… | |
… | |
62 | |
62 | |
63 | // unloop was called, so exit |
63 | // unloop was called, so exit |
64 | return 0; |
64 | return 0; |
65 | } |
65 | } |
66 | |
66 | |
67 | =head1 DESCRIPTION |
67 | =head1 ABOUT THIS DOCUMENT |
|
|
68 | |
|
|
69 | This document documents the libev software package. |
68 | |
70 | |
69 | The newest version of this document is also available as an html-formatted |
71 | The newest version of this document is also available as an html-formatted |
70 | web page you might find easier to navigate when reading it for the first |
72 | web page you might find easier to navigate when reading it for the first |
71 | time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>. |
73 | time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>. |
|
|
74 | |
|
|
75 | While this document tries to be as complete as possible in documenting |
|
|
76 | libev, its usage and the rationale behind its design, it is not a tutorial |
|
|
77 | on event-based programming, nor will it introduce event-based programming |
|
|
78 | with libev. |
|
|
79 | |
|
|
80 | Familarity with event based programming techniques in general is assumed |
|
|
81 | throughout this document. |
|
|
82 | |
|
|
83 | =head1 ABOUT LIBEV |
72 | |
84 | |
73 | Libev is an event loop: you register interest in certain events (such as a |
85 | Libev is an event loop: you register interest in certain events (such as a |
74 | file descriptor being readable or a timeout occurring), and it will manage |
86 | file descriptor being readable or a timeout occurring), and it will manage |
75 | these event sources and provide your program with events. |
87 | these event sources and provide your program with events. |
76 | |
88 | |
… | |
… | |
110 | name C<loop> (which is always of type C<ev_loop *>) will not have |
122 | name C<loop> (which is always of type C<ev_loop *>) will not have |
111 | this argument. |
123 | this argument. |
112 | |
124 | |
113 | =head2 TIME REPRESENTATION |
125 | =head2 TIME REPRESENTATION |
114 | |
126 | |
115 | Libev represents time as a single floating point number, representing the |
127 | Libev represents time as a single floating point number, representing |
116 | (fractional) number of seconds since the (POSIX) epoch (somewhere near |
128 | the (fractional) number of seconds since the (POSIX) epoch (somewhere |
117 | the beginning of 1970, details are complicated, don't ask). This type is |
129 | near the beginning of 1970, details are complicated, don't ask). This |
118 | called C<ev_tstamp>, which is what you should use too. It usually aliases |
130 | type is called C<ev_tstamp>, which is what you should use too. It usually |
119 | to the C<double> type in C, and when you need to do any calculations on |
131 | aliases to the C<double> type in C. When you need to do any calculations |
120 | it, you should treat it as some floating point value. Unlike the name |
132 | on it, you should treat it as some floating point value. Unlike the name |
121 | component C<stamp> might indicate, it is also used for time differences |
133 | component C<stamp> might indicate, it is also used for time differences |
122 | throughout libev. |
134 | throughout libev. |
123 | |
135 | |
124 | =head1 ERROR HANDLING |
136 | =head1 ERROR HANDLING |
125 | |
137 | |
… | |
… | |
609 | |
621 | |
610 | This value can sometimes be useful as a generation counter of sorts (it |
622 | This value can sometimes be useful as a generation counter of sorts (it |
611 | "ticks" the number of loop iterations), as it roughly corresponds with |
623 | "ticks" the number of loop iterations), as it roughly corresponds with |
612 | C<ev_prepare> and C<ev_check> calls. |
624 | C<ev_prepare> and C<ev_check> calls. |
613 | |
625 | |
|
|
626 | =item unsigned int ev_loop_depth (loop) |
|
|
627 | |
|
|
628 | Returns the number of times C<ev_loop> was entered minus the number of |
|
|
629 | times C<ev_loop> was exited, in other words, the recursion depth. |
|
|
630 | |
|
|
631 | Outside C<ev_loop>, this number is zero. In a callback, this number is |
|
|
632 | C<1>, unless C<ev_loop> was invoked recursively (or from another thread), |
|
|
633 | in which case it is higher. |
|
|
634 | |
|
|
635 | Leaving C<ev_loop> abnormally (setjmp/longjmp, cancelling the thread |
|
|
636 | etc.), doesn't count as exit. |
|
|
637 | |
614 | =item unsigned int ev_backend (loop) |
638 | =item unsigned int ev_backend (loop) |
615 | |
639 | |
616 | Returns one of the C<EVBACKEND_*> flags indicating the event backend in |
640 | Returns one of the C<EVBACKEND_*> flags indicating the event backend in |
617 | use. |
641 | use. |
618 | |
642 | |
… | |
… | |
632 | |
656 | |
633 | This function is rarely useful, but when some event callback runs for a |
657 | This function is rarely useful, but when some event callback runs for a |
634 | very long time without entering the event loop, updating libev's idea of |
658 | very long time without entering the event loop, updating libev's idea of |
635 | the current time is a good idea. |
659 | the current time is a good idea. |
636 | |
660 | |
637 | See also "The special problem of time updates" in the C<ev_timer> section. |
661 | See also L<The special problem of time updates> in the C<ev_timer> section. |
638 | |
662 | |
639 | =item ev_suspend (loop) |
663 | =item ev_suspend (loop) |
640 | |
664 | |
641 | =item ev_resume (loop) |
665 | =item ev_resume (loop) |
642 | |
666 | |
… | |
… | |
799 | |
823 | |
800 | By setting a higher I<io collect interval> you allow libev to spend more |
