ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.173 by root, Thu Aug 7 19:24:56 2008 UTC vs.
Revision 1.240 by root, Sat Apr 25 14:12:48 2009 UTC

…		…
9	=head2 EXAMPLE PROGRAM	9	=head2 EXAMPLE PROGRAM
10		10
11	// a single header file is required	11	// a single header file is required
12	#include <ev.h>	12	#include <ev.h>
13		13
		14	#include <stdio.h> // for puts
		15
14	// every watcher type has its own typedef'd struct	16	// every watcher type has its own typedef'd struct
15	// with the name ev_<type>	17	// with the name ev_TYPE
16	ev_io stdin_watcher;	18	ev_io stdin_watcher;
17	ev_timer timeout_watcher;	19	ev_timer timeout_watcher;
18		20
19	// all watcher callbacks have a similar signature	21	// all watcher callbacks have a similar signature
20	// this callback is called when data is readable on stdin	22	// this callback is called when data is readable on stdin
21	static void	23	static void
22	stdin_cb (EV_P_ struct ev_io *w, int revents)	24	stdin_cb (EV_P_ ev_io *w, int revents)
23	{	25	{
24	puts ("stdin ready");	26	puts ("stdin ready");
25	// for one-shot events, one must manually stop the watcher	27	// for one-shot events, one must manually stop the watcher
26	// with its corresponding stop function.	28	// with its corresponding stop function.
27	ev_io_stop (EV_A_ w);	29	ev_io_stop (EV_A_ w);
…		…
30	ev_unloop (EV_A_ EVUNLOOP_ALL);	32	ev_unloop (EV_A_ EVUNLOOP_ALL);
31	}	33	}
32		34
33	// another callback, this time for a time-out	35	// another callback, this time for a time-out
34	static void	36	static void
35	timeout_cb (EV_P_ struct ev_timer *w, int revents)	37	timeout_cb (EV_P_ ev_timer *w, int revents)
36	{	38	{
37	puts ("timeout");	39	puts ("timeout");
38	// this causes the innermost ev_loop to stop iterating	40	// this causes the innermost ev_loop to stop iterating
39	ev_unloop (EV_A_ EVUNLOOP_ONE);	41	ev_unloop (EV_A_ EVUNLOOP_ONE);
40	}	42	}
…		…
60		62
61	// unloop was called, so exit	63	// unloop was called, so exit
62	return 0;	64	return 0;
63	}	65	}
64		66
65	=head1 DESCRIPTION	67	=head1 ABOUT THIS DOCUMENT
		68
		69	This document documents the libev software package.
66		70
67	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
68	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
69	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
70		84
71	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
72	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
73	these event sources and provide your program with events.	87	these event sources and provide your program with events.
74		88
…		…
103	Libev is very configurable. In this manual the default (and most common)	117	Libev is very configurable. In this manual the default (and most common)
104	configuration will be described, which supports multiple event loops. For	118	configuration will be described, which supports multiple event loops. For
105	more info about various configuration options please have a look at	119	more info about various configuration options please have a look at
106	B<EMBED> section in this manual. If libev was configured without support	120	B<EMBED> section in this manual. If libev was configured without support
107	for multiple event loops, then all functions taking an initial argument of	121	for multiple event loops, then all functions taking an initial argument of
108	name C<loop> (which is always of type C<struct ev_loop *>) will not have	122	name C<loop> (which is always of type C<ev_loop *>) will not have
109	this argument.	123	this argument.
110		124
111	=head2 TIME REPRESENTATION	125	=head2 TIME REPRESENTATION
112		126
113	Libev represents time as a single floating point number, representing the	127	Libev represents time as a single floating point number, representing
114	(fractional) number of seconds since the (POSIX) epoch (somewhere near	128	the (fractional) number of seconds since the (POSIX) epoch (somewhere
115	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
116	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
117	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
118	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
119	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
120	throughout libev.	134	throughout libev.
121		135
122	=head1 ERROR HANDLING	136	=head1 ERROR HANDLING
123		137
…		…
214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	228	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
215	recommended ones.	229	recommended ones.
216		230
217	See the description of C<ev_embed> watchers for more info.	231	See the description of C<ev_embed> watchers for more info.
218		232
219	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	233	=item ev_set_allocator (void *(*cb)(void *ptr, long size)) [NOT REENTRANT]
220		234
221	Sets the allocation function to use (the prototype is similar - the	235	Sets the allocation function to use (the prototype is similar - the
222	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is	236	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
223	used to allocate and free memory (no surprises here). If it returns zero	237	used to allocate and free memory (no surprises here). If it returns zero
224	when memory needs to be allocated (C<size != 0>), the library might abort	238	when memory needs to be allocated (C<size != 0>), the library might abort
…		…
250	}	264	}
251		265
252	...	266	...
253	ev_set_allocator (persistent_realloc);	267	ev_set_allocator (persistent_realloc);
254		268
255	=item ev_set_syserr_cb (void (*cb)(const char *msg));	269	=item ev_set_syserr_cb (void (*cb)(const char *msg)); [NOT REENTRANT]
256		270
257	Set the callback function to call on a retryable system call error (such	271	Set the callback function to call on a retryable system call error (such
258	as failed select, poll, epoll_wait). The message is a printable string	272	as failed select, poll, epoll_wait). The message is a printable string
259	indicating the system call or subsystem causing the problem. If this	273	indicating the system call or subsystem causing the problem. If this
260	callback is set, then libev will expect it to remedy the situation, no	274	callback is set, then libev will expect it to remedy the situation, no
…		…
276		290
277	=back	291	=back
278		292
279	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	293	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
280		294
281	An event loop is described by a C<struct ev_loop *>. The library knows two	295	An event loop is described by a C<struct ev_loop *> (the C<struct>
282	types of such loops, the I<default> loop, which supports signals and child	296	is I<not> optional in this case, as there is also an C<ev_loop>
283	events, and dynamically created loops which do not.	297	I<function>).
		298
		299	The library knows two types of such loops, the I<default> loop, which
		300	supports signals and child events, and dynamically created loops which do
		301	not.
284		302
285	=over 4	303	=over 4
286		304
287	=item struct ev_loop *ev_default_loop (unsigned int flags)	305	=item struct ev_loop *ev_default_loop (unsigned int flags)
288		306
…		…
294	If you don't know what event loop to use, use the one returned from this	312	If you don't know what event loop to use, use the one returned from this
295	function.	313	function.
296		314
297	Note that this function is I<not> thread-safe, so if you want to use it	315	Note that this function is I<not> thread-safe, so if you want to use it
298	from multiple threads, you have to lock (note also that this is unlikely,	316	from multiple threads, you have to lock (note also that this is unlikely,
299	as loops cannot bes hared easily between threads anyway).	317	as loops cannot be shared easily between threads anyway).
300		318
301	The default loop is the only loop that can handle C<ev_signal> and	319	The default loop is the only loop that can handle C<ev_signal> and
302	C<ev_child> watchers, and to do this, it always registers a handler	320	C<ev_child> watchers, and to do this, it always registers a handler
303	for C<SIGCHLD>. If this is a problem for your application you can either	321	for C<SIGCHLD>. If this is a problem for your application you can either
304	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you	322	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
…		…
359	writing a server, you should C<accept ()> in a loop to accept as many	377	writing a server, you should C<accept ()> in a loop to accept as many
360	connections as possible during one iteration. You might also want to have	378	connections as possible during one iteration. You might also want to have
361	a look at C<ev_set_io_collect_interval ()> to increase the amount of	379	a look at C<ev_set_io_collect_interval ()> to increase the amount of
362	readiness notifications you get per iteration.	380	readiness notifications you get per iteration.
363		381
		382	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		383	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		384	C<exceptfds> set on that platform).
		385
364	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	386	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
365		387
366	And this is your standard poll(2) backend. It's more complicated	388	And this is your standard poll(2) backend. It's more complicated
367	than select, but handles sparse fds better and has no artificial	389	than select, but handles sparse fds better and has no artificial
368	limit on the number of fds you can use (except it will slow down	390	limit on the number of fds you can use (except it will slow down
369	considerably with a lot of inactive fds). It scales similarly to select,	391	considerably with a lot of inactive fds). It scales similarly to select,
370	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for	392	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
371	performance tips.	393	performance tips.
372		394
		395	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		396	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
		397
373	=item C<EVBACKEND_EPOLL> (value 4, Linux)	398	=item C<EVBACKEND_EPOLL> (value 4, Linux)
374		399
375	For few fds, this backend is a bit little slower than poll and select,	400	For few fds, this backend is a bit little slower than poll and select,
376	but it scales phenomenally better. While poll and select usually scale	401	but it scales phenomenally better. While poll and select usually scale
377	like O(total_fds) where n is the total number of fds (or the highest fd),	402	like O(total_fds) where n is the total number of fds (or the highest fd),
378	epoll scales either O(1) or O(active_fds). The epoll design has a number	403	epoll scales either O(1) or O(active_fds).
379	of shortcomings, such as silently dropping events in some hard-to-detect	404
380	cases and requiring a system call per fd change, no fork support and bad	405	The epoll mechanism deserves honorable mention as the most misdesigned
381	support for dup.	406	of the more advanced event mechanisms: mere annoyances include silently
		407	dropping file descriptors, requiring a system call per change per file
		408	descriptor (and unnecessary guessing of parameters), problems with dup and
		409	so on. The biggest issue is fork races, however - if a program forks then
		410	I<both> parent and child process have to recreate the epoll set, which can
		411	take considerable time (one syscall per file descriptor) and is of course
		412	hard to detect.
		413
		414	Epoll is also notoriously buggy - embedding epoll fds I<should> work, but
		415	of course I<doesn't>, and epoll just loves to report events for totally
		416	I<different> file descriptors (even already closed ones, so one cannot
		417	even remove them from the set) than registered in the set (especially
		418	on SMP systems). Libev tries to counter these spurious notifications by
		419	employing an additional generation counter and comparing that against the
		420	events to filter out spurious ones, recreating the set when required.
382		421
383	While stopping, setting and starting an I/O watcher in the same iteration	422	While stopping, setting and starting an I/O watcher in the same iteration
384	will result in some caching, there is still a system call per such incident	423	will result in some caching, there is still a system call per such
385	(because the fd could point to a different file description now), so its	424	incident (because the same I<file descriptor> could point to a different
386	best to avoid that. Also, C<dup ()>'ed file descriptors might not work	425	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
387	very well if you register events for both fds.	426	file descriptors might not work very well if you register events for both
388		427	file descriptors.
389	Please note that epoll sometimes generates spurious notifications, so you
390	need to use non-blocking I/O or other means to avoid blocking when no data
391	(or space) is available.
392		428
393	Best performance from this backend is achieved by not unregistering all	429	Best performance from this backend is achieved by not unregistering all
394	watchers for a file descriptor until it has been closed, if possible, i.e.	430	watchers for a file descriptor until it has been closed, if possible,
395	keep at least one watcher active per fd at all times.	431	i.e. keep at least one watcher active per fd at all times. Stopping and
		432	starting a watcher (without re-setting it) also usually doesn't cause
		433	extra overhead. A fork can both result in spurious notifications as well
		434	as in libev having to destroy and recreate the epoll object, which can
		435	take considerable time and thus should be avoided.
		436
		437	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		438	faster than epoll for maybe up to a hundred file descriptors, depending on
		439	the usage. So sad.
396		440
397	While nominally embeddable in other event loops, this feature is broken in	441	While nominally embeddable in other event loops, this feature is broken in
398	all kernel versions tested so far.	442	all kernel versions tested so far.
		443
		444	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		445	C<EVBACKEND_POLL>.
399		446
400	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	447	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
401		448
402	Kqueue deserves special mention, as at the time of this writing, it	449	Kqueue deserves special mention, as at the time of this writing, it
403	was broken on all BSDs except NetBSD (usually it doesn't work reliably	450	was broken on all BSDs except NetBSD (usually it doesn't work reliably
404	with anything but sockets and pipes, except on Darwin, where of course	451	with anything but sockets and pipes, except on Darwin, where of course
405	it's completely useless). For this reason it's not being "auto-detected"	452	it's completely useless). Unlike epoll, however, whose brokenness
		453	is by design, these kqueue bugs can (and eventually will) be fixed
		454	without API changes to existing programs. For this reason it's not being
406	unless you explicitly specify it explicitly in the flags (i.e. using	455	"auto-detected" unless you explicitly specify it in the flags (i.e. using
407	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)	456	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
408	system like NetBSD.	457	system like NetBSD.
409		458
410	You still can embed kqueue into a normal poll or select backend and use it	459	You still can embed kqueue into a normal poll or select backend and use it
411	only for sockets (after having made sure that sockets work with kqueue on	460	only for sockets (after having made sure that sockets work with kqueue on
…		…
413		462
414	It scales in the same way as the epoll backend, but the interface to the	463	It scales in the same way as the epoll backend, but the interface to the
415	kernel is more efficient (which says nothing about its actual speed, of	464	kernel is more efficient (which says nothing about its actual speed, of
416	course). While stopping, setting and starting an I/O watcher does never	465	course). While stopping, setting and starting an I/O watcher does never
417	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to	466	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
418	two event changes per incident, support for C<fork ()> is very bad and it	467	two event changes per incident. Support for C<fork ()> is very bad (but
419	drops fds silently in similarly hard-to-detect cases.	468	sane, unlike epoll) and it drops fds silently in similarly hard-to-detect
		469	cases
420		470
421	This backend usually performs well under most conditions.	471	This backend usually performs well under most conditions.
422		472
423	While nominally embeddable in other event loops, this doesn't work	473	While nominally embeddable in other event loops, this doesn't work
424	everywhere, so you might need to test for this. And since it is broken	474	everywhere, so you might need to test for this. And since it is broken
425	almost everywhere, you should only use it when you have a lot of sockets	475	almost everywhere, you should only use it when you have a lot of sockets
426	(for which it usually works), by embedding it into another event loop	476	(for which it usually works), by embedding it into another event loop
427	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for	477	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
428	sockets.	478	also broken on OS X)) and, did I mention it, using it only for sockets.
		479
		480	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		481	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		482	C<NOTE_EOF>.
429		483
430	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	484	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
431		485
432	This is not implemented yet (and might never be, unless you send me an	486	This is not implemented yet (and might never be, unless you send me an
433	implementation). According to reports, C</dev/poll> only supports sockets	487	implementation). According to reports, C</dev/poll> only supports sockets
…		…
446	While this backend scales well, it requires one system call per active	500	While this backend scales well, it requires one system call per active
447	file descriptor per loop iteration. For small and medium numbers of file	501	file descriptor per loop iteration. For small and medium numbers of file
448	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	502	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
449	might perform better.	503	might perform better.
450		504
451	On the positive side, ignoring the spurious readiness notifications, this	505	On the positive side, with the exception of the spurious readiness
452	backend actually performed to specification in all tests and is fully	506	notifications, this backend actually performed fully to specification
453	embeddable, which is a rare feat among the OS-specific backends.	507	in all tests and is fully embeddable, which is a rare feat among the
		508	OS-specific backends (I vastly prefer correctness over speed hacks).
		509
		510	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		511	C<EVBACKEND_POLL>.
454		512
455	=item C<EVBACKEND_ALL>	513	=item C<EVBACKEND_ALL>
456		514
457	Try all backends (even potentially broken ones that wouldn't be tried	515	Try all backends (even potentially broken ones that wouldn't be tried
458	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	516	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
…		…
464		522
465	If one or more of these are or'ed into the flags value, then only these	523	If one or more of these are or'ed into the flags value, then only these
466	backends will be tried (in the reverse order as listed here). If none are	524	backends will be tried (in the reverse order as listed here). If none are
467	specified, all backends in C<ev_recommended_backends ()> will be tried.	525	specified, all backends in C<ev_recommended_backends ()> will be tried.
468		526
469	The most typical usage is like this:	527	Example: This is the most typical usage.
470		528
471	if (!ev_default_loop (0))	529	if (!ev_default_loop (0))
472	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	530	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
473		531
474	Restrict libev to the select and poll backends, and do not allow	532	Example: Restrict libev to the select and poll backends, and do not allow
475	environment settings to be taken into account:	533	environment settings to be taken into account:
476		534
477	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);	535	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
478		536
479	Use whatever libev has to offer, but make sure that kqueue is used if	537	Example: Use whatever libev has to offer, but make sure that kqueue is
480	available (warning, breaks stuff, best use only with your own private	538	used if available (warning, breaks stuff, best use only with your own
481	event loop and only if you know the OS supports your types of fds):	539	private event loop and only if you know the OS supports your types of
		540	fds):
482		541
483	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);	542	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
484		543
485	=item struct ev_loop *ev_loop_new (unsigned int flags)	544	=item struct ev_loop *ev_loop_new (unsigned int flags)
486		545
…		…
507	responsibility to either stop all watchers cleanly yourself I<before>	566	responsibility to either stop all watchers cleanly yourself I<before>
508	calling this function, or cope with the fact afterwards (which is usually	567	calling this function, or cope with the fact afterwards (which is usually
509	the easiest thing, you can just ignore the watchers and/or C<free ()> them	568	the easiest thing, you can just ignore the watchers and/or C<free ()> them
510	for example).	569	for example).
511		570
512	Note that certain global state, such as signal state, will not be freed by	571	Note that certain global state, such as signal state (and installed signal
513	this function, and related watchers (such as signal and child watchers)	572	handlers), will not be freed by this function, and related watchers (such
514	would need to be stopped manually.	573	as signal and child watchers) would need to be stopped manually.
515		574
516	In general it is not advisable to call this function except in the	575	In general it is not advisable to call this function except in the
517	rare occasion where you really need to free e.g. the signal handling	576	rare occasion where you really need to free e.g. the signal handling
518	pipe fds. If you need dynamically allocated loops it is better to use	577	pipe fds. If you need dynamically allocated loops it is better to use
519	C<ev_loop_new> and C<ev_loop_destroy>).	578	C<ev_loop_new> and C<ev_loop_destroy>).
…		…
544		603
545	=item ev_loop_fork (loop)	604	=item ev_loop_fork (loop)
546		605
547	Like C<ev_default_fork>, but acts on an event loop created by	606	Like C<ev_default_fork>, but acts on an event loop created by
548	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	607	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
549	after fork, and how you do this is entirely your own problem.	608	after fork that you want to re-use in the child, and how you do this is
		609	entirely your own problem.
550		610
551	=item int ev_is_default_loop (loop)	611	=item int ev_is_default_loop (loop)
552		612
553	Returns true when the given loop actually is the default loop, false otherwise.	613	Returns true when the given loop is, in fact, the default loop, and false
		614	otherwise.
554		615
555	=item unsigned int ev_loop_count (loop)	616	=item unsigned int ev_loop_count (loop)
556		617
557	Returns the count of loop iterations for the loop, which is identical to	618	Returns the count of loop iterations for the loop, which is identical to
558	the number of times libev did poll for new events. It starts at C<0> and	619	the number of times libev did poll for new events. It starts at C<0> and
…		…
573	received events and started processing them. This timestamp does not	634	received events and started processing them. This timestamp does not
574	change as long as callbacks are being processed, and this is also the base	635	change as long as callbacks are being processed, and this is also the base
575	time used for relative timers. You can treat it as the timestamp of the	636	time used for relative timers. You can treat it as the timestamp of the
576	event occurring (or more correctly, libev finding out about it).	637	event occurring (or more correctly, libev finding out about it).
577		638
		639	=item ev_now_update (loop)
		640
		641	Establishes the current time by querying the kernel, updating the time
		642	returned by C<ev_now ()> in the progress. This is a costly operation and
		643	is usually done automatically within C<ev_loop ()>.
		644
		645	This function is rarely useful, but when some event callback runs for a
		646	very long time without entering the event loop, updating libev's idea of
		647	the current time is a good idea.
