1SELECT(2)                  Linux Programmer's Manual                 SELECT(2)
2
3
4

NAME

6       select,  pselect,  FD_CLR,  FD_ISSET, FD_SET, FD_ZERO - synchronous I/O
7       multiplexing
8

SYNOPSIS

10       /* According to POSIX.1-2001, POSIX.1-2008 */
11       #include <sys/select.h>
12
13       /* According to earlier standards */
14       #include <sys/time.h>
15       #include <sys/types.h>
16       #include <unistd.h>
17
18       int select(int nfds, fd_set *readfds, fd_set *writefds,
19                  fd_set *exceptfds, struct timeval *timeout);
20
21       void FD_CLR(int fd, fd_set *set);
22       int  FD_ISSET(int fd, fd_set *set);
23       void FD_SET(int fd, fd_set *set);
24       void FD_ZERO(fd_set *set);
25
26       #include <sys/select.h>
27
28       int pselect(int nfds, fd_set *readfds, fd_set *writefds,
29                   fd_set *exceptfds, const struct timespec *timeout,
30                   const sigset_t *sigmask);
31
32   Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
33
34       pselect(): _POSIX_C_SOURCE >= 200112L
35

DESCRIPTION

37       select() and  pselect()  allow  a  program  to  monitor  multiple  file
38       descriptors,  waiting  until one or more of the file descriptors become
39       "ready" for some class of I/O operation (e.g., input possible).  A file
40       descriptor  is  considered  ready if it is possible to perform a corre‐
41       sponding  I/O  operation  (e.g.,  read(2),  or  a  sufficiently   small
42       write(2)) without blocking.
43
44       select()  can  monitor only file descriptors numbers that are less than
45       FD_SETSIZE; poll(2) does not have this limitation.  See BUGS.
46
47       The operation of select() and pselect() is identical, other than  these
48       three differences:
49
50       (i)    select()  uses  a timeout that is a struct timeval (with seconds
51              and microseconds), while pselect() uses a struct timespec  (with
52              seconds and nanoseconds).
53
54       (ii)   select()  may  update  the timeout argument to indicate how much
55              time was left.  pselect() does not change this argument.
56
57       (iii)  select() has no  sigmask  argument,  and  behaves  as  pselect()
58              called with NULL sigmask.
59
60       Three  independent  sets  of  file  descriptors  are watched.  The file
61       descriptors listed in readfds will be  watched  to  see  if  characters
62       become available for reading (more precisely, to see if a read will not
63       block; in particular, a file descriptor is also ready on  end-of-file).
64       The  file  descriptors  in  writefds will be watched to see if space is
65       available for write (though a large write may still block).   The  file
66       descriptors  in  exceptfds  will be watched for exceptional conditions.
67       (For examples of some exceptional conditions,  see  the  discussion  of
68       POLLPRI in poll(2).)
69
70       On exit, each of the file descriptor sets is modified in place to indi‐
71       cate which file descriptors actually changed status.  (Thus,  if  using
72       select()  within  a  loop,  the  sets must be reinitialized before each
73       call.)
74
75       Each of the three file descriptor sets may be specified as NULL  if  no
76       file  descriptors  are  to  be  watched  for the corresponding class of
77       events.
78
79       Four macros are provided to manipulate the sets.   FD_ZERO()  clears  a
80       set.  FD_SET() and FD_CLR() add and remove a given file descriptor from
81       a set.  FD_ISSET() tests to see if a file descriptor  is  part  of  the
82       set; this is useful after select() returns.
83
84       nfds  should  be  set to the highest-numbered file descriptor in any of
85       the three sets, plus 1.  The indicated file descriptors in each set are
86       checked, up to this limit (but see BUGS).
87
88       The  timeout argument specifies the interval that select() should block
89       waiting for a file descriptor to become ready.   The  call  will  block
90       until either:
91
92       *  a file descriptor becomes ready;
93
94       *  the call is interrupted by a signal handler; or
95
96       *  the timeout expires.
97
98       Note  that  the timeout interval will be rounded up to the system clock
99       granularity, and kernel scheduling delays mean that the blocking inter‐
100       val  may  overrun  by  a  small  amount.  If both fields of the timeval
101       structure are zero, then select() returns immediately.  (This is useful
102       for  polling.)   If  timeout  is  NULL (no timeout), select() can block
103       indefinitely.
104
105       sigmask is a pointer to a signal mask (see sigprocmask(2));  if  it  is
106       not  NULL, then pselect() first replaces the current signal mask by the
107       one pointed to by sigmask, then does the "select"  function,  and  then
108       restores the original signal mask.
109
110       Other than the difference in the precision of the timeout argument, the
111       following pselect() call:
112
113           ready = pselect(nfds, &readfds, &writefds, &exceptfds,
114                           timeout, &sigmask);
115
116       is equivalent to atomically executing the following calls:
117
118           sigset_t origmask;
119
120           pthread_sigmask(SIG_SETMASK, &sigmask, &origmask);
121           ready = select(nfds, &readfds, &writefds, &exceptfds, timeout);
122           pthread_sigmask(SIG_SETMASK, &origmask, NULL);
123
124       The reason that pselect() is needed is that if one wants  to  wait  for
125       either  a  signal  or  for  a  file descriptor to become ready, then an
126       atomic test is needed to prevent race conditions.  (Suppose the  signal
127       handler  sets  a  global  flag and returns.  Then a test of this global
128       flag followed by a call of select() could hang indefinitely if the sig‐
129       nal arrived just after the test but just before the call.  By contrast,
130       pselect() allows one to first block signals, handle  the  signals  that
131       have  come  in,  then call pselect() with the desired sigmask, avoiding
132       the race.)
133
134   The timeout
135       The time structures involved are defined in <sys/time.h> and look like
136
137           struct timeval {
138               long    tv_sec;         /* seconds */
139               long    tv_usec;        /* microseconds */
140           };
141
142       and
143
144           struct timespec {
145               long    tv_sec;         /* seconds */
146               long    tv_nsec;        /* nanoseconds */
147           };
148
149       (However, see below on the POSIX.1 versions.)
150
151       Some code calls select() with all three sets empty, nfds  zero,  and  a
152       non-NULL  timeout as a fairly portable way to sleep with subsecond pre‐
153       cision.
154
155       On Linux, select() modifies timeout to reflect the amount of  time  not
156       slept;  most  other  implementations  do not do this.  (POSIX.1 permits
157       either behavior.)  This causes problems  both  when  Linux  code  which
158       reads  timeout  is  ported to other operating systems, and when code is
159       ported to Linux that reuses a struct timeval for multiple select()s  in
160       a  loop  without  reinitializing  it.  Consider timeout to be undefined
161       after select() returns.
162

