xref: /aosp_15_r20/system/libprocinfo/include/procinfo/process_map.h (revision e7c5e80fc9b28c04f5db9de8d2855377d05126c5)

/*
 * Copyright (C) 2018 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#pragma once

#include <inttypes.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mman.h>
#include <sys/types.h>

#include <functional>
#include <string>
#include <vector>

#include <android-base/file.h>
#include <android-base/strings.h>

namespace android {
namespace procinfo {

/*
 * The populated fields of MapInfo corresponds to the following fields of an entry
 * in /proc/<pid>/maps:
 *
 * <start>     -<end>         ...   <pgoff>        ...   <inode>    <name>
 * 790b07dc6000-790b07dd9000  r--p  00000000       fe:09 21068208   /system/lib64/foo.so
 *                               |
 *                               |
 *                               |___ p - private (!<shared>)
 *                                    s - <shared>
 */
struct MapInfo {
  uint64_t start;
  uint64_t end;
  // NOTE: It should not be assumed the virtual addresses in range [start,end] all
  //       correspond to valid offsets on the backing file.
  //       See: MappedFileSize().
  uint16_t flags;
  uint64_t pgoff;
  ino_t inode;
  std::string name;
  bool shared;

  // With MTE globals, segments are remapped as anonymous mappings. They're
  // named specifically to preserve offsets and as much of the basename as
  // possible. For example,
  // "[anon:mt:/data/local/tmp/debuggerd_test/arm64/debuggerd_test64+108000]" is
  // the name of anonymized mapping for debuggerd_test64 of the segment starting
  // at 0x108000. The kernel only supports 80 characters (excluding the '[anon:'
  // prefix and ']' suffix, but including the null terminator), and in those
  // instances, we maintain the offset and as much of the basename as possible
  // by left-truncation. For example:
  // "[anon:mt:/data/nativetest64/bionic-unit-tests/bionic-loader-test-libs/libdlext_test.so+e000]"
  // would become:
  // "[anon:mt:...ivetest64/bionic-unit-tests/bionic-loader-test-libs/libdlext_test.so+e000]".
  // For mappings under MTE globals, we thus post-process the name to extract the page offset, and
  // canonicalize the name.
  static constexpr const char* kMtePrefix = "[anon:mt:";
  static constexpr size_t kMtePrefixLength = sizeof(kMtePrefix) - 1;

  void MaybeExtractMemtagGlobalsInfo() {
    if (!this->name.starts_with(kMtePrefix)) return;
    if (this->name.back() != ']') return;

    size_t offset_to_plus = this->name.rfind('+');
    if (offset_to_plus == std::string::npos) return;
    if (sscanf(this->name.c_str() + offset_to_plus + 1, "%" SCNx64 "]", &this->pgoff) != 1) return;

    this->name =
        std::string(this->name.begin() + kMtePrefixLength + 2, this->name.begin() + offset_to_plus);
  }

  MapInfo(uint64_t start, uint64_t end, uint16_t flags, uint64_t pgoff, ino_t inode,
          const char* name, bool shared)
      : start(start),
        end(end),
        flags(flags),
        pgoff(pgoff),
        inode(inode),
        name(name),
        shared(shared) {
    MaybeExtractMemtagGlobalsInfo();
  }

  MapInfo(const MapInfo& params)
      : start(params.start),
        end(params.end),
        flags(params.flags),
        pgoff(params.pgoff),
        inode(params.inode),
        name(params.name),
        shared(params.shared) {}
};

typedef std::function<void(const MapInfo&)> MapInfoCallback;
typedef std::function<void(uint64_t start, uint64_t end, uint16_t flags, uint64_t pgoff,
                      ino_t inode, const char* name, bool shared)> MapInfoParamsCallback;

static inline bool PassSpace(char** p) {
  if (**p != ' ') {
    return false;
  }
  while (**p == ' ') {
    (*p)++;
  }
  return true;
}

static inline bool PassXdigit(char** p) {
  if (!isxdigit(**p)) {
    return false;
  }
  do {
    (*p)++;
  } while (isxdigit(**p));
  return true;
}

// Parses the given line p pointing at proc/<pid>/maps content buffer and returns true on success
// and false on failure parsing. The first new line character of line will be replaced by the
// null character and *next_line will point to the character after the null.
//
// Example of how a parsed line look line:
// 00400000-00409000 r-xp 00000000 fc:00 426998  /usr/lib/gvfs/gvfsd-http
static inline bool ParseMapsFileLine(char* p, uint64_t& start_addr, uint64_t& end_addr, uint16_t& flags,
                      uint64_t& pgoff, ino_t& inode, char** name, bool& shared, char** next_line) {
  // Make the first new line character null.
  *next_line = strchr(p, '\n');
  if (*next_line != nullptr) {
    **next_line = '\0';
    (*next_line)++;
  }

