1SELECT_TUT(2)              Linux Programmer's Manual             SELECT_TUT(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 */
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 *utimeout);
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 *ntimeout,
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 || _XOPEN_SOURCE >= 600
35

DESCRIPTION

37       select() (or pselect()) is used to efficiently  monitor  multiple  file
38       descriptors, to see if any of them is, or becomes, "ready"; that is, to
39       see whether I/O becomes possible, or  an  "exceptional  condition"  has
40       occurred on any of the descriptors.
41
42       Its  principal arguments are three "sets" of file descriptors: readfds,
43       writefds, and exceptfds.  Each set is declared as type fd_set, and  its
44       contents  can  be  manipulated  with  the  macros FD_CLR(), FD_ISSET(),
45       FD_SET(), and FD_ZERO().  A newly declared set should first be  cleared
46       using  FD_ZERO().  select() modifies the contents of the sets according
47       to the rules described below; after calling select() you can test if  a
48       file  descriptor  is  still present in a set with the FD_ISSET() macro.
49       FD_ISSET() returns nonzero if a specified file descriptor is present in
50       a set and zero if it is not.  FD_CLR() removes a file descriptor from a
51       set.
52
53   Arguments
54       readfds
55              This set is watched to see if data is available for reading from
56              any  of  its  file  descriptors.   After  select() has returned,
57              readfds will be cleared of all file descriptors except for those
58              that are immediately available for reading.
59
60       writefds
61              This  set  is  watched to see if there is space to write data to
62              any of its  file  descriptors.   After  select()  has  returned,
63              writefds  will  be  cleared  of  all file descriptors except for
64              those that are immediately available for writing.
65
66       exceptfds
67              This set is watched for "exceptional conditions".  In  practice,
68              only  one such exceptional condition is common: the availability
69              of out-of-band (OOB) data for reading from a  TCP  socket.   See
70              recv(2),  send(2),  and  tcp(7) for more details about OOB data.
71              (One other less common case where select(2) indicates an  excep‐
72              tional  condition  occurs  with pseudo-terminals in packet mode;
73              see tty_ioctl(4).)  After select() has returned, exceptfds  will
74              be cleared of all file descriptors except for those for which an
75              exceptional condition has occurred.
76
77       nfds   This is an integer  one  more  than  the  maximum  of  any  file
78              descriptor  in  any  of  the sets.  In other words, while adding
79              file descriptors to each of the sets,  you  must  calculate  the
80              maximum  integer value of all of them, then increment this value
81              by one, and then pass this as nfds.
82
83       utimeout
84              This is the longest time select()  may  wait  before  returning,
85              even  if  nothing interesting happened.  If this value is passed
86              as NULL, then select() blocks indefinitely waiting  for  a  file
87              descriptor  to  become  ready.  utimeout can be set to zero sec‐
88              onds, which causes select() to return immediately, with informa‐
89              tion  about the readiness of file descriptors at the time of the
90              call.  The structure struct timeval is defined as:
91
92                  struct timeval {
93                      time_t tv_sec;    /* seconds */
94                      long tv_usec;     /* microseconds */
95                  };
96
97       ntimeout
98              This argument for pselect() has the same  meaning  as  utimeout,
99              but struct timespec has nanosecond precision as follows:
100
101                  struct timespec {
102                      long tv_sec;    /* seconds */
103                      long tv_nsec;   /* nanoseconds */
104                  };
105
106       sigmask
107              This  argument  holds  a  set  of signals that the kernel should
108              unblock (i.e., remove  from  the  signal  mask  of  the  calling
109              thread),  while  the caller is blocked inside the pselect() call
110              (see sigaddset(3) and sigprocmask(2)).  It may be NULL, in which
111              case  the call does not modify the signal mask on entry and exit
112              to the function.  In this case, pselect() will then behave  just
113              like select().
114
115   Combining Signal and Data Events
116       pselect() is useful if you are waiting for a signal as well as for file
117       descriptor(s) to become ready for I/O.  Programs that  receive  signals
118       normally  use  the  signal  handler  only  to raise a global flag.  The
119       global flag will indicate that the event must be processed in the  main
120       loop  of  the program.  A signal will cause the select() (or pselect())
121       call to return with errno set to EINTR.  This behavior is essential  so
122       that  signals  can be processed in the main loop of the program, other‐
123       wise select() would block indefinitely.  Now,  somewhere  in  the  main
124       loop  will  be a conditional to check the global flag.  So we must ask:
125       what if a signal arrives after the conditional, but before the select()
126       call?   The  answer  is  that  select()  would block indefinitely, even
127       though an event is actually pending.  This race condition is solved  by
128       the  pselect() call.  This call can be used to set the signal mask to a
129       set of signals that are only to be received within the pselect()  call.
130       For  instance,  let us say that the event in question was the exit of a
131       child process.  Before the start of  the  main  loop,  we  would  block
132       SIGCHLD  using sigprocmask(2).  Our pselect() call would enable SIGCHLD
133       by using an empty signal mask.  Our program would look like:
134
135       static volatile sig_atomic_t got_SIGCHLD = 0;
136
137       static void
138       child_sig_handler(int sig)
