1WAIT(2) Linux Programmer's Manual WAIT(2)
2
6 wait, waitpid, waitid - wait for process to change state
7
9 #include <sys/types.h>
10 #include <sys/wait.h>
11 pid_t wait(int *wstatus);
13 pid_t waitpid(pid_t pid, int *wstatus, int options);
15 int waitid(idtype_t idtype, id_t id, siginfo_t *infop, int options);
17 /* This is the glibc and POSIX interface; see
18 NOTES for information on the raw system call. */
19 Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
21 waitid():
23 Since glibc 2.26: _XOPEN_SOURCE >= 500 ||
24 _POSIX_C_SOURCE >= 200809L
25 Glibc 2.25 and earlier:
26 _XOPEN_SOURCE
27 || /* Since glibc 2.12: */ _POSIX_C_SOURCE >= 200809L
28 || /* Glibc versions <= 2.19: */ _BSD_SOURCE
29
31 All of these system calls are used to wait for state changes in a child
32 of the calling process, and obtain information about the child whose
33 state has changed. A state change is considered to be: the child ter‐
34 minated; the child was stopped by a signal; or the child was resumed by
35 a signal. In the case of a terminated child, performing a wait allows
36 the system to release the resources associated with the child; if a
37 wait is not performed, then the terminated child remains in a "zombie"
38 state (see NOTES below).
39 If a child has already changed state, then these calls return immedi‐
41 ately. Otherwise, they block until either a child changes state or a
42 signal handler interrupts the call (assuming that system calls are not
43 automatically restarted using the SA_RESTART flag of sigaction(2)). In
44 the remainder of this page, a child whose state has changed and which
45 has not yet been waited upon by one of these system calls is termed
46 waitable.
47 wait() and waitpid()
49 The wait() system call suspends execution of the calling process until
50 one of its children terminates. The call wait(&wstatus) is equivalent
51 to:
52 waitpid(-1, &wstatus, 0);
54 The waitpid() system call suspends execution of the calling process
56 until a child specified by pid argument has changed state. By default,
57 waitpid() waits only for terminated children, but this behavior is mod‐
58 ifiable via the options argument, as described below.
59 The value of pid can be:
61 < -1 meaning wait for any child process whose process group ID is
63 equal to the absolute value of pid.
64 -1 meaning wait for any child process.
66 0 meaning wait for any child process whose process group ID is
68 equal to that of the calling process.
69 > 0 meaning wait for the child whose process ID is equal to the
71 value of pid.
72 The value of options is an OR of zero or more of the following con‐
74 stants:
75 WNOHANG return immediately if no child has exited.
77 WUNTRACED also return if a child has stopped (but not traced via
79 ptrace(2)). Status for traced children which have stopped
80 is provided even if this option is not specified.
81 WCONTINUED (since Linux 2.6.10)
83 also return if a stopped child has been resumed by delivery
84 of SIGCONT.
85 (For Linux-only options, see below.)
87 If wstatus is not NULL, wait() and waitpid() store status information
89 in the int to which it points. This integer can be inspected with the
90 following macros (which take the integer itself as an argument, not a
91 pointer to it, as is done in wait() and waitpid()!):
92 WIFEXITED(wstatus)
94 returns true if the child terminated normally, that is, by call‐
95 ing exit(3) or _exit(2), or by returning from main().
96 WEXITSTATUS(wstatus)
98 returns the exit status of the child. This consists of the
99 least significant 8 bits of the status argument that the child
100 specified in a call to exit(3) or _exit(2) or as the argument
101 for a return statement in main(). This macro should be employed
102 only if WIFEXITED returned true.
103 WIFSIGNALED(wstatus)
105 returns true if the child process was terminated by a signal.
106 WTERMSIG(wstatus)
108 returns the number of the signal that caused the child process
109 to terminate. This macro should be employed only if WIFSIGNALED
110 returned true.
111 WCOREDUMP(wstatus)
113 returns true if the child produced a core dump. This macro
114 should be employed only if WIFSIGNALED returned true.
115 This macro is not specified in POSIX.1-2001 and is not available
117 on some UNIX implementations (e.g., AIX, SunOS). Therefore,
118 enclose its use inside #ifdef WCOREDUMP ... #endif.
119 WIFSTOPPED(wstatus)
121 returns true if the child process was stopped by delivery of a
122 signal; this is possible only if the call was done using WUN‐
123 TRACED or when the child is being traced (see ptrace(2)).
124 WSTOPSIG(wstatus)
126 returns the number of the signal which caused the child to stop.
127 This macro should be employed only if WIFSTOPPED returned true.
128 WIFCONTINUED(wstatus)
130 (since Linux 2.6.10) returns true if the child process was
131 resumed by delivery of SIGCONT.
132 waitid()
134 The waitid() system call (available since Linux 2.6.9) provides more
135 precise control over which child state changes to wait for.
