|
|
1 | =encoding utf-8 |
|
|
2 | |
1 | =head1 NAME |
3 | =head1 NAME |
2 | |
4 | |
3 | libev - a high performance full-featured event loop written in C |
5 | libev - a high performance full-featured event loop written in C |
4 | |
6 | |
5 | =head1 SYNOPSIS |
7 | =head1 SYNOPSIS |
… | |
… | |
82 | |
84 | |
83 | =head1 WHAT TO READ WHEN IN A HURRY |
85 | =head1 WHAT TO READ WHEN IN A HURRY |
84 | |
86 | |
85 | This manual tries to be very detailed, but unfortunately, this also makes |
87 | This manual tries to be very detailed, but unfortunately, this also makes |
86 | it very long. If you just want to know the basics of libev, I suggest |
88 | it very long. If you just want to know the basics of libev, I suggest |
87 | reading L<ANATOMY OF A WATCHER>, then the L<EXAMPLE PROGRAM> above and |
89 | reading L</ANATOMY OF A WATCHER>, then the L</EXAMPLE PROGRAM> above and |
88 | look up the missing functions in L<GLOBAL FUNCTIONS> and the C<ev_io> and |
90 | look up the missing functions in L</GLOBAL FUNCTIONS> and the C<ev_io> and |
89 | C<ev_timer> sections in L<WATCHER TYPES>. |
91 | C<ev_timer> sections in L</WATCHER TYPES>. |
90 | |
92 | |
91 | =head1 ABOUT LIBEV |
93 | =head1 ABOUT LIBEV |
92 | |
94 | |
93 | Libev is an event loop: you register interest in certain events (such as a |
95 | Libev is an event loop: you register interest in certain events (such as a |
94 | file descriptor being readable or a timeout occurring), and it will manage |
96 | file descriptor being readable or a timeout occurring), and it will manage |
… | |
… | |
396 | |
398 | |
397 | If this flag bit is or'ed into the flag value (or the program runs setuid |
399 | If this flag bit is or'ed into the flag value (or the program runs setuid |
398 | or setgid) then libev will I<not> look at the environment variable |
400 | or setgid) then libev will I<not> look at the environment variable |
399 | C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will |
401 | C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will |
400 | override the flags completely if it is found in the environment. This is |
402 | override the flags completely if it is found in the environment. This is |
401 | useful to try out specific backends to test their performance, or to work |
403 | useful to try out specific backends to test their performance, to work |
402 | around bugs. |
404 | around bugs, or to make libev threadsafe (accessing environment variables |
|
|
405 | cannot be done in a threadsafe way, but usually it works if no other |
|
|
406 | thread modifies them). |
403 | |
407 | |
404 | =item C<EVFLAG_FORKCHECK> |
408 | =item C<EVFLAG_FORKCHECK> |
405 | |
409 | |
406 | Instead of calling C<ev_loop_fork> manually after a fork, you can also |
410 | Instead of calling C<ev_loop_fork> manually after a fork, you can also |
407 | make libev check for a fork in each iteration by enabling this flag. |
411 | make libev check for a fork in each iteration by enabling this flag. |
… | |
… | |
569 | kernel is more efficient (which says nothing about its actual speed, of |
573 | kernel is more efficient (which says nothing about its actual speed, of |
570 | course). While stopping, setting and starting an I/O watcher does never |
574 | course). While stopping, setting and starting an I/O watcher does never |
571 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
575 | cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to |
572 | two event changes per incident. Support for C<fork ()> is very bad (you |
576 | two event changes per incident. Support for C<fork ()> is very bad (you |
573 | might have to leak fd's on fork, but it's more sane than epoll) and it |
577 | might have to leak fd's on fork, but it's more sane than epoll) and it |
574 | drops fds silently in similarly hard-to-detect cases |
578 | drops fds silently in similarly hard-to-detect cases. |
575 | |
579 | |
576 | This backend usually performs well under most conditions. |
580 | This backend usually performs well under most conditions. |
577 | |
581 | |
578 | While nominally embeddable in other event loops, this doesn't work |
582 | While nominally embeddable in other event loops, this doesn't work |
579 | everywhere, so you might need to test for this. And since it is broken |
583 | everywhere, so you might need to test for this. And since it is broken |
… | |
… | |
684 | reinitialise the kernel state for backends that have one. Despite the |
688 | reinitialise the kernel state for backends that have one. Despite the |
685 | name, you can call it anytime, but it makes most sense after forking, in |
689 | name, you can call it anytime, but it makes most sense after forking, in |
686 | the child process. You I<must> call it (or use C<EVFLAG_FORKCHECK>) in the |
690 | the child process. You I<must> call it (or use C<EVFLAG_FORKCHECK>) in the |
687 | child before resuming or calling C<ev_run>. |
691 | child before resuming or calling C<ev_run>. |
688 | |
692 | |
689 | Again, you I<have> to call it on I<any> loop that you want to re-use after |
693 | Again, you I<have> to call it on I<any> loop that you want to re-use after |
690 | a fork, I<even if you do not plan to use the loop in the parent>. This is |
694 | a fork, I<even if you do not plan to use the loop in the parent>. This is |
691 | because some kernel interfaces *cough* I<kqueue> *cough* do funny things |
695 | because some kernel interfaces *cough* I<kqueue> *cough* do funny things |
692 | during fork. |
696 | during fork. |
693 | |
697 | |
694 | On the other hand, you only need to call this function in the child |
