… | |
… | |
124 | |
124 | |
125 | It supports the following flags: |
125 | It supports the following flags: |
126 | |
126 | |
127 | =over 4 |
127 | =over 4 |
128 | |
128 | |
129 | =item EVFLAG_AUTO |
129 | =item C<EVFLAG_AUTO> |
130 | |
130 | |
131 | The default flags value. Use this if you have no clue (it's the right |
131 | The default flags value. Use this if you have no clue (it's the right |
132 | thing, believe me). |
132 | thing, believe me). |
133 | |
133 | |
134 | =item EVFLAG_NOENV |
134 | =item C<EVFLAG_NOENV> |
135 | |
135 | |
136 | If this flag bit is ored into the flag value (or the program runs setuid |
136 | If this flag bit is ored into the flag value (or the program runs setuid |
137 | or setgid) then libev will I<not> look at the environment variable |
137 | or setgid) then libev will I<not> look at the environment variable |
138 | C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will |
138 | C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will |
139 | override the flags completely if it is found in the environment. This is |
139 | override the flags completely if it is found in the environment. This is |
140 | useful to try out specific backends to test their performance, or to work |
140 | useful to try out specific backends to test their performance, or to work |
141 | around bugs. |
141 | around bugs. |
142 | |
142 | |
143 | =item EVMETHOD_SELECT (portable select backend) |
143 | =item C<EVMETHOD_SELECT> (portable select backend) |
144 | |
144 | |
145 | =item EVMETHOD_POLL (poll backend, available everywhere except on windows) |
145 | =item C<EVMETHOD_POLL> (poll backend, available everywhere except on windows) |
146 | |
146 | |
147 | =item EVMETHOD_EPOLL (linux only) |
147 | =item C<EVMETHOD_EPOLL> (linux only) |
148 | |
148 | |
149 | =item EVMETHOD_KQUEUE (some bsds only) |
149 | =item C<EVMETHOD_KQUEUE> (some bsds only) |
150 | |
150 | |
151 | =item EVMETHOD_DEVPOLL (solaris 8 only) |
151 | =item C<EVMETHOD_DEVPOLL> (solaris 8 only) |
152 | |
152 | |
153 | =item EVMETHOD_PORT (solaris 10 only) |
153 | =item C<EVMETHOD_PORT> (solaris 10 only) |
154 | |
154 | |
155 | If one or more of these are ored into the flags value, then only these |
155 | If one or more of these are ored into the flags value, then only these |
156 | backends will be tried (in the reverse order as given here). If one are |
156 | backends will be tried (in the reverse order as given here). If one are |
157 | specified, any backend will do. |
157 | specified, any backend will do. |
158 | |
158 | |
… | |
… | |
260 | |
260 | |
261 | =head1 ANATOMY OF A WATCHER |
261 | =head1 ANATOMY OF A WATCHER |
262 | |
262 | |
263 | A watcher is a structure that you create and register to record your |
263 | A watcher is a structure that you create and register to record your |
264 | interest in some event. For instance, if you want to wait for STDIN to |
264 | interest in some event. For instance, if you want to wait for STDIN to |
265 | become readable, you would create an ev_io watcher for that: |
265 | become readable, you would create an C<ev_io> watcher for that: |
266 | |
266 | |
267 | static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents) |
267 | static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents) |
268 | { |
268 | { |
269 | ev_io_stop (w); |
269 | ev_io_stop (w); |
270 | ev_unloop (loop, EVUNLOOP_ALL); |
270 | ev_unloop (loop, EVUNLOOP_ALL); |
… | |
… | |
314 | (you can receive multiple events at the same time). The possible bit masks |
314 | (you can receive multiple events at the same time). The possible bit masks |
315 | are: |
315 | are: |
316 | |
316 | |
317 | =over 4 |
317 | =over 4 |
318 | |
318 | |
319 | =item EV_READ |
319 | =item C<EV_READ> |
320 | |
320 | |
321 | =item EV_WRITE |
321 | =item C<EV_WRITE> |
322 | |
322 | |
323 | The file descriptor in the ev_io watcher has become readable and/or |
323 | The file descriptor in the C<ev_io> watcher has become readable and/or |
324 | writable. |
324 | writable. |
325 | |
325 | |
326 | =item EV_TIMEOUT |
326 | =item C<EV_TIMEOUT> |
327 | |
327 | |
328 | The ev_timer watcher has timed out. |
328 | The C<ev_timer> watcher has timed out. |
329 | |
329 | |
330 | =item EV_PERIODIC |
330 | =item C<EV_PERIODIC> |
331 | |
331 | |
332 | The ev_periodic watcher has timed out. |
332 | The C<ev_periodic> watcher has timed out. |
333 | |
333 | |
334 | =item EV_SIGNAL |
334 | =item C<EV_SIGNAL> |
335 | |
335 | |