		648
		649	See also L<The special problem of time updates> in the C<ev_timer> section.
		650
		651	=item ev_suspend (loop)
		652
		653	=item ev_resume (loop)
		654
		655	These two functions suspend and resume a loop, for use when the loop is
		656	not used for a while and timeouts should not be processed.
		657
		658	A typical use case would be an interactive program such as a game: When
		659	the user presses C<^Z> to suspend the game and resumes it an hour later it
		660	would be best to handle timeouts as if no time had actually passed while
		661	the program was suspended. This can be achieved by calling C<ev_suspend>
		662	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		663	C<ev_resume> directly afterwards to resume timer processing.
		664
		665	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		666	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		667	will be rescheduled (that is, they will lose any events that would have
		668	occured while suspended).
		669
		670	After calling C<ev_suspend> you B<must not> call I<any> function on the
		671	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		672	without a previous call to C<ev_suspend>.
		673
		674	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		675	event loop time (see C<ev_now_update>).
		676
578	=item ev_loop (loop, int flags)	677	=item ev_loop (loop, int flags)
579		678
580	Finally, this is it, the event handler. This function usually is called	679	Finally, this is it, the event handler. This function usually is called
581	after you initialised all your watchers and you want to start handling	680	after you initialised all your watchers and you want to start handling
582	events.	681	events.
…		…
584	If the flags argument is specified as C<0>, it will not return until	683	If the flags argument is specified as C<0>, it will not return until
585	either no event watchers are active anymore or C<ev_unloop> was called.	684	either no event watchers are active anymore or C<ev_unloop> was called.
586		685
587	Please note that an explicit C<ev_unloop> is usually better than	686	Please note that an explicit C<ev_unloop> is usually better than
588	relying on all watchers to be stopped when deciding when a program has	687	relying on all watchers to be stopped when deciding when a program has
589	finished (especially in interactive programs), but having a program that	688	finished (especially in interactive programs), but having a program
590	automatically loops as long as it has to and no longer by virtue of	689	that automatically loops as long as it has to and no longer by virtue
591	relying on its watchers stopping correctly is a thing of beauty.	690	of relying on its watchers stopping correctly, that is truly a thing of
		691	beauty.
592		692
593	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	693	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
594	those events and any outstanding ones, but will not block your process in	694	those events and any already outstanding ones, but will not block your
595	case there are no events and will return after one iteration of the loop.	695	process in case there are no events and will return after one iteration of
		696	the loop.
596		697
597	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	698	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
598	necessary) and will handle those and any outstanding ones. It will block	699	necessary) and will handle those and any already outstanding ones. It
599	your process until at least one new event arrives, and will return after	700	will block your process until at least one new event arrives (which could
600	one iteration of the loop. This is useful if you are waiting for some	701	be an event internal to libev itself, so there is no guarantee that a
601	external event in conjunction with something not expressible using other	702	user-registered callback will be called), and will return after one
		703	iteration of the loop.
		704
		705	This is useful if you are waiting for some external event in conjunction
		706	with something not expressible using other libev watchers (i.e. "roll your
602	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	707	own C<ev_loop>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
603	usually a better approach for this kind of thing.	708	usually a better approach for this kind of thing.
604		709
605	Here are the gory details of what C<ev_loop> does:	710	Here are the gory details of what C<ev_loop> does:
606		711
607	- Before the first iteration, call any pending watchers.	712	- Before the first iteration, call any pending watchers.
…		…
617	any active watchers at all will result in not sleeping).	722	any active watchers at all will result in not sleeping).
618	- Sleep if the I/O and timer collect interval say so.	723	- Sleep if the I/O and timer collect interval say so.
619	- Block the process, waiting for any events.	724	- Block the process, waiting for any events.
620	- Queue all outstanding I/O (fd) events.	725	- Queue all outstanding I/O (fd) events.
621	- Update the "event loop time" (ev_now ()), and do time jump adjustments.	726	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
622	- Queue all outstanding timers.	727	- Queue all expired timers.
623	- Queue all outstanding periodics.	728	- Queue all expired periodics.
624	- Unless any events are pending now, queue all idle watchers.	729	- Unless any events are pending now, queue all idle watchers.
625	- Queue all check watchers.	730	- Queue all check watchers.
626	- Call all queued watchers in reverse order (i.e. check watchers first).	731	- Call all queued watchers in reverse order (i.e. check watchers first).
627	Signals and child watchers are implemented as I/O watchers, and will	732	Signals and child watchers are implemented as I/O watchers, and will
628	be handled here by queueing them when their watcher gets executed.	733	be handled here by queueing them when their watcher gets executed.
…		…
645	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	750	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
646	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	751	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
647		752
648	This "unloop state" will be cleared when entering C<ev_loop> again.	753	This "unloop state" will be cleared when entering C<ev_loop> again.
649		754
		755	It is safe to call C<ev_unloop> from otuside any C<ev_loop> calls.
		756
650	=item ev_ref (loop)	757	=item ev_ref (loop)
651		758
652	=item ev_unref (loop)	759	=item ev_unref (loop)
653		760
654	Ref/unref can be used to add or remove a reference count on the event	761	Ref/unref can be used to add or remove a reference count on the event
655	loop: Every watcher keeps one reference, and as long as the reference	762	loop: Every watcher keeps one reference, and as long as the reference
656	count is nonzero, C<ev_loop> will not return on its own. If you have	763	count is nonzero, C<ev_loop> will not return on its own.
		764
657	a watcher you never unregister that should not keep C<ev_loop> from	765	If you have a watcher you never unregister that should not keep C<ev_loop>
658	returning, ev_unref() after starting, and ev_ref() before stopping it. For	766	from returning, call ev_unref() after starting, and ev_ref() before
		767	stopping it.
		768
659	example, libev itself uses this for its internal signal pipe: It is not	769	As an example, libev itself uses this for its internal signal pipe: It
660	visible to the libev user and should not keep C<ev_loop> from exiting if	770	is not visible to the libev user and should not keep C<ev_loop> from
661	no event watchers registered by it are active. It is also an excellent	771	exiting if no event watchers registered by it are active. It is also an
662	way to do this for generic recurring timers or from within third-party	772	excellent way to do this for generic recurring timers or from within
663	libraries. Just remember to I<unref after start> and I<ref before stop>	773	third-party libraries. Just remember to I<unref after start> and I<ref
664	(but only if the watcher wasn't active before, or was active before,	774	before stop> (but only if the watcher wasn't active before, or was active
665	respectively).	775	before, respectively. Note also that libev might stop watchers itself
		776	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		777	in the callback).
666		778
667	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	779	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
668	running when nothing else is active.	780	running when nothing else is active.
669		781
670	struct ev_signal exitsig;	782	ev_signal exitsig;
671	ev_signal_init (&exitsig, sig_cb, SIGINT);	783	ev_signal_init (&exitsig, sig_cb, SIGINT);
672	ev_signal_start (loop, &exitsig);	784	ev_signal_start (loop, &exitsig);
673	evf_unref (loop);	785	evf_unref (loop);
674		786
675	Example: For some weird reason, unregister the above signal handler again.	787	Example: For some weird reason, unregister the above signal handler again.
…		…
689	Setting these to a higher value (the C<interval> I<must> be >= C<0>)	801	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
690	allows libev to delay invocation of I/O and timer/periodic callbacks	802	allows libev to delay invocation of I/O and timer/periodic callbacks
691	to increase efficiency of loop iterations (or to increase power-saving	803	to increase efficiency of loop iterations (or to increase power-saving
692	opportunities).	804	opportunities).
693		805
694	The background is that sometimes your program runs just fast enough to	806	The idea is that sometimes your program runs just fast enough to handle
695	handle one (or very few) event(s) per loop iteration. While this makes	807	one (or very few) event(s) per loop iteration. While this makes the
696	the program responsive, it also wastes a lot of CPU time to poll for new	808	program responsive, it also wastes a lot of CPU time to poll for new
697	events, especially with backends like C<select ()> which have a high	809	events, especially with backends like C<select ()> which have a high
698	overhead for the actual polling but can deliver many events at once.	810	overhead for the actual polling but can deliver many events at once.
699		811
700	By setting a higher I<io collect interval> you allow libev to spend more	812	By setting a higher I<io collect interval> you allow libev to spend more
701	time collecting I/O events, so you can handle more events per iteration,	813	time collecting I/O events, so you can handle more events per iteration,
…		…
703	C<ev_timer>) will be not affected. Setting this to a non-null value will	815	C<ev_timer>) will be not affected. Setting this to a non-null value will
704	introduce an additional C<ev_sleep ()> call into most loop iterations.	816	introduce an additional C<ev_sleep ()> call into most loop iterations.
705		817
706	Likewise, by setting a higher I<timeout collect interval> you allow libev	818	Likewise, by setting a higher I<timeout collect interval> you allow libev
707	to spend more time collecting timeouts, at the expense of increased	819	to spend more time collecting timeouts, at the expense of increased
708	latency (the watcher callback will be called later). C<ev_io> watchers	820	latency/jitter/inexactness (the watcher callback will be called
709	will not be affected. Setting this to a non-null value will not introduce	821	later). C<ev_io> watchers will not be affected. Setting this to a non-null
710	any overhead in libev.	822	value will not introduce any overhead in libev.
711		823
712	Many (busy) programs can usually benefit by setting the I/O collect	824	Many (busy) programs can usually benefit by setting the I/O collect
713	interval to a value near C<0.1> or so, which is often enough for	825	interval to a value near C<0.1> or so, which is often enough for
714	interactive servers (of course not for games), likewise for timeouts. It	826	interactive servers (of course not for games), likewise for timeouts. It
715	usually doesn't make much sense to set it to a lower value than C<0.01>,	827	usually doesn't make much sense to set it to a lower value than C<0.01>,
…		…
723	they fire on, say, one-second boundaries only.	835	they fire on, say, one-second boundaries only.
724		836
725	=item ev_loop_verify (loop)	837	=item ev_loop_verify (loop)
726		838
727	This function only does something when C<EV_VERIFY> support has been	839	This function only does something when C<EV_VERIFY> support has been
728	compiled in. It tries to go through all internal structures and checks	840	compiled in, which is the default for non-minimal builds. It tries to go
729	them for validity. If anything is found to be inconsistent, it will print	841	through all internal structures and checks them for validity. If anything
730	an error message to standard error and call C<abort ()>.	842	is found to be inconsistent, it will print an error message to standard
		843	error and call C<abort ()>.
731		844
732	This can be used to catch bugs inside libev itself: under normal	845	This can be used to catch bugs inside libev itself: under normal
733	circumstances, this function will never abort as of course libev keeps its	846	circumstances, this function will never abort as of course libev keeps its
734	data structures consistent.	847	data structures consistent.
735		848
736	=back	849	=back
737		850
738		851
739	=head1 ANATOMY OF A WATCHER	852	=head1 ANATOMY OF A WATCHER
740		853
		854	In the following description, uppercase C<TYPE> in names stands for the
		855	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		856	watchers and C<ev_io_start> for I/O watchers.
		857
741	A watcher is a structure that you create and register to record your	858	A watcher is a structure that you create and register to record your
742	interest in some event. For instance, if you want to wait for STDIN to	859	interest in some event. For instance, if you want to wait for STDIN to
743	become readable, you would create an C<ev_io> watcher for that:	860	become readable, you would create an C<ev_io> watcher for that:
744		861
745	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	862	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
746	{	863	{
747	ev_io_stop (w);	864	ev_io_stop (w);
748	ev_unloop (loop, EVUNLOOP_ALL);	865	ev_unloop (loop, EVUNLOOP_ALL);
749	}	866	}
750		867
751	struct ev_loop *loop = ev_default_loop (0);	868	struct ev_loop *loop = ev_default_loop (0);
		869
752	struct ev_io stdin_watcher;	870	ev_io stdin_watcher;
		871
753	ev_init (&stdin_watcher, my_cb);	872	ev_init (&stdin_watcher, my_cb);
754	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	873	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
755	ev_io_start (loop, &stdin_watcher);	874	ev_io_start (loop, &stdin_watcher);
		875
756	ev_loop (loop, 0);	876	ev_loop (loop, 0);
757		877
758	As you can see, you are responsible for allocating the memory for your	878	As you can see, you are responsible for allocating the memory for your
759	watcher structures (and it is usually a bad idea to do this on the stack,	879	watcher structures (and it is I<usually> a bad idea to do this on the
760	although this can sometimes be quite valid).	880	stack).
		881
		882	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		883	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
761		884
762	Each watcher structure must be initialised by a call to C<ev_init	885	Each watcher structure must be initialised by a call to C<ev_init
763	(watcher *, callback)>, which expects a callback to be provided. This	886	(watcher *, callback)>, which expects a callback to be provided. This
764	callback gets invoked each time the event occurs (or, in the case of I/O	887	callback gets invoked each time the event occurs (or, in the case of I/O
765	watchers, each time the event loop detects that the file descriptor given	888	watchers, each time the event loop detects that the file descriptor given
766	is readable and/or writable).	889	is readable and/or writable).
767		890
768	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	891	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
769	with arguments specific to this watcher type. There is also a macro	892	macro to configure it, with arguments specific to the watcher type. There
770	to combine initialisation and setting in one call: C<< ev_<type>_init	893	is also a macro to combine initialisation and setting in one call: C<<
771	(watcher *, callback, ...) >>.	894	ev_TYPE_init (watcher *, callback, ...) >>.
772		895
773	To make the watcher actually watch out for events, you have to start it	896	To make the watcher actually watch out for events, you have to start it
774	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	897	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
775	*) >>), and you can stop watching for events at any time by calling the	898	*) >>), and you can stop watching for events at any time by calling the
776	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	899	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
777		900
778	As long as your watcher is active (has been started but not stopped) you	901	As long as your watcher is active (has been started but not stopped) you
779	must not touch the values stored in it. Most specifically you must never	902	must not touch the values stored in it. Most specifically you must never
780	reinitialise it or call its C<set> macro.	903	reinitialise it or call its C<ev_TYPE_set> macro.
781		904
782	Each and every callback receives the event loop pointer as first, the	905	Each and every callback receives the event loop pointer as first, the
783	registered watcher structure as second, and a bitset of received events as	906	registered watcher structure as second, and a bitset of received events as
784	third argument.	907	third argument.
785		908
…		…
843		966
844	=item C<EV_ASYNC>	967	=item C<EV_ASYNC>
845		968
846	The given async watcher has been asynchronously notified (see C<ev_async>).	969	The given async watcher has been asynchronously notified (see C<ev_async>).
847		970
		971	=item C<EV_CUSTOM>
		972
		973	Not ever sent (or otherwise used) by libev itself, but can be freely used
		974	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		975
848	=item C<EV_ERROR>	976	=item C<EV_ERROR>
849		977
850	An unspecified error has occurred, the watcher has been stopped. This might	978	An unspecified error has occurred, the watcher has been stopped. This might
851	happen because the watcher could not be properly started because libev	979	happen because the watcher could not be properly started because libev
852	ran out of memory, a file descriptor was found to be closed or any other	980	ran out of memory, a file descriptor was found to be closed or any other
		981	problem. Libev considers these application bugs.
		982
853	problem. You best act on it by reporting the problem and somehow coping	983	You best act on it by reporting the problem and somehow coping with the
854	with the watcher being stopped.	984	watcher being stopped. Note that well-written programs should not receive
		985	an error ever, so when your watcher receives it, this usually indicates a
		986	bug in your program.
855		987
856	Libev will usually signal a few "dummy" events together with an error,	988	Libev will usually signal a few "dummy" events together with an error, for
857	for example it might indicate that a fd is readable or writable, and if	989	example it might indicate that a fd is readable or writable, and if your
858	your callbacks is well-written it can just attempt the operation and cope	990	callbacks is well-written it can just attempt the operation and cope with
859	with the error from read() or write(). This will not work in multi-threaded	991	the error from read() or write(). This will not work in multi-threaded
860	programs, though, so beware.	992	programs, though, as the fd could already be closed and reused for another
		993	thing, so beware.
861		994
862	=back	995	=back
863		996
864	=head2 GENERIC WATCHER FUNCTIONS	997	=head2 GENERIC WATCHER FUNCTIONS
865
866	In the following description, C<TYPE> stands for the watcher type,
867	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
868		998
869	=over 4	999	=over 4
870		1000
871	=item C<ev_init> (ev_TYPE *watcher, callback)	1001	=item C<ev_init> (ev_TYPE *watcher, callback)
872		1002
…		…
878	which rolls both calls into one.	1008	which rolls both calls into one.
879		1009
880	You can reinitialise a watcher at any time as long as it has been stopped	1010	You can reinitialise a watcher at any time as long as it has been stopped
881	(or never started) and there are no pending events outstanding.	1011	(or never started) and there are no pending events outstanding.
882		1012
883	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	1013	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
884	int revents)>.	1014	int revents)>.
		1015
		1016	Example: Initialise an C<ev_io> watcher in two steps.
		1017
		1018	ev_io w;
		1019	ev_init (&w, my_cb);
		1020	ev_io_set (&w, STDIN_FILENO, EV_READ);
885		1021
886	=item C<ev_TYPE_set> (ev_TYPE *, [args])	1022	=item C<ev_TYPE_set> (ev_TYPE *, [args])
887		1023
888	This macro initialises the type-specific parts of a watcher. You need to	1024	This macro initialises the type-specific parts of a watcher. You need to
889	call C<ev_init> at least once before you call this macro, but you can	1025	call C<ev_init> at least once before you call this macro, but you can
…		…
892	difference to the C<ev_init> macro).	1028	difference to the C<ev_init> macro).
893		1029
894	Although some watcher types do not have type-specific arguments	1030	Although some watcher types do not have type-specific arguments
895	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	1031	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
896		1032
		1033	See C<ev_init>, above, for an example.
		1034
897	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	1035	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
898		1036
899	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	1037	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
900	calls into a single call. This is the most convenient method to initialise	1038	calls into a single call. This is the most convenient method to initialise
901	a watcher. The same limitations apply, of course.	1039	a watcher. The same limitations apply, of course.
902		1040
		1041	Example: Initialise and set an C<ev_io> watcher in one step.
		1042
		1043	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1044
903	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	1045	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
904		1046
905	Starts (activates) the given watcher. Only active watchers will receive	1047	Starts (activates) the given watcher. Only active watchers will receive
906	events. If the watcher is already active nothing will happen.	1048	events. If the watcher is already active nothing will happen.
907		1049
		1050	Example: Start the C<ev_io> watcher that is being abused as example in this
		1051	whole section.
		1052
		1053	ev_io_start (EV_DEFAULT_UC, &w);
		1054
908	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	1055	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
909		1056
910	Stops the given watcher again (if active) and clears the pending	1057	Stops the given watcher if active, and clears the pending status (whether
		1058	the watcher was active or not).
		1059
911	status. It is possible that stopped watchers are pending (for example,	1060	It is possible that stopped watchers are pending - for example,
912	non-repeating timers are being stopped when they become pending), but	1061	non-repeating timers are being stopped when they become pending - but
913	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	1062	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
914	you want to free or reuse the memory used by the watcher it is therefore a	1063	pending. If you want to free or reuse the memory used by the watcher it is
915	good idea to always call its C<ev_TYPE_stop> function.	1064	therefore a good idea to always call its C<ev_TYPE_stop> function.
