ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.181 by root, Fri Sep 19 03:47:50 2008 UTC vs.
Revision 1.319 by root, Fri Oct 22 10:09:12 2010 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);
28		30
29	// this causes all nested ev_loop's to stop iterating	31	// this causes all nested ev_run's to stop iterating
30	ev_unloop (EV_A_ EVUNLOOP_ALL);	32	ev_break (EV_A_ EVBREAK_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_run to stop iterating
39	ev_unloop (EV_A_ EVUNLOOP_ONE);	41	ev_break (EV_A_ EVBREAK_ONE);
40	}	42	}
41		43
42	int	44	int
43	main (void)	45	main (void)
44	{	46	{
…		…
54	// simple non-repeating 5.5 second timeout	56	// simple non-repeating 5.5 second timeout
55	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	57	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
56	ev_timer_start (loop, &timeout_watcher);	58	ev_timer_start (loop, &timeout_watcher);
57		59
58	// now wait for events to arrive	60	// now wait for events to arrive
59	ev_loop (loop, 0);	61	ev_run (loop, 0);
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	Familiarity 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
…		…
84	=head2 FEATURES	98	=head2 FEATURES
85		99
86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the	100	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms	101	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface	102	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers	103	(for C<ev_stat>), Linux eventfd/signalfd (for faster and cleaner
90	with customised rescheduling (C<ev_periodic>), synchronous signals	104	inter-thread wakeup (C<ev_async>)/signal handling (C<ev_signal>)) relative
91	(C<ev_signal>), process status change events (C<ev_child>), and event	105	timers (C<ev_timer>), absolute timers with customised rescheduling
92	watchers dealing with the event loop mechanism itself (C<ev_idle>,	106	(C<ev_periodic>), synchronous signals (C<ev_signal>), process status
93	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as	107	change events (C<ev_child>), and event watchers dealing with the event
94	file watchers (C<ev_stat>) and even limited support for fork events	108	loop mechanism itself (C<ev_idle>, C<ev_embed>, C<ev_prepare> and
95	(C<ev_fork>).	109	C<ev_check> watchers) as well as file watchers (C<ev_stat>) and even
		110	limited support for fork events (C<ev_fork>).
96		111
97	It also is quite fast (see this	112	It also is quite fast (see this
98	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent	113	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
99	for example).	114	for example).
100		115
…		…
108	name C<loop> (which is always of type C<struct ev_loop *>) will not have	123	name C<loop> (which is always of type C<struct ev_loop *>) will not have
109	this argument.	124	this argument.
110		125
111	=head2 TIME REPRESENTATION	126	=head2 TIME REPRESENTATION
112		127
113	Libev represents time as a single floating point number, representing the	128	Libev represents time as a single floating point number, representing
114	(fractional) number of seconds since the (POSIX) epoch (somewhere near	129	the (fractional) number of seconds since the (POSIX) epoch (in practice
115	the beginning of 1970, details are complicated, don't ask). This type is	130	somewhere near the beginning of 1970, details are complicated, don't
116	called C<ev_tstamp>, which is what you should use too. It usually aliases	131	ask). This type is called C<ev_tstamp>, which is what you should use
117	to the C<double> type in C, and when you need to do any calculations on	132	too. It usually aliases to the C<double> type in C. When you need to do
118	it, you should treat it as some floating point value. Unlike the name	133	any calculations on it, you should treat it as some floating point value.
		134
119	component C<stamp> might indicate, it is also used for time differences	135	Unlike the name component C<stamp> might indicate, it is also used for
120	throughout libev.	136	time differences (e.g. delays) throughout libev.
121		137
122	=head1 ERROR HANDLING	138	=head1 ERROR HANDLING
123		139
124	Libev knows three classes of errors: operating system errors, usage errors	140	Libev knows three classes of errors: operating system errors, usage errors
125	and internal errors (bugs).	141	and internal errors (bugs).
…		…
149		165
150	=item ev_tstamp ev_time ()	166	=item ev_tstamp ev_time ()
151		167
152	Returns the current time as libev would use it. Please note that the	168	Returns the current time as libev would use it. Please note that the
153	C<ev_now> function is usually faster and also often returns the timestamp	169	C<ev_now> function is usually faster and also often returns the timestamp
154	you actually want to know.	170	you actually want to know. Also interetsing is the combination of
		171	C<ev_update_now> and C<ev_now>.
155		172
156	=item ev_sleep (ev_tstamp interval)	173	=item ev_sleep (ev_tstamp interval)
157		174
158	Sleep for the given interval: The current thread will be blocked until	175	Sleep for the given interval: The current thread will be blocked until
159	either it is interrupted or the given time interval has passed. Basically	176	either it is interrupted or the given time interval has passed. Basically
…		…
176	as this indicates an incompatible change. Minor versions are usually	193	as this indicates an incompatible change. Minor versions are usually
177	compatible to older versions, so a larger minor version alone is usually	194	compatible to older versions, so a larger minor version alone is usually
178	not a problem.	195	not a problem.
179		196
180	Example: Make sure we haven't accidentally been linked against the wrong	197	Example: Make sure we haven't accidentally been linked against the wrong
181	version.	198	version (note, however, that this will not detect ABI mismatches :).
182		199
183	assert (("libev version mismatch",	200	assert (("libev version mismatch",
184	ev_version_major () == EV_VERSION_MAJOR	201	ev_version_major () == EV_VERSION_MAJOR
185	&& ev_version_minor () >= EV_VERSION_MINOR));	202	&& ev_version_minor () >= EV_VERSION_MINOR));
186		203
…		…
197	assert (("sorry, no epoll, no sex",	214	assert (("sorry, no epoll, no sex",
198	ev_supported_backends () & EVBACKEND_EPOLL));	215	ev_supported_backends () & EVBACKEND_EPOLL));
199		216
200	=item unsigned int ev_recommended_backends ()	217	=item unsigned int ev_recommended_backends ()
201		218
202	Return the set of all backends compiled into this binary of libev and also	219	Return the set of all backends compiled into this binary of libev and
203	recommended for this platform. This set is often smaller than the one	220	also recommended for this platform, meaning it will work for most file
		221	descriptor types. This set is often smaller than the one returned by
204	returned by C<ev_supported_backends>, as for example kqueue is broken on	222	C<ev_supported_backends>, as for example kqueue is broken on most BSDs
205	most BSDs and will not be auto-detected unless you explicitly request it	223	and will not be auto-detected unless you explicitly request it (assuming
206	(assuming you know what you are doing). This is the set of backends that	224	you know what you are doing). This is the set of backends that libev will
207	libev will probe for if you specify no backends explicitly.	225	probe for if you specify no backends explicitly.
208		226
209	=item unsigned int ev_embeddable_backends ()	227	=item unsigned int ev_embeddable_backends ()
210		228
211	Returns the set of backends that are embeddable in other event loops. This	229	Returns the set of backends that are embeddable in other event loops. This
212	is the theoretical, all-platform, value. To find which backends	230	value is platform-specific but can include backends not available on the
213	might be supported on the current system, you would need to look at	231	current system. To find which embeddable backends might be supported on
214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	232	the current system, you would need to look at C<ev_embeddable_backends ()
215	recommended ones.	233	& ev_supported_backends ()>, likewise for recommended ones.
216		234
217	See the description of C<ev_embed> watchers for more info.	235	See the description of C<ev_embed> watchers for more info.