698 | On the other hand, you only need to call this function in the child |
… | |
… | |
764 | |
768 | |
765 | This function is rarely useful, but when some event callback runs for a |
769 | This function is rarely useful, but when some event callback runs for a |
766 | very long time without entering the event loop, updating libev's idea of |
770 | very long time without entering the event loop, updating libev's idea of |
767 | the current time is a good idea. |
771 | the current time is a good idea. |
768 | |
772 | |
769 | See also L<The special problem of time updates> in the C<ev_timer> section. |
773 | See also L</The special problem of time updates> in the C<ev_timer> section. |
770 | |
774 | |
771 | =item ev_suspend (loop) |
775 | =item ev_suspend (loop) |
772 | |
776 | |
773 | =item ev_resume (loop) |
777 | =item ev_resume (loop) |
774 | |
778 | |
… | |
… | |
1318 | |
1322 | |
1319 | =item callback ev_cb (ev_TYPE *watcher) |
1323 | =item callback ev_cb (ev_TYPE *watcher) |
1320 | |
1324 | |
1321 | Returns the callback currently set on the watcher. |
1325 | Returns the callback currently set on the watcher. |
1322 | |
1326 | |
1323 | =item ev_cb_set (ev_TYPE *watcher, callback) |
1327 | =item ev_set_cb (ev_TYPE *watcher, callback) |
1324 | |
1328 | |
1325 | Change the callback. You can change the callback at virtually any time |
1329 | Change the callback. You can change the callback at virtually any time |
1326 | (modulo threads). |
1330 | (modulo threads). |
1327 | |
1331 | |
1328 | =item ev_set_priority (ev_TYPE *watcher, int priority) |
1332 | =item ev_set_priority (ev_TYPE *watcher, int priority) |
… | |
… | |
1346 | or might not have been clamped to the valid range. |
1350 | or might not have been clamped to the valid range. |
1347 | |
1351 | |
1348 | The default priority used by watchers when no priority has been set is |
1352 | The default priority used by watchers when no priority has been set is |
1349 | always C<0>, which is supposed to not be too high and not be too low :). |
1353 | always C<0>, which is supposed to not be too high and not be too low :). |
1350 | |
1354 | |
1351 | See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of |
1355 | See L</WATCHER PRIORITY MODELS>, below, for a more thorough treatment of |
1352 | priorities. |
1356 | priorities. |
1353 | |
1357 | |
1354 | =item ev_invoke (loop, ev_TYPE *watcher, int revents) |
1358 | =item ev_invoke (loop, ev_TYPE *watcher, int revents) |
1355 | |
1359 | |
1356 | Invoke the C<watcher> with the given C<loop> and C<revents>. Neither |
1360 | Invoke the C<watcher> with the given C<loop> and C<revents>. Neither |
… | |
… | |
1381 | See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related |
1385 | See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related |
1382 | functions that do not need a watcher. |
1386 | functions that do not need a watcher. |
1383 | |
1387 | |
1384 | =back |
1388 | =back |
1385 | |
1389 | |
1386 | See also the L<ASSOCIATING CUSTOM DATA WITH A WATCHER> and L<BUILDING YOUR |
1390 | See also the L</ASSOCIATING CUSTOM DATA WITH A WATCHER> and L</BUILDING YOUR |
1387 | OWN COMPOSITE WATCHERS> idioms. |
1391 | OWN COMPOSITE WATCHERS> idioms. |
1388 | |
1392 | |
1389 | =head2 WATCHER STATES |
1393 | =head2 WATCHER STATES |
1390 | |
1394 | |
1391 | There are various watcher states mentioned throughout this manual - |
1395 | There are various watcher states mentioned throughout this manual - |
… | |
… | |
1393 | transition between them will be described in more detail - and while these |
1397 | transition between them will be described in more detail - and while these |
1394 | rules might look complicated, they usually do "the right thing". |
1398 | rules might look complicated, they usually do "the right thing". |
1395 | |
1399 | |
1396 | =over 4 |
1400 | =over 4 |
1397 | |
1401 | |
1398 | =item initialiased |
1402 | =item initialised |
1399 | |
1403 | |
1400 | Before a watcher can be registered with the event loop it has to be |
1404 | Before a watcher can be registered with the event loop it has to be |
1401 | initialised. This can be done with a call to C<ev_TYPE_init>, or calls to |
1405 | initialised. This can be done with a call to C<ev_TYPE_init>, or calls to |
1402 | C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function. |
1406 | C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function. |
1403 | |
1407 | |
… | |
… | |
2136 | =item If the timer is repeating, make the C<repeat> value the new timeout |
2140 | =item If the timer is repeating, make the C<repeat> value the new timeout |
2137 | and start the timer, if necessary. |
2141 | and start the timer, if necessary. |
2138 | |
2142 | |
2139 | =back |
2143 | =back |
2140 | |
2144 | |
2141 | This sounds a bit complicated, see L<Be smart about timeouts>, above, for a |
2145 | This sounds a bit complicated, see L</Be smart about timeouts>, above, for a |
2142 | usage example. |
2146 | usage example. |
2143 | |
2147 | |
2144 | =item ev_tstamp ev_timer_remaining (loop, ev_timer *) |
2148 | =item ev_tstamp ev_timer_remaining (loop, ev_timer *) |
2145 | |
2149 | |
2146 | Returns the remaining time until a timer fires. If the timer is active, |
2150 | Returns the remaining time until a timer fires. If the timer is active, |
… | |
… | |
2389 | |
2393 | |
2390 | ev_periodic hourly_tick; |
2394 | ev_periodic hourly_tick; |
2391 | ev_periodic_init (&hourly_tick, clock_cb, |
2395 | ev_periodic_init (&hourly_tick, clock_cb, |
2392 | fmod (ev_now (loop), 3600.), 3600., 0); |
2396 | fmod (ev_now (loop), 3600.), 3600., 0); |
2393 | ev_periodic_start (loop, &hourly_tick); |