824 | By setting a higher I<io collect interval> you allow libev to spend more |
801 | time collecting I/O events, so you can handle more events per iteration, |
825 | time collecting I/O events, so you can handle more events per iteration, |
802 | at the cost of increasing latency. Timeouts (both C<ev_periodic> and |
826 | at the cost of increasing latency. Timeouts (both C<ev_periodic> and |
803 | C<ev_timer>) will be not affected. Setting this to a non-null value will |
827 | C<ev_timer>) will be not affected. Setting this to a non-null value will |
804 | introduce an additional C<ev_sleep ()> call into most loop iterations. |
828 | introduce an additional C<ev_sleep ()> call into most loop iterations. The |
|
|
829 | sleep time ensures that libev will not poll for I/O events more often then |
|
|
830 | once per this interval, on average. |
805 | |
831 | |
806 | Likewise, by setting a higher I<timeout collect interval> you allow libev |
832 | Likewise, by setting a higher I<timeout collect interval> you allow libev |
807 | to spend more time collecting timeouts, at the expense of increased |
833 | to spend more time collecting timeouts, at the expense of increased |
808 | latency/jitter/inexactness (the watcher callback will be called |
834 | latency/jitter/inexactness (the watcher callback will be called |
809 | later). C<ev_io> watchers will not be affected. Setting this to a non-null |
835 | later). C<ev_io> watchers will not be affected. Setting this to a non-null |
… | |
… | |
811 | |
837 | |
812 | Many (busy) programs can usually benefit by setting the I/O collect |
838 | Many (busy) programs can usually benefit by setting the I/O collect |
813 | interval to a value near C<0.1> or so, which is often enough for |
839 | interval to a value near C<0.1> or so, which is often enough for |
814 | interactive servers (of course not for games), likewise for timeouts. It |
840 | interactive servers (of course not for games), likewise for timeouts. It |
815 | usually doesn't make much sense to set it to a lower value than C<0.01>, |
841 | usually doesn't make much sense to set it to a lower value than C<0.01>, |
816 | as this approaches the timing granularity of most systems. |
842 | as this approaches the timing granularity of most systems. Note that if |
|
|
843 | you do transactions with the outside world and you can't increase the |
|
|
844 | parallelity, then this setting will limit your transaction rate (if you |
|
|
845 | need to poll once per transaction and the I/O collect interval is 0.01, |
|
|
846 | then you can't do more than 100 transations per second). |
817 | |
847 | |
818 | Setting the I<timeout collect interval> can improve the opportunity for |
848 | Setting the I<timeout collect interval> can improve the opportunity for |
819 | saving power, as the program will "bundle" timer callback invocations that |
849 | saving power, as the program will "bundle" timer callback invocations that |
820 | are "near" in time together, by delaying some, thus reducing the number of |
850 | are "near" in time together, by delaying some, thus reducing the number of |
821 | times the process sleeps and wakes up again. Another useful technique to |
851 | times the process sleeps and wakes up again. Another useful technique to |
822 | reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure |
852 | reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure |
823 | they fire on, say, one-second boundaries only. |
853 | they fire on, say, one-second boundaries only. |
|
|
854 | |
|
|
855 | Example: we only need 0.1s timeout granularity, and we wish not to poll |
|
|
856 | more often than 100 times per second: |
|
|
857 | |
|
|
858 | ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1); |
|
|
859 | ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01); |
824 | |
860 | |
825 | =item ev_loop_verify (loop) |
861 | =item ev_loop_verify (loop) |
826 | |
862 | |
827 | This function only does something when C<EV_VERIFY> support has been |
863 | This function only does something when C<EV_VERIFY> support has been |
828 | compiled in, which is the default for non-minimal builds. It tries to go |
864 | compiled in, which is the default for non-minimal builds. It tries to go |
… | |
… | |
1083 | integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI> |
1119 | integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI> |
1084 | (default: C<-2>). Pending watchers with higher priority will be invoked |
1120 | (default: C<-2>). Pending watchers with higher priority will be invoked |
1085 | before watchers with lower priority, but priority will not keep watchers |
1121 | before watchers with lower priority, but priority will not keep watchers |
1086 | from being executed (except for C<ev_idle> watchers). |
1122 | from being executed (except for C<ev_idle> watchers). |
1087 | |
1123 | |
1088 | This means that priorities are I<only> used for ordering callback |
|
|
1089 | invocation after new events have been received. This is useful, for |
|
|
1090 | example, to reduce latency after idling, or more often, to bind two |
|
|
1091 | watchers on the same event and make sure one is called first. |
|
|
1092 | |
|
|
1093 | If you need to suppress invocation when higher priority events are pending |
1124 | If you need to suppress invocation when higher priority events are pending |
1094 | you need to look at C<ev_idle> watchers, which provide this functionality. |
1125 | you need to look at C<ev_idle> watchers, which provide this functionality. |
1095 | |
