/thirdparty/breakpad/common/mac/macho_id.cc

  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
 36extern "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
 52namespace MacFileUtilities {
 53
 54using google_breakpad::MD5Init;
 55using google_breakpad::MD5Update;
 56using google_breakpad::MD5Final;
 57
 58MachoID::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
 67MachoID::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
 76MachoID::~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
 87void 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
127void MachoID::UpdateMD5(unsigned char *bytes, size_t size) {
128  MD5Update(&md5_context_, bytes, size);
129}
130
131void 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
156bool MachoID::UUIDCommand(int cpu_type, unsigned char bytes[16]) {
157  struct breakpad_uuid_command uuid_cmd;
158  uuid_cmd.cmd = 0;
159  if (!WalkHeader(cpu_type, UUIDWalkerCB, &uuid_cmd))
160    return false;
161
162  // If we found the command, we'll have initialized the uuid_command
163  // structure
164  if (uuid_cmd.cmd == LC_UUID) {
165    memcpy(bytes, uuid_cmd.uuid, sizeof(uuid_cmd.uuid));
166    return true;
167  }
168
169  return false;
170}
171
172bool MachoID::IDCommand(int cpu_type, unsigned char identifier[16]) {
173  struct dylib_command dylib_cmd;
174  dylib_cmd.cmd = 0;
175  if (!WalkHeader(cpu_type, IDWalkerCB, &dylib_cmd))
176    return false;
177
178  // If we found the command, we'll have initialized the dylib_command
179  // structure
180  if (dylib_cmd.cmd == LC_ID_DYLIB) {
181    // Take the hashed filename, version, and compatability version bytes
182    // to form the first 12 bytes, pad the rest with zeros
183
184    // create a crude hash of the filename to generate the first 4 bytes
185    identifier[0] = 0;
186    identifier[1] = 0;
187    identifier[2] = 0;
188    identifier[3] = 0;
189
190    for (int j = 0, i = (int)strlen(path_)-1; i>=0 && path_[i]!='/'; ++j, --i) {
191      identifier[j%4] += path_[i];
192    }
193
194    identifier[4] = (dylib_cmd.dylib.current_version >> 24) & 0xFF;
195    identifier[5] = (dylib_cmd.dylib.current_version >> 16) & 0xFF;
196    identifier[6] = (dylib_cmd.dylib.current_version >> 8) & 0xFF;
197    identifier[7] = dylib_cmd.dylib.current_version & 0xFF;
198    identifier[8] = (dylib_cmd.dylib.compatibility_version >> 24) & 0xFF;
199    identifier[9] = (dylib_cmd.dylib.compatibility_version >> 16) & 0xFF;
200    identifier[10] = (dylib_cmd.dylib.compatibility_version >> 8) & 0xFF;
201    identifier[11] = dylib_cmd.dylib.compatibility_version & 0xFF;
202    identifier[12] = (cpu_type >> 24) & 0xFF;
203    identifier[13] = (cpu_type >> 16) & 0xFF;
204    identifier[14] = (cpu_type >> 8) & 0xFF;
205    identifier[15] = cpu_type & 0xFF;
206
207    return true;
208  }
209
210  return false;
211}
212
213uint32_t MachoID::Adler32(int cpu_type) {
214  update_function_ = &MachoID::UpdateCRC;
215  crc_ = 0;
216
217  if (!WalkHeader(cpu_type, WalkerCB, this))
218    return 0;
219
220  return crc_;
221}
222
223bool MachoID::MD5(int cpu_type, unsigned char identifier[16]) {
224  update_function_ = &MachoID::UpdateMD5;
225
226  MD5Init(&md5_context_);
227
228  if (!WalkHeader(cpu_type, WalkerCB, this))
229    return false;
230
231  MD5Final(identifier, &md5_context_);
232  return true;
233}
234
235bool MachoID::WalkHeader(int cpu_type,
236                         MachoWalker::LoadCommandCallback callback,
237                         void *context) {
238  if (memory_) {
239    MachoWalker walker(memory_, memory_size_, callback, context);
240    return walker.WalkHeader(cpu_type);
241  } else {
242    MachoWalker walker(path_, callback, context);
243    return walker.WalkHeader(cpu_type);
244  }
245}
246
247// static