916		1065
917	=item bool ev_is_active (ev_TYPE *watcher)	1066	=item bool ev_is_active (ev_TYPE *watcher)
918		1067
919	Returns a true value iff the watcher is active (i.e. it has been started	1068	Returns a true value iff the watcher is active (i.e. it has been started
920	and not yet been stopped). As long as a watcher is active you must not modify	1069	and not yet been stopped). As long as a watcher is active you must not modify
…		…
946	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>	1095	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
947	(default: C<-2>). Pending watchers with higher priority will be invoked	1096	(default: C<-2>). Pending watchers with higher priority will be invoked
948	before watchers with lower priority, but priority will not keep watchers	1097	before watchers with lower priority, but priority will not keep watchers
949	from being executed (except for C<ev_idle> watchers).	1098	from being executed (except for C<ev_idle> watchers).
950		1099
951	This means that priorities are I<only> used for ordering callback
952	invocation after new events have been received. This is useful, for
953	example, to reduce latency after idling, or more often, to bind two
954	watchers on the same event and make sure one is called first.
955
956	If you need to suppress invocation when higher priority events are pending	1100	If you need to suppress invocation when higher priority events are pending
957	you need to look at C<ev_idle> watchers, which provide this functionality.	1101	you need to look at C<ev_idle> watchers, which provide this functionality.
958		1102
959	You I<must not> change the priority of a watcher as long as it is active or	1103	You I<must not> change the priority of a watcher as long as it is active or
960	pending.	1104	pending.
961		1105
		1106	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1107	fine, as long as you do not mind that the priority value you query might
		1108	or might not have been clamped to the valid range.
		1109
962	The default priority used by watchers when no priority has been set is	1110	The default priority used by watchers when no priority has been set is
963	always C<0>, which is supposed to not be too high and not be too low :).	1111	always C<0>, which is supposed to not be too high and not be too low :).
964		1112
965	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is	1113	See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
966	fine, as long as you do not mind that the priority value you query might	1114	priorities.
967	or might not have been adjusted to be within valid range.
968		1115
969	=item ev_invoke (loop, ev_TYPE *watcher, int revents)	1116	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
970		1117
971	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither	1118	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
972	C<loop> nor C<revents> need to be valid as long as the watcher callback	1119	C<loop> nor C<revents> need to be valid as long as the watcher callback
973	can deal with that fact.	1120	can deal with that fact, as both are simply passed through to the
		1121	callback.
974		1122
975	=item int ev_clear_pending (loop, ev_TYPE *watcher)	1123	=item int ev_clear_pending (loop, ev_TYPE *watcher)
976		1124
977	If the watcher is pending, this function returns clears its pending status	1125	If the watcher is pending, this function clears its pending status and
978	and returns its C<revents> bitset (as if its callback was invoked). If the	1126	returns its C<revents> bitset (as if its callback was invoked). If the
979	watcher isn't pending it does nothing and returns C<0>.	1127	watcher isn't pending it does nothing and returns C<0>.
980		1128
		1129	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1130	callback to be invoked, which can be accomplished with this function.
		1131
981	=back	1132	=back
982		1133
983		1134
984	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1135	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
985		1136
986	Each watcher has, by default, a member C<void *data> that you can change	1137	Each watcher has, by default, a member C<void *data> that you can change
987	and read at any time, libev will completely ignore it. This can be used	1138	and read at any time: libev will completely ignore it. This can be used
988	to associate arbitrary data with your watcher. If you need more data and	1139	to associate arbitrary data with your watcher. If you need more data and
989	don't want to allocate memory and store a pointer to it in that data	1140	don't want to allocate memory and store a pointer to it in that data
990	member, you can also "subclass" the watcher type and provide your own	1141	member, you can also "subclass" the watcher type and provide your own
991	data:	1142	data:
992		1143
993	struct my_io	1144	struct my_io
994	{	1145	{
995	struct ev_io io;	1146	ev_io io;
996	int otherfd;	1147	int otherfd;
997	void *somedata;	1148	void *somedata;
998	struct whatever *mostinteresting;	1149	struct whatever *mostinteresting;
999	}	1150	};
		1151
		1152	...
		1153	struct my_io w;
		1154	ev_io_init (&w.io, my_cb, fd, EV_READ);
1000		1155
1001	And since your callback will be called with a pointer to the watcher, you	1156	And since your callback will be called with a pointer to the watcher, you
1002	can cast it back to your own type:	1157	can cast it back to your own type:
1003		1158
1004	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1159	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
1005	{	1160	{
1006	struct my_io *w = (struct my_io *)w_;	1161	struct my_io *w = (struct my_io *)w_;
1007	...	1162	...
1008	}	1163	}
1009		1164
1010	More interesting and less C-conformant ways of casting your callback type	1165	More interesting and less C-conformant ways of casting your callback type
1011	instead have been omitted.	1166	instead have been omitted.
1012		1167
1013	Another common scenario is having some data structure with multiple	1168	Another common scenario is to use some data structure with multiple
1014	watchers:	1169	embedded watchers:
1015		1170
1016	struct my_biggy	1171	struct my_biggy
1017	{	1172	{
1018	int some_data;	1173	int some_data;
1019	ev_timer t1;	1174	ev_timer t1;
1020	ev_timer t2;	1175	ev_timer t2;
1021	}	1176	}
1022		1177
1023	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1178	In this case getting the pointer to C<my_biggy> is a bit more
1024	you need to use C<offsetof>:	1179	complicated: Either you store the address of your C<my_biggy> struct
		1180	in the C<data> member of the watcher (for woozies), or you need to use
		1181	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1182	programmers):
1025		1183
1026	#include <stddef.h>	1184	#include <stddef.h>
1027		1185
1028	static void	1186	static void
1029	t1_cb (EV_P_ struct ev_timer *w, int revents)	1187	t1_cb (EV_P_ ev_timer *w, int revents)
1030	{	1188	{
1031	struct my_biggy big = (struct my_biggy *	1189	struct my_biggy big = (struct my_biggy *
1032	(((char *)w) - offsetof (struct my_biggy, t1));	1190	(((char *)w) - offsetof (struct my_biggy, t1));
1033	}	1191	}
1034		1192
1035	static void	1193	static void
1036	t2_cb (EV_P_ struct ev_timer *w, int revents)	1194	t2_cb (EV_P_ ev_timer *w, int revents)
1037	{	1195	{
1038	struct my_biggy big = (struct my_biggy *	1196	struct my_biggy big = (struct my_biggy *
1039	(((char *)w) - offsetof (struct my_biggy, t2));	1197	(((char *)w) - offsetof (struct my_biggy, t2));
1040	}	1198	}
		1199
		1200	=head2 WATCHER PRIORITY MODELS
		1201
		1202	Many event loops support I<watcher priorities>, which are usually small
		1203	integers that influence the ordering of event callback invocation
		1204	between watchers in some way, all else being equal.
		1205
		1206	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1207	description for the more technical details such as the actual priority
		1208	range.
		1209
		1210	There are two common ways how these these priorities are being interpreted
		1211	by event loops:
		1212
		1213	In the more common lock-out model, higher priorities "lock out" invocation
		1214	of lower priority watchers, which means as long as higher priority
		1215	watchers receive events, lower priority watchers are not being invoked.
		1216
		1217	The less common only-for-ordering model uses priorities solely to order
		1218	callback invocation within a single event loop iteration: Higher priority
		1219	watchers are invoked before lower priority ones, but they all get invoked
		1220	before polling for new events.
		1221
		1222	Libev uses the second (only-for-ordering) model for all its watchers
		1223	except for idle watchers (which use the lock-out model).
		1224
		1225	The rationale behind this is that implementing the lock-out model for
		1226	watchers is not well supported by most kernel interfaces, and most event
		1227	libraries will just poll for the same events again and again as long as
		1228	their callbacks have not been executed, which is very inefficient in the
		1229	common case of one high-priority watcher locking out a mass of lower
		1230	priority ones.
		1231
		1232	Static (ordering) priorities are most useful when you have two or more
		1233	watchers handling the same resource: a typical usage example is having an
		1234	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1235	timeouts. Under load, data might be received while the program handles
		1236	other jobs, but since timers normally get invoked first, the timeout
		1237	handler will be executed before checking for data. In that case, giving
		1238	the timer a lower priority than the I/O watcher ensures that I/O will be
		1239	handled first even under adverse conditions (which is usually, but not
		1240	always, what you want).
		1241
		1242	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1243	will only be executed when no same or higher priority watchers have
		1244	received events, they can be used to implement the "lock-out" model when
		1245	required.
		1246
		1247	For example, to emulate how many other event libraries handle priorities,
		1248	you can associate an C<ev_idle> watcher to each such watcher, and in
		1249	the normal watcher callback, you just start the idle watcher. The real
		1250	processing is done in the idle watcher callback. This causes libev to
		1251	continously poll and process kernel event data for the watcher, but when
		1252	the lock-out case is known to be rare (which in turn is rare :), this is
		1253	workable.
		1254
		1255	Usually, however, the lock-out model implemented that way will perform
		1256	miserably under the type of load it was designed to handle. In that case,
		1257	it might be preferable to stop the real watcher before starting the
		1258	idle watcher, so the kernel will not have to process the event in case
		1259	the actual processing will be delayed for considerable time.
		1260
		1261	Here is an example of an I/O watcher that should run at a strictly lower
		1262	priority than the default, and which should only process data when no
		1263	other events are pending:
		1264
		1265	ev_idle idle; // actual processing watcher
		1266	ev_io io; // actual event watcher
		1267
		1268	static void
		1269	io_cb (EV_P_ ev_io *w, int revents)
		1270	{
		1271	// stop the I/O watcher, we received the event, but
		1272	// are not yet ready to handle it.
		1273	ev_io_stop (EV_A_ w);
		1274
		1275	// start the idle watcher to ahndle the actual event.
		1276	// it will not be executed as long as other watchers
		1277	// with the default priority are receiving events.
		1278	ev_idle_start (EV_A_ &idle);
		1279	}
		1280
		1281	static void
		1282	idle-cb (EV_P_ ev_idle *w, int revents)
		1283	{
		1284	// actual processing
		1285	read (STDIN_FILENO, ...);
		1286
		1287	// have to start the I/O watcher again, as
		1288	// we have handled the event
		1289	ev_io_start (EV_P_ &io);
		1290	}
		1291
		1292	// initialisation
		1293	ev_idle_init (&idle, idle_cb);
		1294	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1295	ev_io_start (EV_DEFAULT_ &io);
		1296
		1297	In the "real" world, it might also be beneficial to start a timer, so that
		1298	low-priority connections can not be locked out forever under load. This
		1299	enables your program to keep a lower latency for important connections
		1300	during short periods of high load, while not completely locking out less
		1301	important ones.
1041		1302
1042		1303
1043	=head1 WATCHER TYPES	1304	=head1 WATCHER TYPES
1044		1305
1045	This section describes each watcher in detail, but will not repeat	1306	This section describes each watcher in detail, but will not repeat
…		…
1069	In general you can register as many read and/or write event watchers per	1330	In general you can register as many read and/or write event watchers per
1070	fd as you want (as long as you don't confuse yourself). Setting all file	1331	fd as you want (as long as you don't confuse yourself). Setting all file
1071	descriptors to non-blocking mode is also usually a good idea (but not	1332	descriptors to non-blocking mode is also usually a good idea (but not
1072	required if you know what you are doing).	1333	required if you know what you are doing).
1073		1334
1074	If you must do this, then force the use of a known-to-be-good backend	1335	If you cannot use non-blocking mode, then force the use of a
1075	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1336	known-to-be-good backend (at the time of this writing, this includes only
1076	C<EVBACKEND_POLL>).	1337	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
		1338	descriptors for which non-blocking operation makes no sense (such as
		1339	files) - libev doesn't guarentee any specific behaviour in that case.
1077		1340
1078	Another thing you have to watch out for is that it is quite easy to	1341	Another thing you have to watch out for is that it is quite easy to
1079	receive "spurious" readiness notifications, that is your callback might	1342	receive "spurious" readiness notifications, that is your callback might
1080	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1343	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
1081	because there is no data. Not only are some backends known to create a	1344	because there is no data. Not only are some backends known to create a
1082	lot of those (for example Solaris ports), it is very easy to get into	1345	lot of those (for example Solaris ports), it is very easy to get into
1083	this situation even with a relatively standard program structure. Thus	1346	this situation even with a relatively standard program structure. Thus
1084	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1347	it is best to always use non-blocking I/O: An extra C<read>(2) returning
1085	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1348	C<EAGAIN> is far preferable to a program hanging until some data arrives.
1086		1349
1087	If you cannot run the fd in non-blocking mode (for example you should not	1350	If you cannot run the fd in non-blocking mode (for example you should
1088	play around with an Xlib connection), then you have to separately re-test	1351	not play around with an Xlib connection), then you have to separately
1089	whether a file descriptor is really ready with a known-to-be good interface	1352	re-test whether a file descriptor is really ready with a known-to-be good
1090	such as poll (fortunately in our Xlib example, Xlib already does this on	1353	interface such as poll (fortunately in our Xlib example, Xlib already
1091	its own, so its quite safe to use).	1354	does this on its own, so its quite safe to use). Some people additionally
		1355	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1356	indefinitely.
		1357
		1358	But really, best use non-blocking mode.
1092		1359
1093	=head3 The special problem of disappearing file descriptors	1360	=head3 The special problem of disappearing file descriptors
1094		1361
1095	Some backends (e.g. kqueue, epoll) need to be told about closing a file	1362	Some backends (e.g. kqueue, epoll) need to be told about closing a file
1096	descriptor (either by calling C<close> explicitly or by any other means,	1363	descriptor (either due to calling C<close> explicitly or any other means,
1097	such as C<dup>). The reason is that you register interest in some file	1364	such as C<dup2>). The reason is that you register interest in some file
1098	descriptor, but when it goes away, the operating system will silently drop	1365	descriptor, but when it goes away, the operating system will silently drop
1099	this interest. If another file descriptor with the same number then is	1366	this interest. If another file descriptor with the same number then is
1100	registered with libev, there is no efficient way to see that this is, in	1367	registered with libev, there is no efficient way to see that this is, in
1101	fact, a different file descriptor.	1368	fact, a different file descriptor.
1102		1369
…		…
1133	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1400	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
1134	C<EVBACKEND_POLL>.	1401	C<EVBACKEND_POLL>.
1135		1402
1136	=head3 The special problem of SIGPIPE	1403	=head3 The special problem of SIGPIPE
1137		1404
1138	While not really specific to libev, it is easy to forget about SIGPIPE:	1405	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
1139	when reading from a pipe whose other end has been closed, your program	1406	when writing to a pipe whose other end has been closed, your program gets
1140	gets send a SIGPIPE, which, by default, aborts your program. For most	1407	sent a SIGPIPE, which, by default, aborts your program. For most programs
1141	programs this is sensible behaviour, for daemons, this is usually	1408	this is sensible behaviour, for daemons, this is usually undesirable.
1142	undesirable.
1143		1409
1144	So when you encounter spurious, unexplained daemon exits, make sure you	1410	So when you encounter spurious, unexplained daemon exits, make sure you
1145	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon	1411	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
1146	somewhere, as that would have given you a big clue).	1412	somewhere, as that would have given you a big clue).
1147		1413
…		…
1153	=item ev_io_init (ev_io *, callback, int fd, int events)	1419	=item ev_io_init (ev_io *, callback, int fd, int events)
1154		1420
1155	=item ev_io_set (ev_io *, int fd, int events)	1421	=item ev_io_set (ev_io *, int fd, int events)
1156		1422
1157	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1423	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
1158	receive events for and events is either C<EV_READ>, C<EV_WRITE> or	1424	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
1159	C<EV_READ | EV_WRITE> to receive the given events.	1425	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
1160		1426
1161	=item int fd [read-only]	1427	=item int fd [read-only]
1162		1428
1163	The file descriptor being watched.	1429	The file descriptor being watched.
1164		1430
…		…
1173	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1439	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
1174	readable, but only once. Since it is likely line-buffered, you could	1440	readable, but only once. Since it is likely line-buffered, you could
1175	attempt to read a whole line in the callback.	1441	attempt to read a whole line in the callback.
1176		1442
1177	static void	1443	static void
1178	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1444	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
1179	{	1445	{
1180	ev_io_stop (loop, w);	1446	ev_io_stop (loop, w);
1181	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1447	.. read from stdin here (or from w->fd) and handle any I/O errors
1182	}	1448	}
1183		1449
1184	...	1450	...
1185	struct ev_loop *loop = ev_default_init (0);	1451	struct ev_loop *loop = ev_default_init (0);
1186	struct ev_io stdin_readable;	1452	ev_io stdin_readable;
1187	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1453	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1188	ev_io_start (loop, &stdin_readable);	1454	ev_io_start (loop, &stdin_readable);
1189	ev_loop (loop, 0);	1455	ev_loop (loop, 0);
1190		1456
1191		1457
…		…
1194	Timer watchers are simple relative timers that generate an event after a	1460	Timer watchers are simple relative timers that generate an event after a
1195	given time, and optionally repeating in regular intervals after that.	1461	given time, and optionally repeating in regular intervals after that.
1196		1462
1197	The timers are based on real time, that is, if you register an event that	1463	The timers are based on real time, that is, if you register an event that
1198	times out after an hour and you reset your system clock to January last	1464	times out after an hour and you reset your system clock to January last
1199	year, it will still time out after (roughly) and hour. "Roughly" because	1465	year, it will still time out after (roughly) one hour. "Roughly" because
1200	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1466	detecting time jumps is hard, and some inaccuracies are unavoidable (the
1201	monotonic clock option helps a lot here).	1467	monotonic clock option helps a lot here).
		1468
		1469	The callback is guaranteed to be invoked only I<after> its timeout has
		1470	passed (not I<at>, so on systems with very low-resolution clocks this
		1471	might introduce a small delay). If multiple timers become ready during the
		1472	same loop iteration then the ones with earlier time-out values are invoked
		1473	before ones with later time-out values (but this is no longer true when a
		1474	callback calls C<ev_loop> recursively).
		1475
		1476	=head3 Be smart about timeouts
		1477
		1478	Many real-world problems involve some kind of timeout, usually for error
		1479	recovery. A typical example is an HTTP request - if the other side hangs,
		1480	you want to raise some error after a while.
		1481
		1482	What follows are some ways to handle this problem, from obvious and
		1483	inefficient to smart and efficient.
		1484
		1485	In the following, a 60 second activity timeout is assumed - a timeout that
		1486	gets reset to 60 seconds each time there is activity (e.g. each time some
		1487	data or other life sign was received).
		1488
		1489	=over 4
		1490
		1491	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1492
		1493	This is the most obvious, but not the most simple way: In the beginning,
		1494	start the watcher:
		1495
		1496	ev_timer_init (timer, callback, 60., 0.);
		1497	ev_timer_start (loop, timer);
		1498
		1499	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1500	and start it again:
		1501
		1502	ev_timer_stop (loop, timer);
		1503	ev_timer_set (timer, 60., 0.);
		1504	ev_timer_start (loop, timer);
		1505
		1506	This is relatively simple to implement, but means that each time there is
		1507	some activity, libev will first have to remove the timer from its internal
		1508	data structure and then add it again. Libev tries to be fast, but it's
		1509	still not a constant-time operation.
		1510
		1511	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1512
		1513	This is the easiest way, and involves using C<ev_timer_again> instead of
		1514	C<ev_timer_start>.
		1515
		1516	To implement this, configure an C<ev_timer> with a C<repeat> value
		1517	of C<60> and then call C<ev_timer_again> at start and each time you
		1518	successfully read or write some data. If you go into an idle state where
		1519	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1520	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1521
		1522	That means you can ignore both the C<ev_timer_start> function and the
		1523	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1524	member and C<ev_timer_again>.