218		236
219	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	237	=item ev_set_allocator (void *(*cb)(void *ptr, long size)) [NOT REENTRANT]
220		238
221	Sets the allocation function to use (the prototype is similar - the	239	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	240	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	241	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	242	when memory needs to be allocated (C<size != 0>), the library might abort
…		…
250	}	268	}
251		269
252	...	270	...
253	ev_set_allocator (persistent_realloc);	271	ev_set_allocator (persistent_realloc);
254		272
255	=item ev_set_syserr_cb (void (*cb)(const char *msg));	273	=item ev_set_syserr_cb (void (*cb)(const char *msg)); [NOT REENTRANT]
256		274
257	Set the callback function to call on a retryable system call error (such	275	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	276	as failed select, poll, epoll_wait). The message is a printable string
259	indicating the system call or subsystem causing the problem. If this	277	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	278	callback is set, then libev will expect it to remedy the situation, no
…		…
276		294
277	=back	295	=back
278		296
279	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	297	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
280		298
281	An event loop is described by a C<struct ev_loop *>. The library knows two	299	An event loop is described by a C<struct ev_loop *> (the C<struct> is
282	types of such loops, the I<default> loop, which supports signals and child	300	I<not> optional in this case unless libev 3 compatibility is disabled, as
283	events, and dynamically created loops which do not.	301	libev 3 had an C<ev_loop> function colliding with the struct name).
		302
		303	The library knows two types of such loops, the I<default> loop, which
		304	supports signals and child events, and dynamically created event loops
		305	which do not.
284		306
285	=over 4	307	=over 4
286		308
287	=item struct ev_loop *ev_default_loop (unsigned int flags)	309	=item struct ev_loop *ev_default_loop (unsigned int flags)
288		310
…		…
294	If you don't know what event loop to use, use the one returned from this	316	If you don't know what event loop to use, use the one returned from this
295	function.	317	function.
296		318
297	Note that this function is I<not> thread-safe, so if you want to use it	319	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,	320	from multiple threads, you have to lock (note also that this is unlikely,
299	as loops cannot bes hared easily between threads anyway).	321	as loops cannot be shared easily between threads anyway).
300		322
301	The default loop is the only loop that can handle C<ev_signal> and	323	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	324	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	325	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	326	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
…		…
326	useful to try out specific backends to test their performance, or to work	348	useful to try out specific backends to test their performance, or to work
327	around bugs.	349	around bugs.
328		350
329	=item C<EVFLAG_FORKCHECK>	351	=item C<EVFLAG_FORKCHECK>
330		352
331	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after	353	Instead of calling C<ev_loop_fork> manually after a fork, you can also
332	a fork, you can also make libev check for a fork in each iteration by	354	make libev check for a fork in each iteration by enabling this flag.
333	enabling this flag.
334		355
335	This works by calling C<getpid ()> on every iteration of the loop,	356	This works by calling C<getpid ()> on every iteration of the loop,
336	and thus this might slow down your event loop if you do a lot of loop	357	and thus this might slow down your event loop if you do a lot of loop
337	iterations and little real work, but is usually not noticeable (on my	358	iterations and little real work, but is usually not noticeable (on my
338	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence	359	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
…		…
344	flag.	365	flag.
345		366
346	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>	367	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
347	environment variable.	368	environment variable.
348		369
		370	=item C<EVFLAG_NOINOTIFY>
		371
		372	When this flag is specified, then libev will not attempt to use the
		373	I<inotify> API for it's C<ev_stat> watchers. Apart from debugging and
		374	testing, this flag can be useful to conserve inotify file descriptors, as
		375	otherwise each loop using C<ev_stat> watchers consumes one inotify handle.
		376
		377	=item C<EVFLAG_SIGNALFD>
		378
		379	When this flag is specified, then libev will attempt to use the
		380	I<signalfd> API for it's C<ev_signal> (and C<ev_child>) watchers. This API
		381	delivers signals synchronously, which makes it both faster and might make
		382	it possible to get the queued signal data. It can also simplify signal
		383	handling with threads, as long as you properly block signals in your
		384	threads that are not interested in handling them.
		385
		386	Signalfd will not be used by default as this changes your signal mask, and
		387	there are a lot of shoddy libraries and programs (glib's threadpool for
		388	example) that can't properly initialise their signal masks.
		389
349	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	390	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
350		391
351	This is your standard select(2) backend. Not I<completely> standard, as	392	This is your standard select(2) backend. Not I<completely> standard, as
352	libev tries to roll its own fd_set with no limits on the number of fds,	393	libev tries to roll its own fd_set with no limits on the number of fds,
353	but if that fails, expect a fairly low limit on the number of fds when	394	but if that fails, expect a fairly low limit on the number of fds when
…		…
377	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and	418	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
378	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.	419	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
379		420
380	=item C<EVBACKEND_EPOLL> (value 4, Linux)	421	=item C<EVBACKEND_EPOLL> (value 4, Linux)
381		422
		423	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
		424	kernels).
		425
382	For few fds, this backend is a bit little slower than poll and select,	426	For few fds, this backend is a bit little slower than poll and select,
383	but it scales phenomenally better. While poll and select usually scale	427	but it scales phenomenally better. While poll and select usually scale
384	like O(total_fds) where n is the total number of fds (or the highest fd),	428	like O(total_fds) where n is the total number of fds (or the highest fd),
385	epoll scales either O(1) or O(active_fds). The epoll design has a number	429	epoll scales either O(1) or O(active_fds).
386	of shortcomings, such as silently dropping events in some hard-to-detect	430
387	cases and requiring a system call per fd change, no fork support and bad	431	The epoll mechanism deserves honorable mention as the most misdesigned
388	support for dup.	432	of the more advanced event mechanisms: mere annoyances include silently
		433	dropping file descriptors, requiring a system call per change per file
		434	descriptor (and unnecessary guessing of parameters), problems with dup and
		435	so on. The biggest issue is fork races, however - if a program forks then
		436	I<both> parent and child process have to recreate the epoll set, which can
		437	take considerable time (one syscall per file descriptor) and is of course
		438	hard to detect.
		439
		440	Epoll is also notoriously buggy - embedding epoll fds I<should> work, but
		441	of course I<doesn't>, and epoll just loves to report events for totally
		442	I<different> file descriptors (even already closed ones, so one cannot
		443	even remove them from the set) than registered in the set (especially
		444	on SMP systems). Libev tries to counter these spurious notifications by
		445	employing an additional generation counter and comparing that against the
		446	events to filter out spurious ones, recreating the set when required. Last
		447	not least, it also refuses to work with some file descriptors which work
		448	perfectly fine with C<select> (files, many character devices...).
389		449
390	While stopping, setting and starting an I/O watcher in the same iteration	450	While stopping, setting and starting an I/O watcher in the same iteration
391	will result in some caching, there is still a system call per such incident	451	will result in some caching, there is still a system call per such
392	(because the fd could point to a different file description now), so its	452	incident (because the same I<file descriptor> could point to a different
393	best to avoid that. Also, C<dup ()>'ed file descriptors might not work	453	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
394	very well if you register events for both fds.	454	file descriptors might not work very well if you register events for both
395		455	file descriptors.
396	Please note that epoll sometimes generates spurious notifications, so you
397	need to use non-blocking I/O or other means to avoid blocking when no data
398	(or space) is available.