1126 | |
1096 | You I<must not> change the priority of a watcher as long as it is active or |
1127 | You I<must not> change the priority of a watcher as long as it is active or |
1097 | pending. |
1128 | pending. |
1098 | |
|
|
1099 | The default priority used by watchers when no priority has been set is |
|
|
1100 | always C<0>, which is supposed to not be too high and not be too low :). |
|
|
1101 | |
1129 | |
1102 | Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is |
1130 | Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is |
1103 | fine, as long as you do not mind that the priority value you query might |
1131 | fine, as long as you do not mind that the priority value you query might |
1104 | or might not have been clamped to the valid range. |
1132 | or might not have been clamped to the valid range. |
|
|
1133 | |
|
|
1134 | The default priority used by watchers when no priority has been set is |
|
|
1135 | always C<0>, which is supposed to not be too high and not be too low :). |
|
|
1136 | |
|
|
1137 | See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of |
|
|
1138 | priorities. |
1105 | |
1139 | |
1106 | =item ev_invoke (loop, ev_TYPE *watcher, int revents) |
1140 | =item ev_invoke (loop, ev_TYPE *watcher, int revents) |
1107 | |
1141 | |
1108 | Invoke the C<watcher> with the given C<loop> and C<revents>. Neither |
1142 | Invoke the C<watcher> with the given C<loop> and C<revents>. Neither |
1109 | C<loop> nor C<revents> need to be valid as long as the watcher callback |
1143 | C<loop> nor C<revents> need to be valid as long as the watcher callback |
… | |
… | |
1174 | #include <stddef.h> |
1208 | #include <stddef.h> |
1175 | |
1209 | |
1176 | static void |
1210 | static void |
1177 | t1_cb (EV_P_ ev_timer *w, int revents) |
1211 | t1_cb (EV_P_ ev_timer *w, int revents) |
1178 | { |
1212 | { |
1179 | struct my_biggy big = (struct my_biggy * |
1213 | struct my_biggy big = (struct my_biggy *) |
1180 | (((char *)w) - offsetof (struct my_biggy, t1)); |
1214 | (((char *)w) - offsetof (struct my_biggy, t1)); |
1181 | } |
1215 | } |
1182 | |
1216 | |
1183 | static void |
1217 | static void |
1184 | t2_cb (EV_P_ ev_timer *w, int revents) |
1218 | t2_cb (EV_P_ ev_timer *w, int revents) |
1185 | { |
1219 | { |
1186 | struct my_biggy big = (struct my_biggy * |
1220 | struct my_biggy big = (struct my_biggy *) |
1187 | (((char *)w) - offsetof (struct my_biggy, t2)); |
1221 | (((char *)w) - offsetof (struct my_biggy, t2)); |
1188 | } |
1222 | } |
|
|
1223 | |
|
|
1224 | =head2 WATCHER PRIORITY MODELS |
|
|
1225 | |
|
|
1226 | Many event loops support I<watcher priorities>, which are usually small |
|
|
1227 | integers that influence the ordering of event callback invocation |
|
|
1228 | between watchers in some way, all else being equal. |
|
|
1229 | |
|
|
1230 | In libev, Watcher priorities can be set using C<ev_set_priority>. See its |
|
|
1231 | description for the more technical details such as the actual priority |
|
|
1232 | range. |
|
|
1233 | |
|
|
1234 | There are two common ways how these these priorities are being interpreted |
|
|
1235 | by event loops: |
|
|
1236 | |
|
|
1237 | In the more common lock-out model, higher priorities "lock out" invocation |
|
|
1238 | of lower priority watchers, which means as long as higher priority |
|
|
1239 | watchers receive events, lower priority watchers are not being invoked. |
|
|
1240 | |
|
|
1241 | The less common only-for-ordering model uses priorities solely to order |
|
|
1242 | callback invocation within a single event loop iteration: Higher priority |
|
|
1243 | watchers are invoked before lower priority ones, but they all get invoked |
|
|
1244 | before polling for new events. |
|
|
1245 | |
|
|
1246 | Libev uses the second (only-for-ordering) model for all its watchers |
|
|
1247 | except for idle watchers (which use the lock-out model). |
|
|
1248 | |
|
|
1249 | The rationale behind this is that implementing the lock-out model for |
|
|
1250 | watchers is not well supported by most kernel interfaces, and most event |
|
|
1251 | libraries will just poll for the same events again and again as long as |
|
|
1252 | their callbacks have not been executed, which is very inefficient in the |
|
|
1253 | common case of one high-priority watcher locking out a mass of lower |
|
|
1254 | priority ones. |
|
|
1255 | |
|
|
1256 | Static (ordering) priorities are most useful when you have two or more |
|
|
1257 | watchers handling the same resource: a typical usage example is having an |
|
|
1258 | C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle |
|
|
1259 | timeouts. Under load, data might be received while the program handles |
|
|
1260 | other jobs, but since timers normally get invoked first, the timeout |
|
|
1261 | handler will be executed before checking for data. In that case, giving |
|
|
1262 | the timer a lower priority than the I/O watcher ensures that I/O will be |
|
|
1263 | handled first even under adverse conditions (which is usually, but not |
|
|
1264 | always, what you want). |
|
|
1265 | |
|
|
1266 | Since idle watchers use the "lock-out" model, meaning that idle watchers |
|
|
1267 | will only be executed when no same or higher priority watchers have |
|
|
1268 | received events, they can be used to implement the "lock-out" model when |
|
|
1269 | required. |
|
|
1270 | |
|
|
1271 | For example, to emulate how many other event libraries handle priorities, |
|
|
1272 | you can associate an C<ev_idle> watcher to each such watcher, and in |
|
|