		1525
		1526	At start:
		1527
		1528	ev_timer_init (timer, callback);
		1529	timer->repeat = 60.;
		1530	ev_timer_again (loop, timer);
		1531
		1532	Each time there is some activity:
		1533
		1534	ev_timer_again (loop, timer);
		1535
		1536	It is even possible to change the time-out on the fly, regardless of
		1537	whether the watcher is active or not:
		1538
		1539	timer->repeat = 30.;
		1540	ev_timer_again (loop, timer);
		1541
		1542	This is slightly more efficient then stopping/starting the timer each time
		1543	you want to modify its timeout value, as libev does not have to completely
		1544	remove and re-insert the timer from/into its internal data structure.
		1545
		1546	It is, however, even simpler than the "obvious" way to do it.
		1547
		1548	=item 3. Let the timer time out, but then re-arm it as required.
		1549
		1550	This method is more tricky, but usually most efficient: Most timeouts are
		1551	relatively long compared to the intervals between other activity - in
		1552	our example, within 60 seconds, there are usually many I/O events with
		1553	associated activity resets.
		1554
		1555	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1556	but remember the time of last activity, and check for a real timeout only
		1557	within the callback:
		1558
		1559	ev_tstamp last_activity; // time of last activity
		1560
		1561	static void
		1562	callback (EV_P_ ev_timer *w, int revents)
		1563	{
		1564	ev_tstamp now = ev_now (EV_A);
		1565	ev_tstamp timeout = last_activity + 60.;
		1566
		1567	// if last_activity + 60. is older than now, we did time out
		1568	if (timeout < now)
		1569	{
		1570	// timeout occured, take action
		1571	}
		1572	else
		1573	{
		1574	// callback was invoked, but there was some activity, re-arm
		1575	// the watcher to fire in last_activity + 60, which is
		1576	// guaranteed to be in the future, so "again" is positive:
		1577	w->repeat = timeout - now;
		1578	ev_timer_again (EV_A_ w);
		1579	}
		1580	}
		1581
		1582	To summarise the callback: first calculate the real timeout (defined
		1583	as "60 seconds after the last activity"), then check if that time has
		1584	been reached, which means something I<did>, in fact, time out. Otherwise
		1585	the callback was invoked too early (C<timeout> is in the future), so
		1586	re-schedule the timer to fire at that future time, to see if maybe we have
		1587	a timeout then.
		1588
		1589	Note how C<ev_timer_again> is used, taking advantage of the
		1590	C<ev_timer_again> optimisation when the timer is already running.
		1591
		1592	This scheme causes more callback invocations (about one every 60 seconds
		1593	minus half the average time between activity), but virtually no calls to
		1594	libev to change the timeout.
		1595
		1596	To start the timer, simply initialise the watcher and set C<last_activity>
		1597	to the current time (meaning we just have some activity :), then call the
		1598	callback, which will "do the right thing" and start the timer:
		1599
		1600	ev_timer_init (timer, callback);
		1601	last_activity = ev_now (loop);
		1602	callback (loop, timer, EV_TIMEOUT);
		1603
		1604	And when there is some activity, simply store the current time in
		1605	C<last_activity>, no libev calls at all:
		1606
		1607	last_actiivty = ev_now (loop);
		1608
		1609	This technique is slightly more complex, but in most cases where the
		1610	time-out is unlikely to be triggered, much more efficient.
		1611
		1612	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1613	callback :) - just change the timeout and invoke the callback, which will
		1614	fix things for you.
		1615
		1616	=item 4. Wee, just use a double-linked list for your timeouts.
		1617
		1618	If there is not one request, but many thousands (millions...), all
		1619	employing some kind of timeout with the same timeout value, then one can
		1620	do even better:
		1621
		1622	When starting the timeout, calculate the timeout value and put the timeout
		1623	at the I<end> of the list.
		1624
		1625	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1626	the list is expected to fire (for example, using the technique #3).
		1627
		1628	When there is some activity, remove the timer from the list, recalculate
		1629	the timeout, append it to the end of the list again, and make sure to
		1630	update the C<ev_timer> if it was taken from the beginning of the list.
		1631
		1632	This way, one can manage an unlimited number of timeouts in O(1) time for
		1633	starting, stopping and updating the timers, at the expense of a major
		1634	complication, and having to use a constant timeout. The constant timeout
		1635	ensures that the list stays sorted.
		1636
		1637	=back
		1638
		1639	So which method the best?
		1640
		1641	Method #2 is a simple no-brain-required solution that is adequate in most
		1642	situations. Method #3 requires a bit more thinking, but handles many cases
		1643	better, and isn't very complicated either. In most case, choosing either
		1644	one is fine, with #3 being better in typical situations.
		1645
		1646	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1647	rather complicated, but extremely efficient, something that really pays
		1648	off after the first million or so of active timers, i.e. it's usually
		1649	overkill :)
		1650
		1651	=head3 The special problem of time updates
		1652
		1653	Establishing the current time is a costly operation (it usually takes at
		1654	least two system calls): EV therefore updates its idea of the current
		1655	time only before and after C<ev_loop> collects new events, which causes a
		1656	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1657	lots of events in one iteration.
1202		1658
1203	The relative timeouts are calculated relative to the C<ev_now ()>	1659	The relative timeouts are calculated relative to the C<ev_now ()>
1204	time. This is usually the right thing as this timestamp refers to the time	1660	time. This is usually the right thing as this timestamp refers to the time
1205	of the event triggering whatever timeout you are modifying/starting. If	1661	of the event triggering whatever timeout you are modifying/starting. If
1206	you suspect event processing to be delayed and you I<need> to base the timeout	1662	you suspect event processing to be delayed and you I<need> to base the
1207	on the current time, use something like this to adjust for this:	1663	timeout on the current time, use something like this to adjust for this:
1208		1664
1209	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1665	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
1210		1666
1211	The callback is guaranteed to be invoked only after its timeout has passed,	1667	If the event loop is suspended for a long time, you can also force an
1212	but if multiple timers become ready during the same loop iteration then	1668	update of the time returned by C<ev_now ()> by calling C<ev_now_update
1213	order of execution is undefined.	1669	()>.
1214		1670
1215	=head3 Watcher-Specific Functions and Data Members	1671	=head3 Watcher-Specific Functions and Data Members
1216		1672
1217	=over 4	1673	=over 4
1218		1674
…		…
1242	If the timer is started but non-repeating, stop it (as if it timed out).	1698	If the timer is started but non-repeating, stop it (as if it timed out).
1243		1699
1244	If the timer is repeating, either start it if necessary (with the	1700	If the timer is repeating, either start it if necessary (with the
1245	C<repeat> value), or reset the running timer to the C<repeat> value.	1701	C<repeat> value), or reset the running timer to the C<repeat> value.
1246		1702
1247	This sounds a bit complicated, but here is a useful and typical	1703	This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
1248	example: Imagine you have a TCP connection and you want a so-called idle	1704	usage example.
1249	timeout, that is, you want to be called when there have been, say, 60
1250	seconds of inactivity on the socket. The easiest way to do this is to
1251	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
1252	C<ev_timer_again> each time you successfully read or write some data. If
1253	you go into an idle state where you do not expect data to travel on the
1254	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
1255	automatically restart it if need be.
1256
1257	That means you can ignore the C<after> value and C<ev_timer_start>
1258	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
1259
1260	ev_timer_init (timer, callback, 0., 5.);
1261	ev_timer_again (loop, timer);
1262	...
1263	timer->again = 17.;
1264	ev_timer_again (loop, timer);
1265	...
1266	timer->again = 10.;
1267	ev_timer_again (loop, timer);
1268
1269	This is more slightly efficient then stopping/starting the timer each time
1270	you want to modify its timeout value.
1271		1705
1272	=item ev_tstamp repeat [read-write]	1706	=item ev_tstamp repeat [read-write]
1273		1707
1274	The current C<repeat> value. Will be used each time the watcher times out	1708	The current C<repeat> value. Will be used each time the watcher times out
1275	or C<ev_timer_again> is called and determines the next timeout (if any),	1709	or C<ev_timer_again> is called, and determines the next timeout (if any),
1276	which is also when any modifications are taken into account.	1710	which is also when any modifications are taken into account.
1277		1711
1278	=back	1712	=back
1279		1713
1280	=head3 Examples	1714	=head3 Examples
1281		1715
1282	Example: Create a timer that fires after 60 seconds.	1716	Example: Create a timer that fires after 60 seconds.
1283		1717
1284	static void	1718	static void
1285	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1719	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
1286	{	1720	{
1287	.. one minute over, w is actually stopped right here	1721	.. one minute over, w is actually stopped right here
1288	}	1722	}
1289		1723
1290	struct ev_timer mytimer;	1724	ev_timer mytimer;
1291	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1725	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1292	ev_timer_start (loop, &mytimer);	1726	ev_timer_start (loop, &mytimer);
1293		1727
1294	Example: Create a timeout timer that times out after 10 seconds of	1728	Example: Create a timeout timer that times out after 10 seconds of
1295	inactivity.	1729	inactivity.
1296		1730
1297	static void	1731	static void
1298	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1732	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
1299	{	1733	{
1300	.. ten seconds without any activity	1734	.. ten seconds without any activity
1301	}	1735	}
1302		1736
1303	struct ev_timer mytimer;	1737	ev_timer mytimer;
1304	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	1738	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1305	ev_timer_again (&mytimer); /* start timer */	1739	ev_timer_again (&mytimer); /* start timer */
1306	ev_loop (loop, 0);	1740	ev_loop (loop, 0);
1307		1741
1308	// and in some piece of code that gets executed on any "activity":	1742	// and in some piece of code that gets executed on any "activity":
…		…
1313	=head2 C<ev_periodic> - to cron or not to cron?	1747	=head2 C<ev_periodic> - to cron or not to cron?
1314		1748
1315	Periodic watchers are also timers of a kind, but they are very versatile	1749	Periodic watchers are also timers of a kind, but they are very versatile
1316	(and unfortunately a bit complex).	1750	(and unfortunately a bit complex).
1317		1751
1318	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1752	Unlike C<ev_timer>, periodic watchers are not based on real time (or
1319	but on wall clock time (absolute time). You can tell a periodic watcher	1753	relative time, the physical time that passes) but on wall clock time
1320	to trigger after some specific point in time. For example, if you tell a	1754	(absolute time, the thing you can read on your calender or clock). The
1321	periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()	1755	difference is that wall clock time can run faster or slower than real
1322	+ 10.>, that is, an absolute time not a delay) and then reset your system	1756	time, and time jumps are not uncommon (e.g. when you adjust your
1323	clock to January of the previous year, then it will take more than year	1757	wrist-watch).
1324	to trigger the event (unlike an C<ev_timer>, which would still trigger
1325	roughly 10 seconds later as it uses a relative timeout).
1326		1758
		1759	You can tell a periodic watcher to trigger after some specific point
		1760	in time: for example, if you tell a periodic watcher to trigger "in 10
		1761	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		1762	not a delay) and then reset your system clock to January of the previous
		1763	year, then it will take a year or more to trigger the event (unlike an
		1764	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		1765	it, as it uses a relative timeout).
		1766
1327	C<ev_periodic>s can also be used to implement vastly more complex timers,	1767	C<ev_periodic> watchers can also be used to implement vastly more complex
1328	such as triggering an event on each "midnight, local time", or other	1768	timers, such as triggering an event on each "midnight, local time", or
1329	complicated, rules.	1769	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		1770	those cannot react to time jumps.
1330		1771
1331	As with timers, the callback is guaranteed to be invoked only when the	1772	As with timers, the callback is guaranteed to be invoked only when the
1332	time (C<at>) has passed, but if multiple periodic timers become ready	1773	point in time where it is supposed to trigger has passed. If multiple
1333	during the same loop iteration then order of execution is undefined.	1774	timers become ready during the same loop iteration then the ones with
		1775	earlier time-out values are invoked before ones with later time-out values
		1776	(but this is no longer true when a callback calls C<ev_loop> recursively).
1334		1777
1335	=head3 Watcher-Specific Functions and Data Members	1778	=head3 Watcher-Specific Functions and Data Members
1336		1779
1337	=over 4	1780	=over 4
1338		1781
1339	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1782	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1340		1783
1341	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1784	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1342		1785
1343	Lots of arguments, lets sort it out... There are basically three modes of	1786	Lots of arguments, let's sort it out... There are basically three modes of
1344	operation, and we will explain them from simplest to complex:	1787	operation, and we will explain them from simplest to most complex:
1345		1788
1346	=over 4	1789	=over 4
1347		1790
1348	=item * absolute timer (at = time, interval = reschedule_cb = 0)	1791	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
1349		1792
1350	In this configuration the watcher triggers an event after the wall clock	1793	In this configuration the watcher triggers an event after the wall clock
1351	time C<at> has passed and doesn't repeat. It will not adjust when a time	1794	time C<offset> has passed. It will not repeat and will not adjust when a
1352	jump occurs, that is, if it is to be run at January 1st 2011 then it will	1795	time jump occurs, that is, if it is to be run at January 1st 2011 then it
1353	run when the system time reaches or surpasses this time.	1796	will be stopped and invoked when the system clock reaches or surpasses
		1797	this point in time.
1354		1798
1355	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	1799	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
1356		1800
1357	In this mode the watcher will always be scheduled to time out at the next	1801	In this mode the watcher will always be scheduled to time out at the next
1358	C<at + N * interval> time (for some integer N, which can also be negative)	1802	C<offset + N * interval> time (for some integer N, which can also be
1359	and then repeat, regardless of any time jumps.	1803	negative) and then repeat, regardless of any time jumps. The C<offset>
		1804	argument is merely an offset into the C<interval> periods.
1360		1805
1361	This can be used to create timers that do not drift with respect to system	1806	This can be used to create timers that do not drift with respect to the
1362	time, for example, here is a C<ev_periodic> that triggers each hour, on	1807	system clock, for example, here is an C<ev_periodic> that triggers each
1363	the hour:	1808	hour, on the hour (with respect to UTC):
1364		1809
1365	ev_periodic_set (&periodic, 0., 3600., 0);	1810	ev_periodic_set (&periodic, 0., 3600., 0);
1366		1811
1367	This doesn't mean there will always be 3600 seconds in between triggers,	1812	This doesn't mean there will always be 3600 seconds in between triggers,
1368	but only that the callback will be called when the system time shows a	1813	but only that the callback will be called when the system time shows a
1369	full hour (UTC), or more correctly, when the system time is evenly divisible	1814	full hour (UTC), or more correctly, when the system time is evenly divisible
1370	by 3600.	1815	by 3600.
1371		1816
1372	Another way to think about it (for the mathematically inclined) is that	1817	Another way to think about it (for the mathematically inclined) is that
1373	C<ev_periodic> will try to run the callback in this mode at the next possible	1818	C<ev_periodic> will try to run the callback in this mode at the next possible
1374	time where C<time = at (mod interval)>, regardless of any time jumps.	1819	time where C<time = offset (mod interval)>, regardless of any time jumps.
1375		1820
1376	For numerical stability it is preferable that the C<at> value is near	1821	For numerical stability it is preferable that the C<offset> value is near
1377	C<ev_now ()> (the current time), but there is no range requirement for	1822	C<ev_now ()> (the current time), but there is no range requirement for
1378	this value, and in fact is often specified as zero.	1823	this value, and in fact is often specified as zero.
1379		1824
1380	Note also that there is an upper limit to how often a timer can fire (CPU	1825	Note also that there is an upper limit to how often a timer can fire (CPU
1381	speed for example), so if C<interval> is very small then timing stability	1826	speed for example), so if C<interval> is very small then timing stability
1382	will of course deteriorate. Libev itself tries to be exact to be about one	1827	will of course deteriorate. Libev itself tries to be exact to be about one
1383	millisecond (if the OS supports it and the machine is fast enough).	1828	millisecond (if the OS supports it and the machine is fast enough).
1384		1829
1385	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)	1830	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
1386		1831
1387	In this mode the values for C<interval> and C<at> are both being	1832	In this mode the values for C<interval> and C<offset> are both being
1388	ignored. Instead, each time the periodic watcher gets scheduled, the	1833	ignored. Instead, each time the periodic watcher gets scheduled, the
1389	reschedule callback will be called with the watcher as first, and the	1834	reschedule callback will be called with the watcher as first, and the
1390	current time as second argument.	1835	current time as second argument.
1391		1836
1392	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1837	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
1393	ever, or make ANY event loop modifications whatsoever>.	1838	or make ANY other event loop modifications whatsoever, unless explicitly
		1839	allowed by documentation here>.
1394		1840
1395	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop	1841	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
1396	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the	1842	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
1397	only event loop modification you are allowed to do).	1843	only event loop modification you are allowed to do).
1398		1844
1399	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic	1845	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
1400	*w, ev_tstamp now)>, e.g.:	1846	*w, ev_tstamp now)>, e.g.:
1401		1847
		1848	static ev_tstamp
1402	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1849	my_rescheduler (ev_periodic *w, ev_tstamp now)
1403	{	1850	{
1404	return now + 60.;	1851	return now + 60.;
1405	}	1852	}
1406		1853
1407	It must return the next time to trigger, based on the passed time value	1854	It must return the next time to trigger, based on the passed time value
…		…
1427	a different time than the last time it was called (e.g. in a crond like	1874	a different time than the last time it was called (e.g. in a crond like
1428	program when the crontabs have changed).	1875	program when the crontabs have changed).
1429		1876
1430	=item ev_tstamp ev_periodic_at (ev_periodic *)	1877	=item ev_tstamp ev_periodic_at (ev_periodic *)
1431		1878
1432	When active, returns the absolute time that the watcher is supposed to	1879	When active, returns the absolute time that the watcher is supposed
1433	trigger next.	1880	to trigger next. This is not the same as the C<offset> argument to
		1881	C<ev_periodic_set>, but indeed works even in interval and manual
		1882	rescheduling modes.
1434		1883
1435	=item ev_tstamp offset [read-write]	1884	=item ev_tstamp offset [read-write]
1436		1885
1437	When repeating, this contains the offset value, otherwise this is the	1886	When repeating, this contains the offset value, otherwise this is the
1438	absolute point in time (the C<at> value passed to C<ev_periodic_set>).	1887	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		1888	although libev might modify this value for better numerical stability).
1439		1889
1440	Can be modified any time, but changes only take effect when the periodic	1890	Can be modified any time, but changes only take effect when the periodic
1441	timer fires or C<ev_periodic_again> is being called.	1891	timer fires or C<ev_periodic_again> is being called.
1442		1892
1443	=item ev_tstamp interval [read-write]	1893	=item ev_tstamp interval [read-write]
1444		1894
1445	The current interval value. Can be modified any time, but changes only	1895	The current interval value. Can be modified any time, but changes only
1446	take effect when the periodic timer fires or C<ev_periodic_again> is being	1896	take effect when the periodic timer fires or C<ev_periodic_again> is being
1447	called.	1897	called.
1448		1898
1449	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]	1899	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
1450		1900
1451	The current reschedule callback, or C<0>, if this functionality is	1901	The current reschedule callback, or C<0>, if this functionality is
1452	switched off. Can be changed any time, but changes only take effect when	1902	switched off. Can be changed any time, but changes only take effect when
1453	the periodic timer fires or C<ev_periodic_again> is being called.	1903	the periodic timer fires or C<ev_periodic_again> is being called.
1454		1904
1455	=back	1905	=back
1456		1906
1457	=head3 Examples	1907	=head3 Examples
1458		1908
1459	Example: Call a callback every hour, or, more precisely, whenever the	1909	Example: Call a callback every hour, or, more precisely, whenever the
1460	system clock is divisible by 3600. The callback invocation times have	1910	system time is divisible by 3600. The callback invocation times have
1461	potentially a lot of jitter, but good long-term stability.	1911	potentially a lot of jitter, but good long-term stability.