2397 | ev_periodic_start (loop, &hourly_tick); |
2394 | |
2398 | |
2395 | |
2399 | |
2396 | =head2 C<ev_signal> - signal me when a signal gets signalled! |
2400 | =head2 C<ev_signal> - signal me when a signal gets signalled! |
2397 | |
2401 | |
2398 | Signal watchers will trigger an event when the process receives a specific |
2402 | Signal watchers will trigger an event when the process receives a specific |
2399 | signal one or more times. Even though signals are very asynchronous, libev |
2403 | signal one or more times. Even though signals are very asynchronous, libev |
… | |
… | |
2409 | only within the same loop, i.e. you can watch for C<SIGINT> in your |
2413 | only within the same loop, i.e. you can watch for C<SIGINT> in your |
2410 | default loop and for C<SIGIO> in another loop, but you cannot watch for |
2414 | default loop and for C<SIGIO> in another loop, but you cannot watch for |
2411 | C<SIGINT> in both the default loop and another loop at the same time. At |
2415 | C<SIGINT> in both the default loop and another loop at the same time. At |
2412 | the moment, C<SIGCHLD> is permanently tied to the default loop. |
2416 | the moment, C<SIGCHLD> is permanently tied to the default loop. |
2413 | |
2417 | |
2414 | When the first watcher gets started will libev actually register something |
2418 | Only after the first watcher for a signal is started will libev actually |
2415 | with the kernel (thus it coexists with your own signal handlers as long as |
2419 | register something with the kernel. It thus coexists with your own signal |
2416 | you don't register any with libev for the same signal). |
2420 | handlers as long as you don't register any with libev for the same signal. |
2417 | |
2421 | |
2418 | If possible and supported, libev will install its handlers with |
2422 | If possible and supported, libev will install its handlers with |
2419 | C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should |
2423 | C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should |
2420 | not be unduly interrupted. If you have a problem with system calls getting |
2424 | not be unduly interrupted. If you have a problem with system calls getting |
2421 | interrupted by signals you can block all signals in an C<ev_check> watcher |
2425 | interrupted by signals you can block all signals in an C<ev_check> watcher |
… | |
… | |
2606 | |
2610 | |
2607 | =head2 C<ev_stat> - did the file attributes just change? |
2611 | =head2 C<ev_stat> - did the file attributes just change? |
2608 | |
2612 | |
2609 | This watches a file system path for attribute changes. That is, it calls |
2613 | This watches a file system path for attribute changes. That is, it calls |
2610 | C<stat> on that path in regular intervals (or when the OS says it changed) |
2614 | C<stat> on that path in regular intervals (or when the OS says it changed) |
2611 | and sees if it changed compared to the last time, invoking the callback if |
2615 | and sees if it changed compared to the last time, invoking the callback |
2612 | it did. |
2616 | if it did. Starting the watcher C<stat>'s the file, so only changes that |
|
|
2617 | happen after the watcher has been started will be reported. |
2613 | |
2618 | |
2614 | The path does not need to exist: changing from "path exists" to "path does |
2619 | The path does not need to exist: changing from "path exists" to "path does |
2615 | not exist" is a status change like any other. The condition "path does not |
2620 | not exist" is a status change like any other. The condition "path does not |
2616 | exist" (or more correctly "path cannot be stat'ed") is signified by the |
2621 | exist" (or more correctly "path cannot be stat'ed") is signified by the |
2617 | C<st_nlink> field being zero (which is otherwise always forced to be at |
2622 | C<st_nlink> field being zero (which is otherwise always forced to be at |
… | |
… | |
2847 | Apart from keeping your process non-blocking (which is a useful |
2852 | Apart from keeping your process non-blocking (which is a useful |
2848 | effect on its own sometimes), idle watchers are a good place to do |
2853 | effect on its own sometimes), idle watchers are a good place to do |
2849 | "pseudo-background processing", or delay processing stuff to after the |
2854 | "pseudo-background processing", or delay processing stuff to after the |
2850 | event loop has handled all outstanding events. |
2855 | event loop has handled all outstanding events. |
2851 | |
2856 | |
|
|
2857 | =head3 Abusing an C<ev_idle> watcher for its side-effect |
|
|
2858 | |
|
|
2859 | As long as there is at least one active idle watcher, libev will never |
|
|
2860 | sleep unnecessarily. Or in other words, it will loop as fast as possible. |
|
|
2861 | For this to work, the idle watcher doesn't need to be invoked at all - the |
|
|
2862 | lowest priority will do. |
|
|
2863 | |
|
|
2864 | This mode of operation can be useful together with an C<ev_check> watcher, |
|
|
2865 | to do something on each event loop iteration - for example to balance load |
|
|
2866 | between different connections. |
|
|
2867 | |
|
|
2868 | See L</Abusing an ev_check watcher for its side-effect> for a longer |
|
|
2869 | example. |
|
|
2870 | |
2852 | =head3 Watcher-Specific Functions and Data Members |
2871 | =head3 Watcher-Specific Functions and Data Members |
2853 | |
2872 | |
2854 | =over 4 |
2873 | =over 4 |
2855 | |
2874 | |
2856 | =item ev_idle_init (ev_idle *, callback) |
2875 | =item ev_idle_init (ev_idle *, callback) |
… | |
… | |
2867 | callback, free it. Also, use no error checking, as usual. |