399		456
400	Best performance from this backend is achieved by not unregistering all	457	Best performance from this backend is achieved by not unregistering all
401	watchers for a file descriptor until it has been closed, if possible, i.e.	458	watchers for a file descriptor until it has been closed, if possible,
402	keep at least one watcher active per fd at all times.	459	i.e. keep at least one watcher active per fd at all times. Stopping and
		460	starting a watcher (without re-setting it) also usually doesn't cause
		461	extra overhead. A fork can both result in spurious notifications as well
		462	as in libev having to destroy and recreate the epoll object, which can
		463	take considerable time and thus should be avoided.
		464
		465	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		466	faster than epoll for maybe up to a hundred file descriptors, depending on
		467	the usage. So sad.
403		468
404	While nominally embeddable in other event loops, this feature is broken in	469	While nominally embeddable in other event loops, this feature is broken in
405	all kernel versions tested so far.	470	all kernel versions tested so far.
406		471
407	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as	472	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
…		…
410	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	475	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
411		476
412	Kqueue deserves special mention, as at the time of this writing, it	477	Kqueue deserves special mention, as at the time of this writing, it
413	was broken on all BSDs except NetBSD (usually it doesn't work reliably	478	was broken on all BSDs except NetBSD (usually it doesn't work reliably
414	with anything but sockets and pipes, except on Darwin, where of course	479	with anything but sockets and pipes, except on Darwin, where of course
415	it's completely useless). For this reason it's not being "auto-detected"	480	it's completely useless). Unlike epoll, however, whose brokenness
		481	is by design, these kqueue bugs can (and eventually will) be fixed
		482	without API changes to existing programs. For this reason it's not being
416	unless you explicitly specify it explicitly in the flags (i.e. using	483	"auto-detected" unless you explicitly specify it in the flags (i.e. using
417	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)	484	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
418	system like NetBSD.	485	system like NetBSD.
419		486
420	You still can embed kqueue into a normal poll or select backend and use it	487	You still can embed kqueue into a normal poll or select backend and use it
421	only for sockets (after having made sure that sockets work with kqueue on	488	only for sockets (after having made sure that sockets work with kqueue on
…		…
423		490
424	It scales in the same way as the epoll backend, but the interface to the	491	It scales in the same way as the epoll backend, but the interface to the
425	kernel is more efficient (which says nothing about its actual speed, of	492	kernel is more efficient (which says nothing about its actual speed, of
426	course). While stopping, setting and starting an I/O watcher does never	493	course). While stopping, setting and starting an I/O watcher does never
427	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to	494	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
428	two event changes per incident, support for C<fork ()> is very bad and it	495	two event changes per incident. Support for C<fork ()> is very bad (but
429	drops fds silently in similarly hard-to-detect cases.	496	sane, unlike epoll) and it drops fds silently in similarly hard-to-detect
		497	cases
430		498
431	This backend usually performs well under most conditions.	499	This backend usually performs well under most conditions.
432		500
433	While nominally embeddable in other event loops, this doesn't work	501	While nominally embeddable in other event loops, this doesn't work
434	everywhere, so you might need to test for this. And since it is broken	502	everywhere, so you might need to test for this. And since it is broken
435	almost everywhere, you should only use it when you have a lot of sockets	503	almost everywhere, you should only use it when you have a lot of sockets
436	(for which it usually works), by embedding it into another event loop	504	(for which it usually works), by embedding it into another event loop
437	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for	505	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
438	sockets.	506	also broken on OS X)) and, did I mention it, using it only for sockets.
439		507
440	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with	508	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
441	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with	509	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
442	C<NOTE_EOF>.	510	C<NOTE_EOF>.
443		511
…		…
460	While this backend scales well, it requires one system call per active	528	While this backend scales well, it requires one system call per active
461	file descriptor per loop iteration. For small and medium numbers of file	529	file descriptor per loop iteration. For small and medium numbers of file
462	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	530	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
463	might perform better.	531	might perform better.
464		532
465	On the positive side, ignoring the spurious readiness notifications, this	533	On the positive side, with the exception of the spurious readiness
466	backend actually performed to specification in all tests and is fully	534	notifications, this backend actually performed fully to specification
467	embeddable, which is a rare feat among the OS-specific backends.	535	in all tests and is fully embeddable, which is a rare feat among the
		536	OS-specific backends (I vastly prefer correctness over speed hacks).
468		537
469	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as	538	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
470	C<EVBACKEND_POLL>.	539	C<EVBACKEND_POLL>.
471		540
472	=item C<EVBACKEND_ALL>	541	=item C<EVBACKEND_ALL>
…		…
477		546
478	It is definitely not recommended to use this flag.	547	It is definitely not recommended to use this flag.
479		548
480	=back	549	=back
481		550
482	If one or more of these are or'ed into the flags value, then only these	551	If one or more of the backend flags are or'ed into the flags value,
483	backends will be tried (in the reverse order as listed here). If none are	552	then only these backends will be tried (in the reverse order as listed
484	specified, all backends in C<ev_recommended_backends ()> will be tried.	553	here). If none are specified, all backends in C<ev_recommended_backends
		554	()> will be tried.
485		555
486	The most typical usage is like this:	556	Example: This is the most typical usage.
487		557
488	if (!ev_default_loop (0))	558	if (!ev_default_loop (0))
489	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	559	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
490		560
491	Restrict libev to the select and poll backends, and do not allow	561	Example: Restrict libev to the select and poll backends, and do not allow
492	environment settings to be taken into account:	562	environment settings to be taken into account:
493		563
494	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);	564	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
495		565
496	Use whatever libev has to offer, but make sure that kqueue is used if	566	Example: Use whatever libev has to offer, but make sure that kqueue is
497	available (warning, breaks stuff, best use only with your own private	567	used if available (warning, breaks stuff, best use only with your own
498	event loop and only if you know the OS supports your types of fds):	568	private event loop and only if you know the OS supports your types of
		569	fds):
499		570
500	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);	571	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
501		572
502	=item struct ev_loop *ev_loop_new (unsigned int flags)	573	=item struct ev_loop *ev_loop_new (unsigned int flags)
503		574
504	Similar to C<ev_default_loop>, but always creates a new event loop that is	575	Similar to C<ev_default_loop>, but always creates a new event loop that is
505	always distinct from the default loop. Unlike the default loop, it cannot	576	always distinct from the default loop.
506	handle signal and child watchers, and attempts to do so will be greeted by
507	undefined behaviour (or a failed assertion if assertions are enabled).
508		577
509	Note that this function I<is> thread-safe, and the recommended way to use	578	Note that this function I<is> thread-safe, and one common way to use
510	libev with threads is indeed to create one loop per thread, and using the	579	libev with threads is indeed to create one loop per thread, and using the
511	default loop in the "main" or "initial" thread.	580	default loop in the "main" or "initial" thread.
512		581
513	Example: Try to create a event loop that uses epoll and nothing else.	582	Example: Try to create a event loop that uses epoll and nothing else.
514		583
…		…
516	if (!epoller)	585	if (!epoller)
517	fatal ("no epoll found here, maybe it hides under your chair");	586	fatal ("no epoll found here, maybe it hides under your chair");
518		587
519	=item ev_default_destroy ()	588	=item ev_default_destroy ()
520		589
521	Destroys the default loop again (frees all memory and kernel state	590	Destroys the default loop (frees all memory and kernel state etc.). None
522	etc.). None of the active event watchers will be stopped in the normal	591	of the active event watchers will be stopped in the normal sense, so
523	sense, so e.g. C<ev_is_active> might still return true. It is your	592	e.g. C<ev_is_active> might still return true. It is your responsibility to
524	responsibility to either stop all watchers cleanly yourself I<before>	593	either stop all watchers cleanly yourself I<before> calling this function,
525	calling this function, or cope with the fact afterwards (which is usually	594	or cope with the fact afterwards (which is usually the easiest thing, you
526	the easiest thing, you can just ignore the watchers and/or C<free ()> them	595	can just ignore the watchers and/or C<free ()> them for example).