1462		1912
1463	static void	1913	static void
1464	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1914	clock_cb (struct ev_loop *loop, ev_io *w, int revents)
1465	{	1915	{
1466	... its now a full hour (UTC, or TAI or whatever your clock follows)	1916	... its now a full hour (UTC, or TAI or whatever your clock follows)
1467	}	1917	}
1468		1918
1469	struct ev_periodic hourly_tick;	1919	ev_periodic hourly_tick;
1470	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1920	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1471	ev_periodic_start (loop, &hourly_tick);	1921	ev_periodic_start (loop, &hourly_tick);
1472		1922
1473	Example: The same as above, but use a reschedule callback to do it:	1923	Example: The same as above, but use a reschedule callback to do it:
1474		1924
1475	#include <math.h>	1925	#include <math.h>
1476		1926
1477	static ev_tstamp	1927	static ev_tstamp
1478	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1928	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
1479	{	1929	{
1480	return fmod (now, 3600.) + 3600.;	1930	return now + (3600. - fmod (now, 3600.));
1481	}	1931	}
1482		1932
1483	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1933	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1484		1934
1485	Example: Call a callback every hour, starting now:	1935	Example: Call a callback every hour, starting now:
1486		1936
1487	struct ev_periodic hourly_tick;	1937	ev_periodic hourly_tick;
1488	ev_periodic_init (&hourly_tick, clock_cb,	1938	ev_periodic_init (&hourly_tick, clock_cb,
1489	fmod (ev_now (loop), 3600.), 3600., 0);	1939	fmod (ev_now (loop), 3600.), 3600., 0);
1490	ev_periodic_start (loop, &hourly_tick);	1940	ev_periodic_start (loop, &hourly_tick);
1491		1941
1492		1942
…		…
1495	Signal watchers will trigger an event when the process receives a specific	1945	Signal watchers will trigger an event when the process receives a specific
1496	signal one or more times. Even though signals are very asynchronous, libev	1946	signal one or more times. Even though signals are very asynchronous, libev
1497	will try it's best to deliver signals synchronously, i.e. as part of the	1947	will try it's best to deliver signals synchronously, i.e. as part of the
1498	normal event processing, like any other event.	1948	normal event processing, like any other event.
1499		1949
		1950	If you want signals asynchronously, just use C<sigaction> as you would
		1951	do without libev and forget about sharing the signal. You can even use
		1952	C<ev_async> from a signal handler to synchronously wake up an event loop.
		1953
1500	You can configure as many watchers as you like per signal. Only when the	1954	You can configure as many watchers as you like per signal. Only when the
1501	first watcher gets started will libev actually register a signal watcher	1955	first watcher gets started will libev actually register a signal handler
1502	with the kernel (thus it coexists with your own signal handlers as long	1956	with the kernel (thus it coexists with your own signal handlers as long as
1503	as you don't register any with libev). Similarly, when the last signal	1957	you don't register any with libev for the same signal). Similarly, when
1504	watcher for a signal is stopped libev will reset the signal handler to	1958	the last signal watcher for a signal is stopped, libev will reset the
1505	SIG_DFL (regardless of what it was set to before).	1959	signal handler to SIG_DFL (regardless of what it was set to before).
1506		1960
1507	If possible and supported, libev will install its handlers with	1961	If possible and supported, libev will install its handlers with
1508	C<SA_RESTART> behaviour enabled, so system calls should not be unduly	1962	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
1509	interrupted. If you have a problem with system calls getting interrupted by	1963	interrupted. If you have a problem with system calls getting interrupted by
1510	signals you can block all signals in an C<ev_check> watcher and unblock	1964	signals you can block all signals in an C<ev_check> watcher and unblock
…		…
1527		1981
1528	=back	1982	=back
1529		1983
1530	=head3 Examples	1984	=head3 Examples
1531		1985
1532	Example: Try to exit cleanly on SIGINT and SIGTERM.	1986	Example: Try to exit cleanly on SIGINT.
1533		1987
1534	static void	1988	static void
1535	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1989	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
1536	{	1990	{
1537	ev_unloop (loop, EVUNLOOP_ALL);	1991	ev_unloop (loop, EVUNLOOP_ALL);
1538	}	1992	}
1539		1993
1540	struct ev_signal signal_watcher;	1994	ev_signal signal_watcher;
1541	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1995	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1542	ev_signal_start (loop, &sigint_cb);	1996	ev_signal_start (loop, &signal_watcher);
1543		1997
1544		1998
1545	=head2 C<ev_child> - watch out for process status changes	1999	=head2 C<ev_child> - watch out for process status changes
1546		2000
1547	Child watchers trigger when your process receives a SIGCHLD in response to	2001	Child watchers trigger when your process receives a SIGCHLD in response to
1548	some child status changes (most typically when a child of yours dies). It	2002	some child status changes (most typically when a child of yours dies or
1549	is permissible to install a child watcher I<after> the child has been	2003	exits). It is permissible to install a child watcher I<after> the child
1550	forked (which implies it might have already exited), as long as the event	2004	has been forked (which implies it might have already exited), as long
1551	loop isn't entered (or is continued from a watcher).	2005	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2006	forking and then immediately registering a watcher for the child is fine,
		2007	but forking and registering a watcher a few event loop iterations later is
		2008	not.
1552		2009
1553	Only the default event loop is capable of handling signals, and therefore	2010	Only the default event loop is capable of handling signals, and therefore
1554	you can only register child watchers in the default event loop.	2011	you can only register child watchers in the default event loop.
1555		2012
1556	=head3 Process Interaction	2013	=head3 Process Interaction
…		…
1617	its completion.	2074	its completion.
1618		2075
1619	ev_child cw;	2076	ev_child cw;
1620		2077
1621	static void	2078	static void
1622	child_cb (EV_P_ struct ev_child *w, int revents)	2079	child_cb (EV_P_ ev_child *w, int revents)
1623	{	2080	{
1624	ev_child_stop (EV_A_ w);	2081	ev_child_stop (EV_A_ w);
1625	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);	2082	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1626	}	2083	}
1627		2084
…		…
1642		2099
1643		2100
1644	=head2 C<ev_stat> - did the file attributes just change?	2101	=head2 C<ev_stat> - did the file attributes just change?
1645		2102
1646	This watches a file system path for attribute changes. That is, it calls	2103	This watches a file system path for attribute changes. That is, it calls
1647	C<stat> regularly (or when the OS says it changed) and sees if it changed	2104	C<stat> on that path in regular intervals (or when the OS says it changed)
1648	compared to the last time, invoking the callback if it did.	2105	and sees if it changed compared to the last time, invoking the callback if
		2106	it did.
1649		2107
1650	The path does not need to exist: changing from "path exists" to "path does	2108	The path does not need to exist: changing from "path exists" to "path does
1651	not exist" is a status change like any other. The condition "path does	2109	not exist" is a status change like any other. The condition "path does not
1652	not exist" is signified by the C<st_nlink> field being zero (which is	2110	exist" (or more correctly "path cannot be stat'ed") is signified by the
1653	otherwise always forced to be at least one) and all the other fields of	2111	C<st_nlink> field being zero (which is otherwise always forced to be at
1654	the stat buffer having unspecified contents.	2112	least one) and all the other fields of the stat buffer having unspecified
		2113	contents.
1655		2114
1656	The path I<should> be absolute and I<must not> end in a slash. If it is	2115	The path I<must not> end in a slash or contain special components such as
		2116	C<.> or C<..>. The path I<should> be absolute: If it is relative and
1657	relative and your working directory changes, the behaviour is undefined.	2117	your working directory changes, then the behaviour is undefined.
1658		2118
1659	Since there is no standard to do this, the portable implementation simply	2119	Since there is no portable change notification interface available, the
1660	calls C<stat (2)> regularly on the path to see if it changed somehow. You	2120	portable implementation simply calls C<stat(2)> regularly on the path
1661	can specify a recommended polling interval for this case. If you specify	2121	to see if it changed somehow. You can specify a recommended polling
1662	a polling interval of C<0> (highly recommended!) then a I<suitable,	2122	interval for this case. If you specify a polling interval of C<0> (highly
1663	unspecified default> value will be used (which you can expect to be around	2123	recommended!) then a I<suitable, unspecified default> value will be used
1664	five seconds, although this might change dynamically). Libev will also	2124	(which you can expect to be around five seconds, although this might
1665	impose a minimum interval which is currently around C<0.1>, but thats	2125	change dynamically). Libev will also impose a minimum interval which is
1666	usually overkill.	2126	currently around C<0.1>, but that's usually overkill.
1667		2127
1668	This watcher type is not meant for massive numbers of stat watchers,	2128	This watcher type is not meant for massive numbers of stat watchers,
1669	as even with OS-supported change notifications, this can be	2129	as even with OS-supported change notifications, this can be
1670	resource-intensive.	2130	resource-intensive.
1671		2131
1672	At the time of this writing, only the Linux inotify interface is	2132	At the time of this writing, the only OS-specific interface implemented
1673	implemented (implementing kqueue support is left as an exercise for the	2133	is the Linux inotify interface (implementing kqueue support is left as an
1674	reader, note, however, that the author sees no way of implementing ev_stat	2134	exercise for the reader. Note, however, that the author sees no way of
1675	semantics with kqueue). Inotify will be used to give hints only and should	2135	implementing C<ev_stat> semantics with kqueue, except as a hint).
1676	not change the semantics of C<ev_stat> watchers, which means that libev
1677	sometimes needs to fall back to regular polling again even with inotify,
1678	but changes are usually detected immediately, and if the file exists there
1679	will be no polling.
1680		2136
1681	=head3 ABI Issues (Largefile Support)	2137	=head3 ABI Issues (Largefile Support)
1682		2138
1683	Libev by default (unless the user overrides this) uses the default	2139	Libev by default (unless the user overrides this) uses the default
1684	compilation environment, which means that on systems with large file	2140	compilation environment, which means that on systems with large file
1685	support disabled by default, you get the 32 bit version of the stat	2141	support disabled by default, you get the 32 bit version of the stat
1686	structure. When using the library from programs that change the ABI to	2142	structure. When using the library from programs that change the ABI to
1687	use 64 bit file offsets the programs will fail. In that case you have to	2143	use 64 bit file offsets the programs will fail. In that case you have to
1688	compile libev with the same flags to get binary compatibility. This is	2144	compile libev with the same flags to get binary compatibility. This is
1689	obviously the case with any flags that change the ABI, but the problem is	2145	obviously the case with any flags that change the ABI, but the problem is
1690	most noticeably disabled with ev_stat and large file support.	2146	most noticeably displayed with ev_stat and large file support.
1691		2147
1692	The solution for this is to lobby your distribution maker to make large	2148	The solution for this is to lobby your distribution maker to make large
1693	file interfaces available by default (as e.g. FreeBSD does) and not	2149	file interfaces available by default (as e.g. FreeBSD does) and not
1694	optional. Libev cannot simply switch on large file support because it has	2150	optional. Libev cannot simply switch on large file support because it has
1695	to exchange stat structures with application programs compiled using the	2151	to exchange stat structures with application programs compiled using the
1696	default compilation environment.	2152	default compilation environment.
1697		2153
1698	=head3 Inotify	2154	=head3 Inotify and Kqueue
1699		2155
1700	When C<inotify (7)> support has been compiled into libev (generally only	2156	When C<inotify (7)> support has been compiled into libev and present at
1701	available on Linux) and present at runtime, it will be used to speed up	2157	runtime, it will be used to speed up change detection where possible. The
1702	change detection where possible. The inotify descriptor will be created lazily	2158	inotify descriptor will be created lazily when the first C<ev_stat>
1703	when the first C<ev_stat> watcher is being started.	2159	watcher is being started.
1704		2160
1705	Inotify presence does not change the semantics of C<ev_stat> watchers	2161	Inotify presence does not change the semantics of C<ev_stat> watchers
1706	except that changes might be detected earlier, and in some cases, to avoid	2162	except that changes might be detected earlier, and in some cases, to avoid
1707	making regular C<stat> calls. Even in the presence of inotify support	2163	making regular C<stat> calls. Even in the presence of inotify support
1708	there are many cases where libev has to resort to regular C<stat> polling.	2164	there are many cases where libev has to resort to regular C<stat> polling,
		2165	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2166	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2167	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2168	xfs are fully working) libev usually gets away without polling.
1709		2169
1710	(There is no support for kqueue, as apparently it cannot be used to	2170	There is no support for kqueue, as apparently it cannot be used to
1711	implement this functionality, due to the requirement of having a file	2171	implement this functionality, due to the requirement of having a file
1712	descriptor open on the object at all times).	2172	descriptor open on the object at all times, and detecting renames, unlinks
		2173	etc. is difficult.
		2174
		2175	=head3 C<stat ()> is a synchronous operation
		2176
		2177	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2178	the process. The exception are C<ev_stat> watchers - those call C<stat
		2179	()>, which is a synchronous operation.
		2180
		2181	For local paths, this usually doesn't matter: unless the system is very
		2182	busy or the intervals between stat's are large, a stat call will be fast,
		2183	as the path data is usually in memory already (except when starting the
		2184	watcher).
		2185
		2186	For networked file systems, calling C<stat ()> can block an indefinite
		2187	time due to network issues, and even under good conditions, a stat call
		2188	often takes multiple milliseconds.
		2189
		2190	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2191	paths, although this is fully supported by libev.
1713		2192
1714	=head3 The special problem of stat time resolution	2193	=head3 The special problem of stat time resolution
1715		2194
1716	The C<stat ()> system call only supports full-second resolution portably, and	2195	The C<stat ()> system call only supports full-second resolution portably,
1717	even on systems where the resolution is higher, many file systems still	2196	and even on systems where the resolution is higher, most file systems
1718	only support whole seconds.	2197	still only support whole seconds.
1719		2198
1720	That means that, if the time is the only thing that changes, you can	2199	That means that, if the time is the only thing that changes, you can
1721	easily miss updates: on the first update, C<ev_stat> detects a change and	2200	easily miss updates: on the first update, C<ev_stat> detects a change and
1722	calls your callback, which does something. When there is another update	2201	calls your callback, which does something. When there is another update
1723	within the same second, C<ev_stat> will be unable to detect it as the stat	2202	within the same second, C<ev_stat> will be unable to detect unless the
1724	data does not change.	2203	stat data does change in other ways (e.g. file size).
1725		2204
1726	The solution to this is to delay acting on a change for slightly more	2205	The solution to this is to delay acting on a change for slightly more
1727	than a second (or till slightly after the next full second boundary), using	2206	than a second (or till slightly after the next full second boundary), using
1728	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);	2207	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
1729	ev_timer_again (loop, w)>).	2208	ev_timer_again (loop, w)>).
…		…
1749	C<path>. The C<interval> is a hint on how quickly a change is expected to	2228	C<path>. The C<interval> is a hint on how quickly a change is expected to
1750	be detected and should normally be specified as C<0> to let libev choose	2229	be detected and should normally be specified as C<0> to let libev choose
1751	a suitable value. The memory pointed to by C<path> must point to the same	2230	a suitable value. The memory pointed to by C<path> must point to the same
1752	path for as long as the watcher is active.	2231	path for as long as the watcher is active.
1753		2232
1754	The callback will receive C<EV_STAT> when a change was detected, relative	2233	The callback will receive an C<EV_STAT> event when a change was detected,
1755	to the attributes at the time the watcher was started (or the last change	2234	relative to the attributes at the time the watcher was started (or the
1756	was detected).	2235	last change was detected).
1757		2236
1758	=item ev_stat_stat (loop, ev_stat *)	2237	=item ev_stat_stat (loop, ev_stat *)
1759		2238
1760	Updates the stat buffer immediately with new values. If you change the	2239	Updates the stat buffer immediately with new values. If you change the
1761	watched path in your callback, you could call this function to avoid	2240	watched path in your callback, you could call this function to avoid
…		…
1844		2323
1845		2324
1846	=head2 C<ev_idle> - when you've got nothing better to do...	2325	=head2 C<ev_idle> - when you've got nothing better to do...
1847		2326
1848	Idle watchers trigger events when no other events of the same or higher	2327	Idle watchers trigger events when no other events of the same or higher
1849	priority are pending (prepare, check and other idle watchers do not	2328	priority are pending (prepare, check and other idle watchers do not count
1850	count).	2329	as receiving "events").
1851		2330
1852	That is, as long as your process is busy handling sockets or timeouts	2331	That is, as long as your process is busy handling sockets or timeouts
1853	(or even signals, imagine) of the same or higher priority it will not be	2332	(or even signals, imagine) of the same or higher priority it will not be
1854	triggered. But when your process is idle (or only lower-priority watchers	2333	triggered. But when your process is idle (or only lower-priority watchers
1855	are pending), the idle watchers are being called once per event loop	2334	are pending), the idle watchers are being called once per event loop
…		…
1866		2345
1867	=head3 Watcher-Specific Functions and Data Members	2346	=head3 Watcher-Specific Functions and Data Members
1868		2347
1869	=over 4	2348	=over 4
1870		2349
1871	=item ev_idle_init (ev_signal *, callback)	2350	=item ev_idle_init (ev_idle *, callback)
1872		2351
1873	Initialises and configures the idle watcher - it has no parameters of any	2352	Initialises and configures the idle watcher - it has no parameters of any
1874	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2353	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1875	believe me.	2354	believe me.
1876		2355
…		…
1880		2359
1881	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	2360	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1882	callback, free it. Also, use no error checking, as usual.	2361	callback, free it. Also, use no error checking, as usual.
1883		2362
1884	static void	2363	static void
1885	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	2364	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
1886	{	2365	{
1887	free (w);	2366	free (w);
1888	// now do something you wanted to do when the program has	2367	// now do something you wanted to do when the program has
1889	// no longer anything immediate to do.	2368	// no longer anything immediate to do.
1890	}	2369	}
1891		2370
1892	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	2371	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
1893	ev_idle_init (idle_watcher, idle_cb);	2372	ev_idle_init (idle_watcher, idle_cb);
1894	ev_idle_start (loop, idle_cb);	2373	ev_idle_start (loop, idle_cb);
1895		2374
1896		2375
1897	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2376	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1898		2377
1899	Prepare and check watchers are usually (but not always) used in tandem:	2378	Prepare and check watchers are usually (but not always) used in pairs:
1900	prepare watchers get invoked before the process blocks and check watchers	2379	prepare watchers get invoked before the process blocks and check watchers
1901	afterwards.	2380	afterwards.
1902		2381
1903	You I<must not> call C<ev_loop> or similar functions that enter	2382	You I<must not> call C<ev_loop> or similar functions that enter
1904	the current event loop from either C<ev_prepare> or C<ev_check>	2383	the current event loop from either C<ev_prepare> or C<ev_check>
…		…
1907	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,	2386	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
1908	C<ev_check> so if you have one watcher of each kind they will always be	2387	C<ev_check> so if you have one watcher of each kind they will always be
1909	called in pairs bracketing the blocking call.	2388	called in pairs bracketing the blocking call.
1910		2389
1911	Their main purpose is to integrate other event mechanisms into libev and	2390	Their main purpose is to integrate other event mechanisms into libev and
1912	their use is somewhat advanced. This could be used, for example, to track	2391	their use is somewhat advanced. They could be used, for example, to track
1913	variable changes, implement your own watchers, integrate net-snmp or a	2392	variable changes, implement your own watchers, integrate net-snmp or a
1914	coroutine library and lots more. They are also occasionally useful if	2393	coroutine library and lots more. They are also occasionally useful if
1915	you cache some data and want to flush it before blocking (for example,	2394	you cache some data and want to flush it before blocking (for example,
1916	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>	2395	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1917	watcher).	2396	watcher).