2886 | callback, free it. Also, use no error checking, as usual. |
2868 | |
2887 | |
2869 | static void |
2888 | static void |
2870 | idle_cb (struct ev_loop *loop, ev_idle *w, int revents) |
2889 | idle_cb (struct ev_loop *loop, ev_idle *w, int revents) |
2871 | { |
2890 | { |
|
|
2891 | // stop the watcher |
|
|
2892 | ev_idle_stop (loop, w); |
|
|
2893 | |
|
|
2894 | // now we can free it |
2872 | free (w); |
2895 | free (w); |
|
|
2896 | |
2873 | // now do something you wanted to do when the program has |
2897 | // now do something you wanted to do when the program has |
2874 | // no longer anything immediate to do. |
2898 | // no longer anything immediate to do. |
2875 | } |
2899 | } |
2876 | |
2900 | |
2877 | ev_idle *idle_watcher = malloc (sizeof (ev_idle)); |
2901 | ev_idle *idle_watcher = malloc (sizeof (ev_idle)); |
… | |
… | |
2879 | ev_idle_start (loop, idle_watcher); |
2903 | ev_idle_start (loop, idle_watcher); |
2880 | |
2904 | |
2881 | |
2905 | |
2882 | =head2 C<ev_prepare> and C<ev_check> - customise your event loop! |
2906 | =head2 C<ev_prepare> and C<ev_check> - customise your event loop! |
2883 | |
2907 | |
2884 | Prepare and check watchers are usually (but not always) used in pairs: |
2908 | Prepare and check watchers are often (but not always) used in pairs: |
2885 | prepare watchers get invoked before the process blocks and check watchers |
2909 | prepare watchers get invoked before the process blocks and check watchers |
2886 | afterwards. |
2910 | afterwards. |
2887 | |
2911 | |
2888 | You I<must not> call C<ev_run> or similar functions that enter |
2912 | You I<must not> call C<ev_run> or similar functions that enter |
2889 | the current event loop from either C<ev_prepare> or C<ev_check> |
2913 | the current event loop from either C<ev_prepare> or C<ev_check> |
… | |
… | |
2917 | with priority higher than or equal to the event loop and one coroutine |
2941 | with priority higher than or equal to the event loop and one coroutine |
2918 | of lower priority, but only once, using idle watchers to keep the event |
2942 | of lower priority, but only once, using idle watchers to keep the event |
2919 | loop from blocking if lower-priority coroutines are active, thus mapping |
2943 | loop from blocking if lower-priority coroutines are active, thus mapping |
2920 | low-priority coroutines to idle/background tasks). |
2944 | low-priority coroutines to idle/background tasks). |
2921 | |
2945 | |
2922 | It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>) |
2946 | When used for this purpose, it is recommended to give C<ev_check> watchers |
2923 | priority, to ensure that they are being run before any other watchers |
2947 | highest (C<EV_MAXPRI>) priority, to ensure that they are being run before |
2924 | after the poll (this doesn't matter for C<ev_prepare> watchers). |
2948 | any other watchers after the poll (this doesn't matter for C<ev_prepare> |
|
|
2949 | watchers). |
2925 | |
2950 | |
2926 | Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not |
2951 | Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not |
2927 | activate ("feed") events into libev. While libev fully supports this, they |
2952 | activate ("feed") events into libev. While libev fully supports this, they |
2928 | might get executed before other C<ev_check> watchers did their job. As |
2953 | might get executed before other C<ev_check> watchers did their job. As |
2929 | C<ev_check> watchers are often used to embed other (non-libev) event |
2954 | C<ev_check> watchers are often used to embed other (non-libev) event |
2930 | loops those other event loops might be in an unusable state until their |
2955 | loops those other event loops might be in an unusable state until their |
2931 | C<ev_check> watcher ran (always remind yourself to coexist peacefully with |
2956 | C<ev_check> watcher ran (always remind yourself to coexist peacefully with |
2932 | others). |
2957 | others). |
|
|
2958 | |
|
|
2959 | =head3 Abusing an C<ev_check> watcher for its side-effect |
|
|
2960 | |
|
|
2961 | C<ev_check> (and less often also C<ev_prepare>) watchers can also be |
|
|
2962 | useful because they are called once per event loop iteration. For |
|
|
2963 | example, if you want to handle a large number of connections fairly, you |
|
|
2964 | normally only do a bit of work for each active connection, and if there |
|
|
2965 | is more work to do, you wait for the next event loop iteration, so other |
|
|
2966 | connections have a chance of making progress. |
|
|
2967 | |
|
|
2968 | Using an C<ev_check> watcher is almost enough: it will be called on the |
|
|
2969 | next event loop iteration. However, that isn't as soon as possible - |
|
|
2970 | without external events, your C<ev_check> watcher will not be invoked. |
|
|
2971 | |
|
|
2972 | This is where C<ev_idle> watchers come in handy - all you need is a |
|
|
2973 | single global idle watcher that is active as long as you have one active |
|
|
2974 | C<ev_check> watcher. The C<ev_idle> watcher makes sure the event loop |
|
|
2975 | will not sleep, and the C<ev_check> watcher makes sure a callback gets |
|
|
2976 | invoked. Neither watcher alone can do that. |
2933 | |
2977 | |
2934 | =head3 Watcher-Specific Functions and Data Members |
2978 | =head3 Watcher-Specific Functions and Data Members |
2935 | |
2979 | |
2936 | =over 4 |
2980 | =over 4 |
2937 | |
2981 | |
… | |
… | |
3138 | |
3182 | |
3139 | =over 4 |
3183 | =over 4 |
3140 | |
3184 | |
3141 | =item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop) |
3185 | =item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop) |