1273 | the normal watcher callback, you just start the idle watcher. The real |
|
|
1274 | processing is done in the idle watcher callback. This causes libev to |
|
|
1275 | continously poll and process kernel event data for the watcher, but when |
|
|
1276 | the lock-out case is known to be rare (which in turn is rare :), this is |
|
|
1277 | workable. |
|
|
1278 | |
|
|
1279 | Usually, however, the lock-out model implemented that way will perform |
|
|
1280 | miserably under the type of load it was designed to handle. In that case, |
|
|
1281 | it might be preferable to stop the real watcher before starting the |
|
|
1282 | idle watcher, so the kernel will not have to process the event in case |
|
|
1283 | the actual processing will be delayed for considerable time. |
|
|
1284 | |
|
|
1285 | Here is an example of an I/O watcher that should run at a strictly lower |
|
|
1286 | priority than the default, and which should only process data when no |
|
|
1287 | other events are pending: |
|
|
1288 | |
|
|
1289 | ev_idle idle; // actual processing watcher |
|
|
1290 | ev_io io; // actual event watcher |
|
|
1291 | |
|
|
1292 | static void |
|
|
1293 | io_cb (EV_P_ ev_io *w, int revents) |
|
|
1294 | { |
|
|
1295 | // stop the I/O watcher, we received the event, but |
|
|
1296 | // are not yet ready to handle it. |
|
|
1297 | ev_io_stop (EV_A_ w); |
|
|
1298 | |
|
|
1299 | // start the idle watcher to ahndle the actual event. |
|
|
1300 | // it will not be executed as long as other watchers |
|
|
1301 | // with the default priority are receiving events. |
|
|
1302 | ev_idle_start (EV_A_ &idle); |
|
|
1303 | } |
|
|
1304 | |
|
|
1305 | static void |
|
|
1306 | idle_cb (EV_P_ ev_idle *w, int revents) |
|
|
1307 | { |
|
|
1308 | // actual processing |
|
|
1309 | read (STDIN_FILENO, ...); |
|
|
1310 | |
|
|
1311 | // have to start the I/O watcher again, as |
|
|
1312 | // we have handled the event |
|
|
1313 | ev_io_start (EV_P_ &io); |
|
|
1314 | } |
|
|
1315 | |
|
|
1316 | // initialisation |
|
|
1317 | ev_idle_init (&idle, idle_cb); |
|
|
1318 | ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ); |
|
|
1319 | ev_io_start (EV_DEFAULT_ &io); |
|
|
1320 | |
|
|
1321 | In the "real" world, it might also be beneficial to start a timer, so that |
|
|
1322 | low-priority connections can not be locked out forever under load. This |
|
|
1323 | enables your program to keep a lower latency for important connections |
|
|
1324 | during short periods of high load, while not completely locking out less |
|
|
1325 | important ones. |
1189 | |
1326 | |
1190 | |
1327 | |
1191 | =head1 WATCHER TYPES |
1328 | =head1 WATCHER TYPES |
1192 | |
1329 | |
1193 | This section describes each watcher in detail, but will not repeat |
1330 | This section describes each watcher in detail, but will not repeat |
… | |
… | |
1219 | descriptors to non-blocking mode is also usually a good idea (but not |
1356 | descriptors to non-blocking mode is also usually a good idea (but not |
1220 | required if you know what you are doing). |
1357 | required if you know what you are doing). |
1221 | |
1358 | |
1222 | If you cannot use non-blocking mode, then force the use of a |
1359 | If you cannot use non-blocking mode, then force the use of a |
1223 | known-to-be-good backend (at the time of this writing, this includes only |
1360 | known-to-be-good backend (at the time of this writing, this includes only |
1224 | C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). |
1361 | C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file |
|
|
1362 | descriptors for which non-blocking operation makes no sense (such as |
|
|
1363 | files) - libev doesn't guarentee any specific behaviour in that case. |
1225 | |
1364 | |
1226 | Another thing you have to watch out for is that it is quite easy to |
1365 | Another thing you have to watch out for is that it is quite easy to |
1227 | receive "spurious" readiness notifications, that is your callback might |
1366 | receive "spurious" readiness notifications, that is your callback might |
1228 | be called with C<EV_READ> but a subsequent C<read>(2) will actually block |
1367 | be called with C<EV_READ> but a subsequent C<read>(2) will actually block |
1229 | because there is no data. Not only are some backends known to create a |
1368 | because there is no data. Not only are some backends known to create a |
… | |
… | |
1350 | year, it will still time out after (roughly) one hour. "Roughly" because |
1489 | year, it will still time out after (roughly) one hour. "Roughly" because |
1351 | detecting time jumps is hard, and some inaccuracies are unavoidable (the |
1490 | detecting time jumps is hard, and some inaccuracies are unavoidable (the |
1352 | monotonic clock option helps a lot here). |
1491 | monotonic clock option helps a lot here). |
1353 | |
1492 | |
1354 | The callback is guaranteed to be invoked only I<after> its timeout has |
1493 | The callback is guaranteed to be invoked only I<after> its timeout has |
1355 | passed. If multiple timers become ready during the same loop iteration |
1494 | passed (not I<at>, so on systems with very low-resolution clocks this |
1356 | then the ones with earlier time-out values are invoked before ones with |
1495 | might introduce a small delay). If multiple timers become ready during the |
1357 | later time-out values (but this is no longer true when a callback calls |
1496 | same loop iteration then the ones with earlier time-out values are invoked |
1358 | C<ev_loop> recursively). |
