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 thread 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 thread un‐
56 til 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 at the time of the call to
69 waitpid().
70 > 0 meaning wait for the child whose process ID is equal to the
72 value of pid.
73 The value of options is an OR of zero or more of the following con‐
75 stants:
76 WNOHANG
78 return immediately if no child has exited.
79 WUNTRACED
81 also return if a child has stopped (but not traced via
82 ptrace(2)). Status for traced children which have stopped is
83 provided even if this option is not specified.
84 WCONTINUED (since Linux 2.6.10)
86 also return if a stopped child has been resumed by delivery of
87 SIGCONT.
88 (For Linux-only options, see below.)
90 If wstatus is not NULL, wait() and waitpid() store status information
92 in the int to which it points. This integer can be inspected with the
93 following macros (which take the integer itself as an argument, not a
94 pointer to it, as is done in wait() and waitpid()!):
95 WIFEXITED(wstatus)
97 returns true if the child terminated normally, that is, by call‐
98 ing exit(3) or _exit(2), or by returning from main().
99 WEXITSTATUS(wstatus)
101 returns the exit status of the child. This consists of the
102 least significant 8 bits of the status argument that the child
103 specified in a call to exit(3) or _exit(2) or as the argument
104 for a return statement in main(). This macro should be employed
105 only if WIFEXITED returned true.
106 WIFSIGNALED(wstatus)
108 returns true if the child process was terminated by a signal.
109 WTERMSIG(wstatus)
111 returns the number of the signal that caused the child process
112 to terminate. This macro should be employed only if WIFSIGNALED
113 returned true.
114 WCOREDUMP(wstatus)
116 returns true if the child produced a core dump (see core(5)).
117 This macro should be employed only if WIFSIGNALED returned true.
118 This macro is not specified in POSIX.1-2001 and is not available
120 on some UNIX implementations (e.g., AIX, SunOS). Therefore, en‐
121 close its use inside #ifdef WCOREDUMP ... #endif.
122 WIFSTOPPED(wstatus)
124 returns true if the child process was stopped by delivery of a
125 signal; this is possible only if the call was done using WUN‐
126 TRACED or when the child is being traced (see ptrace(2)).
127 WSTOPSIG(wstatus)
129 returns the number of the signal which caused the child to stop.
130 This macro should be employed only if WIFSTOPPED returned true.
131 WIFCONTINUED(wstatus)
133 (since Linux 2.6.10) returns true if the child process was re‐
134 sumed by delivery of SIGCONT.
135 waitid()
137 The waitid() system call (available since Linux 2.6.9) provides more
138 precise control over which child state changes to wait for.
139 The idtype and id arguments select the child(ren) to wait for, as fol‐
141 lows:
142 idtype == P_PID
144 Wait for the child whose process ID matches id.
145 idtype == P_PIDFD (since Linux 5.4)
147 Wait for the child referred to by the PID file descriptor speci‐
148 fied in id. (See pidfd_open(2) for further information on PID
149 file descriptors.)
150 idtype == P_PGID
152 Wait for any child whose process group ID matches id. Since
153 Linux 5.4, if id is zero, then wait for any child that is in the
154 same process group as the caller's process group at the time of
155 the call.
156 idtype == P_ALL
158 Wait for any child; id is ignored.
159 The child state changes to wait for are specified by ORing one or more
161 of the following flags in options:
162 WEXITED
164 Wait for children that have terminated.
165 WSTOPPED
167 Wait for children that have been stopped by delivery of a sig‐
168 nal.
169 WCONTINUED
171 Wait for (previously stopped) children that have been resumed by
172 delivery of SIGCONT.
173 The following flags may additionally be ORed in options:
175 WNOHANG
177 As for waitpid().
178 WNOWAIT
180 Leave the child in a waitable state; a later wait call can be
181 used to again retrieve the child status information.
182 Upon successful return, waitid() fills in the following fields of the
184 siginfo_t structure pointed to by infop:
185 si_pid The process ID of the child.
187 si_uid The real user ID of the child. (This field is not set on most
189 other implementations.)
190 si_signo
192 Always set to SIGCHLD.
193 si_status
195 Either the exit status of the child, as given to _exit(2) (or
196 exit(3)), or the signal that caused the child to terminate,
197 stop, or continue. The si_code field can be used to determine
198 how to interpret this field.