139       {
140           got_SIGCHLD = 1;
141       }
142
143       int
144       main(int argc, char *argv[])
145       {
146           sigset_t sigmask, empty_mask;
147           struct sigaction sa;
148           fd_set readfds, writefds, exceptfds;
149           int r;
150
151           sigemptyset(&sigmask);
152           sigaddset(&sigmask, SIGCHLD);
153           if (sigprocmask(SIG_BLOCK, &sigmask, NULL) == -1) {
154               perror("sigprocmask");
155               exit(EXIT_FAILURE);
156           }
157
158           sa.sa_flags = 0;
159           sa.sa_handler = child_sig_handler;
160           sigemptyset(&sa.sa_mask);
161           if (sigaction(SIGCHLD, &sa, NULL) == -1) {
162               perror("sigaction");
163               exit(EXIT_FAILURE);
164           }
165
166           sigemptyset(&empty_mask);
167
168           for (;;) {          /* main loop */
169               /* Initialize readfds, writefds, and exceptfds
170                  before the pselect() call. (Code omitted.) */
171
172               r = pselect(nfds, &readfds, &writefds, &exceptfds,
173                           NULL, &empty_mask);
174               if (r == -1 && errno != EINTR) {
175                   /* Handle error */
176               }
177
178               if (got_SIGCHLD) {
179                   got_SIGCHLD = 0;
180
181                   /* Handle signalled event here; e.g., wait() for all
182                      terminated children. (Code omitted.) */
183               }
184
185               /* main body of program */
186           }
187       }
188
189   Practical
190       So what is the point of select()?  Can't I just read and  write  to  my
191       descriptors  whenever I want?  The point of select() is that it watches
192       multiple descriptors at the same time and properly puts the process  to
193       sleep  if there is no activity.  Unix programmers often find themselves
194       in a position where they have to handle I/O from  more  than  one  file
195       descriptor  where  the  data  flow may be intermittent.  If you were to
196       merely create a sequence of read(2) and write(2) calls, you would  find
197       that  one  of  your  calls  may  block  waiting for data from/to a file
198       descriptor, while another file descriptor is unused  though  ready  for
199       I/O.  select() efficiently copes with this situation.
200
201   Select Law
202       Many people who try to use select() come across behavior that is diffi‐
203       cult to understand and produces nonportable or borderline results.  For
204       instance,  the  above  program is carefully written not to block at any
205       point, even though it does not set its file descriptors to  nonblocking
206       mode.   It  is  easy  to  introduce  subtle errors that will remove the
207       advantage of using select(), so here is a list of essentials  to  watch
208       for when using select().
209
210       1.  You should always try to use select() without a timeout.  Your pro‐
211           gram should have nothing to do if there is no data available.  Code
212           that  depends  on timeouts is not usually portable and is difficult
213           to debug.
214
215       2.  The value nfds  must  be  properly  calculated  for  efficiency  as
216           explained above.
217
218       3.  No file descriptor must be added to any set if you do not intend to
219           check its result after the select()  call,  and  respond  appropri‐
220           ately.  See next rule.
221
222       4.  After  select() returns, all file descriptors in all sets should be
223           checked to see if they are ready.
224
225       5.  The functions read(2), recv(2), write(2), and send(2) do not neces‐
226           sarily  read/write the full amount of data that you have requested.
227           If they do read/write the full amount, it's because you have a  low
228           traffic load and a fast stream.  This is not always going to be the
229           case.  You should cope with the case of your functions only  manag‐
230           ing to send or receive a single byte.
231
232       6.  Never  read/write  only  in  single  bytes at a time unless you are
233           really sure that you have a small amount of data to process.  It is
234           extremely  inefficient  not  to  read/write as much data as you can
235           buffer each time.  The buffers in the example below are 1024  bytes
236           although they could easily be made larger.
237
238       7.  The  functions  read(2),  recv(2), write(2), and send(2) as well as
239           the select() call can return -1 with errno set to  EINTR,  or  with
240           errno  set to EAGAIN (EWOULDBLOCK).  These results must be properly
241           managed (not done properly above).  If your program is not going to
242           receive  any  signals,  then it is unlikely you will get EINTR.  If
243           your program does not set nonblocking I/O, you will not get EAGAIN.
244
245       8.  Never call read(2), recv(2), write(2), or  send(2)  with  a  buffer
246           length of zero.
247
248       9.  If  the functions read(2), recv(2), write(2), and send(2) fail with
249           errors other than those listed in 7., or one of the input functions
250           returns  0,  indicating  end of file, then you should not pass that
251           descriptor to select() again.  In the example below,  I  close  the
252           descriptor  immediately,  and then set it to -1 to prevent it being
253           included in a set.
254
255       10. The timeout value  must  be  initialized  with  each  new  call  to
256           select(),  since some operating systems modify the structure.  pse‐
257           lect() however does not modify its timeout structure.
258
259       11. Since select() modifies its file descriptor sets, if  the  call  is
260           being  used  in  a loop, then the sets must be reinitialized before
261           each call.
262
263   Usleep Emulation
264       On systems that do not have a usleep(3) function, you can call select()
265       with a finite timeout and no file descriptors as follows:
266
267           struct timeval tv;
268           tv.tv_sec = 0;
269           tv.tv_usec = 200000;  /* 0.2 seconds */
270           select(0, NULL, NULL, NULL, &tv);
271
272       This is only guaranteed to work on Unix systems, however.
273