3142 | |
3186 | |
3143 | =item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop) |
3187 | =item ev_embed_set (ev_embed *, struct ev_loop *embedded_loop) |
3144 | |
3188 | |
3145 | Configures the watcher to embed the given loop, which must be |
3189 | Configures the watcher to embed the given loop, which must be |
3146 | embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be |
3190 | embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be |
3147 | invoked automatically, otherwise it is the responsibility of the callback |
3191 | invoked automatically, otherwise it is the responsibility of the callback |
3148 | to invoke it (it will continue to be called until the sweep has been done, |
3192 | to invoke it (it will continue to be called until the sweep has been done, |
… | |
… | |
3169 | used). |
3213 | used). |
3170 | |
3214 | |
3171 | struct ev_loop *loop_hi = ev_default_init (0); |
3215 | struct ev_loop *loop_hi = ev_default_init (0); |
3172 | struct ev_loop *loop_lo = 0; |
3216 | struct ev_loop *loop_lo = 0; |
3173 | ev_embed embed; |
3217 | ev_embed embed; |
3174 | |
3218 | |
3175 | // see if there is a chance of getting one that works |
3219 | // see if there is a chance of getting one that works |
3176 | // (remember that a flags value of 0 means autodetection) |
3220 | // (remember that a flags value of 0 means autodetection) |
3177 | loop_lo = ev_embeddable_backends () & ev_recommended_backends () |
3221 | loop_lo = ev_embeddable_backends () & ev_recommended_backends () |
3178 | ? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ()) |
3222 | ? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ()) |
3179 | : 0; |
3223 | : 0; |
… | |
… | |
3193 | C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too). |
3237 | C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too). |
3194 | |
3238 | |
3195 | struct ev_loop *loop = ev_default_init (0); |
3239 | struct ev_loop *loop = ev_default_init (0); |
3196 | struct ev_loop *loop_socket = 0; |
3240 | struct ev_loop *loop_socket = 0; |
3197 | ev_embed embed; |
3241 | ev_embed embed; |
3198 | |
3242 | |
3199 | if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE) |
3243 | if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE) |
3200 | if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE)) |
3244 | if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE)) |
3201 | { |
3245 | { |
3202 | ev_embed_init (&embed, 0, loop_socket); |
3246 | ev_embed_init (&embed, 0, loop_socket); |
3203 | ev_embed_start (loop, &embed); |
3247 | ev_embed_start (loop, &embed); |
… | |
… | |
3211 | |
3255 | |
3212 | =head2 C<ev_fork> - the audacity to resume the event loop after a fork |
3256 | =head2 C<ev_fork> - the audacity to resume the event loop after a fork |
3213 | |
3257 | |
3214 | Fork watchers are called when a C<fork ()> was detected (usually because |
3258 | Fork watchers are called when a C<fork ()> was detected (usually because |
3215 | whoever is a good citizen cared to tell libev about it by calling |
3259 | whoever is a good citizen cared to tell libev about it by calling |
3216 | C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the |
3260 | C<ev_loop_fork>). The invocation is done before the event loop blocks next |
3217 | event loop blocks next and before C<ev_check> watchers are being called, |
3261 | and before C<ev_check> watchers are being called, and only in the child |
3218 | and only in the child after the fork. If whoever good citizen calling |
3262 | after the fork. If whoever good citizen calling C<ev_default_fork> cheats |
3219 | C<ev_default_fork> cheats and calls it in the wrong process, the fork |
3263 | and calls it in the wrong process, the fork handlers will be invoked, too, |
3220 | handlers will be invoked, too, of course. |
3264 | of course. |
3221 | |
3265 | |
3222 | =head3 The special problem of life after fork - how is it possible? |
3266 | =head3 The special problem of life after fork - how is it possible? |
3223 | |
3267 | |
3224 | Most uses of C<fork()> consist of forking, then some simple calls to set |
3268 | Most uses of C<fork()> consist of forking, then some simple calls to set |
3225 | up/change the process environment, followed by a call to C<exec()>. This |
3269 | up/change the process environment, followed by a call to C<exec()>. This |
… | |
… | |
3619 | already been invoked. |
3663 | already been invoked. |
3620 | |
3664 | |
3621 | A common way around all these issues is to make sure that |
3665 | A common way around all these issues is to make sure that |
3622 | C<start_new_request> I<always> returns before the callback is invoked. If |
3666 | C<start_new_request> I<always> returns before the callback is invoked. If |
3623 | C<start_new_request> immediately knows the result, it can artificially |
3667 | C<start_new_request> immediately knows the result, it can artificially |
3624 | delay invoking the callback by e.g. using a C<prepare> or C<idle> watcher |
3668 | delay invoking the callback by using a C<prepare> or C<idle> watcher for |
3625 | for example, or more sneakily, by reusing an existing (stopped) watcher |
3669 | example, or more sneakily, by reusing an existing (stopped) watcher and |
3626 | and pushing it into the pending queue: |
3670 | pushing it into the pending queue: |
3627 | |
3671 | |
3628 | ev_set_cb (watcher, callback); |
3672 | ev_set_cb (watcher, callback); |
3629 | ev_feed_event (EV_A_ watcher, 0); |
3673 | ev_feed_event (EV_A_ watcher, 0); |
3630 | |
3674 | |
3631 | This way, C<start_new_request> can safely return before the callback is |