527	for example).
528		596
529	Note that certain global state, such as signal state, will not be freed by	597	Note that certain global state, such as signal state (and installed signal
530	this function, and related watchers (such as signal and child watchers)	598	handlers), will not be freed by this function, and related watchers (such
531	would need to be stopped manually.	599	as signal and child watchers) would need to be stopped manually.
532		600
533	In general it is not advisable to call this function except in the	601	In general it is not advisable to call this function except in the
534	rare occasion where you really need to free e.g. the signal handling	602	rare occasion where you really need to free e.g. the signal handling
535	pipe fds. If you need dynamically allocated loops it is better to use	603	pipe fds. If you need dynamically allocated loops it is better to use
536	C<ev_loop_new> and C<ev_loop_destroy>).	604	C<ev_loop_new> and C<ev_loop_destroy>.
537		605
538	=item ev_loop_destroy (loop)	606	=item ev_loop_destroy (loop)
539		607
540	Like C<ev_default_destroy>, but destroys an event loop created by an	608	Like C<ev_default_destroy>, but destroys an event loop created by an
541	earlier call to C<ev_loop_new>.	609	earlier call to C<ev_loop_new>.
542		610
543	=item ev_default_fork ()	611	=item ev_default_fork ()
544		612
545	This function sets a flag that causes subsequent C<ev_loop> iterations	613	This function sets a flag that causes subsequent C<ev_run> iterations
546	to reinitialise the kernel state for backends that have one. Despite the	614	to reinitialise the kernel state for backends that have one. Despite the
547	name, you can call it anytime, but it makes most sense after forking, in	615	name, you can call it anytime, but it makes most sense after forking, in
548	the child process (or both child and parent, but that again makes little	616	the child process (or both child and parent, but that again makes little
549	sense). You I<must> call it in the child before using any of the libev	617	sense). You I<must> call it in the child before using any of the libev
550	functions, and it will only take effect at the next C<ev_loop> iteration.	618	functions, and it will only take effect at the next C<ev_run> iteration.
		619
		620	Again, you I<have> to call it on I<any> loop that you want to re-use after
		621	a fork, I<even if you do not plan to use the loop in the parent>. This is
		622	because some kernel interfaces *cough* I<kqueue> *cough* do funny things
		623	during fork.
551		624
552	On the other hand, you only need to call this function in the child	625	On the other hand, you only need to call this function in the child
553	process if and only if you want to use the event library in the child. If	626	process if and only if you want to use the event loop in the child. If
554	you just fork+exec, you don't have to call it at all.	627	you just fork+exec or create a new loop in the child, you don't have to
		628	call it at all (in fact, C<epoll> is so badly broken that it makes a
		629	difference, but libev will usually detect this case on its own and do a
		630	costly reset of the backend).
555		631
556	The function itself is quite fast and it's usually not a problem to call	632	The function itself is quite fast and it's usually not a problem to call
557	it just in case after a fork. To make this easy, the function will fit in	633	it just in case after a fork. To make this easy, the function will fit in
558	quite nicely into a call to C<pthread_atfork>:	634	quite nicely into a call to C<pthread_atfork>:
559		635
…		…
561		637
562	=item ev_loop_fork (loop)	638	=item ev_loop_fork (loop)
563		639
564	Like C<ev_default_fork>, but acts on an event loop created by	640	Like C<ev_default_fork>, but acts on an event loop created by
565	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	641	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
566	after fork, and how you do this is entirely your own problem.	642	after fork that you want to re-use in the child, and how you keep track of
		643	them is entirely your own problem.
567		644
568	=item int ev_is_default_loop (loop)	645	=item int ev_is_default_loop (loop)
569		646
570	Returns true when the given loop actually is the default loop, false otherwise.	647	Returns true when the given loop is, in fact, the default loop, and false
		648	otherwise.
571		649
572	=item unsigned int ev_loop_count (loop)	650	=item unsigned int ev_iteration (loop)
573		651
574	Returns the count of loop iterations for the loop, which is identical to	652	Returns the current iteration count for the event loop, which is identical
575	the number of times libev did poll for new events. It starts at C<0> and	653	to the number of times libev did poll for new events. It starts at C<0>
576	happily wraps around with enough iterations.	654	and happily wraps around with enough iterations.
577		655
578	This value can sometimes be useful as a generation counter of sorts (it	656	This value can sometimes be useful as a generation counter of sorts (it
579	"ticks" the number of loop iterations), as it roughly corresponds with	657	"ticks" the number of loop iterations), as it roughly corresponds with
580	C<ev_prepare> and C<ev_check> calls.	658	C<ev_prepare> and C<ev_check> calls - and is incremented between the
		659	prepare and check phases.
		660
		661	=item unsigned int ev_depth (loop)
		662
		663	Returns the number of times C<ev_run> was entered minus the number of
		664	times C<ev_run> was exited, in other words, the recursion depth.
		665
		666	Outside C<ev_run>, this number is zero. In a callback, this number is
		667	C<1>, unless C<ev_run> was invoked recursively (or from another thread),
		668	in which case it is higher.
		669
		670	Leaving C<ev_run> abnormally (setjmp/longjmp, cancelling the thread
		671	etc.), doesn't count as "exit" - consider this as a hint to avoid such
		672	ungentleman-like behaviour unless it's really convenient.
581		673
582	=item unsigned int ev_backend (loop)	674	=item unsigned int ev_backend (loop)
583		675
584	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	676	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
585	use.	677	use.
…		…
594		686
595	=item ev_now_update (loop)	687	=item ev_now_update (loop)
596		688
597	Establishes the current time by querying the kernel, updating the time	689	Establishes the current time by querying the kernel, updating the time
598	returned by C<ev_now ()> in the progress. This is a costly operation and	690	returned by C<ev_now ()> in the progress. This is a costly operation and
599	is usually done automatically within C<ev_loop ()>.	691	is usually done automatically within C<ev_run ()>.
600		692
601	This function is rarely useful, but when some event callback runs for a	693	This function is rarely useful, but when some event callback runs for a
602	very long time without entering the event loop, updating libev's idea of	694	very long time without entering the event loop, updating libev's idea of
603	the current time is a good idea.	695	the current time is a good idea.
604		696
605	See also "The special problem of time updates" in the C<ev_timer> section.	697	See also L<The special problem of time updates> in the C<ev_timer> section.
606		698
		699	=item ev_suspend (loop)
		700
		701	=item ev_resume (loop)
		702
		703	These two functions suspend and resume an event loop, for use when the
		704	loop is not used for a while and timeouts should not be processed.
		705
		706	A typical use case would be an interactive program such as a game: When
		707	the user presses C<^Z> to suspend the game and resumes it an hour later it
		708	would be best to handle timeouts as if no time had actually passed while
		709	the program was suspended. This can be achieved by calling C<ev_suspend>
		710	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		711	C<ev_resume> directly afterwards to resume timer processing.
		712
		713	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		714	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		715	will be rescheduled (that is, they will lose any events that would have
		716	occurred while suspended).
		717
		718	After calling C<ev_suspend> you B<must not> call I<any> function on the
		719	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		720	without a previous call to C<ev_suspend>.