  char* end;
  // start_addr
  start_addr = strtoull(p, &end, 16);
  if (end == p || *end != '-') {
    return false;
  }
  p = end + 1;
  // end_addr
  end_addr = strtoull(p, &end, 16);
  if (end == p) {
    return false;
  }
  p = end;
  if (!PassSpace(&p)) {
    return false;
  }
  // flags
  flags = 0;
  if (*p == 'r') {
    flags |= PROT_READ;
  } else if (*p != '-') {
    return false;
  }
  p++;
  if (*p == 'w') {
    flags |= PROT_WRITE;
  } else if (*p != '-') {
    return false;
  }
  p++;
  if (*p == 'x') {
    flags |= PROT_EXEC;
  } else if (*p != '-') {
    return false;
  }
  p++;
  if (*p != 'p' && *p != 's') {
    return false;
  }
  shared = *p == 's';

  p++;
  if (!PassSpace(&p)) {
    return false;
  }
  // pgoff
  pgoff = strtoull(p, &end, 16);
  if (end == p) {
    return false;
  }
  p = end;
  if (!PassSpace(&p)) {
    return false;
  }
  // major:minor
  if (!PassXdigit(&p) || *p++ != ':' || !PassXdigit(&p) || !PassSpace(&p)) {
    return false;
  }
  // inode
  inode = strtoull(p, &end, 10);
  if (end == p) {
    return false;
  }
  p = end;

  if (*p != '\0' && !PassSpace(&p)) {
    return false;
  }

  // Assumes that the first new character was replaced with null.
  *name = p;

  return true;
}

inline bool ReadMapFileContent(char* content, const MapInfoParamsCallback& callback) {
  uint64_t start_addr;
  uint64_t end_addr;
  uint16_t flags;
  uint64_t pgoff;
  ino_t inode;
  char* line_start = content;
  char* next_line;
  char* name;
  bool shared;

  while (line_start != nullptr && *line_start != '\0') {
    bool parsed = ParseMapsFileLine(line_start, start_addr, end_addr, flags, pgoff,
                                    inode, &name, shared, &next_line);
    if (!parsed) {
      return false;
    }

    line_start = next_line;
    callback(start_addr, end_addr, flags, pgoff, inode, name, shared);
  }
  return true;
}

inline bool ReadMapFileContent(char* content, const MapInfoCallback& callback) {
  uint64_t start_addr;
  uint64_t end_addr;
  uint16_t flags;
  uint64_t pgoff;
  ino_t inode;
  char* line_start = content;
  char* next_line;
  char* name;
  bool shared;

  while (line_start != nullptr && *line_start != '\0') {
    bool parsed = ParseMapsFileLine(line_start, start_addr, end_addr, flags, pgoff,
                                    inode, &name, shared, &next_line);
    if (!parsed) {
      return false;
    }

    line_start = next_line;
    callback(MapInfo(start_addr, end_addr, flags, pgoff, inode, name, shared));
  }
  return true;
}

inline bool ReadMapFile(const std::string& map_file,
                const MapInfoCallback& callback) {
  std::string content;
  if (!android::base::ReadFileToString(map_file, &content)) {
    return false;
  }
  return ReadMapFileContent(&content[0], callback);
}


inline bool ReadMapFile(const std::string& map_file, const MapInfoParamsCallback& callback,
                        std::string& mapsBuffer) {
  if (!android::base::ReadFileToString(map_file, &mapsBuffer)) {
    return false;
  }
  return ReadMapFileContent(&mapsBuffer[0], callback);
}

inline bool ReadMapFile(const std::string& map_file,
                const MapInfoParamsCallback& callback) {
  std::string content;
  return ReadMapFile(map_file, callback, content);
}

inline bool ReadProcessMaps(pid_t pid, const MapInfoCallback& callback) {
  return ReadMapFile("/proc/" + std::to_string(pid) + "/maps", callback);
}

inline bool ReadProcessMaps(pid_t pid, const MapInfoParamsCallback& callback,
                            std::string& mapsBuffer) {
  return ReadMapFile("/proc/" + std::to_string(pid) + "/maps", callback, mapsBuffer);
}

inline bool ReadProcessMaps(pid_t pid, const MapInfoParamsCallback& callback) {
  std::string content;
  return ReadProcessMaps(pid, callback, content);
}

inline bool ReadProcessMaps(pid_t pid, std::vector<MapInfo>* maps) {
  return ReadProcessMaps(pid, [&](const MapInfo& mapinfo) { maps->emplace_back(mapinfo); });
}