1497 | before ones of the same priority with later time-out values (but this is |
|
|
1498 | no longer true when a callback calls C<ev_loop> recursively). |
1359 | |
1499 | |
1360 | =head3 Be smart about timeouts |
1500 | =head3 Be smart about timeouts |
1361 | |
1501 | |
1362 | Many real-world problems involve some kind of timeout, usually for error |
1502 | Many real-world problems involve some kind of timeout, usually for error |
1363 | recovery. A typical example is an HTTP request - if the other side hangs, |
1503 | recovery. A typical example is an HTTP request - if the other side hangs, |
… | |
… | |
1407 | C<after> argument to C<ev_timer_set>, and only ever use the C<repeat> |
1547 | C<after> argument to C<ev_timer_set>, and only ever use the C<repeat> |
1408 | member and C<ev_timer_again>. |
1548 | member and C<ev_timer_again>. |
1409 | |
1549 | |
1410 | At start: |
1550 | At start: |
1411 | |
1551 | |
1412 | ev_timer_init (timer, callback); |
1552 | ev_init (timer, callback); |
1413 | timer->repeat = 60.; |
1553 | timer->repeat = 60.; |
1414 | ev_timer_again (loop, timer); |
1554 | ev_timer_again (loop, timer); |
1415 | |
1555 | |
1416 | Each time there is some activity: |
1556 | Each time there is some activity: |
1417 | |
1557 | |
… | |
… | |
1479 | |
1619 | |
1480 | To start the timer, simply initialise the watcher and set C<last_activity> |
1620 | To start the timer, simply initialise the watcher and set C<last_activity> |
1481 | to the current time (meaning we just have some activity :), then call the |
1621 | to the current time (meaning we just have some activity :), then call the |
1482 | callback, which will "do the right thing" and start the timer: |
1622 | callback, which will "do the right thing" and start the timer: |
1483 | |
1623 | |
1484 | ev_timer_init (timer, callback); |
1624 | ev_init (timer, callback); |
1485 | last_activity = ev_now (loop); |
1625 | last_activity = ev_now (loop); |
1486 | callback (loop, timer, EV_TIMEOUT); |
1626 | callback (loop, timer, EV_TIMEOUT); |
1487 | |
1627 | |
1488 | And when there is some activity, simply store the current time in |
1628 | And when there is some activity, simply store the current time in |
1489 | C<last_activity>, no libev calls at all: |
1629 | C<last_activity>, no libev calls at all: |
… | |
… | |
1582 | If the timer is started but non-repeating, stop it (as if it timed out). |
1722 | If the timer is started but non-repeating, stop it (as if it timed out). |
1583 | |
1723 | |
1584 | If the timer is repeating, either start it if necessary (with the |
1724 | If the timer is repeating, either start it if necessary (with the |
1585 | C<repeat> value), or reset the running timer to the C<repeat> value. |
1725 | C<repeat> value), or reset the running timer to the C<repeat> value. |
1586 | |
1726 | |
1587 | This sounds a bit complicated, see "Be smart about timeouts", above, for a |
1727 | This sounds a bit complicated, see L<Be smart about timeouts>, above, for a |
1588 | usage example. |
1728 | usage example. |
1589 | |
1729 | |
1590 | =item ev_tstamp repeat [read-write] |
1730 | =item ev_tstamp repeat [read-write] |
1591 | |
1731 | |
1592 | The current C<repeat> value. Will be used each time the watcher times out |
1732 | The current C<repeat> value. Will be used each time the watcher times out |
… | |
… | |
1886 | some child status changes (most typically when a child of yours dies or |
2026 | some child status changes (most typically when a child of yours dies or |
1887 | exits). It is permissible to install a child watcher I<after> the child |
2027 | exits). It is permissible to install a child watcher I<after> the child |
1888 | has been forked (which implies it might have already exited), as long |
2028 | has been forked (which implies it might have already exited), as long |
1889 | as the event loop isn't entered (or is continued from a watcher), i.e., |
2029 | as the event loop isn't entered (or is continued from a watcher), i.e., |
1890 | forking and then immediately registering a watcher for the child is fine, |
2030 | forking and then immediately registering a watcher for the child is fine, |
1891 | but forking and registering a watcher a few event loop iterations later is |
2031 | but forking and registering a watcher a few event loop iterations later or |
1892 | not. |
2032 | in the next callback invocation is not. |
1893 | |
2033 | |
1894 | Only the default event loop is capable of handling signals, and therefore |
2034 | Only the default event loop is capable of handling signals, and therefore |
1895 | you can only register child watchers in the default event loop. |
2035 | you can only register child watchers in the default event loop. |
|
|
2036 | |
|
|
2037 | Due to some design glitches inside libev, child watchers will always be |
|
|
2038 | handled at maximum priority (their priority is set to C<EV_MAXPRI> by |
|
|
2039 | libev) |
1896 | |
2040 | |
1897 | =head3 Process Interaction |
2041 | =head3 Process Interaction |
1898 | |
2042 | |
1899 | Libev grabs C<SIGCHLD> as soon as the default event loop is |
2043 | Libev grabs C<SIGCHLD> as soon as the default event loop is |
1900 | initialised. This is necessary to guarantee proper behaviour even if |
2044 | initialised. This is necessary to guarantee proper behaviour even if |
… | |
… | |
2252 | // no longer anything immediate to do. |
2396 | // no longer anything immediate to do. |
2253 | } |
2397 | } |
2254 | |
2398 | |
2255 | ev_idle *idle_watcher = malloc (sizeof (ev_idle)); |
2399 | ev_idle *idle_watcher = malloc (sizeof (ev_idle)); |