1918		2397
1919	This is done by examining in each prepare call which file descriptors need	2398	This is done by examining in each prepare call which file descriptors
1920	to be watched by the other library, registering C<ev_io> watchers for	2399	need to be watched by the other library, registering C<ev_io> watchers
1921	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2400	for them and starting an C<ev_timer> watcher for any timeouts (many
1922	provide just this functionality). Then, in the check watcher you check for	2401	libraries provide exactly this functionality). Then, in the check watcher,
1923	any events that occurred (by checking the pending status of all watchers	2402	you check for any events that occurred (by checking the pending status
1924	and stopping them) and call back into the library. The I/O and timer	2403	of all watchers and stopping them) and call back into the library. The
1925	callbacks will never actually be called (but must be valid nevertheless,	2404	I/O and timer callbacks will never actually be called (but must be valid
1926	because you never know, you know?).	2405	nevertheless, because you never know, you know?).
1927		2406
1928	As another example, the Perl Coro module uses these hooks to integrate	2407	As another example, the Perl Coro module uses these hooks to integrate
1929	coroutines into libev programs, by yielding to other active coroutines	2408	coroutines into libev programs, by yielding to other active coroutines
1930	during each prepare and only letting the process block if no coroutines	2409	during each prepare and only letting the process block if no coroutines
1931	are ready to run (it's actually more complicated: it only runs coroutines	2410	are ready to run (it's actually more complicated: it only runs coroutines
…		…
1934	loop from blocking if lower-priority coroutines are active, thus mapping	2413	loop from blocking if lower-priority coroutines are active, thus mapping
1935	low-priority coroutines to idle/background tasks).	2414	low-priority coroutines to idle/background tasks).
1936		2415
1937	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	2416	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
1938	priority, to ensure that they are being run before any other watchers	2417	priority, to ensure that they are being run before any other watchers
		2418	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2419
1939	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	2420	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
1940	too) should not activate ("feed") events into libev. While libev fully	2421	activate ("feed") events into libev. While libev fully supports this, they
1941	supports this, they might get executed before other C<ev_check> watchers	2422	might get executed before other C<ev_check> watchers did their job. As
1942	did their job. As C<ev_check> watchers are often used to embed other	2423	C<ev_check> watchers are often used to embed other (non-libev) event
1943	(non-libev) event loops those other event loops might be in an unusable	2424	loops those other event loops might be in an unusable state until their
1944	state until their C<ev_check> watcher ran (always remind yourself to	2425	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
1945	coexist peacefully with others).	2426	others).
1946		2427
1947	=head3 Watcher-Specific Functions and Data Members	2428	=head3 Watcher-Specific Functions and Data Members
1948		2429
1949	=over 4	2430	=over 4
1950		2431
…		…
1952		2433
1953	=item ev_check_init (ev_check *, callback)	2434	=item ev_check_init (ev_check *, callback)
1954		2435
1955	Initialises and configures the prepare or check watcher - they have no	2436	Initialises and configures the prepare or check watcher - they have no
1956	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2437	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1957	macros, but using them is utterly, utterly and completely pointless.	2438	macros, but using them is utterly, utterly, utterly and completely
		2439	pointless.
1958		2440
1959	=back	2441	=back
1960		2442
1961	=head3 Examples	2443	=head3 Examples
1962		2444
…		…
1975		2457
1976	static ev_io iow [nfd];	2458	static ev_io iow [nfd];
1977	static ev_timer tw;	2459	static ev_timer tw;
1978		2460
1979	static void	2461	static void
1980	io_cb (ev_loop *loop, ev_io *w, int revents)	2462	io_cb (struct ev_loop *loop, ev_io *w, int revents)
1981	{	2463	{
1982	}	2464	}
1983		2465
1984	// create io watchers for each fd and a timer before blocking	2466	// create io watchers for each fd and a timer before blocking
1985	static void	2467	static void
1986	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2468	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
1987	{	2469	{
1988	int timeout = 3600000;	2470	int timeout = 3600000;
1989	struct pollfd fds [nfd];	2471	struct pollfd fds [nfd];
1990	// actual code will need to loop here and realloc etc.	2472	// actual code will need to loop here and realloc etc.
1991	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2473	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
…		…
2006	}	2488	}
2007	}	2489	}
2008		2490
2009	// stop all watchers after blocking	2491	// stop all watchers after blocking
2010	static void	2492	static void
2011	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2493	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
2012	{	2494	{
2013	ev_timer_stop (loop, &tw);	2495	ev_timer_stop (loop, &tw);
2014		2496
2015	for (int i = 0; i < nfd; ++i)	2497	for (int i = 0; i < nfd; ++i)
2016	{	2498	{
…		…
2055	}	2537	}
2056		2538
2057	// do not ever call adns_afterpoll	2539	// do not ever call adns_afterpoll
2058		2540
2059	Method 4: Do not use a prepare or check watcher because the module you	2541	Method 4: Do not use a prepare or check watcher because the module you
2060	want to embed is too inflexible to support it. Instead, you can override	2542	want to embed is not flexible enough to support it. Instead, you can
2061	their poll function. The drawback with this solution is that the main	2543	override their poll function. The drawback with this solution is that the
2062	loop is now no longer controllable by EV. The C<Glib::EV> module does	2544	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
2063	this.	2545	this approach, effectively embedding EV as a client into the horrible
		2546	libglib event loop.
2064		2547
2065	static gint	2548	static gint
2066	event_poll_func (GPollFD *fds, guint nfds, gint timeout)	2549	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
2067	{	2550	{
2068	int got_events = 0;	2551	int got_events = 0;
…		…
2099	prioritise I/O.	2582	prioritise I/O.
2100		2583
2101	As an example for a bug workaround, the kqueue backend might only support	2584	As an example for a bug workaround, the kqueue backend might only support
2102	sockets on some platform, so it is unusable as generic backend, but you	2585	sockets on some platform, so it is unusable as generic backend, but you
2103	still want to make use of it because you have many sockets and it scales	2586	still want to make use of it because you have many sockets and it scales
2104	so nicely. In this case, you would create a kqueue-based loop and embed it	2587	so nicely. In this case, you would create a kqueue-based loop and embed
2105	into your default loop (which might use e.g. poll). Overall operation will	2588	it into your default loop (which might use e.g. poll). Overall operation
2106	be a bit slower because first libev has to poll and then call kevent, but	2589	will be a bit slower because first libev has to call C<poll> and then
2107	at least you can use both at what they are best.	2590	C<kevent>, but at least you can use both mechanisms for what they are
		2591	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
2108		2592
2109	As for prioritising I/O: rarely you have the case where some fds have	2593	As for prioritising I/O: under rare circumstances you have the case where
2110	to be watched and handled very quickly (with low latency), and even	2594	some fds have to be watched and handled very quickly (with low latency),
2111	priorities and idle watchers might have too much overhead. In this case	2595	and even priorities and idle watchers might have too much overhead. In
2112	you would put all the high priority stuff in one loop and all the rest in	2596	this case you would put all the high priority stuff in one loop and all
2113	a second one, and embed the second one in the first.	2597	the rest in a second one, and embed the second one in the first.
2114		2598
2115	As long as the watcher is active, the callback will be invoked every time	2599	As long as the watcher is active, the callback will be invoked every
2116	there might be events pending in the embedded loop. The callback must then	2600	time there might be events pending in the embedded loop. The callback
2117	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke	2601	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
2118	their callbacks (you could also start an idle watcher to give the embedded	2602	sweep and invoke their callbacks (the callback doesn't need to invoke the
2119	loop strictly lower priority for example). You can also set the callback	2603	C<ev_embed_sweep> function directly, it could also start an idle watcher
2120	to C<0>, in which case the embed watcher will automatically execute the	2604	to give the embedded loop strictly lower priority for example).
2121	embedded loop sweep.
2122		2605
2123	As long as the watcher is started it will automatically handle events. The	2606	You can also set the callback to C<0>, in which case the embed watcher
2124	callback will be invoked whenever some events have been handled. You can	2607	will automatically execute the embedded loop sweep whenever necessary.
2125	set the callback to C<0> to avoid having to specify one if you are not
2126	interested in that.
2127		2608
2128	Also, there have not currently been made special provisions for forking:	2609	Fork detection will be handled transparently while the C<ev_embed> watcher
2129	when you fork, you not only have to call C<ev_loop_fork> on both loops,	2610	is active, i.e., the embedded loop will automatically be forked when the
2130	but you will also have to stop and restart any C<ev_embed> watchers	2611	embedding loop forks. In other cases, the user is responsible for calling
2131	yourself.	2612	C<ev_loop_fork> on the embedded loop.
2132		2613
2133	Unfortunately, not all backends are embeddable, only the ones returned by	2614	Unfortunately, not all backends are embeddable: only the ones returned by
2134	C<ev_embeddable_backends> are, which, unfortunately, does not include any	2615	C<ev_embeddable_backends> are, which, unfortunately, does not include any
2135	portable one.	2616	portable one.
2136		2617
2137	So when you want to use this feature you will always have to be prepared	2618	So when you want to use this feature you will always have to be prepared
2138	that you cannot get an embeddable loop. The recommended way to get around	2619	that you cannot get an embeddable loop. The recommended way to get around
2139	this is to have a separate variables for your embeddable loop, try to	2620	this is to have a separate variables for your embeddable loop, try to
2140	create it, and if that fails, use the normal loop for everything.	2621	create it, and if that fails, use the normal loop for everything.
		2622
		2623	=head3 C<ev_embed> and fork
		2624
		2625	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2626	automatically be applied to the embedded loop as well, so no special
		2627	fork handling is required in that case. When the watcher is not running,
		2628	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2629	as applicable.
2141		2630
2142	=head3 Watcher-Specific Functions and Data Members	2631	=head3 Watcher-Specific Functions and Data Members
2143		2632
2144	=over 4	2633	=over 4
2145		2634
…		…
2173	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be	2662	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
2174	used).	2663	used).
2175		2664
2176	struct ev_loop *loop_hi = ev_default_init (0);	2665	struct ev_loop *loop_hi = ev_default_init (0);
2177	struct ev_loop *loop_lo = 0;	2666	struct ev_loop *loop_lo = 0;
2178	struct ev_embed embed;	2667	ev_embed embed;
2179		2668
2180	// see if there is a chance of getting one that works	2669	// see if there is a chance of getting one that works
2181	// (remember that a flags value of 0 means autodetection)	2670	// (remember that a flags value of 0 means autodetection)
2182	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()	2671	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
2183	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())	2672	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
…		…
2197	kqueue implementation). Store the kqueue/socket-only event loop in	2686	kqueue implementation). Store the kqueue/socket-only event loop in
2198	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).	2687	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
2199		2688
2200	struct ev_loop *loop = ev_default_init (0);	2689	struct ev_loop *loop = ev_default_init (0);
2201	struct ev_loop *loop_socket = 0;	2690	struct ev_loop *loop_socket = 0;
2202	struct ev_embed embed;	2691	ev_embed embed;
2203		2692
2204	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)	2693	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
2205	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))	2694	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
2206	{	2695	{
2207	ev_embed_init (&embed, 0, loop_socket);	2696	ev_embed_init (&embed, 0, loop_socket);
…		…
2222	event loop blocks next and before C<ev_check> watchers are being called,	2711	event loop blocks next and before C<ev_check> watchers are being called,
2223	and only in the child after the fork. If whoever good citizen calling	2712	and only in the child after the fork. If whoever good citizen calling
2224	C<ev_default_fork> cheats and calls it in the wrong process, the fork	2713	C<ev_default_fork> cheats and calls it in the wrong process, the fork
2225	handlers will be invoked, too, of course.	2714	handlers will be invoked, too, of course.
2226		2715
		2716	=head3 The special problem of life after fork - how is it possible?
		2717
		2718	Most uses of C<fork()> consist of forking, then some simple calls to ste
		2719	up/change the process environment, followed by a call to C<exec()>. This
		2720	sequence should be handled by libev without any problems.
		2721
		2722	This changes when the application actually wants to do event handling
		2723	in the child, or both parent in child, in effect "continuing" after the
		2724	fork.
		2725
		2726	The default mode of operation (for libev, with application help to detect
		2727	forks) is to duplicate all the state in the child, as would be expected
		2728	when I<either> the parent I<or> the child process continues.
		2729
		2730	When both processes want to continue using libev, then this is usually the
		2731	wrong result. In that case, usually one process (typically the parent) is
		2732	supposed to continue with all watchers in place as before, while the other
		2733	process typically wants to start fresh, i.e. without any active watchers.
		2734
		2735	The cleanest and most efficient way to achieve that with libev is to
		2736	simply create a new event loop, which of course will be "empty", and
		2737	use that for new watchers. This has the advantage of not touching more
		2738	memory than necessary, and thus avoiding the copy-on-write, and the
		2739	disadvantage of having to use multiple event loops (which do not support
		2740	signal watchers).
		2741
		2742	When this is not possible, or you want to use the default loop for
		2743	other reasons, then in the process that wants to start "fresh", call
		2744	C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying
		2745	the default loop will "orphan" (not stop) all registered watchers, so you
		2746	have to be careful not to execute code that modifies those watchers. Note
		2747	also that in that case, you have to re-register any signal watchers.
		2748
2227	=head3 Watcher-Specific Functions and Data Members	2749	=head3 Watcher-Specific Functions and Data Members
2228		2750
2229	=over 4	2751	=over 4
2230		2752
2231	=item ev_fork_init (ev_signal *, callback)	2753	=item ev_fork_init (ev_signal *, callback)
…		…
2263	is that the author does not know of a simple (or any) algorithm for a	2785	is that the author does not know of a simple (or any) algorithm for a
2264	multiple-writer-single-reader queue that works in all cases and doesn't	2786	multiple-writer-single-reader queue that works in all cases and doesn't
2265	need elaborate support such as pthreads.	2787	need elaborate support such as pthreads.
2266		2788
2267	That means that if you want to queue data, you have to provide your own	2789	That means that if you want to queue data, you have to provide your own
2268	queue. But at least I can tell you would implement locking around your	2790	queue. But at least I can tell you how to implement locking around your
2269	queue:	2791	queue:
2270		2792
2271	=over 4	2793	=over 4
2272		2794
2273	=item queueing from a signal handler context	2795	=item queueing from a signal handler context
2274		2796
2275	To implement race-free queueing, you simply add to the queue in the signal	2797	To implement race-free queueing, you simply add to the queue in the signal
2276	handler but you block the signal handler in the watcher callback. Here is an example that does that for	2798	handler but you block the signal handler in the watcher callback. Here is
2277	some fictitious SIGUSR1 handler:	2799	an example that does that for some fictitious SIGUSR1 handler:
2278		2800
2279	static ev_async mysig;	2801	static ev_async mysig;
2280		2802
2281	static void	2803	static void
2282	sigusr1_handler (void)	2804	sigusr1_handler (void)
…		…
2348	=over 4	2870	=over 4
2349		2871
2350	=item ev_async_init (ev_async *, callback)	2872	=item ev_async_init (ev_async *, callback)
2351		2873
2352	Initialises and configures the async watcher - it has no parameters of any	2874	Initialises and configures the async watcher - it has no parameters of any
2353	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,	2875	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
2354	believe me.	2876	trust me.
2355		2877
2356	=item ev_async_send (loop, ev_async *)	2878	=item ev_async_send (loop, ev_async *)
2357		2879
2358	Sends/signals/activates the given C<ev_async> watcher, that is, feeds	2880	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
2359	an C<EV_ASYNC> event on the watcher into the event loop. Unlike	2881	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
2360	C<ev_feed_event>, this call is safe to do in other threads, signal or	2882	C<ev_feed_event>, this call is safe to do from other threads, signal or
2361	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding	2883	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
2362	section below on what exactly this means).	2884	section below on what exactly this means).
2363		2885
		2886	Note that, as with other watchers in libev, multiple events might get
		2887	compressed into a single callback invocation (another way to look at this
		2888	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
		2889	reset when the event loop detects that).
		2890
2364	This call incurs the overhead of a system call only once per loop iteration,	2891	This call incurs the overhead of a system call only once per event loop
2365	so while the overhead might be noticeable, it doesn't apply to repeated	2892	iteration, so while the overhead might be noticeable, it doesn't apply to
2366	calls to C<ev_async_send>.	2893	repeated calls to C<ev_async_send> for the same event loop.
2367		2894
2368	=item bool = ev_async_pending (ev_async *)	2895	=item bool = ev_async_pending (ev_async *)
2369		2896
2370	Returns a non-zero value when C<ev_async_send> has been called on the	2897	Returns a non-zero value when C<ev_async_send> has been called on the
2371	watcher but the event has not yet been processed (or even noted) by the	2898	watcher but the event has not yet been processed (or even noted) by the
…		…
2374	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When	2901	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
2375	the loop iterates next and checks for the watcher to have become active,	2902	the loop iterates next and checks for the watcher to have become active,
2376	it will reset the flag again. C<ev_async_pending> can be used to very	2903	it will reset the flag again. C<ev_async_pending> can be used to very
2377	quickly check whether invoking the loop might be a good idea.	2904	quickly check whether invoking the loop might be a good idea.
2378		2905
2379	Not that this does I<not> check whether the watcher itself is pending, only	2906	Not that this does I<not> check whether the watcher itself is pending,
2380	whether it has been requested to make this watcher pending.	2907	only whether it has been requested to make this watcher pending: there
		2908	is a time window between the event loop checking and resetting the async
		2909	notification, and the callback being invoked.
2381		2910
2382	=back	2911	=back
2383		2912
2384		2913
2385	=head1 OTHER FUNCTIONS	2914	=head1 OTHER FUNCTIONS
…		…
2389	=over 4	2918	=over 4
2390		2919
2391	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	2920	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2392		2921
2393	This function combines a simple timer and an I/O watcher, calls your	2922	This function combines a simple timer and an I/O watcher, calls your
2394	callback on whichever event happens first and automatically stop both	2923	callback on whichever event happens first and automatically stops both
2395	watchers. This is useful if you want to wait for a single event on an fd	2924	watchers. This is useful if you want to wait for a single event on an fd
2396	or timeout without having to allocate/configure/start/stop/free one or	2925	or timeout without having to allocate/configure/start/stop/free one or
2397	more watchers yourself.	2926	more watchers yourself.
2398		2927
2399	If C<fd> is less than 0, then no I/O watcher will be started and events	2928	If C<fd> is less than 0, then no I/O watcher will be started and the
2400	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2929	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2401	C<events> set will be created and started.	2930	the given C<fd> and C<events> set will be created and started.
2402		2931
2403	If C<timeout> is less than 0, then no timeout watcher will be	2932	If C<timeout> is less than 0, then no timeout watcher will be
2404	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2933	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
2405	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2934	repeat = 0) will be started. C<0> is a valid timeout.
2406	dubious value.
2407		2935
2408	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	2936	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
2409	passed an C<revents> set like normal event callbacks (a combination of	2937	passed an C<revents> set like normal event callbacks (a combination of
2410	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2938	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
2411	value passed to C<ev_once>:	2939	value passed to C<ev_once>. Note that it is possible to receive I<both>
		2940	a timeout and an io event at the same time - you probably should give io
		2941	events precedence.
		2942
		2943	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
2412		2944
2413	static void stdin_ready (int revents, void *arg)	2945	static void stdin_ready (int revents, void *arg)
2414	{	2946	{
		2947	if (revents & EV_READ)
		2948	/* stdin might have data for us, joy! */;
2415	if (revents & EV_TIMEOUT)	2949	else if (revents & EV_TIMEOUT)
2416	/* doh, nothing entered */;	2950	/* doh, nothing entered */;
2417	else if (revents & EV_READ)
2418	/* stdin might have data for us, joy! */;
2419	}	2951	}
2420		2952
2421	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2953	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2422		2954
2423	=item ev_feed_event (ev_loop *, watcher *, int revents)	2955	=item ev_feed_event (struct ev_loop *, watcher *, int revents)
2424		2956
2425	Feeds the given event set into the event loop, as if the specified event	2957	Feeds the given event set into the event loop, as if the specified event
2426	had happened for the specified watcher (which must be a pointer to an	2958	had happened for the specified watcher (which must be a pointer to an
2427	initialised but not necessarily started event watcher).	2959	initialised but not necessarily started event watcher).
