man2/fork.2

   1 .\" Copyright (C) 2006 Michael Kerrisk <mtk.manpages@gmail.com>
   2 .\" A few fragments remain from an earlier (1992) page by
   3 .\" Drew Eckhardt (drew@cs.colorado.edu),
   4 .\"
   5 .\" %%%LICENSE_START(VERBATIM)
   6 .\" Permission is granted to make and distribute verbatim copies of this
   7 .\" manual provided the copyright notice and this permission notice are
   8 .\" preserved on all copies.
   9 .\"
  10 .\" Permission is granted to copy and distribute modified versions of this
  11 .\" manual under the conditions for verbatim copying, provided that the
  12 .\" entire resulting derived work is distributed under the terms of a
  13 .\" permission notice identical to this one.
  14 .\"
  15 .\" Since the Linux kernel and libraries are constantly changing, this
  16 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  17 .\" responsibility for errors or omissions, or for damages resulting from
  18 .\" the use of the information contained herein.  The author(s) may not
  19 .\" have taken the same level of care in the production of this manual,
  20 .\" which is licensed free of charge, as they might when working
  21 .\" professionally.
  22 .\"
  23 .\" Formatted or processed versions of this manual, if unaccompanied by
  24 .\" the source, must acknowledge the copyright and authors of this work.
  25 .\" %%%LICENSE_END
  26 .\"
  27 .\" Modified by Michael Haardt (michael@moria.de)
  28 .\" Modified Sat Jul 24 13:22:07 1993 by Rik Faith (faith@cs.unc.edu)
  29 .\" Modified 21 Aug 1994 by Michael Chastain (mec@shell.portal.com):
  30 .\"   Referenced 'clone(2)'.
  31 .\" Modified 1995-06-10, 1996-04-18, 1999-11-01, 2000-12-24
  32 .\"   by Andries Brouwer (aeb@cwi.nl)
  33 .\" Modified, 27 May 2004, Michael Kerrisk <mtk.manpages@gmail.com>
  34 .\"     Added notes on capability requirements
  35 .\" 2006-09-04, Michael Kerrisk
  36 .\"     Greatly expanded, to describe all attributes that differ
  37 .\"     parent and child.
  38 .\"
  39 .TH FORK 2 2017-09-15 "Linux" "Linux Programmer's Manual"
  40 .SH NAME
  41 fork \- create a child process
  42 .SH SYNOPSIS
  43 .B #include <sys/types.h>
  44 .br
  45 .B #include <unistd.h>
  46 .PP
  47 .B pid_t fork(void);
  48 .SH DESCRIPTION
  49 .BR fork ()
  50 creates a new process by duplicating the calling process.
  51 The new process is referred to as the
  52 .I child
  53 process.
  54 The calling process is referred to as the
  55 .I parent
  56 process.
  57 .PP
  58 The child process and the parent process run in separate memory spaces.
  59 At the time of
  60 .BR fork ()
  61 both memory spaces have the same content.
  62 Memory writes, file mappings
  63 .RB ( mmap (2)),
  64 and unmappings
  65 .RB ( munmap (2))
  66 performed by one of the processes do not affect the other.
  67 .PP
  68 The child process is an exact duplicate of the parent
  69 process except for the following points:
  70 .IP * 3
  71 The child has its own unique process ID,
  72 and this PID does not match the ID of any existing process group
  73 .RB ( setpgid (2))
  74 or session.
  75 .IP *
  76 The child's parent process ID is the same as the parent's process ID.
  77 .IP *
  78 The child does not inherit its parent's memory locks
  79 .RB ( mlock (2),
  80 .BR mlockall (2)).
  81 .IP *
  82 Process resource utilizations
  83 .RB ( getrusage (2))
  84 and CPU time counters
  85 .RB ( times (2))
  86 are reset to zero in the child.
  87 .IP *
  88 The child's set of pending signals is initially empty
  89 .RB ( sigpending (2)).
  90 .IP *
  91 The child does not inherit semaphore adjustments from its parent
  92 .RB ( semop (2)).
  93 .IP *
  94 The child does not inherit process-associated record locks from its parent
  95 .RB ( fcntl (2)).