3675 | This way, C<start_new_request> can safely return before the callback is |
… | |
… | |
3639 | |
3683 | |
3640 | This brings the problem of exiting - a callback might want to finish the |
3684 | This brings the problem of exiting - a callback might want to finish the |
3641 | main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but |
3685 | main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but |
3642 | a modal "Are you sure?" dialog is still waiting), or just the nested one |
3686 | a modal "Are you sure?" dialog is still waiting), or just the nested one |
3643 | and not the main one (e.g. user clocked "Ok" in a modal dialog), or some |
3687 | and not the main one (e.g. user clocked "Ok" in a modal dialog), or some |
3644 | other combination: In these cases, C<ev_break> will not work alone. |
3688 | other combination: In these cases, a simple C<ev_break> will not work. |
3645 | |
3689 | |
3646 | The solution is to maintain "break this loop" variable for each C<ev_run> |
3690 | The solution is to maintain "break this loop" variable for each C<ev_run> |
3647 | invocation, and use a loop around C<ev_run> until the condition is |
3691 | invocation, and use a loop around C<ev_run> until the condition is |
3648 | triggered, using C<EVRUN_ONCE>: |
3692 | triggered, using C<EVRUN_ONCE>: |
3649 | |
3693 | |
… | |
… | |
3835 | called): |
3879 | called): |
3836 | |
3880 | |
3837 | void |
3881 | void |
3838 | wait_for_event (ev_watcher *w) |
3882 | wait_for_event (ev_watcher *w) |
3839 | { |
3883 | { |
3840 | ev_cb_set (w) = current_coro; |
3884 | ev_set_cb (w, current_coro); |
3841 | switch_to (libev_coro); |
3885 | switch_to (libev_coro); |
3842 | } |
3886 | } |
3843 | |
3887 | |
3844 | That basically suspends the coroutine inside C<wait_for_event> and |
3888 | That basically suspends the coroutine inside C<wait_for_event> and |
3845 | continues the libev coroutine, which, when appropriate, switches back to |
3889 | continues the libev coroutine, which, when appropriate, switches back to |
… | |
… | |
3848 | You can do similar tricks if you have, say, threads with an event queue - |
3892 | You can do similar tricks if you have, say, threads with an event queue - |
3849 | instead of storing a coroutine, you store the queue object and instead of |
3893 | instead of storing a coroutine, you store the queue object and instead of |
3850 | switching to a coroutine, you push the watcher onto the queue and notify |
3894 | switching to a coroutine, you push the watcher onto the queue and notify |
3851 | any waiters. |
3895 | any waiters. |
3852 | |
3896 | |
3853 | To embed libev, see L<EMBEDDING>, but in short, it's easiest to create two |
3897 | To embed libev, see L</EMBEDDING>, but in short, it's easiest to create two |
3854 | files, F<my_ev.h> and F<my_ev.c> that include the respective libev files: |
3898 | files, F<my_ev.h> and F<my_ev.c> that include the respective libev files: |
3855 | |
3899 | |
3856 | // my_ev.h |
3900 | // my_ev.h |
3857 | #define EV_CB_DECLARE(type) struct my_coro *cb; |
3901 | #define EV_CB_DECLARE(type) struct my_coro *cb; |
3858 | #define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb); |
3902 | #define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb); |
… | |
… | |
3910 | libev sources can be compiled as C++. Therefore, code that uses the C API |
3954 | libev sources can be compiled as C++. Therefore, code that uses the C API |
3911 | will work fine. |
3955 | will work fine. |
3912 | |
3956 | |
3913 | Proper exception specifications might have to be added to callbacks passed |
3957 | Proper exception specifications might have to be added to callbacks passed |
3914 | to libev: exceptions may be thrown only from watcher callbacks, all |
3958 | to libev: exceptions may be thrown only from watcher callbacks, all |
3915 | other callbacks (allocator, syserr, loop acquire/release and periodioc |
3959 | other callbacks (allocator, syserr, loop acquire/release and periodic |
3916 | reschedule callbacks) must not throw exceptions, and might need a C<throw |
3960 | reschedule callbacks) must not throw exceptions, and might need a C<throw |
3917 | ()> specification. If you have code that needs to be compiled as both C |
3961 | ()> specification. If you have code that needs to be compiled as both C |
3918 | and C++ you can use the C<EV_THROW> macro for this: |
3962 | and C++ you can use the C<EV_THROW> macro for this: |
3919 | |
3963 | |
3920 | static void |
3964 | static void |
… | |
… | |
3940 | Libev comes with some simplistic wrapper classes for C++ that mainly allow |
3984 | Libev comes with some simplistic wrapper classes for C++ that mainly allow |
3941 | you to use some convenience methods to start/stop watchers and also change |
3985 | you to use some convenience methods to start/stop watchers and also change |
3942 | the callback model to a model using method callbacks on objects. |
3986 | the callback model to a model using method callbacks on objects. |
3943 | |
3987 | |
3944 | To use it, |
3988 | To use it, |
3945 | |
3989 | |
3946 | #include <ev++.h> |
3990 | #include <ev++.h> |
3947 | |
3991 | |
3948 | This automatically includes F<ev.h> and puts all of its definitions (many |
3992 | This automatically includes F<ev.h> and puts all of its definitions (many |
3949 | of them macros) into the global namespace. All C++ specific things are |
3993 | of them macros) into the global namespace. All C++ specific things are |
3950 | put into the C<ev> namespace. It should support all the same embedding |
3994 | put into the C<ev> namespace. It should support all the same embedding |