336 | The signal specified in the ev_signal watcher has been received by a thread. |
336 | The signal specified in the C<ev_signal> watcher has been received by a thread. |
337 | |
337 | |
338 | =item EV_CHILD |
338 | =item C<EV_CHILD> |
339 | |
339 | |
340 | The pid specified in the ev_child watcher has received a status change. |
340 | The pid specified in the C<ev_child> watcher has received a status change. |
341 | |
341 | |
342 | =item EV_IDLE |
342 | =item C<EV_IDLE> |
343 | |
343 | |
344 | The ev_idle watcher has determined that you have nothing better to do. |
344 | The C<ev_idle> watcher has determined that you have nothing better to do. |
345 | |
345 | |
346 | =item EV_PREPARE |
346 | =item C<EV_PREPARE> |
347 | |
347 | |
348 | =item EV_CHECK |
348 | =item C<EV_CHECK> |
349 | |
349 | |
350 | All ev_prepare watchers are invoked just I<before> C<ev_loop> starts |
350 | All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts |
351 | to gather new events, and all ev_check watchers are invoked just after |
351 | to gather new events, and all C<ev_check> watchers are invoked just after |
352 | C<ev_loop> has gathered them, but before it invokes any callbacks for any |
352 | C<ev_loop> has gathered them, but before it invokes any callbacks for any |
353 | received events. Callbacks of both watcher types can start and stop as |
353 | received events. Callbacks of both watcher types can start and stop as |
354 | many watchers as they want, and all of them will be taken into account |
354 | many watchers as they want, and all of them will be taken into account |
355 | (for example, a ev_prepare watcher might start an idle watcher to keep |
355 | (for example, a C<ev_prepare> watcher might start an idle watcher to keep |
356 | C<ev_loop> from blocking). |
356 | C<ev_loop> from blocking). |
357 | |
357 | |
358 | =item EV_ERROR |
358 | =item C<EV_ERROR> |
359 | |
359 | |
360 | An unspecified error has occured, the watcher has been stopped. This might |
360 | An unspecified error has occured, the watcher has been stopped. This might |
361 | happen because the watcher could not be properly started because libev |
361 | happen because the watcher could not be properly started because libev |
362 | ran out of memory, a file descriptor was found to be closed or any other |
362 | ran out of memory, a file descriptor was found to be closed or any other |
363 | problem. You best act on it by reporting the problem and somehow coping |
363 | problem. You best act on it by reporting the problem and somehow coping |
… | |
… | |
404 | =head1 WATCHER TYPES |
404 | =head1 WATCHER TYPES |
405 | |
405 | |
406 | This section describes each watcher in detail, but will not repeat |
406 | This section describes each watcher in detail, but will not repeat |
407 | information given in the last section. |
407 | information given in the last section. |
408 | |
408 | |
409 | =head2 struct ev_io - is my file descriptor readable or writable |
409 | =head2 C<ev_io> - is this file descriptor readable or writable |
410 | |
410 | |
411 | I/O watchers check whether a file descriptor is readable or writable |
411 | I/O watchers check whether a file descriptor is readable or writable |
412 | in each iteration of the event loop (This behaviour is called |
412 | in each iteration of the event loop (This behaviour is called |
413 | level-triggering because you keep receiving events as long as the |
413 | level-triggering because you keep receiving events as long as the |
414 | condition persists. Remember you cna stop the watcher if you don't want to |
414 | condition persists. Remember you cna stop the watcher if you don't want to |
… | |
… | |
432 | |
432 | |
433 | =item ev_io_init (ev_io *, callback, int fd, int events) |
433 | =item ev_io_init (ev_io *, callback, int fd, int events) |
434 | |
434 | |
435 | =item ev_io_set (ev_io *, int fd, int events) |
435 | =item ev_io_set (ev_io *, int fd, int events) |
436 | |
436 | |
437 | Configures an ev_io watcher. The fd is the file descriptor to rceeive |
437 | Configures an C<ev_io> watcher. The fd is the file descriptor to rceeive |
438 | events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ | |
438 | events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ | |
439 | EV_WRITE> to receive the given events. |
439 | EV_WRITE> to receive the given events. |
440 | |
440 | |
441 | =back |
441 | =back |
442 | |
442 | |
443 | =head2 struct ev_timer - relative and optionally recurring timeouts |