  96 (On the other hand, it does inherit
  97 .BR fcntl (2)
  98 open file description locks and
  99 .BR flock (2)
 100 locks from its parent.)
 101 .IP *
 102 The child does not inherit timers from its parent
 103 .RB ( setitimer (2),
 104 .BR alarm (2),
 105 .BR timer_create (2)).
 106 .IP *
 107 The child does not inherit outstanding asynchronous I/O operations
 108 from its parent
 109 .RB ( aio_read (3),
 110 .BR aio_write (3)),
 111 nor does it inherit any asynchronous I/O contexts from its parent (see
 112 .BR io_setup (2)).
 113 .PP
 114 The process attributes in the preceding list are all specified
 115 in POSIX.1.
 116 The parent and child also differ with respect to the following
 117 Linux-specific process attributes:
 118 .IP * 3
 119 The child does not inherit directory change notifications (dnotify)
 120 from its parent
 121 (see the description of
 122 .B F_NOTIFY
 123 in
 124 .BR fcntl (2)).
 125 .IP *
 126 The
 127 .BR prctl (2)
 128 .B PR_SET_PDEATHSIG
 129 setting is reset so that the child does not receive a signal
 130 when its parent terminates.
 131 .IP *
 132 The default timer slack value is set to the parent's
 133 current timer slack value.
 134 See the description of
 135 .BR PR_SET_TIMERSLACK
 136 in
 137 .BR prctl (2).
 138 .IP *
 139 Memory mappings that have been marked with the
 140 .BR madvise (2)
 141 .B MADV_DONTFORK
 142 flag are not inherited across a
 143 .BR fork ().
 144 .IP *
 145 Memory in address ranges that have been marked with the
 146 .BR madvise (2)
 147 .B MADV_WIPEONFORK
 148 flag is zeroed in the child after a
 149 .BR fork ().
 150 (The
 151 .B MADV_WIPEONFORK
 152 setting remains in place for those address ranges in the child.)
 153 .IP *
 154 The termination signal of the child is always
 155 .B SIGCHLD
 156 (see
 157 .BR clone (2)).
 158 .IP *
 159 The port access permission bits set by
 160 .BR ioperm (2)
 161 are not inherited by the child;
 162 the child must turn on any bits that it requires using
 163 .BR ioperm (2).
 164 .PP
 165 Note the following further points:
 166 .IP * 3
 167 The child process is created with a single thread\(emthe
 168 one that called
 169 .BR fork ().
 170 The entire virtual address space of the parent is replicated in the child,
 171 including the states of mutexes, condition variables,
 172 and other pthreads objects; the use of
 173 .BR pthread_atfork (3)
 174 may be helpful for dealing with problems that this can cause.
 175 .IP *
 176 After a
 177 .BR fork ()
 178 in a multithreaded program,
 179 the child can safely call only async-signal-safe functions (see
 180 .BR signal-safety (7))
 181 until such time as it calls
 182 .BR execve (2).
 183 .IP *
 184 The child inherits copies of the parent's set of open file descriptors.
 185 Each file descriptor in the child refers to the same
 186 open file description (see
 187 .BR open (2))
 188 as the corresponding file descriptor in the parent.
 189 This means that the two file descriptors share open file status flags,
 190 file offset,
 191 and signal-driven I/O attributes (see the description of
 192 .B F_SETOWN
 193 and
 194 .B F_SETSIG
 195 in
 196 .BR fcntl (2)).
 197 .IP *
 198 The child inherits copies of the parent's set of open message
 199 queue descriptors (see
 200 .BR mq_overview (7)).
 201 Each file descriptor in the child refers to the same
 202 open message queue description
 203 as the corresponding file descriptor in the parent.
 204 This means that the two file descriptors share the same flags
 205 .RI ( mq_flags ).
 206 .IP *
 207 The child inherits copies of the parent's set of open directory streams (see
 208 .BR opendir (3)).
 209 POSIX.1 says that the corresponding directory streams
 210 in the parent and child
 211 .I may
 212 share the directory stream positioning;
 213 on Linux/glibc they do not.