136 The idtype and id arguments select the child(ren) to wait for, as fol‐
138 lows:
139 idtype == P_PID
141 Wait for the child whose process ID matches id.
142 idtype == P_PGID
144 Wait for any child whose process group ID matches id.
145 idtype == P_ALL
147 Wait for any child; id is ignored.
148 The child state changes to wait for are specified by ORing one or more
150 of the following flags in options:
151 WEXITED Wait for children that have terminated.
153 WSTOPPED Wait for children that have been stopped by delivery of a
155 signal.
156 WCONTINUED Wait for (previously stopped) children that have been
158 resumed by delivery of SIGCONT.
159 The following flags may additionally be ORed in options:
161 WNOHANG As for waitpid().
163 WNOWAIT Leave the child in a waitable state; a later wait call can
165 be used to again retrieve the child status information.
166 Upon successful return, waitid() fills in the following fields of the
168 siginfo_t structure pointed to by infop:
169 si_pid The process ID of the child.
171 si_uid The real user ID of the child. (This field is not set on
173 most other implementations.)
174 si_signo Always set to SIGCHLD.
176 si_status Either the exit status of the child, as given to _exit(2)
178 (or exit(3)), or the signal that caused the child to termi‐
179 nate, stop, or continue. The si_code field can be used to
180 determine how to interpret this field.
181 si_code Set to one of: CLD_EXITED (child called _exit(2));
183 CLD_KILLED (child killed by signal); CLD_DUMPED (child
184 killed by signal, and dumped core); CLD_STOPPED (child
185 stopped by signal); CLD_TRAPPED (traced child has trapped);
186 or CLD_CONTINUED (child continued by SIGCONT).
187 If WNOHANG was specified in options and there were no children in a
189 waitable state, then waitid() returns 0 immediately and the state of
190 the siginfo_t structure pointed to by infop depends on the implementa‐
191 tion. To (portably) distinguish this case from that where a child was
192 in a waitable state, zero out the si_pid field before the call and
193 check for a nonzero value in this field after the call returns.
194 POSIX.1-2008 Technical Corrigendum 1 (2013) adds the requirement that
196 when WNOHANG is specified in options and there were no children in a
197 waitable state, then waitid() should zero out the si_pid and si_signo
198 fields of the structure. On Linux and other implementations that
199 adhere to this requirement, it is not necessary to zero out the si_pid
200 field before calling waitid(). However, not all implementations follow
201 the POSIX.1 specification on this point.
202
204 wait(): on success, returns the process ID of the terminated child; on
205 error, -1 is returned.
206 waitpid(): on success, returns the process ID of the child whose state
208 has changed; if WNOHANG was specified and one or more child(ren) speci‐
209 fied by pid exist, but have not yet changed state, then 0 is returned.
210 On error, -1 is returned.
211 waitid(): returns 0 on success or if WNOHANG was specified and no
213 child(ren) specified by id has yet changed state; on error, -1 is
214 returned.
215 Each of these calls sets errno to an appropriate value in the case of
217 an error.
218
220 ECHILD (for wait()) The calling process does not have any unwaited-for
221 children.
222 ECHILD (for waitpid() or waitid()) The process specified by pid (wait‐
224 pid()) or idtype and id (waitid()) does not exist or is not a
225 child of the calling process. (This can happen for one's own
226 child if the action for SIGCHLD is set to SIG_IGN. See also the
227 Linux Notes section about threads.)
228 EINTR WNOHANG was not set and an unblocked signal or a SIGCHLD was
230 caught; see signal(7).
231 EINVAL The options argument was invalid.
233
235 SVr4, 4.3BSD, POSIX.1-2001.
236
238 A child that terminates, but has not been waited for becomes a "zom‐
239 bie". The kernel maintains a minimal set of information about the zom‐
240 bie process (PID, termination status, resource usage information) in
241 order to allow the parent to later perform a wait to obtain information
242 about the child. As long as a zombie is not removed from the system
243 via a wait, it will consume a slot in the kernel process table, and if
244 this table fills, it will not be possible to create further processes.
245 If a parent process terminates, then its "zombie" children (if any) are
246 adopted by init(1), (or by the nearest "subreaper" process as defined
247 through the use of the prctl(2) PR_SET_CHILD_SUBREAPER operation);
248 init(1) automatically performs a wait to remove the zombies.
249 POSIX.1-2001 specifies that if the disposition of SIGCHLD is set to
251 SIG_IGN or the SA_NOCLDWAIT flag is set for SIGCHLD (see sigaction(2)),
252 then children that terminate do not become zombies and a call to wait()
253 or waitpid() will block until all children have terminated, and then
254 fail with errno set to ECHILD. (The original POSIX standard left the
255 behavior of setting SIGCHLD to SIG_IGN unspecified. Note that even
256 though the default disposition of SIGCHLD is "ignore", explicitly set‐
257 ting the disposition to SIG_IGN results in different treatment of zom‐
258 bie process children.)
259 Linux 2.6 conforms to the POSIX requirements. However, Linux 2.4 (and
261 earlier) does not: if a wait() or waitpid() call is made while SIGCHLD
262 is being ignored, the call behaves just as though SIGCHLD were not
263 being ignored, that is, the call blocks until the next child terminates
264 and then returns the process ID and status of that child.