// Reads maps file and executes given callback for each mapping
// Warning: buffer should not be modified asynchronously while this function executes
template <class CallbackType>
inline bool ReadMapFileAsyncSafe(const char* map_file, void* buffer, size_t buffer_size,
                                 const CallbackType& callback) {
  if (buffer == nullptr || buffer_size == 0) {
    return false;
  }

  int fd = open(map_file, O_RDONLY | O_CLOEXEC);
  if (fd == -1) {
    return false;
  }

  char* char_buffer = reinterpret_cast<char*>(buffer);
  size_t start = 0;
  size_t read_bytes = 0;
  char* line = nullptr;
  bool read_complete = false;
  while (true) {
    ssize_t bytes =
        TEMP_FAILURE_RETRY(read(fd, char_buffer + read_bytes, buffer_size - read_bytes - 1));
    if (bytes <= 0) {
      if (read_bytes == 0) {
        close(fd);
        return bytes == 0;
      }
      // Treat the last piece of data as the last line.
      char_buffer[start + read_bytes] = '\n';
      bytes = 1;
      read_complete = true;
    }
    read_bytes += bytes;

    while (read_bytes > 0) {
      char* newline = reinterpret_cast<char*>(memchr(&char_buffer[start], '\n', read_bytes));
      if (newline == nullptr) {
        break;
      }
      *newline = '\0';
      line = &char_buffer[start];
      start = newline - char_buffer + 1;
      read_bytes -= newline - line + 1;

      // Ignore the return code, errors are okay.
      ReadMapFileContent(line, callback);
    }

    if (read_complete) {
      close(fd);
      return true;
    }

    if (start == 0 && read_bytes == buffer_size - 1) {
      // The buffer provided is too small to contain this line, give up
      // and indicate failure.
      close(fd);
      return false;
    }

    // Copy any leftover data to the front  of the buffer.
    if (start > 0) {
      if (read_bytes > 0) {
        memmove(char_buffer, &char_buffer[start], read_bytes);
      }
      start = 0;
    }
  }
}

/**
 * A file memory mapping can be created such that it is only partially
 * backed by the underlying file. i.e. the mapping size is larger than
 * the file size.
 *
 * On builds that support larger than 4KB page-size, the common assumption
 * that a file mapping is entirely backed by the underlying file, is
 * more likely to be false.
 *
 * If an access to a region of the mapping beyond the end of the file
 * occurs, there are 2 situations:
 *     1) The access is between the end of the file and the next page
 *        boundary. The kernel will facilitate this although there is
 *        no file here.
 *        Note: That writing this region does not persist any data to
 *        the actual backing file.
 *     2) The access is beyond the first page boundary after the end
 *        of the file. This will cause a filemap_fault which does not
 *        correspond to a valid file offset and the kernel will return
 *        a SIGBUS.
 *        See return value SIGBUS at:
 *        https://man7.org/linux/man-pages/man2/mmap.2.html#RETURN_VALUE
 *
 * Userspace programs that parse /proc/<pid>/maps or /proc/<pid>/smaps
 * to determine the extent of memory mappings which they then use as
 * arguments to other syscalls or directly access; should be aware of
 * the second case above (2) and not assume that file mappings are
 * entirely back by the underlying file.
 *
 * This is especially important for operations that would cause a
 * page-fault on the range described in (2). In this case userspace
 * should either handle the signal or use the range backed by the
 * underlying file for the desired operation.
 *
 *
 * MappedFileSize() - Returns the size of the memory map backed
 *                    by the underlying file; or 0 if not file-backed.
 * @start_addr   - start address of the memory map.
 * @end_addr     - end address of the memory map.
 * @file_offset  - file offset of the backing file corresponding to the
 *                 start of the memory map.
 * @file_size    - size of the file (<file_path>) in bytes.
 *
 * NOTE: The arguments corresponds to the following fields of an entry
 * in /proc/<pid>/maps:
 *
 * <start_addr>-< end_addr >  ...   <file_offset>  ...   ...        <file_path>
 * 790b07dc6000-790b07dd9000  r--p  00000000       fe:09 21068208   /system/lib64/foo.so
 *
 * NOTE: Clients of this API should be aware that, although unlikely,
 * it is possible for @file_size to change under us and race with
 * the checks in MappedFileSize().
 * Users should avoid concurrent modifications of @file_size, or
 * use appropriate locking according to the usecase.
 */
inline uint64_t MappedFileSize(uint64_t start_addr, uint64_t end_addr,
                               uint64_t file_offset, uint64_t file_size) {
    uint64_t len = end_addr - start_addr;

    // This VMA may have been split from a larger file mapping; or the
    // file may have been resized since the mapping was created.
    if (file_offset > file_size) {
        return 0;
    }

    // Mapping exceeds file_size ?
    if ((file_offset + len) > file_size) {
        return file_size - file_offset;
    }

    return len;
}

/*
 * MappedFileSize() - Returns the size of the memory map backed
 *                    by the underlying file; or 0 if not file-backed.
 */
inline uint64_t MappedFileSize(const MapInfo& map) {
    // Anon mapping or device?
    if (map.name.empty() || map.name[0] != '/' ||
          android::base::StartsWith(map.name, "/dev/")) {
        return 0;
    }

    struct stat file_stat;
    if (stat(map.name.c_str(), &file_stat) != 0) {
        return 0;
    }

    return MappedFileSize(map.start, map.end, map.pgoff, file_stat.st_size);
}

} /* namespace procinfo */
} /* namespace android */