443 | =head2 C<ev_timer> - relative and optionally recurring timeouts |
444 | |
444 | |
445 | Timer watchers are simple relative timers that generate an event after a |
445 | Timer watchers are simple relative timers that generate an event after a |
446 | given time, and optionally repeating in regular intervals after that. |
446 | given time, and optionally repeating in regular intervals after that. |
447 | |
447 | |
448 | The timers are based on real time, that is, if you register an event that |
448 | The timers are based on real time, that is, if you register an event that |
… | |
… | |
488 | |
488 | |
489 | This sounds a bit complicated, but here is a useful and typical |
489 | This sounds a bit complicated, but here is a useful and typical |
490 | example: Imagine you have a tcp connection and you want a so-called idle |
490 | example: Imagine you have a tcp connection and you want a so-called idle |
491 | timeout, that is, you want to be called when there have been, say, 60 |
491 | timeout, that is, you want to be called when there have been, say, 60 |
492 | seconds of inactivity on the socket. The easiest way to do this is to |
492 | seconds of inactivity on the socket. The easiest way to do this is to |
493 | configure an ev_timer with after=repeat=60 and calling ev_timer_again each |
493 | configure an C<ev_timer> with after=repeat=60 and calling ev_timer_again each |
494 | time you successfully read or write some data. If you go into an idle |
494 | time you successfully read or write some data. If you go into an idle |
495 | state where you do not expect data to travel on the socket, you can stop |
495 | state where you do not expect data to travel on the socket, you can stop |
496 | the timer, and again will automatically restart it if need be. |
496 | the timer, and again will automatically restart it if need be. |
497 | |
497 | |
498 | =back |
498 | =back |
499 | |
499 | |
500 | =head2 ev_periodic - to cron or not to cron it |
500 | =head2 C<ev_periodic> - to cron or not to cron it |
501 | |
501 | |
502 | Periodic watchers are also timers of a kind, but they are very versatile |
502 | Periodic watchers are also timers of a kind, but they are very versatile |
503 | (and unfortunately a bit complex). |
503 | (and unfortunately a bit complex). |
504 | |
504 | |
505 | Unlike ev_timer's, they are not based on real time (or relative time) |
505 | Unlike C<ev_timer>'s, they are not based on real time (or relative time) |
506 | but on wallclock time (absolute time). You can tell a periodic watcher |
506 | but on wallclock time (absolute time). You can tell a periodic watcher |
507 | to trigger "at" some specific point in time. For example, if you tell a |
507 | to trigger "at" some specific point in time. For example, if you tell a |
508 | periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () |
508 | periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now () |
509 | + 10.>) and then reset your system clock to the last year, then it will |
509 | + 10.>) and then reset your system clock to the last year, then it will |
510 | take a year to trigger the event (unlike an ev_timer, which would trigger |
510 | take a year to trigger the event (unlike an C<ev_timer>, which would trigger |
511 | roughly 10 seconds later and of course not if you reset your system time |
511 | roughly 10 seconds later and of course not if you reset your system time |
512 | again). |
512 | again). |
513 | |
513 | |
514 | They can also be used to implement vastly more complex timers, such as |
514 | They can also be used to implement vastly more complex timers, such as |
515 | triggering an event on eahc midnight, local time. |
515 | triggering an event on eahc midnight, local time. |
… | |
… | |
544 | |
544 | |
545 | ev_periodic_set (&periodic, 0., 3600., 0); |
545 | ev_periodic_set (&periodic, 0., 3600., 0); |
546 | |
546 | |
547 | This doesn't mean there will always be 3600 seconds in between triggers, |
547 | This doesn't mean there will always be 3600 seconds in between triggers, |
548 | but only that the the callback will be called when the system time shows a |
548 | but only that the the callback will be called when the system time shows a |
549 | full hour (UTC), or more correct, when the system time is evenly divisible |
549 | full hour (UTC), or more correctly, when the system time is evenly divisible |
550 | by 3600. |
550 | by 3600. |