265 Linux notes
267 In the Linux kernel, a kernel-scheduled thread is not a distinct con‐
268 struct from a process. Instead, a thread is simply a process that is
269 created using the Linux-unique clone(2) system call; other routines
270 such as the portable pthread_create(3) call are implemented using
271 clone(2). Before Linux 2.4, a thread was just a special case of a
272 process, and as a consequence one thread could not wait on the children
273 of another thread, even when the latter belongs to the same thread
274 group. However, POSIX prescribes such functionality, and since Linux
275 2.4 a thread can, and by default will, wait on children of other
276 threads in the same thread group.
277 The following Linux-specific options are for use with children created
279 using clone(2); they can also, since Linux 4.7, be used with waitid():
280 __WCLONE
282 Wait for "clone" children only. If omitted, then wait for "non-
283 clone" children only. (A "clone" child is one which delivers no
284 signal, or a signal other than SIGCHLD to its parent upon termi‐
285 nation.) This option is ignored if __WALL is also specified.
286 __WALL (since Linux 2.4)
288 Wait for all children, regardless of type ("clone" or "non-
289 clone").
290 __WNOTHREAD (since Linux 2.4)
292 Do not wait for children of other threads in the same thread
293 group. This was the default before Linux 2.4.
294 Since Linux 4.7, the __WALL flag is automatically implied if the child
296 is being ptraced.
297 C library/kernel differences
299 wait() is actually a library function that (in glibc) is implemented as
300 a call to wait4(2).
301 On some architectures, there is no waitpid() system call; instead, this
303 interface is implemented via a C library wrapper function that calls
304 wait4(2).
305 The raw waitid() system call takes a fifth argument, of type struct
307 rusage *. If this argument is non-NULL, then it is used to return
308 resource usage information about the child, in the same manner as
309 wait4(2). See getrusage(2) for details.
310
312 According to POSIX.1-2008, an application calling waitid() must ensure
313 that infop points to a siginfo_t structure (i.e., that it is a non-null
314 pointer). On Linux, if infop is NULL, waitid() succeeds, and returns
315 the process ID of the waited-for child. Applications should avoid
316 relying on this inconsistent, nonstandard, and unnecessary feature.
317
319 The following program demonstrates the use of fork(2) and waitpid().
320 The program creates a child process. If no command-line argument is
321 supplied to the program, then the child suspends its execution using
322 pause(2), to allow the user to send signals to the child. Otherwise,
323 if a command-line argument is supplied, then the child exits immedi‐
324 ately, using the integer supplied on the command line as the exit sta‐
325 tus. The parent process executes a loop that monitors the child using
326 waitpid(), and uses the W*() macros described above to analyze the wait
327 status value.
328 The following shell session demonstrates the use of the program:
330 $ ./a.out &
332 Child PID is 32360
333 [1] 32359
334 $ kill -STOP 32360
335 stopped by signal 19
336 $ kill -CONT 32360
337 continued
338 $ kill -TERM 32360
339 killed by signal 15
340 [1]+ Done ./a.out
341 $
342 Program source
344 #include <sys/wait.h>
346 #include <stdlib.h>
347 #include <unistd.h>
348 #include <stdio.h>
349 int
351 main(int argc, char *argv[])
352 {
353 pid_t cpid, w;
354 int wstatus;
355 cpid = fork();
357 if (cpid == -1) {
358 perror("fork");
359 exit(EXIT_FAILURE);
360 }
361 if (cpid == 0) { /* Code executed by child */
363 printf("Child PID is %ld\n", (long) getpid());
364 if (argc == 1)
365 pause(); /* Wait for signals */
366 _exit(atoi(argv[1]));
367 } else { /* Code executed by parent */
369 do {
370 w = waitpid(cpid, &wstatus, WUNTRACED | WCONTINUED);
371 if (w == -1) {
372 perror("waitpid");
373 exit(EXIT_FAILURE);
374 }
375 if (WIFEXITED(wstatus)) {
377 printf("exited, status=%d\n", WEXITSTATUS(wstatus));
378 } else if (WIFSIGNALED(wstatus)) {
379 printf("killed by signal %d\n", WTERMSIG(wstatus));
380 } else if (WIFSTOPPED(wstatus)) {
381 printf("stopped by signal %d\n", WSTOPSIG(wstatus));
382 } else if (WIFCONTINUED(wstatus)) {
383 printf("continued\n");
384 }
385 } while (!WIFEXITED(wstatus) && !WIFSIGNALED(wstatus));
386 exit(EXIT_SUCCESS);
387 }
388 }
389
391 _exit(2), clone(2), fork(2), kill(2), ptrace(2), sigaction(2), sig‐
392 nal(2), wait4(2), pthread_create(3), credentials(7), signal(7)
393
395 This page is part of release 4.15 of the Linux man-pages project. A
396 description of the project, information about reporting bugs, and the
397 latest version of this page, can be found at
398 https://www.kernel.org/doc/man-pages/.
399Linux 2017-09-15 WAIT(2)