1EXIT(3) Linux Programmer's Manual EXIT(3)
2
6 exit - cause normal process termination
7
9 #include <stdlib.h>
10 noreturn void exit(int status);
12
14 The exit() function causes normal process termination and the least
15 significant byte of status (i.e., status & 0xFF) is returned to the
16 parent (see wait(2)).
17 All functions registered with atexit(3) and on_exit(3) are called, in
19 the reverse order of their registration. (It is possible for one of
20 these functions to use atexit(3) or on_exit(3) to register an addi‐
21 tional function to be executed during exit processing; the new regis‐
22 tration is added to the front of the list of functions that remain to
23 be called.) If one of these functions does not return (e.g., it calls
24 _exit(2), or kills itself with a signal), then none of the remaining
25 functions is called, and further exit processing (in particular, flush‐
26 ing of stdio(3) streams) is abandoned. If a function has been regis‐
27 tered multiple times using atexit(3) or on_exit(3), then it is called
28 as many times as it was registered.
29 All open stdio(3) streams are flushed and closed. Files created by
31 tmpfile(3) are removed.
32 The C standard specifies two constants, EXIT_SUCCESS and EXIT_FAILURE,
34 that may be passed to exit() to indicate successful or unsuccessful
35 termination, respectively.
36
38 The exit() function does not return.
39
41 For an explanation of the terms used in this section, see at‐
42 tributes(7).
43 ┌────────────────────────────────┬───────────────┬─────────────────────┐
45 │Interface │ Attribute │ Value │
46 ├────────────────────────────────┼───────────────┼─────────────────────┤
47 │exit() │ Thread safety │ MT-Unsafe race:exit │
48 └────────────────────────────────┴───────────────┴─────────────────────┘
49 The exit() function uses a global variable that is not protected, so it
51 is not thread-safe.
52
54 POSIX.1-2001, POSIX.1-2008, C89, C99, SVr4, 4.3BSD.
55
57 The behavior is undefined if one of the functions registered using
58 atexit(3) and on_exit(3) calls either exit() or longjmp(3). Note that
59 a call to execve(2) removes registrations created using atexit(3) and
60 on_exit(3).
61 The use of EXIT_SUCCESS and EXIT_FAILURE is slightly more portable (to
63 non-UNIX environments) than the use of 0 and some nonzero value like 1
64 or -1. In particular, VMS uses a different convention.
65 BSD has attempted to standardize exit codes (which some C libraries
67 such as the GNU C library have also adopted); see the file <sysex‐
68 its.h>.
69 After exit(), the exit status must be transmitted to the parent
71 process. There are three cases:
72 • If the parent has set SA_NOCLDWAIT, or has set the SIGCHLD handler
74 to SIG_IGN, the status is discarded and the child dies immediately.
75 • If the parent was waiting on the child, it is notified of the exit
77 status and the child dies immediately.
78 • Otherwise, the child becomes a "zombie" process: most of the process
80 resources are recycled, but a slot containing minimal information
81 about the child process (termination status, resource usage statis‐
82 tics) is retained in process table. This allows the parent to sub‐
83 sequently use waitpid(2) (or similar) to learn the termination sta‐
84 tus of the child; at that point the zombie process slot is released.
85 If the implementation supports the SIGCHLD signal, this signal is sent
87 to the parent. If the parent has set SA_NOCLDWAIT, it is undefined
88 whether a SIGCHLD signal is sent.
89 Signals sent to other processes
91 If the exiting process is a session leader and its controlling terminal
92 is the controlling terminal of the session, then each process in the
93 foreground process group of this controlling terminal is sent a SIGHUP
94 signal, and the terminal is disassociated from this session, allowing
95 it to be acquired by a new controlling process.
96 If the exit of the process causes a process group to become orphaned,
98 and if any member of the newly orphaned process group is stopped, then
99 a SIGHUP signal followed by a SIGCONT signal will be sent to each
100 process in this process group. See setpgid(2) for an explanation of
101 orphaned process groups.
102 Except in the above cases, where the signalled processes may be chil‐
104 dren of the terminating process, termination of a process does not in
105 general cause a signal to be sent to children of that process. Howev‐
106 er, a process can use the prctl(2) PR_SET_PDEATHSIG operation to ar‐
107 range that it receives a signal if its parent terminates.
108
110 _exit(2), get_robust_list(2), setpgid(2), wait(2), atexit(3), on_ex‐
111 it(3), tmpfile(3)
112
114 This page is part of release 5.12 of the Linux man-pages project. A
115 description of the project, information about reporting bugs, and the
116 latest version of this page, can be found at
117 https://www.kernel.org/doc/man-pages/.
118Linux 2021-03-22 EXIT(3)