		721
		722	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		723	event loop time (see C<ev_now_update>).
		724
607	=item ev_loop (loop, int flags)	725	=item ev_run (loop, int flags)
608		726
609	Finally, this is it, the event handler. This function usually is called	727	Finally, this is it, the event handler. This function usually is called
610	after you initialised all your watchers and you want to start handling	728	after you have initialised all your watchers and you want to start
611	events.	729	handling events. It will ask the operating system for any new events, call
		730	the watcher callbacks, an then repeat the whole process indefinitely: This
		731	is why event loops are called I<loops>.
612		732
613	If the flags argument is specified as C<0>, it will not return until	733	If the flags argument is specified as C<0>, it will keep handling events
614	either no event watchers are active anymore or C<ev_unloop> was called.	734	until either no event watchers are active anymore or C<ev_break> was
		735	called.
615		736
616	Please note that an explicit C<ev_unloop> is usually better than	737	Please note that an explicit C<ev_break> is usually better than
617	relying on all watchers to be stopped when deciding when a program has	738	relying on all watchers to be stopped when deciding when a program has
618	finished (especially in interactive programs), but having a program that	739	finished (especially in interactive programs), but having a program
619	automatically loops as long as it has to and no longer by virtue of	740	that automatically loops as long as it has to and no longer by virtue
620	relying on its watchers stopping correctly is a thing of beauty.	741	of relying on its watchers stopping correctly, that is truly a thing of
		742	beauty.
621		743
622	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	744	A flags value of C<EVRUN_NOWAIT> will look for new events, will handle
623	those events and any outstanding ones, but will not block your process in	745	those events and any already outstanding ones, but will not wait and
624	case there are no events and will return after one iteration of the loop.	746	block your process in case there are no events and will return after one
		747	iteration of the loop. This is sometimes useful to poll and handle new
		748	events while doing lengthy calculations, to keep the program responsive.
625		749
626	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	750	A flags value of C<EVRUN_ONCE> will look for new events (waiting if
627	necessary) and will handle those and any outstanding ones. It will block	751	necessary) and will handle those and any already outstanding ones. It
628	your process until at least one new event arrives, and will return after	752	will block your process until at least one new event arrives (which could
629	one iteration of the loop. This is useful if you are waiting for some	753	be an event internal to libev itself, so there is no guarantee that a
630	external event in conjunction with something not expressible using other	754	user-registered callback will be called), and will return after one
		755	iteration of the loop.
		756
		757	This is useful if you are waiting for some external event in conjunction
		758	with something not expressible using other libev watchers (i.e. "roll your
631	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	759	own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
632	usually a better approach for this kind of thing.	760	usually a better approach for this kind of thing.
633		761
634	Here are the gory details of what C<ev_loop> does:	762	Here are the gory details of what C<ev_run> does:
635		763
		764	- Increment loop depth.
		765	- Reset the ev_break status.
636	- Before the first iteration, call any pending watchers.	766	- Before the first iteration, call any pending watchers.
		767	LOOP:
637	* If EVFLAG_FORKCHECK was used, check for a fork.	768	- If EVFLAG_FORKCHECK was used, check for a fork.
638	- If a fork was detected (by any means), queue and call all fork watchers.	769	- If a fork was detected (by any means), queue and call all fork watchers.
639	- Queue and call all prepare watchers.	770	- Queue and call all prepare watchers.
		771	- If ev_break was called, goto FINISH.
640	- If we have been forked, detach and recreate the kernel state	772	- If we have been forked, detach and recreate the kernel state
641	as to not disturb the other process.	773	as to not disturb the other process.
642	- Update the kernel state with all outstanding changes.	774	- Update the kernel state with all outstanding changes.
643	- Update the "event loop time" (ev_now ()).	775	- Update the "event loop time" (ev_now ()).
644	- Calculate for how long to sleep or block, if at all	776	- Calculate for how long to sleep or block, if at all
645	(active idle watchers, EVLOOP_NONBLOCK or not having	777	(active idle watchers, EVRUN_NOWAIT or not having
646	any active watchers at all will result in not sleeping).	778	any active watchers at all will result in not sleeping).
647	- Sleep if the I/O and timer collect interval say so.	779	- Sleep if the I/O and timer collect interval say so.
		780	- Increment loop iteration counter.
648	- Block the process, waiting for any events.	781	- Block the process, waiting for any events.
649	- Queue all outstanding I/O (fd) events.	782	- Queue all outstanding I/O (fd) events.
650	- Update the "event loop time" (ev_now ()), and do time jump adjustments.	783	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
651	- Queue all outstanding timers.	784	- Queue all expired timers.
652	- Queue all outstanding periodics.	785	- Queue all expired periodics.
653	- Unless any events are pending now, queue all idle watchers.	786	- Queue all idle watchers with priority higher than that of pending events.
654	- Queue all check watchers.	787	- Queue all check watchers.
655	- Call all queued watchers in reverse order (i.e. check watchers first).	788	- Call all queued watchers in reverse order (i.e. check watchers first).
656	Signals and child watchers are implemented as I/O watchers, and will	789	Signals and child watchers are implemented as I/O watchers, and will
657	be handled here by queueing them when their watcher gets executed.	790	be handled here by queueing them when their watcher gets executed.
658	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	791	- If ev_break has been called, or EVRUN_ONCE or EVRUN_NOWAIT
659	were used, or there are no active watchers, return, otherwise	792	were used, or there are no active watchers, goto FINISH, otherwise
660	continue with step *.	793	continue with step LOOP.
		794	FINISH:
		795	- Reset the ev_break status iff it was EVBREAK_ONE.
		796	- Decrement the loop depth.
		797	- Return.
661		798
662	Example: Queue some jobs and then loop until no events are outstanding	799	Example: Queue some jobs and then loop until no events are outstanding
663	anymore.	800	anymore.
664		801
665	... queue jobs here, make sure they register event watchers as long	802	... queue jobs here, make sure they register event watchers as long
666	... as they still have work to do (even an idle watcher will do..)	803	... as they still have work to do (even an idle watcher will do..)
667	ev_loop (my_loop, 0);	804	ev_run (my_loop, 0);
668	... jobs done or somebody called unloop. yeah!	805	... jobs done or somebody called unloop. yeah!
669		806
670	=item ev_unloop (loop, how)	807	=item ev_break (loop, how)
671		808
672	Can be used to make a call to C<ev_loop> return early (but only after it	809	Can be used to make a call to C<ev_run> return early (but only after it
673	has processed all outstanding events). The C<how> argument must be either	810	has processed all outstanding events). The C<how> argument must be either
674	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	811	C<EVBREAK_ONE>, which will make the innermost C<ev_run> call return, or
675	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	812	C<EVBREAK_ALL>, which will make all nested C<ev_run> calls return.
676		813
677	This "unloop state" will be cleared when entering C<ev_loop> again.	814	This "unloop state" will be cleared when entering C<ev_run> again.
		815
		816	It is safe to call C<ev_break> from outside any C<ev_run> calls. ##TODO##
678		817
679	=item ev_ref (loop)	818	=item ev_ref (loop)
680		819
681	=item ev_unref (loop)	820	=item ev_unref (loop)
682		821
683	Ref/unref can be used to add or remove a reference count on the event	822	Ref/unref can be used to add or remove a reference count on the event
684	loop: Every watcher keeps one reference, and as long as the reference	823	loop: Every watcher keeps one reference, and as long as the reference
685	count is nonzero, C<ev_loop> will not return on its own. If you have	824	count is nonzero, C<ev_run> will not return on its own.