RETURN VALUE

164       On success, select() and pselect() return the number of  file  descrip‐
165       tors  contained  in  the  three  returned descriptor sets (that is, the
166       total number of bits that are  set  in  readfds,  writefds,  exceptfds)
167       which  may  be  zero if the timeout expires before anything interesting
168       happens.  On error, -1 is returned, and errno is set  to  indicate  the
169       error;  the  file  descriptor  sets are unmodified, and timeout becomes
170       undefined.
171

ERRORS

173       EBADF  An invalid file descriptor was given in one of the sets.   (Per‐
174              haps  a file descriptor that was already closed, or one on which
175              an error has occurred.)  However, see BUGS.
176
177       EINTR  A signal was caught; see signal(7).
178
179       EINVAL nfds is negative or exceeds  the  RLIMIT_NOFILE  resource  limit
180              (see getrlimit(2)).
181
182       EINVAL The value contained within timeout is invalid.
183
184       ENOMEM Unable to allocate memory for internal tables.
185

VERSIONS

187       pselect()  was  added  to  Linux in kernel 2.6.16.  Prior to this, pse‐
188       lect() was emulated in glibc (but see BUGS).
189

CONFORMING TO

191       select() conforms to POSIX.1-2001, POSIX.1-2008, and  4.4BSD  (select()
192       first  appeared in 4.2BSD).  Generally portable to/from non-BSD systems
193       supporting clones of the BSD socket  layer  (including  System V  vari‐
194       ants).   However,  note  that  the  System V variant typically sets the
195       timeout variable before exit, but the BSD variant does not.
196
197       pselect() is defined in POSIX.1g, and in POSIX.1-2001 and POSIX.1-2008.
198