… | |
… | |
4053 | void operator() (ev::io &w, int revents) |
4097 | void operator() (ev::io &w, int revents) |
4054 | { |
4098 | { |
4055 | ... |
4099 | ... |
4056 | } |
4100 | } |
4057 | } |
4101 | } |
4058 | |
4102 | |
4059 | myfunctor f; |
4103 | myfunctor f; |
4060 | |
4104 | |
4061 | ev::io w; |
4105 | ev::io w; |
4062 | w.set (&f); |
4106 | w.set (&f); |
4063 | |
4107 | |
… | |
… | |
4081 | Associates a different C<struct ev_loop> with this watcher. You can only |
4125 | Associates a different C<struct ev_loop> with this watcher. You can only |
4082 | do this when the watcher is inactive (and not pending either). |
4126 | do this when the watcher is inactive (and not pending either). |
4083 | |
4127 | |
4084 | =item w->set ([arguments]) |
4128 | =item w->set ([arguments]) |
4085 | |
4129 | |
4086 | Basically the same as C<ev_TYPE_set>, with the same arguments. Either this |
4130 | Basically the same as C<ev_TYPE_set> (except for C<ev::embed> watchers>), |
4087 | method or a suitable start method must be called at least once. Unlike the |
4131 | with the same arguments. Either this method or a suitable start method |
4088 | C counterpart, an active watcher gets automatically stopped and restarted |
4132 | must be called at least once. Unlike the C counterpart, an active watcher |
4089 | when reconfiguring it with this method. |
4133 | gets automatically stopped and restarted when reconfiguring it with this |
|
|
4134 | method. |
|
|
4135 | |
|
|
4136 | For C<ev::embed> watchers this method is called C<set_embed>, to avoid |
|
|
4137 | clashing with the C<set (loop)> method. |
4090 | |
4138 | |
4091 | =item w->start () |
4139 | =item w->start () |
4092 | |
4140 | |
4093 | Starts the watcher. Note that there is no C<loop> argument, as the |
4141 | Starts the watcher. Note that there is no C<loop> argument, as the |
4094 | constructor already stores the event loop. |
4142 | constructor already stores the event loop. |
… | |
… | |
4197 | =item Lua |
4245 | =item Lua |
4198 | |
4246 | |
4199 | Brian Maher has written a partial interface to libev for lua (at the |
4247 | Brian Maher has written a partial interface to libev for lua (at the |
4200 | time of this writing, only C<ev_io> and C<ev_timer>), to be found at |
4248 | time of this writing, only C<ev_io> and C<ev_timer>), to be found at |
4201 | L<http://github.com/brimworks/lua-ev>. |
4249 | L<http://github.com/brimworks/lua-ev>. |
|
|
4250 | |
|
|
4251 | =item Javascript |
|
|
4252 | |
|
|
4253 | Node.js (L<http://nodejs.org>) uses libev as the underlying event library. |
|
|
4254 | |
|
|
4255 | =item Others |
|
|
4256 | |
|
|
4257 | There are others, and I stopped counting. |
4202 | |
4258 | |
4203 | =back |
4259 | =back |
4204 | |
4260 | |
4205 | |
4261 | |
4206 | =head1 MACRO MAGIC |
4262 | =head1 MACRO MAGIC |
… | |
… | |
4505 | |
4561 | |
4506 | If programs implement their own fd to handle mapping on win32, then this |
4562 | If programs implement their own fd to handle mapping on win32, then this |
4507 | macro can be used to override the C<close> function, useful to unregister |
4563 | macro can be used to override the C<close> function, useful to unregister |
4508 | file descriptors again. Note that the replacement function has to close |
4564 | file descriptors again. Note that the replacement function has to close |
4509 | the underlying OS handle. |
4565 | the underlying OS handle. |
|
|
4566 | |
|
|
4567 | =item EV_USE_WSASOCKET |
|
|
4568 | |
|
|
4569 | If defined to be C<1>, libev will use C<WSASocket> to create its internal |
|
|
4570 | communication socket, which works better in some environments. Otherwise, |
|
|
4571 | the normal C<socket> function will be used, which works better in other |
|
|
4572 | environments. |
4510 | |
4573 | |
4511 | =item EV_USE_POLL |
4574 | =item EV_USE_POLL |
4512 | |
4575 | |
4513 | If defined to be C<1>, libev will compile in support for the C<poll>(2) |
4576 | If defined to be C<1>, libev will compile in support for the C<poll>(2) |
4514 | backend. Otherwise it will be enabled on non-win32 platforms. It |
4577 | backend. Otherwise it will be enabled on non-win32 platforms. It |
… | |
… | |
4559 | different cpus (or different cpu cores). This reduces dependencies |
4622 | different cpus (or different cpu cores). This reduces dependencies |
4560 | and makes libev faster. |
4623 | and makes libev faster. |
4561 | |
4624 | |
4562 | =item EV_NO_THREADS |
4625 | =item EV_NO_THREADS |
4563 | |
4626 | |
4564 | If defined to be C<1>, libev will assume that it will never be called |
4627 | If defined to be C<1>, libev will assume that it will never be called from |
4565 | from different threads, which is a stronger assumption than C<EV_NO_SMP>, |
4628 | different threads (that includes signal handlers), which is a stronger |
4566 | above. This reduces dependencies and makes libev faster. |
4629 | assumption than C<EV_NO_SMP>, above. This reduces dependencies and makes |
|
|
4630 | libev faster. |
4567 | |
4631 | |
4568 | =item EV_ATOMIC_T |
4632 | =item EV_ATOMIC_T |
4569 | |
4633 | |
4570 | Libev requires an integer type (suitable for storing C<0> or C<1>) whose |
4634 | Libev requires an integer type (suitable for storing C<0> or C<1>) whose |
4571 | access is atomic and serialised with respect to other threads or signal |
4635 | access is atomic with respect to other threads or signal contexts. No |
4572 | contexts. No such type is easily found in the C language, so you can |
4636 | such type is easily found in the C language, so you can provide your own |