199 si_code
201 Set to one of: CLD_EXITED (child called _exit(2)); CLD_KILLED
202 (child killed by signal); CLD_DUMPED (child killed by signal,
203 and dumped core); CLD_STOPPED (child stopped by signal);
204 CLD_TRAPPED (traced child has trapped); or CLD_CONTINUED (child
205 continued by SIGCONT).
206 If WNOHANG was specified in options and there were no children in a
208 waitable state, then waitid() returns 0 immediately and the state of
209 the siginfo_t structure pointed to by infop depends on the implementa‐
210 tion. To (portably) distinguish this case from that where a child was
211 in a waitable state, zero out the si_pid field before the call and
212 check for a nonzero value in this field after the call returns.
213 POSIX.1-2008 Technical Corrigendum 1 (2013) adds the requirement that
215 when WNOHANG is specified in options and there were no children in a
216 waitable state, then waitid() should zero out the si_pid and si_signo
217 fields of the structure. On Linux and other implementations that ad‐
218 here to this requirement, it is not necessary to zero out the si_pid
219 field before calling waitid(). However, not all implementations follow
220 the POSIX.1 specification on this point.
221
223 wait(): on success, returns the process ID of the terminated child; on
224 error, -1 is returned.
225 waitpid(): on success, returns the process ID of the child whose state
227 has changed; if WNOHANG was specified and one or more child(ren) speci‐
228 fied by pid exist, but have not yet changed state, then 0 is returned.
229 On error, -1 is returned.
230 waitid(): returns 0 on success or if WNOHANG was specified and no
232 child(ren) specified by id has yet changed state; on error, -1 is re‐
233 turned.
234 Each of these calls sets errno to an appropriate value in the case of
236 an error.
237
239 ECHILD (for wait()) The calling process does not have any unwaited-for
240 children.
241 ECHILD (for waitpid() or waitid()) The process specified by pid (wait‐
243 pid()) or idtype and id (waitid()) does not exist or is not a
244 child of the calling process. (This can happen for one's own
245 child if the action for SIGCHLD is set to SIG_IGN. See also the
246 Linux Notes section about threads.)
247 EINTR WNOHANG was not set and an unblocked signal or a SIGCHLD was
249 caught; see signal(7).
250 EINVAL The options argument was invalid.
252
254 SVr4, 4.3BSD, POSIX.1-2001.
255
257 A child that terminates, but has not been waited for becomes a "zom‐
258 bie". The kernel maintains a minimal set of information about the zom‐
259 bie process (PID, termination status, resource usage information) in
260 order to allow the parent to later perform a wait to obtain information
261 about the child. As long as a zombie is not removed from the system
262 via a wait, it will consume a slot in the kernel process table, and if
263 this table fills, it will not be possible to create further processes.
264 If a parent process terminates, then its "zombie" children (if any) are
265 adopted by init(1), (or by the nearest "subreaper" process as defined
266 through the use of the prctl(2) PR_SET_CHILD_SUBREAPER operation);
267 init(1) automatically performs a wait to remove the zombies.
268 POSIX.1-2001 specifies that if the disposition of SIGCHLD is set to
270 SIG_IGN or the SA_NOCLDWAIT flag is set for SIGCHLD (see sigaction(2)),
271 then children that terminate do not become zombies and a call to wait()
272 or waitpid() will block until all children have terminated, and then
273 fail with errno set to ECHILD. (The original POSIX standard left the
274 behavior of setting SIGCHLD to SIG_IGN unspecified. Note that even
275 though the default disposition of SIGCHLD is "ignore", explicitly set‐
276 ting the disposition to SIG_IGN results in different treatment of zom‐
277 bie process children.)
278 Linux 2.6 conforms to the POSIX requirements. However, Linux 2.4 (and
280 earlier) does not: if a wait() or waitpid() call is made while SIGCHLD
281 is being ignored, the call behaves just as though SIGCHLD were not be‐
282 ing ignored, that is, the call blocks until the next child terminates
283 and then returns the process ID and status of that child.
284 Linux notes
286 In the Linux kernel, a kernel-scheduled thread is not a distinct con‐
287 struct from a process. Instead, a thread is simply a process that is
288 created using the Linux-unique clone(2) system call; other routines
289 such as the portable pthread_create(3) call are implemented using
290 clone(2). Before Linux 2.4, a thread was just a special case of a
291 process, and as a consequence one thread could not wait on the children
292 of another thread, even when the latter belongs to the same thread
293 group. However, POSIX prescribes such functionality, and since Linux
294 2.4 a thread can, and by default will, wait on children of other
295 threads in the same thread group.