NOTES

200       An fd_set is a fixed size buffer.  Executing FD_CLR() or FD_SET()  with
201       a value of fd that is negative or is equal to or larger than FD_SETSIZE
202       will result in undefined behavior.  Moreover, POSIX requires fd to be a
203       valid file descriptor.
204
205       The  operation  of select() and pselect() is not affected by the O_NON‐
206       BLOCK flag.
207
208       On some other UNIX systems, select() can fail with the error EAGAIN  if
209       the  system  fails  to  allocate kernel-internal resources, rather than
210       ENOMEM as Linux does.  POSIX specifies this error for poll(2), but  not
211       for select().  Portable programs may wish to check for EAGAIN and loop,
212       just as with EINTR.
213
214       On systems that lack pselect(), reliable  (and  more  portable)  signal
215       trapping can be achieved using the self-pipe trick.  In this technique,
216       a signal handler writes a byte to a pipe whose other end  is  monitored
217       by  select()  in  the  main  program.  (To avoid possibly blocking when
218       writing to a pipe that may be full or reading from a pipe that  may  be
219       empty,  nonblocking  I/O  is  used when reading from and writing to the
220       pipe.)
221
222       Concerning the types involved, the classical situation is that the  two
223       fields  of  a timeval structure are typed as long (as shown above), and
224       the structure is defined in <sys/time.h>.  The POSIX.1 situation is
225
226           struct timeval {
227               time_t         tv_sec;     /* seconds */
228               suseconds_t    tv_usec;    /* microseconds */
229           };
230
231       where the structure is defined in <sys/select.h>  and  the  data  types
232       time_t and suseconds_t are defined in <sys/types.h>.
233
234       Concerning  prototypes,  the  classical  situation  is  that one should
235       include <time.h> for select().   The  POSIX.1  situation  is  that  one
236       should include <sys/select.h> for select() and pselect().
237
238       Under  glibc  2.0,  <sys/select.h>  gives  the wrong prototype for pse‐
239       lect().  Under glibc 2.1 to 2.2.1, it gives pselect() when  _GNU_SOURCE
240       is  defined.   Since  glibc 2.2.2, the requirements are as shown in the
241       SYNOPSIS.
242
243   Correspondence between select() and poll() notifications
244       Within the Linux kernel source, we find the following definitions which
245       show the correspondence between the readable, writable, and exceptional
246       condition notifications of select() and the  event  notifications  pro‐
247       vided by poll(2) and epoll(7):
248
249           #define POLLIN_SET  (EPOLLRDNORM | EPOLLRDBAND | EPOLLIN |
250                                EPOLLHUP | EPOLLERR)
251                              /* Ready for reading */
252           #define POLLOUT_SET (EPOLLWRBAND | EPOLLWRNORM | EPOLLOUT |
253                                EPOLLERR)
254                              /* Ready for writing */
255           #define POLLEX_SET  (EPOLLPRI)
256                              /* Exceptional condition */
257
258   Multithreaded applications
259       If  a  file descriptor being monitored by select() is closed in another
260       thread, the result is unspecified.   On  some  UNIX  systems,  select()
261       unblocks  and  returns,  with an indication that the file descriptor is
262       ready (a subsequent I/O operation  will  likely  fail  with  an  error,
263       unless  another  process  reopens  file  descriptor  between  the  time
264       select() returned and the I/O operation is performed).  On  Linux  (and
265       some  other systems), closing the file descriptor in another thread has
266       no effect on select().  In summary, any application that  relies  on  a
267       particular behavior in this scenario must be considered buggy.
268
269   C library/kernel differences
270       The  Linux kernel allows file descriptor sets of arbitrary size, deter‐
271       mining the length of the sets to be checked from  the  value  of  nfds.
272       However, in the glibc implementation, the fd_set type is fixed in size.
273       See also BUGS.
274
275       The pselect() interface described in this page is implemented by glibc.
276       The underlying Linux system call is named pselect6().  This system call
277       has somewhat different behavior from the glibc wrapper function.
278
279       The Linux pselect6() system call modifies its timeout  argument.   How‐
280       ever,  the  glibc wrapper function hides this behavior by using a local
281       variable for the timeout argument that is passed to  the  system  call.
282       Thus,  the  glibc  pselect() function does not modify its timeout argu‐
283       ment; this is the behavior required by POSIX.1-2001.
284
285       The final argument of the pselect6() system call is  not  a  sigset_t *
286       pointer, but is instead a structure of the form:
287
288           struct {
289               const kernel_sigset_t *ss;   /* Pointer to signal set */
290               size_t ss_len;               /* Size (in bytes) of object
291                                               pointed to by 'ss' */
292           };
293
294       This  allows the system call to obtain both a pointer to the signal set
295       and its size, while allowing for the fact that most architectures  sup‐
296       port a maximum of 6 arguments to a system call.  See sigprocmask(2) for
297       a discussion of the difference between the kernel and  libc  notion  of
298       the signal set.
299