4573 | provide your own type that you know is safe for your purposes. It is used |
4637 | type that you know is safe for your purposes. It is used both for signal |
4574 | both for signal handler "locking" as well as for signal and thread safety |
4638 | handler "locking" as well as for signal and thread safety in C<ev_async> |
4575 | in C<ev_async> watchers. |
4639 | watchers. |
4576 | |
4640 | |
4577 | In the absence of this define, libev will use C<sig_atomic_t volatile> |
4641 | In the absence of this define, libev will use C<sig_atomic_t volatile> |
4578 | (from F<signal.h>), which is usually good enough on most platforms, |
4642 | (from F<signal.h>), which is usually good enough on most platforms. |
4579 | although strictly speaking using a type that also implies a memory fence |
|
|
4580 | is required. |
|
|
4581 | |
4643 | |
4582 | =item EV_H (h) |
4644 | =item EV_H (h) |
4583 | |
4645 | |
4584 | The name of the F<ev.h> header file used to include it. The default if |
4646 | The name of the F<ev.h> header file used to include it. The default if |
4585 | undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be |
4647 | undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be |
… | |
… | |
4953 | default loop and triggering an C<ev_async> watcher from the default loop |
5015 | default loop and triggering an C<ev_async> watcher from the default loop |
4954 | watcher callback into the event loop interested in the signal. |
5016 | watcher callback into the event loop interested in the signal. |
4955 | |
5017 | |
4956 | =back |
5018 | =back |
4957 | |
5019 | |
4958 | See also L<THREAD LOCKING EXAMPLE>. |
5020 | See also L</THREAD LOCKING EXAMPLE>. |
4959 | |
5021 | |
4960 | =head3 COROUTINES |
5022 | =head3 COROUTINES |
4961 | |
5023 | |
4962 | Libev is very accommodating to coroutines ("cooperative threads"): |
5024 | Libev is very accommodating to coroutines ("cooperative threads"): |
4963 | libev fully supports nesting calls to its functions from different |
5025 | libev fully supports nesting calls to its functions from different |
… | |
… | |
5254 | thread" or will block signals process-wide, both behaviours would |
5316 | thread" or will block signals process-wide, both behaviours would |
5255 | be compatible with libev. Interaction between C<sigprocmask> and |
5317 | be compatible with libev. Interaction between C<sigprocmask> and |
5256 | C<pthread_sigmask> could complicate things, however. |
5318 | C<pthread_sigmask> could complicate things, however. |
5257 | |
5319 | |
5258 | The most portable way to handle signals is to block signals in all threads |
5320 | The most portable way to handle signals is to block signals in all threads |
5259 | except the initial one, and run the default loop in the initial thread as |
5321 | except the initial one, and run the signal handling loop in the initial |
5260 | well. |
5322 | thread as well. |
5261 | |
5323 | |
5262 | =item C<long> must be large enough for common memory allocation sizes |
5324 | =item C<long> must be large enough for common memory allocation sizes |
5263 | |
5325 | |
5264 | To improve portability and simplify its API, libev uses C<long> internally |
5326 | To improve portability and simplify its API, libev uses C<long> internally |
5265 | instead of C<size_t> when allocating its data structures. On non-POSIX |
5327 | instead of C<size_t> when allocating its data structures. On non-POSIX |
… | |
… | |
5369 | =over 4 |
5431 | =over 4 |
5370 | |
5432 | |
5371 | =item C<EV_COMPAT3> backwards compatibility mechanism |
5433 | =item C<EV_COMPAT3> backwards compatibility mechanism |
5372 | |
5434 | |
5373 | The backward compatibility mechanism can be controlled by |
5435 | The backward compatibility mechanism can be controlled by |
5374 | C<EV_COMPAT3>. See L<PREPROCESSOR SYMBOLS/MACROS> in the L<EMBEDDING> |
5436 | C<EV_COMPAT3>. See L</"PREPROCESSOR SYMBOLS/MACROS"> in the L</EMBEDDING> |
5375 | section. |
5437 | section. |
5376 | |
5438 | |
5377 | =item C<ev_default_destroy> and C<ev_default_fork> have been removed |
5439 | =item C<ev_default_destroy> and C<ev_default_fork> have been removed |
5378 | |
5440 | |
5379 | These calls can be replaced easily by their C<ev_loop_xxx> counterparts: |
5441 | These calls can be replaced easily by their C<ev_loop_xxx> counterparts: |
… | |
… | |
5422 | =over 4 |
5484 | =over 4 |
5423 | |
5485 | |
5424 | =item active |
5486 | =item active |
5425 | |
5487 | |
5426 | A watcher is active as long as it has been started and not yet stopped. |
5488 | A watcher is active as long as it has been started and not yet stopped. |
5427 | See L<WATCHER STATES> for details. |
5489 | See L</WATCHER STATES> for details. |
5428 | |
5490 | |
5429 | =item application |
5491 | =item application |
5430 | |
5492 | |
5431 | In this document, an application is whatever is using libev. |
5493 | In this document, an application is whatever is using libev. |
5432 | |
5494 | |
… | |
… | |
5468 | watchers and events. |
5530 | watchers and events. |
5469 | |
5531 | |
5470 | =item pending |
5532 | =item pending |
5471 | |
5533 | |
5472 | A watcher is pending as soon as the corresponding event has been |
5534 | A watcher is pending as soon as the corresponding event has been |
5473 | detected. See L<WATCHER STATES> for details. |
5535 | detected. See L</WATCHER STATES> for details. |
5474 | |
5536 | |
5475 | =item real time |
5537 | =item real time |
5476 | |
5538 | |
5477 | The physical time that is observed. It is apparently strictly monotonic :) |
5539 | The physical time that is observed. It is apparently strictly monotonic :) |
5478 | |
5540 | |