551 | |
551 | |
552 | Another way to think about it (for the mathematically inclined) is that |
552 | Another way to think about it (for the mathematically inclined) is that |
553 | ev_periodic will try to run the callback in this mode at the next possible |
553 | C<ev_periodic> will try to run the callback in this mode at the next possible |
554 | time where C<time = at (mod interval)>, regardless of any time jumps. |
554 | time where C<time = at (mod interval)>, regardless of any time jumps. |
555 | |
555 | |
556 | =item * manual reschedule mode (reschedule_cb = callback) |
556 | =item * manual reschedule mode (reschedule_cb = callback) |
557 | |
557 | |
558 | In this mode the values for C<interval> and C<at> are both being |
558 | In this mode the values for C<interval> and C<at> are both being |
559 | ignored. Instead, each time the periodic watcher gets scheduled, the |
559 | ignored. Instead, each time the periodic watcher gets scheduled, the |
560 | reschedule callback will be called with the watcher as first, and the |
560 | reschedule callback will be called with the watcher as first, and the |
561 | current time as second argument. |
561 | current time as second argument. |
562 | |
562 | |
563 | NOTE: I<This callback MUST NOT stop or destroy the periodic or any other |
563 | NOTE: I<This callback MUST NOT stop or destroy the periodic or any other |
564 | periodic watcher, ever, or make any event loop modificstions>. If you need |
564 | periodic watcher, ever, or make any event loop modifications>. If you need |
565 | to stop it, return 1e30 (or so, fudge fudge) and stop it afterwards. |
565 | to stop it, return C<now + 1e30> (or so, fudge fudge) and stop it afterwards. |
566 | |
566 | |
|
|
567 | Also, I<< this callback must always return a time that is later than the |
|
|
568 | passed C<now> value >>. Not even C<now> itself will be ok. |
|
|
569 | |
567 | Its prototype is c<ev_tstamp (*reschedule_cb)(struct ev_periodic *w, |
570 | Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w, |
568 | ev_tstamp now)>, e.g.: |
571 | ev_tstamp now)>, e.g.: |
569 | |
572 | |
570 | static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now) |
573 | static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now) |
571 | { |
574 | { |
572 | return now + 60.; |
575 | return now + 60.; |
… | |
… | |
591 | a different time than the last time it was called (e.g. in a crond like |
594 | a different time than the last time it was called (e.g. in a crond like |
592 | program when the crontabs have changed). |
595 | program when the crontabs have changed). |
593 | |
596 | |
594 | =back |
597 | =back |
595 | |
598 | |
596 | =head2 ev_signal - signal me when a signal gets signalled |
599 | =head2 C<ev_signal> - signal me when a signal gets signalled |
597 | |
600 | |
598 | Signal watchers will trigger an event when the process receives a specific |
601 | Signal watchers will trigger an event when the process receives a specific |
599 | signal one or more times. Even though signals are very asynchronous, libev |
602 | signal one or more times. Even though signals are very asynchronous, libev |
600 | will try it's best to deliver signals synchronously, i.e. as part of the |
603 | will try it's best to deliver signals synchronously, i.e. as part of the |
601 | normal event processing, like any other event. |
604 | normal event processing, like any other event. |
… | |
… | |
616 | Configures the watcher to trigger on the given signal number (usually one |
619 | Configures the watcher to trigger on the given signal number (usually one |
617 | of the C<SIGxxx> constants). |
620 | of the C<SIGxxx> constants). |
618 | |
621 | |
619 | =back |
622 | =back |
620 | |
623 | |
621 | =head2 ev_child - wait for pid status changes |
624 | =head2 C<ev_child> - wait for pid status changes |
622 | |
625 | |
623 | Child watchers trigger when your process receives a SIGCHLD in response to |
626 | Child watchers trigger when your process receives a SIGCHLD in response to |
624 | some child status changes (most typically when a child of yours dies). |
627 | some child status changes (most typically when a child of yours dies). |
625 | |
628 | |
626 | =over 4 |
629 | =over 4 |
… | |
… | |
635 | the status word (use the macros from C<sys/wait.h>). The C<rpid> member |
638 | the status word (use the macros from C<sys/wait.h>). The C<rpid> member |
636 | contains the pid of the process causing the status change. |