BUGS

301       POSIX allows an implementation to define an upper limit, advertised via
302       the constant FD_SETSIZE, on the range of file descriptors that  can  be
303       specified  in a file descriptor set.  The Linux kernel imposes no fixed
304       limit, but the glibc implementation makes  fd_set  a  fixed-size  type,
305       with  FD_SETSIZE  defined  as  1024,  and  the  FD_*() macros operating
306       according to that limit.  To  monitor  file  descriptors  greater  than
307       1023, use poll(2) instead.
308
309       The  implementation  of  the fd_set arguments as value-result arguments
310       means that they must be reinitialized on each call to  select().   This
311       design  error  is  avoided  by  poll(2),  which uses separate structure
312       fields for the input and output of the call.
313
314       According to POSIX, select() should check all specified  file  descrip‐
315       tors  in  the three file descriptor sets, up to the limit nfds-1.  How‐
316       ever, the current implementation ignores any file descriptor  in  these
317       sets  that  is greater than the maximum file descriptor number that the
318       process currently has open.  According to POSIX, any such file descrip‐
319       tor  that  is  specified  in one of the sets should result in the error
320       EBADF.
321
322       Glibc 2.0 provided a version of pselect() that did not take  a  sigmask
323       argument.
324
325       Starting  with  version  2.1,  glibc provided an emulation of pselect()
326       that was implemented using sigprocmask(2) and select().  This implemen‐
327       tation  remained  vulnerable  to the very race condition that pselect()
328       was designed to prevent.  Modern versions of glibc use the  (race-free)
329       pselect() system call on kernels where it is provided.
330
331       Under Linux, select() may report a socket file descriptor as "ready for
332       reading", while nevertheless a subsequent read blocks.  This could  for
333       example  happen  when  data  has arrived but upon examination has wrong
334       checksum and is discarded.  There may be other circumstances in which a
335       file  descriptor is spuriously reported as ready.  Thus it may be safer
336       to use O_NONBLOCK on sockets that should not block.
337
338       On Linux, select() also modifies timeout if the call is interrupted  by
339       a signal handler (i.e., the EINTR error return).  This is not permitted
340       by POSIX.1.  The Linux pselect() system call has the same behavior, but
341       the glibc wrapper hides this behavior by internally copying the timeout
342       to a local variable and passing that variable to the system call.
343

EXAMPLE

345       #include <stdio.h>
346       #include <stdlib.h>
347       #include <sys/time.h>
348       #include <sys/types.h>
349       #include <unistd.h>
350
351       int
352       main(void)
353       {
354           fd_set rfds;
355           struct timeval tv;
356           int retval;
357
358           /* Watch stdin (fd 0) to see when it has input. */
359
360           FD_ZERO(&rfds);
361           FD_SET(0, &rfds);
362
363           /* Wait up to five seconds. */
364
365           tv.tv_sec = 5;
366           tv.tv_usec = 0;
367
368           retval = select(1, &rfds, NULL, NULL, &tv);
369           /* Don't rely on the value of tv now! */
370
371           if (retval == -1)
372               perror("select()");
373           else if (retval)
374               printf("Data is available now.\n");
375               /* FD_ISSET(0, &rfds) will be true. */
376           else
377               printf("No data within five seconds.\n");
378
379           exit(EXIT_SUCCESS);
380       }
381

SEE ALSO

383       accept(2), connect(2), poll(2), read(2),  recv(2),  restart_syscall(2),
384       send(2), sigprocmask(2), write(2), epoll(7), time(7)
385
386       For a tutorial with discussion and examples, see select_tut(2).
387

COLOPHON

389       This  page  is  part of release 5.04 of the Linux man-pages project.  A
390       description of the project, information about reporting bugs,  and  the
391       latest     version     of     this    page,    can    be    found    at
392       https://www.kernel.org/doc/man-pages/.
393
394
395
396Linux                             2019-11-19                         SELECT(2)
Impressum