2428		2960
2429	=item ev_feed_fd_event (ev_loop *, int fd, int revents)	2961	=item ev_feed_fd_event (struct ev_loop *, int fd, int revents)
2430		2962
2431	Feed an event on the given fd, as if a file descriptor backend detected	2963	Feed an event on the given fd, as if a file descriptor backend detected
2432	the given events it.	2964	the given events it.
2433		2965
2434	=item ev_feed_signal_event (ev_loop *loop, int signum)	2966	=item ev_feed_signal_event (struct ev_loop *loop, int signum)
2435		2967
2436	Feed an event as if the given signal occurred (C<loop> must be the default	2968	Feed an event as if the given signal occurred (C<loop> must be the default
2437	loop!).	2969	loop!).
2438		2970
2439	=back	2971	=back
…		…
2561		3093
2562	myclass obj;	3094	myclass obj;
2563	ev::io iow;	3095	ev::io iow;
2564	iow.set <myclass, &myclass::io_cb> (&obj);	3096	iow.set <myclass, &myclass::io_cb> (&obj);
2565		3097
		3098	=item w->set (object *)
		3099
		3100	This is an B<experimental> feature that might go away in a future version.
		3101
		3102	This is a variation of a method callback - leaving out the method to call
		3103	will default the method to C<operator ()>, which makes it possible to use
		3104	functor objects without having to manually specify the C<operator ()> all
		3105	the time. Incidentally, you can then also leave out the template argument
		3106	list.
		3107
		3108	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3109	int revents)>.
		3110
		3111	See the method-C<set> above for more details.
		3112
		3113	Example: use a functor object as callback.
		3114
		3115	struct myfunctor
		3116	{
		3117	void operator() (ev::io &w, int revents)
		3118	{
		3119	...
		3120	}
		3121	}
		3122
		3123	myfunctor f;
		3124
		3125	ev::io w;
		3126	w.set (&f);
		3127
2566	=item w->set<function> (void *data = 0)	3128	=item w->set<function> (void *data = 0)
2567		3129
2568	Also sets a callback, but uses a static method or plain function as	3130	Also sets a callback, but uses a static method or plain function as
2569	callback. The optional C<data> argument will be stored in the watcher's	3131	callback. The optional C<data> argument will be stored in the watcher's
2570	C<data> member and is free for you to use.	3132	C<data> member and is free for you to use.
2571		3133
2572	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.	3134	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
2573		3135
2574	See the method-C<set> above for more details.	3136	See the method-C<set> above for more details.
2575		3137
2576	Example:	3138	Example: Use a plain function as callback.
2577		3139
2578	static void io_cb (ev::io &w, int revents) { }	3140	static void io_cb (ev::io &w, int revents) { }
2579	iow.set <io_cb> ();	3141	iow.set <io_cb> ();
2580		3142
2581	=item w->set (struct ev_loop *)	3143	=item w->set (struct ev_loop *)
…		…
2619	Example: Define a class with an IO and idle watcher, start one of them in	3181	Example: Define a class with an IO and idle watcher, start one of them in
2620	the constructor.	3182	the constructor.
2621		3183
2622	class myclass	3184	class myclass
2623	{	3185	{
2624	ev::io io; void io_cb (ev::io &w, int revents);	3186	ev::io io ; void io_cb (ev::io &w, int revents);
2625	ev:idle idle void idle_cb (ev::idle &w, int revents);	3187	ev::idle idle; void idle_cb (ev::idle &w, int revents);
2626		3188
2627	myclass (int fd)	3189	myclass (int fd)
2628	{	3190	{
2629	io .set <myclass, &myclass::io_cb > (this);	3191	io .set <myclass, &myclass::io_cb > (this);
2630	idle.set <myclass, &myclass::idle_cb> (this);	3192	idle.set <myclass, &myclass::idle_cb> (this);
…		…
2646	=item Perl	3208	=item Perl
2647		3209
2648	The EV module implements the full libev API and is actually used to test	3210	The EV module implements the full libev API and is actually used to test
2649	libev. EV is developed together with libev. Apart from the EV core module,	3211	libev. EV is developed together with libev. Apart from the EV core module,
2650	there are additional modules that implement libev-compatible interfaces	3212	there are additional modules that implement libev-compatible interfaces
2651	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the	3213	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
2652	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).	3214	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3215	and C<EV::Glib>).
2653		3216
2654	It can be found and installed via CPAN, its homepage is at	3217	It can be found and installed via CPAN, its homepage is at
2655	L<http://software.schmorp.de/pkg/EV>.	3218	L<http://software.schmorp.de/pkg/EV>.
2656		3219
2657	=item Python	3220	=item Python
2658		3221
2659	Python bindings can be found at L<http://code.google.com/p/pyev/>. It	3222	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
2660	seems to be quite complete and well-documented. Note, however, that the	3223	seems to be quite complete and well-documented.
2661	patch they require for libev is outright dangerous as it breaks the ABI
2662	for everybody else, and therefore, should never be applied in an installed
2663	libev (if python requires an incompatible ABI then it needs to embed
2664	libev).
2665		3224
2666	=item Ruby	3225	=item Ruby
2667		3226
2668	Tony Arcieri has written a ruby extension that offers access to a subset	3227	Tony Arcieri has written a ruby extension that offers access to a subset
2669	of the libev API and adds file handle abstractions, asynchronous DNS and	3228	of the libev API and adds file handle abstractions, asynchronous DNS and
2670	more on top of it. It can be found via gem servers. Its homepage is at	3229	more on top of it. It can be found via gem servers. Its homepage is at
2671	L<http://rev.rubyforge.org/>.	3230	L<http://rev.rubyforge.org/>.
2672		3231
		3232	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		3233	makes rev work even on mingw.
		3234
		3235	=item Haskell
		3236
		3237	A haskell binding to libev is available at
		3238	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		3239
2673	=item D	3240	=item D
2674		3241
2675	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to	3242	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
2676	be found at L<http://proj.llucax.com.ar/wiki/evd>.	3243	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		3244
		3245	=item Ocaml
		3246
		3247	Erkki Seppala has written Ocaml bindings for libev, to be found at
		3248	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
2677		3249
2678	=back	3250	=back
2679		3251
2680		3252
2681	=head1 MACRO MAGIC	3253	=head1 MACRO MAGIC
…		…
2782		3354
2783	#define EV_STANDALONE 1	3355	#define EV_STANDALONE 1
2784	#include "ev.h"	3356	#include "ev.h"
2785		3357
2786	Both header files and implementation files can be compiled with a C++	3358	Both header files and implementation files can be compiled with a C++
2787	compiler (at least, thats a stated goal, and breakage will be treated	3359	compiler (at least, that's a stated goal, and breakage will be treated
2788	as a bug).	3360	as a bug).
2789		3361
2790	You need the following files in your source tree, or in a directory	3362	You need the following files in your source tree, or in a directory
2791	in your include path (e.g. in libev/ when using -Ilibev):	3363	in your include path (e.g. in libev/ when using -Ilibev):
2792		3364
…		…
2836		3408
2837	=head2 PREPROCESSOR SYMBOLS/MACROS	3409	=head2 PREPROCESSOR SYMBOLS/MACROS
2838		3410
2839	Libev can be configured via a variety of preprocessor symbols you have to	3411	Libev can be configured via a variety of preprocessor symbols you have to
2840	define before including any of its files. The default in the absence of	3412	define before including any of its files. The default in the absence of
2841	autoconf is noted for every option.	3413	autoconf is documented for every option.
2842		3414
2843	=over 4	3415	=over 4
2844		3416
2845	=item EV_STANDALONE	3417	=item EV_STANDALONE
2846		3418
…		…
2848	keeps libev from including F<config.h>, and it also defines dummy	3420	keeps libev from including F<config.h>, and it also defines dummy
2849	implementations for some libevent functions (such as logging, which is not	3421	implementations for some libevent functions (such as logging, which is not
2850	supported). It will also not define any of the structs usually found in	3422	supported). It will also not define any of the structs usually found in
2851	F<event.h> that are not directly supported by the libev core alone.	3423	F<event.h> that are not directly supported by the libev core alone.
2852		3424
		3425	In stanbdalone mode, libev will still try to automatically deduce the
		3426	configuration, but has to be more conservative.
		3427
2853	=item EV_USE_MONOTONIC	3428	=item EV_USE_MONOTONIC
2854		3429
2855	If defined to be C<1>, libev will try to detect the availability of the	3430	If defined to be C<1>, libev will try to detect the availability of the
2856	monotonic clock option at both compile time and runtime. Otherwise no use	3431	monotonic clock option at both compile time and runtime. Otherwise no
2857	of the monotonic clock option will be attempted. If you enable this, you	3432	use of the monotonic clock option will be attempted. If you enable this,
2858	usually have to link against librt or something similar. Enabling it when	3433	you usually have to link against librt or something similar. Enabling it
2859	the functionality isn't available is safe, though, although you have	3434	when the functionality isn't available is safe, though, although you have
2860	to make sure you link against any libraries where the C<clock_gettime>	3435	to make sure you link against any libraries where the C<clock_gettime>
2861	function is hiding in (often F<-lrt>).	3436	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
2862		3437
2863	=item EV_USE_REALTIME	3438	=item EV_USE_REALTIME
2864		3439
2865	If defined to be C<1>, libev will try to detect the availability of the	3440	If defined to be C<1>, libev will try to detect the availability of the
2866	real-time clock option at compile time (and assume its availability at	3441	real-time clock option at compile time (and assume its availability
2867	runtime if successful). Otherwise no use of the real-time clock option will	3442	at runtime if successful). Otherwise no use of the real-time clock
2868	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	3443	option will be attempted. This effectively replaces C<gettimeofday>
2869	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the	3444	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
2870	note about libraries in the description of C<EV_USE_MONOTONIC>, though.	3445	correctness. See the note about libraries in the description of
		3446	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		3447	C<EV_USE_CLOCK_SYSCALL>.
		3448
		3449	=item EV_USE_CLOCK_SYSCALL
		3450
		3451	If defined to be C<1>, libev will try to use a direct syscall instead
		3452	of calling the system-provided C<clock_gettime> function. This option
		3453	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		3454	unconditionally pulls in C<libpthread>, slowing down single-threaded
		3455	programs needlessly. Using a direct syscall is slightly slower (in
		3456	theory), because no optimised vdso implementation can be used, but avoids
		3457	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		3458	higher, as it simplifies linking (no need for C<-lrt>).
2871		3459
2872	=item EV_USE_NANOSLEEP	3460	=item EV_USE_NANOSLEEP
2873		3461
2874	If defined to be C<1>, libev will assume that C<nanosleep ()> is available	3462	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
2875	and will use it for delays. Otherwise it will use C<select ()>.	3463	and will use it for delays. Otherwise it will use C<select ()>.
…		…
2891		3479
2892	=item EV_SELECT_USE_FD_SET	3480	=item EV_SELECT_USE_FD_SET
2893		3481
2894	If defined to C<1>, then the select backend will use the system C<fd_set>	3482	If defined to C<1>, then the select backend will use the system C<fd_set>
2895	structure. This is useful if libev doesn't compile due to a missing	3483	structure. This is useful if libev doesn't compile due to a missing
2896	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on	3484	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
2897	exotic systems. This usually limits the range of file descriptors to some	3485	on exotic systems. This usually limits the range of file descriptors to
2898	low limit such as 1024 or might have other limitations (winsocket only	3486	some low limit such as 1024 or might have other limitations (winsocket
2899	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	3487	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
2900	influence the size of the C<fd_set> used.	3488	configures the maximum size of the C<fd_set>.
2901		3489
2902	=item EV_SELECT_IS_WINSOCKET	3490	=item EV_SELECT_IS_WINSOCKET
2903		3491
2904	When defined to C<1>, the select backend will assume that	3492	When defined to C<1>, the select backend will assume that
2905	select/socket/connect etc. don't understand file descriptors but	3493	select/socket/connect etc. don't understand file descriptors but
…		…
3016	When doing priority-based operations, libev usually has to linearly search	3604	When doing priority-based operations, libev usually has to linearly search
3017	all the priorities, so having many of them (hundreds) uses a lot of space	3605	all the priorities, so having many of them (hundreds) uses a lot of space
3018	and time, so using the defaults of five priorities (-2 .. +2) is usually	3606	and time, so using the defaults of five priorities (-2 .. +2) is usually
3019	fine.	3607	fine.
3020		3608
3021	If your embedding application does not need any priorities, defining these both to	3609	If your embedding application does not need any priorities, defining these
3022	C<0> will save some memory and CPU.	3610	both to C<0> will save some memory and CPU.
3023		3611
3024	=item EV_PERIODIC_ENABLE	3612	=item EV_PERIODIC_ENABLE
3025		3613
3026	If undefined or defined to be C<1>, then periodic timers are supported. If	3614	If undefined or defined to be C<1>, then periodic timers are supported. If
3027	defined to be C<0>, then they are not. Disabling them saves a few kB of	3615	defined to be C<0>, then they are not. Disabling them saves a few kB of
…		…
3034	code.	3622	code.
3035		3623
3036	=item EV_EMBED_ENABLE	3624	=item EV_EMBED_ENABLE
3037		3625
3038	If undefined or defined to be C<1>, then embed watchers are supported. If	3626	If undefined or defined to be C<1>, then embed watchers are supported. If
3039	defined to be C<0>, then they are not.	3627	defined to be C<0>, then they are not. Embed watchers rely on most other
		3628	watcher types, which therefore must not be disabled.
3040		3629
3041	=item EV_STAT_ENABLE	3630	=item EV_STAT_ENABLE
3042		3631
3043	If undefined or defined to be C<1>, then stat watchers are supported. If	3632	If undefined or defined to be C<1>, then stat watchers are supported. If
3044	defined to be C<0>, then they are not.	3633	defined to be C<0>, then they are not.
…		…
3076	two).	3665	two).
3077		3666
3078	=item EV_USE_4HEAP	3667	=item EV_USE_4HEAP
3079		3668
3080	Heaps are not very cache-efficient. To improve the cache-efficiency of the	3669	Heaps are not very cache-efficient. To improve the cache-efficiency of the
3081	timer and periodics heap, libev uses a 4-heap when this symbol is defined	3670	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
3082	to C<1>. The 4-heap uses more complicated (longer) code but has	3671	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
3083	noticeably faster performance with many (thousands) of watchers.	3672	faster performance with many (thousands) of watchers.
3084		3673
3085	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>	3674	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
3086	(disabled).	3675	(disabled).
3087		3676
3088	=item EV_HEAP_CACHE_AT	3677	=item EV_HEAP_CACHE_AT
3089		3678
3090	Heaps are not very cache-efficient. To improve the cache-efficiency of the	3679	Heaps are not very cache-efficient. To improve the cache-efficiency of the
3091	timer and periodics heap, libev can cache the timestamp (I<at>) within	3680	timer and periodics heaps, libev can cache the timestamp (I<at>) within
3092	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),	3681	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
3093	which uses 8-12 bytes more per watcher and a few hundred bytes more code,	3682	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
3094	but avoids random read accesses on heap changes. This improves performance	3683	but avoids random read accesses on heap changes. This improves performance
3095	noticeably with with many (hundreds) of watchers.	3684	noticeably with many (hundreds) of watchers.
3096		3685
3097	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>	3686	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
3098	(disabled).	3687	(disabled).
3099		3688
3100	=item EV_VERIFY	3689	=item EV_VERIFY
…		…
3106	called once per loop, which can slow down libev. If set to C<3>, then the	3695	called once per loop, which can slow down libev. If set to C<3>, then the
3107	verification code will be called very frequently, which will slow down	3696	verification code will be called very frequently, which will slow down
3108	libev considerably.	3697	libev considerably.
3109		3698
3110	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be	3699	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
3111	C<0.>	3700	C<0>.
3112		3701
3113	=item EV_COMMON	3702	=item EV_COMMON
3114		3703
3115	By default, all watchers have a C<void *data> member. By redefining	3704	By default, all watchers have a C<void *data> member. By redefining
3116	this macro to a something else you can include more and other types of	3705	this macro to a something else you can include more and other types of
…		…
3133	and the way callbacks are invoked and set. Must expand to a struct member	3722	and the way callbacks are invoked and set. Must expand to a struct member
3134	definition and a statement, respectively. See the F<ev.h> header file for	3723	definition and a statement, respectively. See the F<ev.h> header file for
3135	their default definitions. One possible use for overriding these is to	3724	their default definitions. One possible use for overriding these is to
3136	avoid the C<struct ev_loop *> as first argument in all cases, or to use	3725	avoid the C<struct ev_loop *> as first argument in all cases, or to use
3137	method calls instead of plain function calls in C++.	3726	method calls instead of plain function calls in C++.
		3727
		3728	=back
3138		3729
3139	=head2 EXPORTED API SYMBOLS	3730	=head2 EXPORTED API SYMBOLS
3140		3731
3141	If you need to re-export the API (e.g. via a DLL) and you need a list of	3732	If you need to re-export the API (e.g. via a DLL) and you need a list of
3142	exported symbols, you can use the provided F<Symbol.*> files which list	3733	exported symbols, you can use the provided F<Symbol.*> files which list
…		…
3189	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3780	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
3190		3781
3191	#include "ev_cpp.h"	3782	#include "ev_cpp.h"
3192	#include "ev.c"	3783	#include "ev.c"
3193		3784
		3785	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
3194		3786
3195	=head1 THREADS AND COROUTINES	3787	=head2 THREADS AND COROUTINES
3196		3788
3197	=head2 THREADS	3789	=head3 THREADS
3198		3790
3199	Libev itself is completely thread-safe, but it uses no locking. This	3791	All libev functions are reentrant and thread-safe unless explicitly
		3792	documented otherwise, but libev implements no locking itself. This means
3200	means that you can use as many loops as you want in parallel, as long as	3793	that you can use as many loops as you want in parallel, as long as there
3201	only one thread ever calls into one libev function with the same loop	3794	are no concurrent calls into any libev function with the same loop
3202	parameter.	3795	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		3796	of course): libev guarantees that different event loops share no data
		3797	structures that need any locking.
3203		3798
3204	Or put differently: calls with different loop parameters can be done in	3799	Or to put it differently: calls with different loop parameters can be done
3205	parallel from multiple threads, calls with the same loop parameter must be	3800	concurrently from multiple threads, calls with the same loop parameter
3206	done serially (but can be done from different threads, as long as only one	3801	must be done serially (but can be done from different threads, as long as
3207	thread ever is inside a call at any point in time, e.g. by using a mutex	3802	only one thread ever is inside a call at any point in time, e.g. by using
3208	per loop).	3803	a mutex per loop).
		3804
		3805	Specifically to support threads (and signal handlers), libev implements
		3806	so-called C<ev_async> watchers, which allow some limited form of
		3807	concurrency on the same event loop, namely waking it up "from the
		3808	outside".
3209		3809
3210	If you want to know which design (one loop, locking, or multiple loops	3810	If you want to know which design (one loop, locking, or multiple loops
3211	without or something else still) is best for your problem, then I cannot	3811	without or something else still) is best for your problem, then I cannot
3212	help you. I can give some generic advice however:	3812	help you, but here is some generic advice:
3213		3813
3214	=over 4	3814	=over 4
3215		3815
3216	=item * most applications have a main thread: use the default libev loop	3816	=item * most applications have a main thread: use the default libev loop
3217	in that thread, or create a separate thread running only the default loop.	3817	in that thread, or create a separate thread running only the default loop.