 214 .SH RETURN VALUE
 215 On success, the PID of the child process is returned in the parent,
 216 and 0 is returned in the child.
 217 On failure, \-1 is returned in the parent,
 218 no child process is created, and
 219 .I errno
 220 is set appropriately.
 221 .SH ERRORS
 222 .TP
 223 .B EAGAIN
 224 .\" NOTE! The following should match the description in pthread_create(3)
 225 A system-imposed limit on the number of threads was encountered.
 226 There are a number of limits that may trigger this error:
 227 .RS
 228 .IP * 3
 229 the
 230 .BR RLIMIT_NPROC
 231 soft resource limit (set via
 232 .BR setrlimit (2)),
 233 which limits the number of processes and threads for a real user ID,
 234 was reached;
 235 .IP *
 236 the kernel's system-wide limit on the number of processes and threads,
 237 .IR /proc/sys/kernel/threads-max ,
 238 was reached (see
 239 .BR proc (5));
 240 .IP *
 241 the maximum number of PIDs,
 242 .IR /proc/sys/kernel/pid_max ,
 243 was reached (see
 244 .BR proc (5));
 245 or
 246 .IP *
 247 the PID limit
 248 .RI ( pids.max )
 249 imposed by the cgroup "process number" (PIDs) controller was reached.
 250 .RE
 251 .TP
 252 .B EAGAIN
 253 The caller is operating under the
 254 .BR SCHED_DEADLINE
 255 scheduling policy and does not have the reset-on-fork flag set.
 256 See
 257 .BR sched (7).
 258 .TP
 259 .B ENOMEM
 260 .BR fork ()
 261 failed to allocate the necessary kernel structures because memory is tight.
 262 .TP
 263 .B ENOMEM
 264 An attempt was made to create a child process in a PID namespace
 265 whose "init" process has terminated.
 266 See
 267 .BR pid_namespaces (7).
 268 .TP
 269 .B ENOSYS
 270 .BR fork ()
 271 is not supported on this platform (for example,
 272 .\" e.g., arm (optionally), blackfin, c6x, frv, h8300, microblaze, xtensa
 273 hardware without a Memory-Management Unit).
 274 .TP
 275 .BR ERESTARTNOINTR " (since Linux 2.6.17)"
 276 .\" commit 4a2c7a7837da1b91468e50426066d988050e4d56
 277 System call was interrupted by a signal and will be restarted.
 278 (This can be seen only during a trace.)
 279 .SH CONFORMING TO
 280 POSIX.1-2001, POSIX.1-2008, SVr4, 4.3BSD.
 281 .SH NOTES
 282 .PP
 283 Under Linux,
 284 .BR fork ()
 285 is implemented using copy-on-write pages, so the only penalty that it incurs
 286 is the time and memory required to duplicate the parent's page tables,
 287 and to create a unique task structure for the child.
 288 .SS C library/kernel differences
 289 Since version 2.3.3,
 290 .\" nptl/sysdeps/unix/sysv/linux/fork.c
 291 rather than invoking the kernel's
 292 .BR fork ()
 293 system call,
 294 the glibc
 295 .BR fork ()
 296 wrapper that is provided as part of the
 297 NPTL threading implementation invokes
 298 .BR clone (2)
 299 with flags that provide the same effect as the traditional system call.
 300 (A call to
 301 .BR fork ()
 302 is equivalent to a call to
 303 .BR clone (2)
 304 specifying
 305 .I flags
 306 as just
 307 .BR SIGCHLD .)
 308 The glibc wrapper invokes any fork handlers that have been
 309 established using
 310 .BR pthread_atfork (3).
 311 .\" and does some magic to ensure that getpid(2) returns the right value.
 312 .SH EXAMPLE
 313 See
 314 .BR pipe (2)
 315 and
 316 .BR wait (2).
 317 .SH SEE ALSO
 318 .BR clone (2),
 319 .BR execve (2),
 320 .BR exit (2),
 321 .BR setrlimit (2),
 322 .BR unshare (2),
 323 .BR vfork (2),
 324 .BR wait (2),
 325 .BR daemon (3),
 326 .BR pthread_atfork (3),
 327 .BR capabilities (7),
 328 .BR credentials (7)