2256 | ev_idle_init (idle_watcher, idle_cb); |
2400 | ev_idle_init (idle_watcher, idle_cb); |
2257 | ev_idle_start (loop, idle_cb); |
2401 | ev_idle_start (loop, idle_watcher); |
2258 | |
2402 | |
2259 | |
2403 | |
2260 | =head2 C<ev_prepare> and C<ev_check> - customise your event loop! |
2404 | =head2 C<ev_prepare> and C<ev_check> - customise your event loop! |
2261 | |
2405 | |
2262 | Prepare and check watchers are usually (but not always) used in pairs: |
2406 | Prepare and check watchers are usually (but not always) used in pairs: |
… | |
… | |
2355 | struct pollfd fds [nfd]; |
2499 | struct pollfd fds [nfd]; |
2356 | // actual code will need to loop here and realloc etc. |
2500 | // actual code will need to loop here and realloc etc. |
2357 | adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ())); |
2501 | adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ())); |
2358 | |
2502 | |
2359 | /* the callback is illegal, but won't be called as we stop during check */ |
2503 | /* the callback is illegal, but won't be called as we stop during check */ |
2360 | ev_timer_init (&tw, 0, timeout * 1e-3); |
2504 | ev_timer_init (&tw, 0, timeout * 1e-3, 0.); |
2361 | ev_timer_start (loop, &tw); |
2505 | ev_timer_start (loop, &tw); |
2362 | |
2506 | |
2363 | // create one ev_io per pollfd |
2507 | // create one ev_io per pollfd |
2364 | for (int i = 0; i < nfd; ++i) |
2508 | for (int i = 0; i < nfd; ++i) |
2365 | { |
2509 | { |
… | |
… | |
2595 | event loop blocks next and before C<ev_check> watchers are being called, |
2739 | event loop blocks next and before C<ev_check> watchers are being called, |
2596 | and only in the child after the fork. If whoever good citizen calling |
2740 | and only in the child after the fork. If whoever good citizen calling |
2597 | C<ev_default_fork> cheats and calls it in the wrong process, the fork |
2741 | C<ev_default_fork> cheats and calls it in the wrong process, the fork |
2598 | handlers will be invoked, too, of course. |
2742 | handlers will be invoked, too, of course. |
2599 | |
2743 | |
|
|
2744 | =head3 The special problem of life after fork - how is it possible? |
|
|
2745 | |
|
|
2746 | Most uses of C<fork()> consist of forking, then some simple calls to ste |
|
|
2747 | up/change the process environment, followed by a call to C<exec()>. This |
|
|
2748 | sequence should be handled by libev without any problems. |
|
|
2749 | |
|
|
2750 | This changes when the application actually wants to do event handling |
|
|
2751 | in the child, or both parent in child, in effect "continuing" after the |
|
|
2752 | fork. |
|
|
2753 | |
|
|
2754 | The default mode of operation (for libev, with application help to detect |
|
|
2755 | forks) is to duplicate all the state in the child, as would be expected |
|
|
2756 | when I<either> the parent I<or> the child process continues. |
|
|
2757 | |
|
|
2758 | When both processes want to continue using libev, then this is usually the |
|
|
2759 | wrong result. In that case, usually one process (typically the parent) is |
|
|
2760 | supposed to continue with all watchers in place as before, while the other |
|
|
2761 | process typically wants to start fresh, i.e. without any active watchers. |
|
|
2762 | |
|
|
2763 | The cleanest and most efficient way to achieve that with libev is to |
|
|
2764 | simply create a new event loop, which of course will be "empty", and |
|
|
2765 | use that for new watchers. This has the advantage of not touching more |
|
|
2766 | memory than necessary, and thus avoiding the copy-on-write, and the |
|
|
2767 | disadvantage of having to use multiple event loops (which do not support |
|
|
2768 | signal watchers). |
|
|
2769 | |
|
|
2770 | When this is not possible, or you want to use the default loop for |
|
|
2771 | other reasons, then in the process that wants to start "fresh", call |
|
|
2772 | C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying |
|
|
2773 | the default loop will "orphan" (not stop) all registered watchers, so you |
|
|
2774 | have to be careful not to execute code that modifies those watchers. Note |
|
|
2775 | also that in that case, you have to re-register any signal watchers. |
|
|
2776 | |
2600 | =head3 Watcher-Specific Functions and Data Members |
2777 | =head3 Watcher-Specific Functions and Data Members |
2601 | |
2778 | |
2602 | =over 4 |
2779 | =over 4 |
2603 | |
2780 | |
2604 | =item ev_fork_init (ev_signal *, callback) |
2781 | =item ev_fork_init (ev_signal *, callback) |
… | |
… | |
3494 | defined to be C<0>, then they are not. |
3671 | defined to be C<0>, then they are not. |
3495 | |
3672 | |
3496 | =item EV_MINIMAL |
3673 | =item EV_MINIMAL |
3497 | |
3674 | |
3498 | If you need to shave off some kilobytes of code at the expense of some |
3675 | If you need to shave off some kilobytes of code at the expense of some |
3499 | speed, define this symbol to C<1>. Currently this is used to override some |
3676 | speed (but with the full API), define this symbol to C<1>. Currently this |
3500 | inlining decisions, saves roughly 30% code size on amd64. It also selects a |
3677 | is used to override some inlining decisions, saves roughly 30% code size |
3501 | much smaller 2-heap for timer management over the default 4-heap. |
3678 | on amd64. It also selects a much smaller 2-heap for timer management over |
|
|
3679 | the default 4-heap. |
|
|
3680 | |
|
|
3681 | You can save even more by disabling watcher types you do not need and |
|
|
3682 | setting C<EV_MAXPRI> == C<EV_MINPRI>. |
|
|
3683 | |