248bool MachoID::WalkerCB(MachoWalker *walker, load_command *cmd, off_t offset,
249                       bool swap, void *context) {
250  MachoID *macho_id = (MachoID *)context;
251
252  if (cmd->cmd == LC_SEGMENT) {
253    struct segment_command seg;
254
255    if (!walker->ReadBytes(&seg, sizeof(seg), offset))
256      return false;
257
258    if (swap)
259      swap_segment_command(&seg, NXHostByteOrder());
260
261    struct mach_header_64 header;
262    off_t header_offset;
263    
264    if (!walker->CurrentHeader(&header, &header_offset))
265      return false;
266        
267    // Process segments that have sections:
268    // (e.g., __TEXT, __DATA, __IMPORT, __OBJC)
269    offset += sizeof(struct segment_command);
270    struct section sec;
271    for (unsigned long i = 0; i < seg.nsects; ++i) {
272      if (!walker->ReadBytes(&sec, sizeof(sec), offset))
273        return false;
274
275      if (swap)
276        swap_section(&sec, 1, NXHostByteOrder());
277
278      // sections of type S_ZEROFILL are "virtual" and contain no data
279      // in the file itself
280      if ((sec.flags & SECTION_TYPE) != S_ZEROFILL && sec.offset != 0)
281        macho_id->Update(walker, header_offset + sec.offset, sec.size);
282
283      offset += sizeof(struct section);
284    }
285  } else if (cmd->cmd == LC_SEGMENT_64) {
286    struct segment_command_64 seg64;
287
288    if (!walker->ReadBytes(&seg64, sizeof(seg64), offset))
289      return false;
290
291    if (swap)
292      breakpad_swap_segment_command_64(&seg64, NXHostByteOrder());
293
294    struct mach_header_64 header;
295    off_t header_offset;
296    
297    if (!walker->CurrentHeader(&header, &header_offset))
298      return false;
299    
300    // Process segments that have sections:
301    // (e.g., __TEXT, __DATA, __IMPORT, __OBJC)
302    offset += sizeof(struct segment_command_64);
303    struct section_64 sec64;
304    for (unsigned long i = 0; i < seg64.nsects; ++i) {
305      if (!walker->ReadBytes(&sec64, sizeof(sec64), offset))
306        return false;
307
308      if (swap)
309        breakpad_swap_section_64(&sec64, 1, NXHostByteOrder());
310
311      // sections of type S_ZEROFILL are "virtual" and contain no data
312      // in the file itself
313      if ((sec64.flags & SECTION_TYPE) != S_ZEROFILL && sec64.offset != 0)
314        macho_id->Update(walker, 
315                         header_offset + sec64.offset, 
316                         (size_t)sec64.size);
317
318      offset += sizeof(struct section_64);
319    }
320  }
321
322  // Continue processing
323  return true;
324}
325
326// static
327bool MachoID::UUIDWalkerCB(MachoWalker *walker, load_command *cmd, off_t offset,
328                           bool swap, void *context) {
329  if (cmd->cmd == LC_UUID) {
330    struct breakpad_uuid_command *uuid_cmd =
331      (struct breakpad_uuid_command *)context;
332
333    if (!walker->ReadBytes(uuid_cmd, sizeof(struct breakpad_uuid_command),
334                           offset))
335      return false;
336
337    if (swap)
338      breakpad_swap_uuid_command(uuid_cmd, NXHostByteOrder());
339
340    return false;
341  }
342
343  // Continue processing
344  return true;
345}
346
347// static
348bool MachoID::IDWalkerCB(MachoWalker *walker, load_command *cmd, off_t offset,
349                         bool swap, void *context) {
350  if (cmd->cmd == LC_ID_DYLIB) {
351    struct dylib_command *dylib_cmd = (struct dylib_command *)context;
352
353    if (!walker->ReadBytes(dylib_cmd, sizeof(struct dylib_command), offset))
354      return false;
355
356    if (swap)
357      swap_dylib_command(dylib_cmd, NXHostByteOrder());
358
359    return false;
360  }
361
362  // Continue processing
363  return true;
364}
365
366}  // namespace MacFileUtilities