RETURN VALUE

275       On success, select() returns the total number of file descriptors still
276       present in the file descriptor sets.
277
278       If select() timed out, then the return value will be  zero.   The  file
279       descriptors set should be all empty (but may not be on some systems).
280
281       A return value of -1 indicates an error, with errno being set appropri‐
282       ately.  In the case of an error, the contents of the returned sets  and
283       the struct timeout contents are undefined and should not be used.  pse‐
284       lect() however never modifies ntimeout.
285

NOTES

287       Generally speaking, all operating systems  that  support  sockets  also
288       support  select().   select()  can  be used to solve many problems in a
289       portable and efficient way that naive programmers try  to  solve  in  a
290       more  complicated  manner using threads, forking, IPCs, signals, memory
291       sharing, and so on.
292
293       The poll(2) system call has the same functionality as select(), and  is
294       somewhat  more  efficient  when monitoring sparse file descriptor sets.
295       It is nowadays widely available, but  historically  was  less  portable
296       than select().
297
298       The  Linux-specific  epoll(7)  API  provides  an interface that is more
299       efficient than select(2) and poll(2) when monitoring large  numbers  of
300       file descriptors.
301

EXAMPLE

303       Here  is  an  example  that  better  demonstrates  the  true utility of
304       select().  The listing below is a TCP forwarding program that  forwards
305       from one TCP port to another.
306
307       #include <stdlib.h>
308       #include <stdio.h>
309       #include <unistd.h>
310       #include <sys/time.h>
311       #include <sys/types.h>
312       #include <string.h>
313       #include <signal.h>
314       #include <sys/socket.h>
315       #include <netinet/in.h>
316       #include <arpa/inet.h>
317       #include <errno.h>
318
319       static int forward_port;
320
321       #undef max
322       #define max(x,y) ((x) > (y) ? (x) : (y))
323
324       static int
325       listen_socket(int listen_port)
326       {
327           struct sockaddr_in a;
328           int s;
329           int yes;
330
331           if ((s = socket(AF_INET, SOCK_STREAM, 0)) == -1) {
332               perror("socket");
333               return -1;
334           }
335           yes = 1;
336           if (setsockopt(s, SOL_SOCKET, SO_REUSEADDR,
337                   (char *) &yes, sizeof(yes)) == -1) {
338               perror("setsockopt");
339               close(s);
340               return -1;
341           }
342           memset(&a, 0, sizeof(a));
343           a.sin_port = htons(listen_port);
344           a.sin_family = AF_INET;
345           if (bind(s, (struct sockaddr *) &a, sizeof(a)) == -1) {
346               perror("bind");
347               close(s);
348               return -1;
349           }
350           printf("accepting connections on port %d\n", listen_port);
351           listen(s, 10);
352           return s;
353       }
354
355       static int
356       connect_socket(int connect_port, char *address)
357       {
358           struct sockaddr_in a;
359           int s;
360
361           if ((s = socket(AF_INET, SOCK_STREAM, 0)) == -1) {
362               perror("socket");
363               close(s);
364               return -1;
365           }
366
367           memset(&a, 0, sizeof(a));
368           a.sin_port = htons(connect_port);
369           a.sin_family = AF_INET;
370
371           if (!inet_aton(address, (struct in_addr *) &a.sin_addr.s_addr)) {
372               perror("bad IP address format");
373               close(s);
374               return -1;
375           }
376
377           if (connect(s, (struct sockaddr *) &a, sizeof(a)) == -1) {
378               perror("connect()");
379               shutdown(s, SHUT_RDWR);
380               close(s);
381               return -1;
382           }
383           return s;
384       }
385
386       #define SHUT_FD1 do {                                \
387                            if (fd1 >= 0) {                 \