|
|
3684 | Defining C<EV_MINIMAL> to C<2> will additionally reduce the core API to |
|
|
3685 | provide a bare-bones event library. See C<ev.h> for details on what parts |
|
|
3686 | of the API are still available, and do not complain if this subset changes |
|
|
3687 | over time. |
3502 | |
3688 | |
3503 | =item EV_PID_HASHSIZE |
3689 | =item EV_PID_HASHSIZE |
3504 | |
3690 | |
3505 | C<ev_child> watchers use a small hash table to distribute workload by |
3691 | C<ev_child> watchers use a small hash table to distribute workload by |
3506 | pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more |
3692 | pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more |
… | |
… | |
3786 | way (note also that glib is the slowest event library known to man). |
3972 | way (note also that glib is the slowest event library known to man). |
3787 | |
3973 | |
3788 | There is no supported compilation method available on windows except |
3974 | There is no supported compilation method available on windows except |
3789 | embedding it into other applications. |
3975 | embedding it into other applications. |
3790 | |
3976 | |
|
|
3977 | Sensible signal handling is officially unsupported by Microsoft - libev |
|
|
3978 | tries its best, but under most conditions, signals will simply not work. |
|
|
3979 | |
3791 | Not a libev limitation but worth mentioning: windows apparently doesn't |
3980 | Not a libev limitation but worth mentioning: windows apparently doesn't |
3792 | accept large writes: instead of resulting in a partial write, windows will |
3981 | accept large writes: instead of resulting in a partial write, windows will |
3793 | either accept everything or return C<ENOBUFS> if the buffer is too large, |
3982 | either accept everything or return C<ENOBUFS> if the buffer is too large, |
3794 | so make sure you only write small amounts into your sockets (less than a |
3983 | so make sure you only write small amounts into your sockets (less than a |
3795 | megabyte seems safe, but this apparently depends on the amount of memory |
3984 | megabyte seems safe, but this apparently depends on the amount of memory |
… | |
… | |
3799 | the abysmal performance of winsockets, using a large number of sockets |
3988 | the abysmal performance of winsockets, using a large number of sockets |
3800 | is not recommended (and not reasonable). If your program needs to use |
3989 | is not recommended (and not reasonable). If your program needs to use |
3801 | more than a hundred or so sockets, then likely it needs to use a totally |
3990 | more than a hundred or so sockets, then likely it needs to use a totally |
3802 | different implementation for windows, as libev offers the POSIX readiness |
3991 | different implementation for windows, as libev offers the POSIX readiness |
3803 | notification model, which cannot be implemented efficiently on windows |
3992 | notification model, which cannot be implemented efficiently on windows |
3804 | (Microsoft monopoly games). |
3993 | (due to Microsoft monopoly games). |
3805 | |
3994 | |
3806 | A typical way to use libev under windows is to embed it (see the embedding |
3995 | A typical way to use libev under windows is to embed it (see the embedding |
3807 | section for details) and use the following F<evwrap.h> header file instead |
3996 | section for details) and use the following F<evwrap.h> header file instead |
3808 | of F<ev.h>: |
3997 | of F<ev.h>: |
3809 | |
3998 | |
… | |
… | |
3845 | |
4034 | |
3846 | Early versions of winsocket's select only supported waiting for a maximum |
4035 | Early versions of winsocket's select only supported waiting for a maximum |
3847 | of C<64> handles (probably owning to the fact that all windows kernels |
4036 | of C<64> handles (probably owning to the fact that all windows kernels |
3848 | can only wait for C<64> things at the same time internally; Microsoft |
4037 | can only wait for C<64> things at the same time internally; Microsoft |
3849 | recommends spawning a chain of threads and wait for 63 handles and the |
4038 | recommends spawning a chain of threads and wait for 63 handles and the |
3850 | previous thread in each. Great). |
4039 | previous thread in each. Sounds great!). |
3851 | |
4040 | |
3852 | Newer versions support more handles, but you need to define C<FD_SETSIZE> |
4041 | Newer versions support more handles, but you need to define C<FD_SETSIZE> |
3853 | to some high number (e.g. C<2048>) before compiling the winsocket select |
4042 | to some high number (e.g. C<2048>) before compiling the winsocket select |
3854 | call (which might be in libev or elsewhere, for example, perl does its own |
4043 | call (which might be in libev or elsewhere, for example, perl and many |
3855 | select emulation on windows). |
4044 | other interpreters do their own select emulation on windows). |
3856 | |
4045 | |
3857 | Another limit is the number of file descriptors in the Microsoft runtime |
4046 | Another limit is the number of file descriptors in the Microsoft runtime |
3858 | libraries, which by default is C<64> (there must be a hidden I<64> fetish |
4047 | libraries, which by default is C<64> (there must be a hidden I<64> |
3859 | or something like this inside Microsoft). You can increase this by calling |
4048 | fetish or something like this inside Microsoft). You can increase this |
3860 | C<_setmaxstdio>, which can increase this limit to C<2048> (another |
4049 | by calling C<_setmaxstdio>, which can increase this limit to C<2048> |
3861 | arbitrary limit), but is broken in many versions of the Microsoft runtime |