296 The following Linux-specific options are for use with children created
298 using clone(2); they can also, since Linux 4.7, be used with waitid():
299 __WCLONE
301 Wait for "clone" children only. If omitted, then wait for "non-
302 clone" children only. (A "clone" child is one which delivers no
303 signal, or a signal other than SIGCHLD to its parent upon termi‐
304 nation.) This option is ignored if __WALL is also specified.
305 __WALL (since Linux 2.4)
307 Wait for all children, regardless of type ("clone" or "non-
308 clone").
309 __WNOTHREAD (since Linux 2.4)
311 Do not wait for children of other threads in the same thread
312 group. This was the default before Linux 2.4.
313 Since Linux 4.7, the __WALL flag is automatically implied if the child
315 is being ptraced.
316 C library/kernel differences
318 wait() is actually a library function that (in glibc) is implemented as
319 a call to wait4(2).
320 On some architectures, there is no waitpid() system call; instead, this
322 interface is implemented via a C library wrapper function that calls
323 wait4(2).
324 The raw waitid() system call takes a fifth argument, of type struct
326 rusage *. If this argument is non-NULL, then it is used to return re‐
327 source usage information about the child, in the same manner as
328 wait4(2). See getrusage(2) for details.
329
331 According to POSIX.1-2008, an application calling waitid() must ensure
332 that infop points to a siginfo_t structure (i.e., that it is a non-null
333 pointer). On Linux, if infop is NULL, waitid() succeeds, and returns
334 the process ID of the waited-for child. Applications should avoid re‐
335 lying on this inconsistent, nonstandard, and unnecessary feature.
336
338 The following program demonstrates the use of fork(2) and waitpid().
339 The program creates a child process. If no command-line argument is
340 supplied to the program, then the child suspends its execution using
341 pause(2), to allow the user to send signals to the child. Otherwise,
342 if a command-line argument is supplied, then the child exits immedi‐
343 ately, using the integer supplied on the command line as the exit sta‐
344 tus. The parent process executes a loop that monitors the child using
345 waitpid(), and uses the W*() macros described above to analyze the wait
346 status value.
347 The following shell session demonstrates the use of the program:
349 $ ./a.out &
351 Child PID is 32360
352 [1] 32359
353 $ kill -STOP 32360
354 stopped by signal 19
355 $ kill -CONT 32360
356 continued
357 $ kill -TERM 32360
358 killed by signal 15
359 [1]+ Done ./a.out
360 $
361 Program source
363 #include <sys/wait.h>
365 #include <stdint.h>
366 #include <stdlib.h>
367 #include <unistd.h>
368 #include <stdio.h>
369 int
371 main(int argc, char *argv[])
372 {
373 pid_t cpid, w;
374 int wstatus;
375 cpid = fork();
377 if (cpid == -1) {
378 perror("fork");
379 exit(EXIT_FAILURE);
380 }
381 if (cpid == 0) { /* Code executed by child */
383 printf("Child PID is %jd\n", (intmax_t) getpid());
384 if (argc == 1)
385 pause(); /* Wait for signals */
386 _exit(atoi(argv[1]));
387 } else { /* Code executed by parent */
389 do {
390 w = waitpid(cpid, &wstatus, WUNTRACED | WCONTINUED);
391 if (w == -1) {
392 perror("waitpid");
393 exit(EXIT_FAILURE);
394 }
395 if (WIFEXITED(wstatus)) {
397 printf("exited, status=%d\n", WEXITSTATUS(wstatus));
398 } else if (WIFSIGNALED(wstatus)) {
399 printf("killed by signal %d\n", WTERMSIG(wstatus));
400 } else if (WIFSTOPPED(wstatus)) {
401 printf("stopped by signal %d\n", WSTOPSIG(wstatus));
402 } else if (WIFCONTINUED(wstatus)) {
403 printf("continued\n");
404 }
405 } while (!WIFEXITED(wstatus) && !WIFSIGNALED(wstatus));
406 exit(EXIT_SUCCESS);
407 }
408 }
409
411 _exit(2), clone(2), fork(2), kill(2), ptrace(2), sigaction(2), sig‐
412 nal(2), wait4(2), pthread_create(3), core(5), credentials(7), signal(7)
413
415 This page is part of release 5.10 of the Linux man-pages project. A
416 description of the project, information about reporting bugs, and the
417 latest version of this page, can be found at
418 https://www.kernel.org/doc/man-pages/.
419Linux 2020-11-01 WAIT(2)