388                                shutdown(fd1, SHUT_RDWR);   \
389                                close(fd1);                 \
390                                fd1 = -1;                   \
391                            }                               \
392                        } while (0)
393
394       #define SHUT_FD2 do {                                \
395                            if (fd2 >= 0) {                 \
396                                shutdown(fd2, SHUT_RDWR);   \
397                                close(fd2);                 \
398                                fd2 = -1;                   \
399                            }                               \
400                        } while (0)
401
402       #define BUF_SIZE 1024
403
404       int
405       main(int argc, char *argv[])
406       {
407           int h;
408           int fd1 = -1, fd2 = -1;
409           char buf1[BUF_SIZE], buf2[BUF_SIZE];
410           int buf1_avail, buf1_written;
411           int buf2_avail, buf2_written;
412
413           if (argc != 4) {
414               fprintf(stderr, "Usage\n\tfwd <listen-port> "
415                        "<forward-to-port> <forward-to-ip-address>\n");
416               exit(EXIT_FAILURE);
417           }
418
419           signal(SIGPIPE, SIG_IGN);
420
421           forward_port = atoi(argv[2]);
422
423           h = listen_socket(atoi(argv[1]));
424           if (h == -1)
425               exit(EXIT_FAILURE);
426
427           for (;;) {
428               int r, nfds = 0;
429               fd_set rd, wr, er;
430
431               FD_ZERO(&rd);
432               FD_ZERO(&wr);
433               FD_ZERO(&er);
434               FD_SET(h, &rd);
435               nfds = max(nfds, h);
436               if (fd1 > 0 && buf1_avail < BUF_SIZE) {
437                   FD_SET(fd1, &rd);
438                   nfds = max(nfds, fd1);
439               }
440               if (fd2 > 0 && buf2_avail < BUF_SIZE) {
441                   FD_SET(fd2, &rd);
442                   nfds = max(nfds, fd2);
443               }
444               if (fd1 > 0 && buf2_avail - buf2_written > 0) {
445                   FD_SET(fd1, &wr);
446                   nfds = max(nfds, fd1);
447               }
448               if (fd2 > 0 && buf1_avail - buf1_written > 0) {
449                   FD_SET(fd2, &wr);
450                   nfds = max(nfds, fd2);
451               }
452               if (fd1 > 0) {
453                   FD_SET(fd1, &er);
454                   nfds = max(nfds, fd1);
455               }
456               if (fd2 > 0) {
457                   FD_SET(fd2, &er);
458                   nfds = max(nfds, fd2);
459               }
460
461               r = select(nfds + 1, &rd, &wr, &er, NULL);
462
463               if (r == -1 && errno == EINTR)
464                   continue;
465
466               if (r == -1) {
467                   perror("select()");
468                   exit(EXIT_FAILURE);
469               }
470
471               if (FD_ISSET(h, &rd)) {
472                   unsigned int l;
473                   struct sockaddr_in client_address;
474
475                   memset(&client_address, 0, l = sizeof(client_address));
476                   r = accept(h, (struct sockaddr *) &client_address, &l);
477                   if (r == -1) {
478                       perror("accept()");
479                   } else {
480                       SHUT_FD1;
481                       SHUT_FD2;
482                       buf1_avail = buf1_written = 0;
483                       buf2_avail = buf2_written = 0;
484                       fd1 = r;
485                       fd2 = connect_socket(forward_port, argv[3]);
486                       if (fd2 == -1)
487                           SHUT_FD1;
488                       else
489                           printf("connect from %s\n",
490                                   inet_ntoa(client_address.sin_addr));
491                   }
492               }
493
494               /* NB: read oob data before normal reads */
495
496               if (fd1 > 0)
497                   if (FD_ISSET(fd1, &er)) {
498                       char c;
499
500                       r = recv(fd1, &c, 1, MSG_OOB);