686	a watcher you never unregister that should not keep C<ev_loop> from	825
687	returning, ev_unref() after starting, and ev_ref() before stopping it. For	826	This is useful when you have a watcher that you never intend to
		827	unregister, but that nevertheless should not keep C<ev_run> from
		828	returning. In such a case, call C<ev_unref> after starting, and C<ev_ref>
		829	before stopping it.
		830
688	example, libev itself uses this for its internal signal pipe: It is not	831	As an example, libev itself uses this for its internal signal pipe: It
689	visible to the libev user and should not keep C<ev_loop> from exiting if	832	is not visible to the libev user and should not keep C<ev_run> from
690	no event watchers registered by it are active. It is also an excellent	833	exiting if no event watchers registered by it are active. It is also an
691	way to do this for generic recurring timers or from within third-party	834	excellent way to do this for generic recurring timers or from within
692	libraries. Just remember to I<unref after start> and I<ref before stop>	835	third-party libraries. Just remember to I<unref after start> and I<ref
693	(but only if the watcher wasn't active before, or was active before,	836	before stop> (but only if the watcher wasn't active before, or was active
694	respectively).	837	before, respectively. Note also that libev might stop watchers itself
		838	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		839	in the callback).
695		840
696	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	841	Example: Create a signal watcher, but keep it from keeping C<ev_run>
697	running when nothing else is active.	842	running when nothing else is active.
698		843
699	struct ev_signal exitsig;	844	ev_signal exitsig;
700	ev_signal_init (&exitsig, sig_cb, SIGINT);	845	ev_signal_init (&exitsig, sig_cb, SIGINT);
701	ev_signal_start (loop, &exitsig);	846	ev_signal_start (loop, &exitsig);
702	evf_unref (loop);	847	evf_unref (loop);
703		848
704	Example: For some weird reason, unregister the above signal handler again.	849	Example: For some weird reason, unregister the above signal handler again.
…		…
718	Setting these to a higher value (the C<interval> I<must> be >= C<0>)	863	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
719	allows libev to delay invocation of I/O and timer/periodic callbacks	864	allows libev to delay invocation of I/O and timer/periodic callbacks
720	to increase efficiency of loop iterations (or to increase power-saving	865	to increase efficiency of loop iterations (or to increase power-saving
721	opportunities).	866	opportunities).
722		867
723	The background is that sometimes your program runs just fast enough to	868	The idea is that sometimes your program runs just fast enough to handle
724	handle one (or very few) event(s) per loop iteration. While this makes	869	one (or very few) event(s) per loop iteration. While this makes the
725	the program responsive, it also wastes a lot of CPU time to poll for new	870	program responsive, it also wastes a lot of CPU time to poll for new
726	events, especially with backends like C<select ()> which have a high	871	events, especially with backends like C<select ()> which have a high
727	overhead for the actual polling but can deliver many events at once.	872	overhead for the actual polling but can deliver many events at once.
728		873
729	By setting a higher I<io collect interval> you allow libev to spend more	874	By setting a higher I<io collect interval> you allow libev to spend more
730	time collecting I/O events, so you can handle more events per iteration,	875	time collecting I/O events, so you can handle more events per iteration,
731	at the cost of increasing latency. Timeouts (both C<ev_periodic> and	876	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
732	C<ev_timer>) will be not affected. Setting this to a non-null value will	877	C<ev_timer>) will be not affected. Setting this to a non-null value will
733	introduce an additional C<ev_sleep ()> call into most loop iterations.	878	introduce an additional C<ev_sleep ()> call into most loop iterations. The
		879	sleep time ensures that libev will not poll for I/O events more often then
		880	once per this interval, on average.
734		881
735	Likewise, by setting a higher I<timeout collect interval> you allow libev	882	Likewise, by setting a higher I<timeout collect interval> you allow libev
736	to spend more time collecting timeouts, at the expense of increased	883	to spend more time collecting timeouts, at the expense of increased
737	latency (the watcher callback will be called later). C<ev_io> watchers	884	latency/jitter/inexactness (the watcher callback will be called
738	will not be affected. Setting this to a non-null value will not introduce	885	later). C<ev_io> watchers will not be affected. Setting this to a non-null
739	any overhead in libev.	886	value will not introduce any overhead in libev.
740		887
741	Many (busy) programs can usually benefit by setting the I/O collect	888	Many (busy) programs can usually benefit by setting the I/O collect
742	interval to a value near C<0.1> or so, which is often enough for	889	interval to a value near C<0.1> or so, which is often enough for
743	interactive servers (of course not for games), likewise for timeouts. It	890	interactive servers (of course not for games), likewise for timeouts. It
744	usually doesn't make much sense to set it to a lower value than C<0.01>,	891	usually doesn't make much sense to set it to a lower value than C<0.01>,
745	as this approaches the timing granularity of most systems.	892	as this approaches the timing granularity of most systems. Note that if
		893	you do transactions with the outside world and you can't increase the
		894	parallelity, then this setting will limit your transaction rate (if you
		895	need to poll once per transaction and the I/O collect interval is 0.01,
		896	then you can't do more than 100 transactions per second).
746		897
747	Setting the I<timeout collect interval> can improve the opportunity for	898	Setting the I<timeout collect interval> can improve the opportunity for
748	saving power, as the program will "bundle" timer callback invocations that	899	saving power, as the program will "bundle" timer callback invocations that
749	are "near" in time together, by delaying some, thus reducing the number of	900	are "near" in time together, by delaying some, thus reducing the number of
750	times the process sleeps and wakes up again. Another useful technique to	901	times the process sleeps and wakes up again. Another useful technique to
751	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure	902	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
752	they fire on, say, one-second boundaries only.	903	they fire on, say, one-second boundaries only.
753		904
		905	Example: we only need 0.1s timeout granularity, and we wish not to poll
		906	more often than 100 times per second:
		907
		908	ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
		909	ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
		910
		911	=item ev_invoke_pending (loop)
		912
		913	This call will simply invoke all pending watchers while resetting their
		914	pending state. Normally, C<ev_run> does this automatically when required,
		915	but when overriding the invoke callback this call comes handy. This
		916	function can be invoked from a watcher - this can be useful for example
		917	when you want to do some lengthy calculation and want to pass further
		918	event handling to another thread (you still have to make sure only one
		919	thread executes within C<ev_invoke_pending> or C<ev_run> of course).
		920
		921	=item int ev_pending_count (loop)
		922
		923	Returns the number of pending watchers - zero indicates that no watchers
		924	are pending.
		925
		926	=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
		927
		928	This overrides the invoke pending functionality of the loop: Instead of
		929	invoking all pending watchers when there are any, C<ev_run> will call
		930	this callback instead. This is useful, for example, when you want to
		931	invoke the actual watchers inside another context (another thread etc.).
		932
		933	If you want to reset the callback, use C<ev_invoke_pending> as new
		934	callback.
		935
		936	=item ev_set_loop_release_cb (loop, void (*release)(EV_P), void (*acquire)(EV_P))
		937
		938	Sometimes you want to share the same loop between multiple threads. This
		939	can be done relatively simply by putting mutex_lock/unlock calls around
		940	each call to a libev function.