639 | contains the pid of the process causing the status change. |
637 | |
640 | |
638 | =back |
641 | =back |
639 | |
642 | |
640 | =head2 ev_idle - when you've got nothing better to do |
643 | =head2 C<ev_idle> - when you've got nothing better to do |
641 | |
644 | |
642 | Idle watchers trigger events when there are no other I/O or timer (or |
645 | Idle watchers trigger events when there are no other I/O or timer (or |
643 | periodic) events pending. That is, as long as your process is busy |
646 | periodic) events pending. That is, as long as your process is busy |
644 | handling sockets or timeouts it will not be called. But when your process |
647 | handling sockets or timeouts it will not be called. But when your process |
645 | is idle all idle watchers are being called again and again - until |
648 | is idle all idle watchers are being called again and again - until |
… | |
… | |
672 | Their main purpose is to integrate other event mechanisms into libev. This |
675 | Their main purpose is to integrate other event mechanisms into libev. This |
673 | could be used, for example, to track variable changes, implement your own |
676 | could be used, for example, to track variable changes, implement your own |
674 | watchers, integrate net-snmp or a coroutine library and lots more. |
677 | watchers, integrate net-snmp or a coroutine library and lots more. |
675 | |
678 | |
676 | This is done by examining in each prepare call which file descriptors need |
679 | This is done by examining in each prepare call which file descriptors need |
677 | to be watched by the other library, registering ev_io watchers for them |
680 | to be watched by the other library, registering C<ev_io> watchers for them |
678 | and starting an ev_timer watcher for any timeouts (many libraries provide |
681 | and starting an C<ev_timer> watcher for any timeouts (many libraries provide |
679 | just this functionality). Then, in the check watcher you check for any |
682 | just this functionality). Then, in the check watcher you check for any |
680 | events that occured (by making your callbacks set soem flags for example) |
683 | events that occured (by making your callbacks set soem flags for example) |
681 | and call back into the library. |
684 | and call back into the library. |
682 | |
685 | |
683 | As another example, the perl Coro module uses these hooks to integrate |
686 | As another example, the perl Coro module uses these hooks to integrate |
… | |
… | |
710 | watchers. This is useful if you want to wait for a single event on an fd |
713 | watchers. This is useful if you want to wait for a single event on an fd |
711 | or timeout without havign to allocate/configure/start/stop/free one or |
714 | or timeout without havign to allocate/configure/start/stop/free one or |
712 | more watchers yourself. |
715 | more watchers yourself. |
713 | |
716 | |
714 | If C<fd> is less than 0, then no I/O watcher will be started and events is |
717 | If C<fd> is less than 0, then no I/O watcher will be started and events is |
715 | ignored. Otherwise, an ev_io watcher for the given C<fd> and C<events> set |
718 | ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and C<events> set |
716 | will be craeted and started. |
719 | will be craeted and started. |
717 | |
720 | |
718 | If C<timeout> is less than 0, then no timeout watcher will be |
721 | If C<timeout> is less than 0, then no timeout watcher will be |
719 | started. Otherwise an ev_timer watcher with after = C<timeout> (and repeat |
722 | started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and repeat |
720 | = 0) will be started. |
723 | = 0) will be started. |
721 | |
724 | |
722 | The callback has the type C<void (*cb)(int revents, void *arg)> and |
725 | The callback has the type C<void (*cb)(int revents, void *arg)> and |
723 | gets passed an events set (normally a combination of EV_ERROR, EV_READ, |
726 | gets passed an events set (normally a combination of C<EV_ERROR>, C<EV_READ>, |
724 | EV_WRITE or EV_TIMEOUT) and the C<arg> value passed to C<ev_once>: |
727 | C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg> value passed to C<ev_once>: |
725 | |
728 | |
726 | static void stdin_ready (int revents, void *arg) |
729 | static void stdin_ready (int revents, void *arg) |
727 | { |
730 | { |
728 | if (revents & EV_TIMEOUT) |
731 | if (revents & EV_TIMEOUT) |
729 | /* doh, nothing entered */ |
732 | /* doh, nothing entered */ |