501                       if (r < 1)
502                           SHUT_FD1;
503                       else
504                           send(fd2, &c, 1, MSG_OOB);
505                   }
506               if (fd2 > 0)
507                   if (FD_ISSET(fd2, &er)) {
508                       char c;
509
510                       r = recv(fd2, &c, 1, MSG_OOB);
511                       if (r < 1)
512                           SHUT_FD2;
513                       else
514                           send(fd1, &c, 1, MSG_OOB);
515                   }
516               if (fd1 > 0)
517                   if (FD_ISSET(fd1, &rd)) {
518                       r = read(fd1, buf1 + buf1_avail,
519                                 BUF_SIZE - buf1_avail);
520                       if (r < 1)
521                           SHUT_FD1;
522                       else
523                           buf1_avail += r;
524                   }
525               if (fd2 > 0)
526                   if (FD_ISSET(fd2, &rd)) {
527                       r = read(fd2, buf2 + buf2_avail,
528                                 BUF_SIZE - buf2_avail);
529                       if (r < 1)
530                           SHUT_FD2;
531                       else
532                           buf2_avail += r;
533                   }
534               if (fd1 > 0)
535                   if (FD_ISSET(fd1, &wr)) {
536                       r = write(fd1, buf2 + buf2_written,
537                                  buf2_avail - buf2_written);
538                       if (r < 1)
539                           SHUT_FD1;
540                       else
541                           buf2_written += r;
542                   }
543               if (fd2 > 0)
544                   if (FD_ISSET(fd2, &wr)) {
545                       r = write(fd2, buf1 + buf1_written,
546                                  buf1_avail - buf1_written);
547                       if (r < 1)
548                           SHUT_FD2;
549                       else
550                           buf1_written += r;
551                   }
552
553               /* check if write data has caught read data */
554
555               if (buf1_written == buf1_avail)
556                   buf1_written = buf1_avail = 0;
557               if (buf2_written == buf2_avail)
558                   buf2_written = buf2_avail = 0;
559
560               /* one side has closed the connection, keep
561                  writing to the other side until empty */
562
563               if (fd1 < 0 && buf1_avail - buf1_written == 0)
564                   SHUT_FD2;
565               if (fd2 < 0 && buf2_avail - buf2_written == 0)
566                   SHUT_FD1;
567           }
568           exit(EXIT_SUCCESS);
569       }
570
571       The  above  program  properly  forwards  most  kinds of TCP connections
572       including OOB signal data transmitted by telnet  servers.   It  handles
573       the  tricky  problem  of having data flow in both directions simultane‐
574       ously.  You might think it more efficient to use  a  fork(2)  call  and
575       devote  a  thread  to  each  stream.  This becomes more tricky than you
576       might suspect.  Another idea is to set nonblocking I/O using  fcntl(2).
577       This  also  has its problems because you end up using inefficient time‐
578       outs.
579
580       The program does not handle more than one simultaneous connection at  a
581       time,  although  it  could  easily be extended to do this with a linked
582       list of buffers — one for each connection.  At the moment, new  connec‐
583       tions cause the current connection to be dropped.
584

SEE ALSO

586       accept(2),  connect(2), ioctl(2), poll(2), read(2), recv(2), select(2),
587       send(2), sigprocmask(2), write(2), sigaddset(3), sigdelset(3),  sigemp‐
588       tyset(3), sigfillset(3), sigismember(3), epoll(7)
589

COLOPHON

591       This  page  is  part of release 3.25 of the Linux man-pages project.  A
592       description of the project, and information about reporting  bugs,  can
593       be found at http://www.kernel.org/doc/man-pages/.
594
595
596
597Linux                             2010-06-10                     SELECT_TUT(2)
Impressum