		941
		942	However, C<ev_run> can run an indefinite time, so it is not feasible
		943	to wait for it to return. One way around this is to wake up the event
		944	loop via C<ev_break> and C<av_async_send>, another way is to set these
		945	I<release> and I<acquire> callbacks on the loop.
		946
		947	When set, then C<release> will be called just before the thread is
		948	suspended waiting for new events, and C<acquire> is called just
		949	afterwards.
		950
		951	Ideally, C<release> will just call your mutex_unlock function, and
		952	C<acquire> will just call the mutex_lock function again.
		953
		954	While event loop modifications are allowed between invocations of
		955	C<release> and C<acquire> (that's their only purpose after all), no
		956	modifications done will affect the event loop, i.e. adding watchers will
		957	have no effect on the set of file descriptors being watched, or the time
		958	waited. Use an C<ev_async> watcher to wake up C<ev_run> when you want it
		959	to take note of any changes you made.
		960
		961	In theory, threads executing C<ev_run> will be async-cancel safe between
		962	invocations of C<release> and C<acquire>.
		963
		964	See also the locking example in the C<THREADS> section later in this
		965	document.
		966
		967	=item ev_set_userdata (loop, void *data)
		968
		969	=item ev_userdata (loop)
		970
		971	Set and retrieve a single C<void *> associated with a loop. When
		972	C<ev_set_userdata> has never been called, then C<ev_userdata> returns
		973	C<0.>
		974
		975	These two functions can be used to associate arbitrary data with a loop,
		976	and are intended solely for the C<invoke_pending_cb>, C<release> and
		977	C<acquire> callbacks described above, but of course can be (ab-)used for
		978	any other purpose as well.
		979
754	=item ev_loop_verify (loop)	980	=item ev_verify (loop)
755		981
756	This function only does something when C<EV_VERIFY> support has been	982	This function only does something when C<EV_VERIFY> support has been
757	compiled in. It tries to go through all internal structures and checks	983	compiled in, which is the default for non-minimal builds. It tries to go
758	them for validity. If anything is found to be inconsistent, it will print	984	through all internal structures and checks them for validity. If anything
759	an error message to standard error and call C<abort ()>.	985	is found to be inconsistent, it will print an error message to standard
		986	error and call C<abort ()>.
760		987
761	This can be used to catch bugs inside libev itself: under normal	988	This can be used to catch bugs inside libev itself: under normal
762	circumstances, this function will never abort as of course libev keeps its	989	circumstances, this function will never abort as of course libev keeps its
763	data structures consistent.	990	data structures consistent.
764		991
765	=back	992	=back
766		993
767		994
768	=head1 ANATOMY OF A WATCHER	995	=head1 ANATOMY OF A WATCHER
769		996
		997	In the following description, uppercase C<TYPE> in names stands for the
		998	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		999	watchers and C<ev_io_start> for I/O watchers.
		1000
770	A watcher is a structure that you create and register to record your	1001	A watcher is an opaque structure that you allocate and register to record
771	interest in some event. For instance, if you want to wait for STDIN to	1002	your interest in some event. To make a concrete example, imagine you want
772	become readable, you would create an C<ev_io> watcher for that:	1003	to wait for STDIN to become readable, you would create an C<ev_io> watcher
		1004	for that:
773		1005
774	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1006	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
775	{	1007	{
776	ev_io_stop (w);	1008	ev_io_stop (w);
777	ev_unloop (loop, EVUNLOOP_ALL);	1009	ev_break (loop, EVBREAK_ALL);
778	}	1010	}
779		1011
780	struct ev_loop *loop = ev_default_loop (0);	1012	struct ev_loop *loop = ev_default_loop (0);
		1013
781	struct ev_io stdin_watcher;	1014	ev_io stdin_watcher;
		1015
782	ev_init (&stdin_watcher, my_cb);	1016	ev_init (&stdin_watcher, my_cb);
783	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	1017	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
784	ev_io_start (loop, &stdin_watcher);	1018	ev_io_start (loop, &stdin_watcher);
		1019
785	ev_loop (loop, 0);	1020	ev_run (loop, 0);
786		1021
787	As you can see, you are responsible for allocating the memory for your	1022	As you can see, you are responsible for allocating the memory for your
788	watcher structures (and it is usually a bad idea to do this on the stack,	1023	watcher structures (and it is I<usually> a bad idea to do this on the
789	although this can sometimes be quite valid).	1024	stack).
790		1025
		1026	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		1027	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
		1028
791	Each watcher structure must be initialised by a call to C<ev_init	1029	Each watcher structure must be initialised by a call to C<ev_init (watcher
792	(watcher *, callback)>, which expects a callback to be provided. This	1030	*, callback)>, which expects a callback to be provided. This callback is
793	callback gets invoked each time the event occurs (or, in the case of I/O	1031	invoked each time the event occurs (or, in the case of I/O watchers, each
794	watchers, each time the event loop detects that the file descriptor given	1032	time the event loop detects that the file descriptor given is readable
795	is readable and/or writable).	1033	and/or writable).
796		1034
797	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	1035	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
798	with arguments specific to this watcher type. There is also a macro	1036	macro to configure it, with arguments specific to the watcher type. There
799	to combine initialisation and setting in one call: C<< ev_<type>_init	1037	is also a macro to combine initialisation and setting in one call: C<<
800	(watcher *, callback, ...) >>.	1038	ev_TYPE_init (watcher *, callback, ...) >>.
801		1039
802	To make the watcher actually watch out for events, you have to start it	1040	To make the watcher actually watch out for events, you have to start it
803	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	1041	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
804	*) >>), and you can stop watching for events at any time by calling the	1042	*) >>), and you can stop watching for events at any time by calling the
805	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	1043	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
806		1044
807	As long as your watcher is active (has been started but not stopped) you	1045	As long as your watcher is active (has been started but not stopped) you
808	must not touch the values stored in it. Most specifically you must never	1046	must not touch the values stored in it. Most specifically you must never
809	reinitialise it or call its C<set> macro.	1047	reinitialise it or call its C<ev_TYPE_set> macro.
810		1048
811	Each and every callback receives the event loop pointer as first, the	1049	Each and every callback receives the event loop pointer as first, the
812	registered watcher structure as second, and a bitset of received events as	1050	registered watcher structure as second, and a bitset of received events as
813	third argument.	1051	third argument.
814		1052
…		…
823	=item C<EV_WRITE>	1061	=item C<EV_WRITE>
824		1062
825	The file descriptor in the C<ev_io> watcher has become readable and/or	1063	The file descriptor in the C<ev_io> watcher has become readable and/or
826	writable.	1064	writable.
827		1065
828	=item C<EV_TIMEOUT>	1066	=item C<EV_TIMER>
829		1067
830	The C<ev_timer> watcher has timed out.	1068	The C<ev_timer> watcher has timed out.
831		1069
832	=item C<EV_PERIODIC>	1070	=item C<EV_PERIODIC>
833		1071
…		…
851		1089
852	=item C<EV_PREPARE>	1090	=item C<EV_PREPARE>
853		1091
854	=item C<EV_CHECK>	1092	=item C<EV_CHECK>
855		1093
856	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts	1094	All C<ev_prepare> watchers are invoked just I<before> C<ev_run> starts
857	to gather new events, and all C<ev_check> watchers are invoked just after	1095	to gather new events, and all C<ev_check> watchers are invoked just after
858	C<ev_loop> has gathered them, but before it invokes any callbacks for any	1096	C<ev_run> has gathered them, but before it invokes any callbacks for any
859	received events. Callbacks of both watcher types can start and stop as	1097	received events. Callbacks of both watcher types can start and stop as
860	many watchers as they want, and all of them will be taken into account	1098	many watchers as they want, and all of them will be taken into account
861	(for example, a C<ev_prepare> watcher might start an idle watcher to keep	1099	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
862	C<ev_loop> from blocking).	1100	C<ev_run> from blocking).
863		1101
864	=item C<EV_EMBED>	1102	=item C<EV_EMBED>
865		1103
866	The embedded event loop specified in the C<ev_embed> watcher needs attention.	1104	The embedded event loop specified in the C<ev_embed> watcher needs attention.