4050 | (another arbitrary limit), but is broken in many versions of the Microsoft |
3862 | libraries. |
|
|
3863 | |
|
|
3864 | This might get you to about C<512> or C<2048> sockets (depending on |
4051 | runtime libraries. This might get you to about C<512> or C<2048> sockets |
3865 | windows version and/or the phase of the moon). To get more, you need to |
4052 | (depending on windows version and/or the phase of the moon). To get more, |
3866 | wrap all I/O functions and provide your own fd management, but the cost of |
4053 | you need to wrap all I/O functions and provide your own fd management, but |
3867 | calling select (O(n²)) will likely make this unworkable. |
4054 | the cost of calling select (O(n²)) will likely make this unworkable. |
3868 | |
4055 | |
3869 | =back |
4056 | =back |
3870 | |
4057 | |
3871 | =head2 PORTABILITY REQUIREMENTS |
4058 | =head2 PORTABILITY REQUIREMENTS |
3872 | |
4059 | |
… | |
… | |
3915 | =item C<double> must hold a time value in seconds with enough accuracy |
4102 | =item C<double> must hold a time value in seconds with enough accuracy |
3916 | |
4103 | |
3917 | The type C<double> is used to represent timestamps. It is required to |
4104 | The type C<double> is used to represent timestamps. It is required to |
3918 | have at least 51 bits of mantissa (and 9 bits of exponent), which is good |
4105 | have at least 51 bits of mantissa (and 9 bits of exponent), which is good |
3919 | enough for at least into the year 4000. This requirement is fulfilled by |
4106 | enough for at least into the year 4000. This requirement is fulfilled by |
3920 | implementations implementing IEEE 754 (basically all existing ones). |
4107 | implementations implementing IEEE 754, which is basically all existing |
|
|
4108 | ones. With IEEE 754 doubles, you get microsecond accuracy until at least |
|
|
4109 | 2200. |
3921 | |
4110 | |
3922 | =back |
4111 | =back |
3923 | |
4112 | |
3924 | If you know of other additional requirements drop me a note. |
4113 | If you know of other additional requirements drop me a note. |
3925 | |
4114 | |
… | |
… | |
3993 | involves iterating over all running async watchers or all signal numbers. |
4182 | involves iterating over all running async watchers or all signal numbers. |
3994 | |
4183 | |
3995 | =back |
4184 | =back |
3996 | |
4185 | |
3997 | |
4186 | |
|
|
4187 | =head1 GLOSSARY |
|
|
4188 | |
|
|
4189 | =over 4 |
|
|
4190 | |
|
|
4191 | =item active |
|
|
4192 | |
|
|
4193 | A watcher is active as long as it has been started (has been attached to |
|
|
4194 | an event loop) but not yet stopped (disassociated from the event loop). |
|
|
4195 | |
|
|
4196 | =item application |
|
|
4197 | |
|
|
4198 | In this document, an application is whatever is using libev. |
|
|
4199 | |
|
|
4200 | =item callback |
|
|
4201 | |
|
|
4202 | The address of a function that is called when some event has been |
|
|
4203 | detected. Callbacks are being passed the event loop, the watcher that |
|
|
4204 | received the event, and the actual event bitset. |
|
|
4205 | |
|
|
4206 | =item callback invocation |
|
|
4207 | |
|
|
4208 | The act of calling the callback associated with a watcher. |
|
|
4209 | |
|
|
4210 | =item event |
|
|
4211 | |
|
|
4212 | A change of state of some external event, such as data now being available |
|
|
4213 | for reading on a file descriptor, time having passed or simply not having |
|
|
4214 | any other events happening anymore. |
|
|
4215 | |
|
|
4216 | In libev, events are represented as single bits (such as C<EV_READ> or |
|
|
4217 | C<EV_TIMEOUT>). |
|
|
4218 | |
|
|
4219 | =item event library |
|
|
4220 | |
|
|
4221 | A software package implementing an event model and loop. |
|
|
4222 | |
|
|
4223 | =item event loop |
|
|
4224 | |
|
|
4225 | An entity that handles and processes external events and converts them |
|
|
4226 | into callback invocations. |
|
|
4227 | |
|
|
4228 | =item event model |
|
|
4229 | |
|
|
4230 | The model used to describe how an event loop handles and processes |
|
|
4231 | watchers and events. |
|
|
4232 | |
|
|
4233 | =item pending |
|
|
4234 | |
|
|
4235 | A watcher is pending as soon as the corresponding event has been detected, |
|
|
4236 | and stops being pending as soon as the watcher will be invoked or its |
|
|
4237 | pending status is explicitly cleared by the application. |
|
|
4238 | |
|
|
4239 | A watcher can be pending, but not active. Stopping a watcher also clears |
|
|
4240 | its pending status. |
|
|
4241 | |
|
|
4242 | =item real time |
|
|
4243 | |
|
|
4244 | The physical time that is observed. It is apparently strictly monotonic :) |
|
|
4245 | |
|
|
4246 | =item wall-clock time |
|
|
4247 | |
|
|
4248 | The time and date as shown on clocks. Unlike real time, it can actually |
|
|
4249 | be wrong and jump forwards and backwards, e.g. when the you adjust your |
|
|
4250 | clock. |
|
|
4251 | |
|
|
4252 | =item watcher |
|
|
4253 | |
|
|
4254 | A data structure that describes interest in certain events. Watchers need |
|
|
4255 | to be started (attached to an event loop) before they can receive events. |
|
|
4256 | |
|
|
4257 | =item watcher invocation |
|
|
4258 | |
|
|
4259 | The act of calling the callback associated with a watcher. |
|
|
4260 | |
|
|
4261 | =back |
|
|
4262 | |
3998 | =head1 AUTHOR |
4263 | =head1 AUTHOR |
3999 | |
4264 | |
4000 | Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson. |
4265 | Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson. |
4001 | |
4266 | |