…		…
3229		3829
3230	Choosing a model is hard - look around, learn, know that usually you can do	3830	Choosing a model is hard - look around, learn, know that usually you can do
3231	better than you currently do :-)	3831	better than you currently do :-)
3232		3832
3233	=item * often you need to talk to some other thread which blocks in the	3833	=item * often you need to talk to some other thread which blocks in the
		3834	event loop.
		3835
3234	event loop - C<ev_async> watchers can be used to wake them up from other	3836	C<ev_async> watchers can be used to wake them up from other threads safely
3235	threads safely (or from signal contexts...).	3837	(or from signal contexts...).
		3838
		3839	An example use would be to communicate signals or other events that only
		3840	work in the default loop by registering the signal watcher with the
		3841	default loop and triggering an C<ev_async> watcher from the default loop
		3842	watcher callback into the event loop interested in the signal.
3236		3843
3237	=back	3844	=back
3238		3845
3239	=head2 COROUTINES	3846	=head3 COROUTINES
3240		3847
3241	Libev is much more accommodating to coroutines ("cooperative threads"):	3848	Libev is very accommodating to coroutines ("cooperative threads"):
3242	libev fully supports nesting calls to it's functions from different	3849	libev fully supports nesting calls to its functions from different
3243	coroutines (e.g. you can call C<ev_loop> on the same loop from two	3850	coroutines (e.g. you can call C<ev_loop> on the same loop from two
3244	different coroutines and switch freely between both coroutines running the	3851	different coroutines, and switch freely between both coroutines running the
3245	loop, as long as you don't confuse yourself). The only exception is that	3852	loop, as long as you don't confuse yourself). The only exception is that
3246	you must not do this from C<ev_periodic> reschedule callbacks.	3853	you must not do this from C<ev_periodic> reschedule callbacks.
3247		3854
3248	Care has been invested into making sure that libev does not keep local	3855	Care has been taken to ensure that libev does not keep local state inside
3249	state inside C<ev_loop>, and other calls do not usually allow coroutine	3856	C<ev_loop>, and other calls do not usually allow for coroutine switches as
3250	switches.	3857	they do not call any callbacks.
3251		3858
		3859	=head2 COMPILER WARNINGS
3252		3860
3253	=head1 COMPLEXITIES	3861	Depending on your compiler and compiler settings, you might get no or a
		3862	lot of warnings when compiling libev code. Some people are apparently
		3863	scared by this.
3254		3864
3255	In this section the complexities of (many of) the algorithms used inside	3865	However, these are unavoidable for many reasons. For one, each compiler
3256	libev will be explained. For complexity discussions about backends see the	3866	has different warnings, and each user has different tastes regarding
3257	documentation for C<ev_default_init>.	3867	warning options. "Warn-free" code therefore cannot be a goal except when
		3868	targeting a specific compiler and compiler-version.
3258		3869
3259	All of the following are about amortised time: If an array needs to be	3870	Another reason is that some compiler warnings require elaborate
3260	extended, libev needs to realloc and move the whole array, but this	3871	workarounds, or other changes to the code that make it less clear and less
3261	happens asymptotically never with higher number of elements, so O(1) might	3872	maintainable.
3262	mean it might do a lengthy realloc operation in rare cases, but on average
3263	it is much faster and asymptotically approaches constant time.
3264		3873
3265	=over 4	3874	And of course, some compiler warnings are just plain stupid, or simply
		3875	wrong (because they don't actually warn about the condition their message
		3876	seems to warn about). For example, certain older gcc versions had some
		3877	warnings that resulted an extreme number of false positives. These have
		3878	been fixed, but some people still insist on making code warn-free with
		3879	such buggy versions.
3266		3880
3267	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3881	While libev is written to generate as few warnings as possible,
		3882	"warn-free" code is not a goal, and it is recommended not to build libev
		3883	with any compiler warnings enabled unless you are prepared to cope with
		3884	them (e.g. by ignoring them). Remember that warnings are just that:
		3885	warnings, not errors, or proof of bugs.
3268		3886
3269	This means that, when you have a watcher that triggers in one hour and
3270	there are 100 watchers that would trigger before that then inserting will
3271	have to skip roughly seven (C<ld 100>) of these watchers.
3272		3887
3273	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	3888	=head2 VALGRIND
3274		3889
3275	That means that changing a timer costs less than removing/adding them	3890	Valgrind has a special section here because it is a popular tool that is
3276	as only the relative motion in the event queue has to be paid for.	3891	highly useful. Unfortunately, valgrind reports are very hard to interpret.
3277		3892
3278	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)	3893	If you think you found a bug (memory leak, uninitialised data access etc.)
		3894	in libev, then check twice: If valgrind reports something like:
3279		3895
3280	These just add the watcher into an array or at the head of a list.	3896	==2274== definitely lost: 0 bytes in 0 blocks.
		3897	==2274== possibly lost: 0 bytes in 0 blocks.
		3898	==2274== still reachable: 256 bytes in 1 blocks.
3281		3899
3282	=item Stopping check/prepare/idle/fork/async watchers: O(1)	3900	Then there is no memory leak, just as memory accounted to global variables
		3901	is not a memleak - the memory is still being referenced, and didn't leak.
3283		3902
3284	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3903	Similarly, under some circumstances, valgrind might report kernel bugs
		3904	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		3905	although an acceptable workaround has been found here), or it might be
		3906	confused.
3285		3907
3286	These watchers are stored in lists then need to be walked to find the	3908	Keep in mind that valgrind is a very good tool, but only a tool. Don't
3287	correct watcher to remove. The lists are usually short (you don't usually	3909	make it into some kind of religion.
3288	have many watchers waiting for the same fd or signal).
3289		3910
3290	=item Finding the next timer in each loop iteration: O(1)	3911	If you are unsure about something, feel free to contact the mailing list
		3912	with the full valgrind report and an explanation on why you think this
		3913	is a bug in libev (best check the archives, too :). However, don't be
		3914	annoyed when you get a brisk "this is no bug" answer and take the chance
		3915	of learning how to interpret valgrind properly.
3291		3916
3292	By virtue of using a binary or 4-heap, the next timer is always found at a	3917	If you need, for some reason, empty reports from valgrind for your project
3293	fixed position in the storage array.	3918	I suggest using suppression lists.
3294		3919
3295	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
3296		3920
3297	A change means an I/O watcher gets started or stopped, which requires	3921	=head1 PORTABILITY NOTES
3298	libev to recalculate its status (and possibly tell the kernel, depending
3299	on backend and whether C<ev_io_set> was used).
3300		3922
3301	=item Activating one watcher (putting it into the pending state): O(1)
3302
3303	=item Priority handling: O(number_of_priorities)
3304
3305	Priorities are implemented by allocating some space for each
3306	priority. When doing priority-based operations, libev usually has to
3307	linearly search all the priorities, but starting/stopping and activating
3308	watchers becomes O(1) w.r.t. priority handling.
3309
3310	=item Sending an ev_async: O(1)
3311
3312	=item Processing ev_async_send: O(number_of_async_watchers)
3313
3314	=item Processing signals: O(max_signal_number)
3315
3316	Sending involves a system call I<iff> there were no other C<ev_async_send>
3317	calls in the current loop iteration. Checking for async and signal events
3318	involves iterating over all running async watchers or all signal numbers.
3319
3320	=back
3321
3322
3323	=head1 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS	3923	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
3324		3924
3325	Win32 doesn't support any of the standards (e.g. POSIX) that libev	3925	Win32 doesn't support any of the standards (e.g. POSIX) that libev
3326	requires, and its I/O model is fundamentally incompatible with the POSIX	3926	requires, and its I/O model is fundamentally incompatible with the POSIX
3327	model. Libev still offers limited functionality on this platform in	3927	model. Libev still offers limited functionality on this platform in
3328	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	3928	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
…		…
3339		3939
3340	Not a libev limitation but worth mentioning: windows apparently doesn't	3940	Not a libev limitation but worth mentioning: windows apparently doesn't
3341	accept large writes: instead of resulting in a partial write, windows will	3941	accept large writes: instead of resulting in a partial write, windows will
3342	either accept everything or return C<ENOBUFS> if the buffer is too large,	3942	either accept everything or return C<ENOBUFS> if the buffer is too large,
3343	so make sure you only write small amounts into your sockets (less than a	3943	so make sure you only write small amounts into your sockets (less than a
3344	megabyte seems safe, but thsi apparently depends on the amount of memory	3944	megabyte seems safe, but this apparently depends on the amount of memory
3345	available).	3945	available).
3346		3946
3347	Due to the many, low, and arbitrary limits on the win32 platform and	3947	Due to the many, low, and arbitrary limits on the win32 platform and
3348	the abysmal performance of winsockets, using a large number of sockets	3948	the abysmal performance of winsockets, using a large number of sockets
3349	is not recommended (and not reasonable). If your program needs to use	3949	is not recommended (and not reasonable). If your program needs to use
…		…
3360	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */	3960	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
3361		3961
3362	#include "ev.h"	3962	#include "ev.h"
3363		3963
3364	And compile the following F<evwrap.c> file into your project (make sure	3964	And compile the following F<evwrap.c> file into your project (make sure
3365	you do I<not> compile the F<ev.c> or any other embedded soruce files!):	3965	you do I<not> compile the F<ev.c> or any other embedded source files!):
3366		3966
3367	#include "evwrap.h"	3967	#include "evwrap.h"
3368	#include "ev.c"	3968	#include "ev.c"
3369		3969
3370	=over 4	3970	=over 4
…		…
3415	wrap all I/O functions and provide your own fd management, but the cost of	4015	wrap all I/O functions and provide your own fd management, but the cost of
3416	calling select (O(n²)) will likely make this unworkable.	4016	calling select (O(n²)) will likely make this unworkable.
3417		4017
3418	=back	4018	=back
3419		4019
3420
3421	=head1 PORTABILITY REQUIREMENTS	4020	=head2 PORTABILITY REQUIREMENTS
3422		4021
3423	In addition to a working ISO-C implementation, libev relies on a few	4022	In addition to a working ISO-C implementation and of course the
3424	additional extensions:	4023	backend-specific APIs, libev relies on a few additional extensions:
3425		4024
3426	=over 4	4025	=over 4
3427		4026
3428	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible	4027	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
3429	calling conventions regardless of C<ev_watcher_type *>.	4028	calling conventions regardless of C<ev_watcher_type *>.
…		…
3435	calls them using an C<ev_watcher *> internally.	4034	calls them using an C<ev_watcher *> internally.
3436		4035
3437	=item C<sig_atomic_t volatile> must be thread-atomic as well	4036	=item C<sig_atomic_t volatile> must be thread-atomic as well
3438		4037
3439	The type C<sig_atomic_t volatile> (or whatever is defined as	4038	The type C<sig_atomic_t volatile> (or whatever is defined as
3440	C<EV_ATOMIC_T>) must be atomic w.r.t. accesses from different	4039	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
3441	threads. This is not part of the specification for C<sig_atomic_t>, but is	4040	threads. This is not part of the specification for C<sig_atomic_t>, but is
3442	believed to be sufficiently portable.	4041	believed to be sufficiently portable.
3443		4042
3444	=item C<sigprocmask> must work in a threaded environment	4043	=item C<sigprocmask> must work in a threaded environment
3445		4044
…		…
3454	except the initial one, and run the default loop in the initial thread as	4053	except the initial one, and run the default loop in the initial thread as
3455	well.	4054	well.
3456		4055
3457	=item C<long> must be large enough for common memory allocation sizes	4056	=item C<long> must be large enough for common memory allocation sizes
3458		4057
3459	To improve portability and simplify using libev, libev uses C<long>	4058	To improve portability and simplify its API, libev uses C<long> internally
3460	internally instead of C<size_t> when allocating its data structures. On	4059	instead of C<size_t> when allocating its data structures. On non-POSIX
3461	non-POSIX systems (Microsoft...) this might be unexpectedly low, but	4060	systems (Microsoft...) this might be unexpectedly low, but is still at
3462	is still at least 31 bits everywhere, which is enough for hundreds of	4061	least 31 bits everywhere, which is enough for hundreds of millions of
3463	millions of watchers.	4062	watchers.
3464		4063
3465	=item C<double> must hold a time value in seconds with enough accuracy	4064	=item C<double> must hold a time value in seconds with enough accuracy
3466		4065
3467	The type C<double> is used to represent timestamps. It is required to	4066	The type C<double> is used to represent timestamps. It is required to
3468	have at least 51 bits of mantissa (and 9 bits of exponent), which is good	4067	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
…		…
3472	=back	4071	=back
3473		4072
3474	If you know of other additional requirements drop me a note.	4073	If you know of other additional requirements drop me a note.
3475		4074
3476		4075
3477	=head1 COMPILER WARNINGS	4076	=head1 ALGORITHMIC COMPLEXITIES
3478		4077
3479	Depending on your compiler and compiler settings, you might get no or a	4078	In this section the complexities of (many of) the algorithms used inside
3480	lot of warnings when compiling libev code. Some people are apparently	4079	libev will be documented. For complexity discussions about backends see
3481	scared by this.	4080	the documentation for C<ev_default_init>.
3482		4081
3483	However, these are unavoidable for many reasons. For one, each compiler	4082	All of the following are about amortised time: If an array needs to be
3484	has different warnings, and each user has different tastes regarding	4083	extended, libev needs to realloc and move the whole array, but this
3485	warning options. "Warn-free" code therefore cannot be a goal except when	4084	happens asymptotically rarer with higher number of elements, so O(1) might
3486	targeting a specific compiler and compiler-version.	4085	mean that libev does a lengthy realloc operation in rare cases, but on
		4086	average it is much faster and asymptotically approaches constant time.
3487		4087
3488	Another reason is that some compiler warnings require elaborate	4088	=over 4
3489	workarounds, or other changes to the code that make it less clear and less
3490	maintainable.
3491		4089
3492	And of course, some compiler warnings are just plain stupid, or simply	4090	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
3493	wrong (because they don't actually warn about the condition their message
3494	seems to warn about).
3495		4091
3496	While libev is written to generate as few warnings as possible,	4092	This means that, when you have a watcher that triggers in one hour and
3497	"warn-free" code is not a goal, and it is recommended not to build libev	4093	there are 100 watchers that would trigger before that, then inserting will
3498	with any compiler warnings enabled unless you are prepared to cope with	4094	have to skip roughly seven (C<ld 100>) of these watchers.
3499	them (e.g. by ignoring them). Remember that warnings are just that:
3500	warnings, not errors, or proof of bugs.
3501		4095
		4096	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
3502		4097
3503	=head1 VALGRIND	4098	That means that changing a timer costs less than removing/adding them,
		4099	as only the relative motion in the event queue has to be paid for.
3504		4100
3505	Valgrind has a special section here because it is a popular tool that is	4101	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
3506	highly useful, but valgrind reports are very hard to interpret.
3507		4102
3508	If you think you found a bug (memory leak, uninitialised data access etc.)	4103	These just add the watcher into an array or at the head of a list.
3509	in libev, then check twice: If valgrind reports something like:
3510		4104
3511	==2274== definitely lost: 0 bytes in 0 blocks.	4105	=item Stopping check/prepare/idle/fork/async watchers: O(1)
3512	==2274== possibly lost: 0 bytes in 0 blocks.
3513	==2274== still reachable: 256 bytes in 1 blocks.
3514		4106
3515	Then there is no memory leak. Similarly, under some circumstances,	4107	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
3516	valgrind might report kernel bugs as if it were a bug in libev, or it
3517	might be confused (it is a very good tool, but only a tool).
3518		4108
3519	If you are unsure about something, feel free to contact the mailing list	4109	These watchers are stored in lists, so they need to be walked to find the
3520	with the full valgrind report and an explanation on why you think this is	4110	correct watcher to remove. The lists are usually short (you don't usually
3521	a bug in libev. However, don't be annoyed when you get a brisk "this is	4111	have many watchers waiting for the same fd or signal: one is typical, two
3522	no bug" answer and take the chance of learning how to interpret valgrind	4112	is rare).
3523	properly.
3524		4113
3525	If you need, for some reason, empty reports from valgrind for your project	4114	=item Finding the next timer in each loop iteration: O(1)
3526	I suggest using suppression lists.
3527		4115
		4116	By virtue of using a binary or 4-heap, the next timer is always found at a
		4117	fixed position in the storage array.
		4118
		4119	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		4120
		4121	A change means an I/O watcher gets started or stopped, which requires
		4122	libev to recalculate its status (and possibly tell the kernel, depending
		4123	on backend and whether C<ev_io_set> was used).
		4124
		4125	=item Activating one watcher (putting it into the pending state): O(1)
		4126
		4127	=item Priority handling: O(number_of_priorities)
		4128
		4129	Priorities are implemented by allocating some space for each
		4130	priority. When doing priority-based operations, libev usually has to
		4131	linearly search all the priorities, but starting/stopping and activating
		4132	watchers becomes O(1) with respect to priority handling.
		4133
		4134	=item Sending an ev_async: O(1)
		4135
		4136	=item Processing ev_async_send: O(number_of_async_watchers)
		4137
		4138	=item Processing signals: O(max_signal_number)
		4139
		4140	Sending involves a system call I<iff> there were no other C<ev_async_send>
		4141	calls in the current loop iteration. Checking for async and signal events
		4142	involves iterating over all running async watchers or all signal numbers.
		4143
		4144	=back
		4145
		4146
		4147	=head1 GLOSSARY
		4148
		4149	=over 4
		4150
		4151	=item active
		4152
		4153	A watcher is active as long as it has been started (has been attached to
		4154	an event loop) but not yet stopped (disassociated from the event loop).
		4155
		4156	=item application
		4157
		4158	In this document, an application is whatever is using libev.
		4159
		4160	=item callback
		4161
		4162	The address of a function that is called when some event has been
		4163	detected. Callbacks are being passed the event loop, the watcher that
		4164	received the event, and the actual event bitset.
		4165
		4166	=item callback invocation
		4167
		4168	The act of calling the callback associated with a watcher.
		4169
		4170	=item event
		4171
		4172	A change of state of some external event, such as data now being available
		4173	for reading on a file descriptor, time having passed or simply not having
		4174	any other events happening anymore.
		4175
		4176	In libev, events are represented as single bits (such as C<EV_READ> or
		4177	C<EV_TIMEOUT>).
		4178
		4179	=item event library
		4180
		4181	A software package implementing an event model and loop.
		4182
		4183	=item event loop
		4184
		4185	An entity that handles and processes external events and converts them
		4186	into callback invocations.
		4187
		4188	=item event model
		4189
		4190	The model used to describe how an event loop handles and processes
		4191	watchers and events.
		4192
		4193	=item pending
		4194
		4195	A watcher is pending as soon as the corresponding event has been detected,
		4196	and stops being pending as soon as the watcher will be invoked or its
		4197	pending status is explicitly cleared by the application.
		4198
		4199	A watcher can be pending, but not active. Stopping a watcher also clears
		4200	its pending status.
		4201
		4202	=item real time
		4203
		4204	The physical time that is observed. It is apparently strictly monotonic :)
		4205
		4206	=item wall-clock time
		4207
		4208	The time and date as shown on clocks. Unlike real time, it can actually
		4209	be wrong and jump forwards and backwards, e.g. when the you adjust your
		4210	clock.
		4211
		4212	=item watcher
		4213
		4214	A data structure that describes interest in certain events. Watchers need
		4215	to be started (attached to an event loop) before they can receive events.
		4216
		4217	=item watcher invocation
		4218
		4219	The act of calling the callback associated with a watcher.
		4220
		4221	=back
3528		4222
3529	=head1 AUTHOR	4223	=head1 AUTHOR
3530		4224
3531	Marc Lehmann <libev@schmorp.de>.	4225	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.
3532		4226

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines