third_party/tcmalloc/chromium/src/symbolize.cc

   1 // Copyright (c) 2009, 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 // ---
  31 // Author: Craig Silverstein
  32 //
  33 // This forks out to pprof to do the actual symbolizing.  We might
  34 // be better off writing our own in C++.
  35
  36 #include "config.h"
  37 #include "symbolize.h"
  38 #include <stdlib.h>
  39 #ifdef HAVE_UNISTD_H
  40 #include <unistd.h>   // for write()
  41 #endif
  42 #ifdef HAVE_SYS_SOCKET_H
  43 #include <sys/socket.h>   // for socketpair() -- needed by Symbolize
  44 #endif
  45 #ifdef HAVE_SYS_WAIT_H
  46 #include <sys/wait.h>   // for wait() -- needed by Symbolize
  47 #endif
  48 #ifdef HAVE_POLL_H
  49 #include <poll.h>
  50 #endif
  51 #ifdef __MACH__
  52 #include <mach-o/dyld.h>   // for GetProgramInvocationName()
  53 #include <limits.h>        // for PATH_MAX
  54 #endif
  55 #if defined(__CYGWIN__) || defined(__CYGWIN32__)
  56 #include <io.h>            // for get_osfhandle()
  57 #endif
  58 #include <string>
  59 #include "base/commandlineflags.h"
  60 #include "base/logging.h"
  61 #include "base/sysinfo.h"
  62
  63 using std::string;
  64 using tcmalloc::DumpProcSelfMaps;   // from sysinfo.h
  65
  66
  67 DEFINE_string(symbolize_pprof,
  68               EnvToString("PPROF_PATH", "pprof"),
  69               "Path to pprof to call for reporting function names.");
  70
  71 // Returns NULL if we're on an OS where we can't get the invocation name.
  72 // Using a static var is ok because we're not called from a thread.
  73 static char* GetProgramInvocationName() {
  74 #if defined(HAVE_PROGRAM_INVOCATION_NAME)
  75   extern char* program_invocation_name;  // gcc provides this
  76   return program_invocation_name;
  77 #elif defined(__MACH__)
  78   // We don't want to allocate memory for this since we may be
  79   // calculating it when memory is corrupted.
  80   static char program_invocation_name[PATH_MAX];
  81   if (program_invocation_name[0] == '\0') {  // first time calculating
  82     uint32_t length = sizeof(program_invocation_name);
  83     if (_NSGetExecutablePath(program_invocation_name, &length))
  84       return NULL;
  85   }
  86   return program_invocation_name;
  87 #else
  88   return NULL;   // figure out a way to get argv[0]
  89 #endif
  90 }
  91
  92 // Prints an error message when you can't run Symbolize().
  93 static void PrintError(const char* reason) {
  94   RAW_LOG(ERROR,
  95           "*** WARNING: Cannot convert addresses to symbols in output below.\n"
  96           "*** Reason: %s\n"
  97           "*** If you cannot fix this, try running pprof directly.\n",
  98           reason);
  99 }
 100
 101 void SymbolTable::Add(const void* addr) {
 102   symbolization_table_[addr] = "";
 103 }
 104
 105 const char* SymbolTable::GetSymbol(const void* addr) {
 106   return symbolization_table_[addr];
 107 }
 108
 109 // Updates symbolization_table with the pointers to symbol names corresponding
 110 // to its keys. The symbol names are stored in out, which is allocated and
 111 // freed by the caller of this routine.
 112 // Note that the forking/etc is not thread-safe or re-entrant.  That's
 113 // ok for the purpose we need -- reporting leaks detected by heap-checker
 114 // -- but be careful if you decide to use this routine for other purposes.
 115 // Returns number of symbols read on error.  If can't symbolize, returns 0
 116 // and emits an error message about why.
 117 int SymbolTable::Symbolize() {
 118 #if !defined(HAVE_UNISTD_H)  || !defined(HAVE_SYS_SOCKET_H) || !defined(HAVE_SYS_WAIT_H)
 119   PrintError("Perftools does not know how to call a sub-process on this O/S");
 120   return 0;
 121 #else
 122   const char* argv0 = GetProgramInvocationName();
 123   if (argv0 == NULL) {  // can't call symbolize if we can't figure out our name
 124     PrintError("Cannot figure out the name of this executable (argv0)");
 125     return 0;
 126   }
 127   if (access(FLAGS_symbolize_pprof, R_OK) != 0) {
 128     PrintError("Cannot find 'pprof' (is PPROF_PATH set correctly?)");
 129     return 0;
 130   }
 131
 132   // All this work is to do two-way communication.  ugh.
 133   int *child_in = NULL;   // file descriptors
 134   int *child_out = NULL;  // for now, we don't worry about child_err
 135   int child_fds[5][2];    // socketpair may be called up to five times below
 136
 137   // The client program may close its stdin and/or stdout and/or stderr
 138   // thus allowing socketpair to reuse file descriptors 0, 1 or 2.
 139   // In this case the communication between the forked processes may be broken
 140   // if either the parent or the child tries to close or duplicate these
 141   // descriptors. The loop below produces two pairs of file descriptors, each
 142   // greater than 2 (stderr).
 143   for (int i = 0; i < 5; i++) {
 144     if (socketpair(AF_UNIX, SOCK_STREAM, 0, child_fds[i]) == -1) {
 145       for (int j = 0; j < i; j++) {
 146         close(child_fds[j][0]);
 147         close(child_fds[j][1]);
 148         PrintError("Cannot create a socket pair");
 149         return 0;
 150       }
 151     } else {
 152       if ((child_fds[i][0] > 2) && (child_fds[i][1] > 2)) {
 153         if (child_in == NULL) {
 154           child_in = child_fds[i];
 155         } else {
 156           child_out = child_fds[i];
 157           for (int j = 0; j < i; j++) {
 158             if (child_fds[j] == child_in) continue;
 159             close(child_fds[j][0]);
 160             close(child_fds[j][1]);
 161           }
 162           break;
 163         }
 164       }
 165     }
 166   }
 167
 168   switch (fork()) {
 169     case -1: {  // error
 170       close(child_in[0]);
 171       close(child_in[1]);
 172       close(child_out[0]);
 173       close(child_out[1]);
 174       PrintError("Unknown error calling fork()");
 175       return 0;
 176     }
 177     case 0: {  // child
 178       close(child_in[1]);   // child uses the 0's, parent uses the 1's
 179       close(child_out[1]);  // child uses the 0's, parent uses the 1's
 180       close(0);
 181       close(1);
 182       if (dup2(child_in[0], 0) == -1) _exit(1);
 183       if (dup2(child_out[0], 1) == -1) _exit(2);
 184       // Unset vars that might cause trouble when we fork
 185       unsetenv("CPUPROFILE");
 186       unsetenv("HEAPPROFILE");
 187       unsetenv("HEAPCHECK");
 188       unsetenv("PERFTOOLS_VERBOSE");
 189       execlp(FLAGS_symbolize_pprof, FLAGS_symbolize_pprof,
 190              "--symbols", argv0, NULL);
 191       _exit(3);  // if execvp fails, it's bad news for us
 192     }
 193     default: {  // parent
 194       close(child_in[0]);   // child uses the 0's, parent uses the 1's
 195       close(child_out[0]);  // child uses the 0's, parent uses the 1's
 196 #ifdef HAVE_POLL_H
 197       // Waiting for 1ms seems to give the OS time to notice any errors.
 198       poll(0, 0, 1);
 199       // For maximum safety, we check to make sure the execlp
 200       // succeeded before trying to write.  (Otherwise we'll get a
 201       // SIGPIPE.)  For systems without poll.h, we'll just skip this
 202       // check, and trust that the user set PPROF_PATH correctly!
 203       struct pollfd pfd = { child_in[1], POLLOUT, 0 };
 204       if (!poll(&pfd, 1, 0) || !(pfd.revents & POLLOUT) ||
 205           (pfd.revents & (POLLHUP|POLLERR))) {
 206         PrintError("Cannot run 'pprof' (is PPROF_PATH set correctly?)");
 207         return 0;
 208       }
 209 #endif
 210 #if defined(__CYGWIN__) || defined(__CYGWIN32__)
 211       // On cygwin, DumpProcSelfMaps() takes a HANDLE, not an fd.  Convert.
 212       const HANDLE symbols_handle = (HANDLE) get_osfhandle(child_in[1]);
 213       DumpProcSelfMaps(symbols_handle);
 214 #else
 215       DumpProcSelfMaps(child_in[1]);  // what pprof expects on stdin
 216 #endif
 217
 218       // Allocate 24 bytes = ("0x" + 8 bytes + "\n" + overhead) for each
 219       // address to feed to pprof.
 220       const int kOutBufSize = 24 * symbolization_table_.size();
 221       char *pprof_buffer = new char[kOutBufSize];
 222       int written = 0;
 223       for (SymbolMap::const_iterator iter = symbolization_table_.begin();
 224            iter != symbolization_table_.end(); ++iter) {
 225         written += snprintf(pprof_buffer + written, kOutBufSize - written,
 226                  // pprof expects format to be 0xXXXXXX
 227                  "0x%" PRIxPTR "\n", reinterpret_cast<uintptr_t>(iter->first));
 228       }
 229       write(child_in[1], pprof_buffer, strlen(pprof_buffer));
 230       close(child_in[1]);             // that's all we need to write
 231
 232       const int kSymbolBufferSize = kSymbolSize * symbolization_table_.size();
 233       int total_bytes_read = 0;
 234       delete[] symbol_buffer_;
 235       symbol_buffer_ = new char[kSymbolBufferSize];
 236       memset(symbol_buffer_, '\0', kSymbolBufferSize);
 237       while (1) {
 238         int bytes_read = read(child_out[1], symbol_buffer_ + total_bytes_read,
 239                               kSymbolBufferSize - total_bytes_read);
 240         if (bytes_read < 0) {
 241           close(child_out[1]);
 242           PrintError("Cannot read data from pprof");
 243           return 0;
 244         } else if (bytes_read == 0) {
 245           close(child_out[1]);
 246           wait(NULL);
 247           break;
 248         } else {
 249           total_bytes_read += bytes_read;
 250         }
 251       }
 252       // We have successfully read the output of pprof into out.  Make sure
 253       // the last symbol is full (we can tell because it ends with a \n).
 254       if (total_bytes_read == 0 || symbol_buffer_[total_bytes_read - 1] != '\n')
 255         return 0;
 256       // make the symbolization_table_ values point to the output vector
 257       SymbolMap::iterator fill = symbolization_table_.begin();
 258       int num_symbols = 0;
 259       const char *current_name = symbol_buffer_;
 260       for (int i = 0; i < total_bytes_read; i++) {
 261         if (symbol_buffer_[i] == '\n') {
 262           fill->second = current_name;
 263           symbol_buffer_[i] = '\0';
 264           current_name = symbol_buffer_ + i + 1;
 265           fill++;
 266           num_symbols++;
 267         }
 268       }
 269       return num_symbols;
 270     }
 271   }
 272   PrintError("Unkown error (should never occur!)");
 273   return 0;  // shouldn't be reachable
 274 #endif
 275 }