man3/exit.3

   1 .\" Copyright (C) 2001 Andries Brouwer <aeb@cwi.nl>.
   2 .\"
   3 .\" %%%LICENSE_START(VERBATIM)
   4 .\" Permission is granted to make and distribute verbatim copies of this
   5 .\" manual provided the copyright notice and this permission notice are
   6 .\" preserved on all copies.
   7 .\"
   8 .\" Permission is granted to copy and distribute modified versions of this
   9 .\" manual under the conditions for verbatim copying, provided that the
  10 .\" entire resulting derived work is distributed under the terms of a
  11 .\" permission notice identical to this one.
  12 .\"
  13 .\" Since the Linux kernel and libraries are constantly changing, this
  14 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  15 .\" responsibility for errors or omissions, or for damages resulting from
  16 .\" the use of the information contained herein.  The author(s) may not
  17 .\" have taken the same level of care in the production of this manual,
  18 .\" which is licensed free of charge, as they might when working
  19 .\" professionally.
  20 .\"
  21 .\" Formatted or processed versions of this manual, if unaccompanied by
  22 .\" the source, must acknowledge the copyright and authors of this work.
  23 .\" %%%LICENSE_END
  24 .\"
  25 .\" FIXME . There are a lot of other process termination actions that
  26 .\" could be listed on this page. See, for example, the list in the
  27 .\" POSIX exit(3p) page.
  28 .\"
  29 .TH EXIT 3  2020-02-09 "Linux" "Linux Programmer's Manual"
  30 .SH NAME
  31 exit \- cause normal process termination
  32 .SH SYNOPSIS
  33 .nf
  34 .B #include <stdlib.h>
  35 .PP
  36 .BI "void exit(int " status );
  37 .fi
  38 .SH DESCRIPTION
  39 The
  40 .BR exit ()
  41 function causes normal process termination and the least significant byte of
  42 .I status
  43 (i.e., \fIstatus & 0xFF\fP) is returned to the parent (see
  44 .BR wait (2)).
  45 .PP
  46 All functions registered with
  47 .BR atexit (3)
  48 and
  49 .BR on_exit (3)
  50 are called, in the reverse order of their registration.
  51 (It is possible for one of these functions to use
  52 .BR atexit (3)
  53 or
  54 .BR on_exit (3)
  55 to register an additional
  56 function to be executed during exit processing;
  57 the new registration is added to the front of the list of functions
  58 that remain to be called.)
  59 If one of these functions does not return
  60 (e.g., it calls
  61 .BR _exit (2),
  62 or kills itself with a signal),
  63 then none of the remaining functions is called,
  64 and further exit processing (in particular, flushing of
  65 .BR stdio (3)
  66 streams) is abandoned.
  67 If a function has been registered multiple times using
  68 .BR atexit (3)
  69 or
  70 .BR on_exit (3),
  71 then it is called as many times as it was registered.
  72 .PP
  73 All open
  74 .BR stdio (3)
  75 streams are flushed and closed.
  76 Files created by
  77 .BR tmpfile (3)
  78 are removed.
  79 .PP
  80 The C standard specifies two constants,
  81 \fBEXIT_SUCCESS\fP and \fBEXIT_FAILURE\fP,
  82 that may be passed to
  83 .BR exit ()
  84 to indicate successful or unsuccessful
  85 termination, respectively.
  86 .SH RETURN VALUE
  87 The
  88 .BR exit ()
  89 function does not return.
  90 .SH ATTRIBUTES
  91 For an explanation of the terms used in this section, see
  92 .BR attributes (7).
  93 .TS
  94 allbox;
  95 lb lb lb
  96 l l l.
  97 Interface       Attribute       Value
  98 T{
  99 .BR exit ()
 100 T}      Thread safety   MT-Unsafe race:exit
 101 .TE
 102 .PP
 103 The
 104 .BR exit ()
 105 function uses a global variable that is not protected,
 106 so it is not thread-safe.
 107 .SH CONFORMING TO
 108 POSIX.1-2001, POSIX.1-2008, C89, C99, SVr4, 4.3BSD.
 109 .SH NOTES
 110 .PP
 111 The behavior is undefined if one of the functions registered using
 112 .BR atexit (3)
 113 and
 114 .BR on_exit (3)
 115 calls either
 116 .BR exit ()
 117 or
 118 .BR longjmp (3).
 119 Note that a call to
 120 .BR execve (2)
 121 removes registrations created using
 122 .BR atexit (3)
 123 and
 124 .BR on_exit (3).
 125 .PP
 126 The use of
 127 .B EXIT_SUCCESS
 128 and
 129 .B EXIT_FAILURE
 130 is slightly more portable
 131 (to non-UNIX environments) than the use of 0 and some nonzero value
 132 like 1 or \-1.
 133 In particular, VMS uses a different convention.
 134 .PP
 135 BSD has attempted to standardize exit codes
 136 (which some C libraries such as the GNU C library have also adopted);
 137 see the file
 138 .IR <sysexits.h> .
 139 .PP
 140 After
 141 .BR exit (),
 142 the exit status must be transmitted to the
 143 parent process.
 144 There are three cases:
 145 .IP \(bu 3
 146 If the parent has set
 147 .BR SA_NOCLDWAIT ,
 148 or has set the
 149 .B SIGCHLD
 150 handler to
 151 .BR SIG_IGN ,
 152 the status is discarded and the child dies immediately.
 153 .IP \(bu
 154 If the parent was waiting on the child,
 155 it is notified of the exit status and the child dies immediately.
 156 .IP \(bu
 157 Otherwise,
 158 the child becomes a "zombie" process:
 159 most of the process resources are recycled,
 160 but a slot containing minimal information about the child process
 161 (termination status, resource usage statistics) is retained in process table.
 162 This allows the parent to subsequently use
 163 .BR waitpid (2)
 164 (or similar) to learn the termination status of the child;
 165 at that point the zombie process slot is released.
 166 .PP
 167 If the implementation supports the
 168 .B SIGCHLD
 169 signal, this signal
 170 is sent to the parent.
 171 If the parent has set
 172 .BR SA_NOCLDWAIT ,
 173 it is undefined whether a
 174 .B SIGCHLD
 175 signal is sent.
 176 .\"
 177 .SS Signals sent to other processes
 178 If the exiting process is a session leader and its controlling terminal
 179 is the controlling terminal of the session, then each process in
 180 the foreground process group of this controlling terminal
 181 is sent a
 182 .B SIGHUP
 183 signal, and the terminal is disassociated
 184 from this session, allowing it to be acquired by a new controlling
 185 process.
 186 .PP
 187 If the exit of the process causes a process group to become orphaned,
 188 and if any member of the newly orphaned process group is stopped,
 189 then a
 190 .B SIGHUP
 191 signal followed by a
 192 .B SIGCONT
 193 signal will be
 194 sent to each process in this process group.
 195 See
 196 .BR setpgid (2)
 197 for an explanation of orphaned process groups.
 198 .PP
 199 Except in the above cases,
 200 where the signalled processes may be children of the terminating process,
 201 termination of a process does
 202 .I not
 203 in general cause a signal to be sent to children of that process.
 204 However, a process can use the
 205 .BR prctl (2)
 206 .B PR_SET_PDEATHSIG
 207 operation to arrange that it receives a signal if its parent terminates.
 208 .SH SEE ALSO
 209 .BR _exit (2),
 210 .BR get_robust_list (2),
 211 .BR setpgid (2),
 212 .BR wait (2),
 213 .BR atexit (3),
 214 .BR on_exit (3),
 215 .BR tmpfile (3)