] = (dylib_cmd.dylib.compatibility_version >> 24) & 0xFF; |
|
203 identifier[9] = (dylib_cmd.dylib.compatibility_version >> 16) & 0xFF; |
|
204 identifier[10] = (dylib_cmd.dylib.compatibility_version >> 8) & 0xFF; |
|
205 identifier[11] = dylib_cmd.dylib.compatibility_version & 0xFF; |
|
206 identifier[12] = (cpu_type >> 24) & 0xFF; |
|
207 identifier[13] = (cpu_type >> 16) & 0xFF; |
|
208 identifier[14] = (cpu_type >> 8) & 0xFF; |
|
209 identifier[15] = cpu_type & 0xFF; |
|
210 |
|
211 return true; |
|
212 } |
|
213 |
|
214 return false; |
|
215 } |
|
216 |
|
217 uint32_t MachoID::Adler32(cpu_type_t cpu_type, cpu_subtype_t cpu_subtype) { |
|
218 update_function_ = &MachoID::UpdateCRC; |
|
219 crc_ = 0; |
|
220 |
|
221 if (!WalkHeader(cpu_type, cpu_subtype, WalkerCB, this)) |
|
222 return 0; |
|
223 |
|
224 return crc_; |
|
225 } |
|
226 |
|
227 bool MachoID::MD5(cpu_type_t cpu_type, cpu_subtype_t cpu_subtype, unsigned char identifier[16]) { |
|
228 update_function_ = &MachoID::UpdateMD5; |
|
229 |
|
230 MD5Init(&md5_context_); |
|
231 |
|
232 if (!WalkHeader(cpu_type, cpu_subtype, WalkerCB, this)) |
|
233 return false; |
|
234 |
|
235 MD5Final(identifier, &md5_context_); |
|
236 return true; |
|
237 } |
|
238 |
|
239 bool MachoID::WalkHeader(cpu_type_t cpu_type, |
|
240 cpu_subtype_t cpu_subtype, |
|
241 MachoWalker::LoadCommandCallback callback, |
|
242 void *context) { |
|
243 if (memory_) { |
|
244 MachoWalker walker(memory_, memory_size_, callback, context); |
|
245 return walker.WalkHeader(cpu_type, cpu_subtype); |
|
246 } else { |
|
247 MachoWalker walker(path_, callback, context); |
|
248 return walker.WalkHeader(cpu_type, cpu_subtype); |
|
249 } |
|
250 } |
|
251 |
|
252 // static |
|
253 bool MachoID::WalkerCB(MachoWalker *walker, load_command *cmd, off_t offset, |
|
254 bool swap, void *context) { |
|
255 MachoID *macho_id = (MachoID *)context; |
|
256 |
|
257 if (cmd->cmd == LC_SEGMENT) { |
|
258 struct segment_command seg; |
|
259 |
|
260 if (!walker->ReadBytes(&seg, sizeof(seg), offset)) |
|
261 return false; |
|
262 |
|
263 if (swap) |
|
264 swap_segment_command(&seg, NXHostByteOrder()); |
|
265 |
|
266 struct mach_header_64 header; |
|
267 off_t header_offset; |
|
268 |
|
269 if (!walker->CurrentHeader(&header, &header_offset)) |
|
270 return false; |
|
271 |
|
272 // Process segments that have sections: |
|
273 // (e.g., __TEXT, __DATA, __IMPORT, __OBJC) |
|
274 offset += sizeof(struct segment_command); |
|
275 struct section sec; |
|
276 for (unsigned long i = 0; i < seg.nsects; ++i) { |
|
277 if (!walker->ReadBytes(&sec, sizeof(sec), offset)) |
|
278 return false; |
|
279 |
|
280 if (swap) |
|
281 swap_section(&sec, 1, NXHostByteOrder()); |
|
282 |
|
283 // sections of type S_ZEROFILL are "virtual" and contain no data |
|
284 // in the file itself |
|
285 if ((sec.flags & SECTION_TYPE) != S_ZEROFILL && sec.offset != 0) |
|
286 macho_id->Update(walker, header_offset + sec.offset, sec.size); |
|
287 |
|
288 offset += sizeof(struct section); |
|
289 } |
|
290 } else if (cmd->cmd == LC_SEGMENT_64) { |
|
291 struct segment_command_64 seg64; |
|
292 |
|
293 if (!walker->ReadBytes(&seg64, sizeof(seg64), offset)) |
|
294 return false; |
|
295 |
|
296 if (swap) |
|
297 breakpad_swap_segment_command_64(&seg64, NXHostByteOrder()); |
|
298 |
|
299 struct mach_header_64 header; |
|
300 off_t header_offset; |
|
301 |
|
302 if (!walker->CurrentHeader(&header, &header_offset)) |
|
303 return false; |
|
304 |
|
305 // Process segments that have sections: |
|
306 // (e.g., __TEXT, __DATA, __IMPORT, __OBJC) |
|
307 offset += sizeof(struct segment_command_64); |
|
308 struct section_64 sec64; |
|
309 for (unsigned long i = 0; i < seg64.nsects; ++i) { |
|
310 if (!walker->ReadBytes(&sec64, sizeof(sec64), offset)) |
|
311 return false; |
|
312 |
|
313 if (swap) |
|
314 breakpad_swap_section_64(&sec64, 1, NXHostByteOrder()); |
|
315 |
|
316 // sections of type S_ZEROFILL are "virtual" and contain no data |
|
317 // in the file itself |
|
318 if ((sec64.flags & SECTION_TYPE) != S_ZEROFILL && sec64.offset != 0) |
|
319 macho_id->Update(walker, |
|
320 header_offset + sec64.offset, |
|
321 (size_t)sec64.size); |
|
322 |
|
323 offset += sizeof(struct section_64); |
|
324 } |
|
325 } |
|
326 |
|
327 // Continue processing |
|
328 return true; |
|
329 } |
|
330 |
|
331 // static |
|
332 bool MachoID::UUIDWalkerCB(MachoWalker *walker, load_command *cmd, off_t offset, |
|
333 bool swap, void *context) { |
|
334 if (cmd->cmd == LC_UUID) { |
|
335 struct breakpad_uuid_command *uuid_cmd = |
|
336 (struct breakpad_uuid_command *)context; |
|
337 |
|
338 if (!walker->ReadBytes(uuid_cmd, sizeof(struct breakpad_uuid_command), |
|
339 offset)) |
|
340 return false; |
|
341 |
|
342 if (swap) |
|
343 breakpad_swap_uuid_command(uuid_cmd, NXHostByteOrder()); |
|
344 |
|
345 return false; |
|
346 } |
|
347 |
|
348 // Continue processing |
|
349 return true; |
|
350 } |
|
351 |
|
352 // static |
|
353 bool MachoID::IDWalkerCB(MachoWalker *walker, load_command *cmd, off_t offset, |
|
354 bool swap, void *context) { |
|
355 if (cmd->cmd == LC_ID_DYLIB) { |
|
356 struct dylib_command *dylib_cmd = (struct dylib_command *)context; |
|
357 |
|
358 if (!walker->ReadBytes(dylib_cmd, sizeof(struct dylib_command), offset)) |
|
359 return false; |
|
360 |
|
361 if (swap) |
|
362 swap_dylib_command(dylib_cmd, NXHostByteOrder()); |
|
363 |
|
364 return false; |
|
365 } |
|
366 |
|
367 // Continue processing |
|
368 return true; |
|
369 } |
|
370 |
|
371 } // namespace MacFileUtilities |