867		1105
…		…
871	C<ev_fork>).	1109	C<ev_fork>).
872		1110
873	=item C<EV_ASYNC>	1111	=item C<EV_ASYNC>
874		1112
875	The given async watcher has been asynchronously notified (see C<ev_async>).	1113	The given async watcher has been asynchronously notified (see C<ev_async>).
		1114
		1115	=item C<EV_CUSTOM>
		1116
		1117	Not ever sent (or otherwise used) by libev itself, but can be freely used
		1118	by libev users to signal watchers (e.g. via C<ev_feed_event>).
876		1119
877	=item C<EV_ERROR>	1120	=item C<EV_ERROR>
878		1121
879	An unspecified error has occurred, the watcher has been stopped. This might	1122	An unspecified error has occurred, the watcher has been stopped. This might
880	happen because the watcher could not be properly started because libev	1123	happen because the watcher could not be properly started because libev
881	ran out of memory, a file descriptor was found to be closed or any other	1124	ran out of memory, a file descriptor was found to be closed or any other
		1125	problem. Libev considers these application bugs.
		1126
882	problem. You best act on it by reporting the problem and somehow coping	1127	You best act on it by reporting the problem and somehow coping with the
883	with the watcher being stopped.	1128	watcher being stopped. Note that well-written programs should not receive
		1129	an error ever, so when your watcher receives it, this usually indicates a
		1130	bug in your program.
884		1131
885	Libev will usually signal a few "dummy" events together with an error,	1132	Libev will usually signal a few "dummy" events together with an error, for
886	for example it might indicate that a fd is readable or writable, and if	1133	example it might indicate that a fd is readable or writable, and if your
887	your callbacks is well-written it can just attempt the operation and cope	1134	callbacks is well-written it can just attempt the operation and cope with
888	with the error from read() or write(). This will not work in multi-threaded	1135	the error from read() or write(). This will not work in multi-threaded
889	programs, though, so beware.	1136	programs, though, as the fd could already be closed and reused for another
		1137	thing, so beware.
890		1138
891	=back	1139	=back
892		1140
		1141	=head2 WATCHER STATES
		1142
		1143	There are various watcher states mentioned throughout this manual -
		1144	active, pending and so on. In this section these states and the rules to
		1145	transition between them will be described in more detail - and while these
		1146	rules might look complicated, they usually do "the right thing".
		1147
		1148	=over 4
		1149
		1150	=item initialiased
		1151
		1152	Before a watcher can be registered with the event looop it has to be
		1153	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to
		1154	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.
		1155
		1156	In this state it is simply some block of memory that is suitable for use
		1157	in an event loop. It can be moved around, freed, reused etc. at will.
		1158
		1159	=item started/running/active
		1160
		1161	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes
		1162	property of the event loop, and is actively waiting for events. While in
		1163	this state it cannot be accessed (except in a few documented ways), moved,
		1164	freed or anything else - the only legal thing is to keep a pointer to it,
		1165	and call libev functions on it that are documented to work on active watchers.
		1166
		1167	=item pending
		1168
		1169	If a watcher is active and libev determines that an event it is interested
		1170	in has occurred (such as a timer expiring), it will become pending. It will
		1171	stay in this pending state until either it is stopped or its callback is
		1172	about to be invoked, so it is not normally pending inside the watcher
		1173	callback.
		1174
		1175	The watcher might or might not be active while it is pending (for example,
		1176	an expired non-repeating timer can be pending but no longer active). If it
		1177	is stopped, it can be freely accessed (e.g. by calling C<ev_TYPE_set>),
		1178	but it is still property of the event loop at this time, so cannot be
		1179	moved, freed or reused. And if it is active the rules described in the
		1180	previous item still apply.
		1181
		1182	It is also possible to feed an event on a watcher that is not active (e.g.
		1183	via C<ev_feed_event>), in which case it becomes pending without being
		1184	active.
		1185
		1186	=item stopped
		1187
		1188	A watcher can be stopped implicitly by libev (in which case it might still
		1189	be pending), or explicitly by calling its C<ev_TYPE_stop> function. The
		1190	latter will clear any pending state the watcher might be in, regardless
		1191	of whether it was active or not, so stopping a watcher explicitly before
		1192	freeing it is often a good idea.
		1193
		1194	While stopped (and not pending) the watcher is essentially in the
		1195	initialised state, that is it can be reused, moved, modified in any way
		1196	you wish.
		1197
		1198	=back
		1199
893	=head2 GENERIC WATCHER FUNCTIONS	1200	=head2 GENERIC WATCHER FUNCTIONS
894
895	In the following description, C<TYPE> stands for the watcher type,
896	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
897		1201
898	=over 4	1202	=over 4
899		1203
900	=item C<ev_init> (ev_TYPE *watcher, callback)	1204	=item C<ev_init> (ev_TYPE *watcher, callback)
901		1205
…		…
907	which rolls both calls into one.	1211	which rolls both calls into one.
908		1212
909	You can reinitialise a watcher at any time as long as it has been stopped	1213	You can reinitialise a watcher at any time as long as it has been stopped
910	(or never started) and there are no pending events outstanding.	1214	(or never started) and there are no pending events outstanding.
911		1215
912	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	1216	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
913	int revents)>.	1217	int revents)>.
914		1218
		1219	Example: Initialise an C<ev_io> watcher in two steps.
		1220
		1221	ev_io w;
		1222	ev_init (&w, my_cb);
		1223	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1224
915	=item C<ev_TYPE_set> (ev_TYPE *, [args])	1225	=item C<ev_TYPE_set> (ev_TYPE *watcher, [args])
916		1226
917	This macro initialises the type-specific parts of a watcher. You need to	1227	This macro initialises the type-specific parts of a watcher. You need to
918	call C<ev_init> at least once before you call this macro, but you can	1228	call C<ev_init> at least once before you call this macro, but you can
919	call C<ev_TYPE_set> any number of times. You must not, however, call this	1229	call C<ev_TYPE_set> any number of times. You must not, however, call this
920	macro on a watcher that is active (it can be pending, however, which is a	1230	macro on a watcher that is active (it can be pending, however, which is a
921	difference to the C<ev_init> macro).	1231	difference to the C<ev_init> macro).
922		1232
923	Although some watcher types do not have type-specific arguments	1233	Although some watcher types do not have type-specific arguments
924	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	1234	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
925		1235
		1236	See C<ev_init>, above, for an example.
		1237
926	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	1238	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
927		1239
928	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	1240	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
929	calls into a single call. This is the most convenient method to initialise	1241	calls into a single call. This is the most convenient method to initialise
930	a watcher. The same limitations apply, of course.	1242	a watcher. The same limitations apply, of course.
931		1243
		1244	Example: Initialise and set an C<ev_io> watcher in one step.
		1245
		1246	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1247
932	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	1248	=item C<ev_TYPE_start> (loop, ev_TYPE *watcher)
933		1249
934	Starts (activates) the given watcher. Only active watchers will receive	1250	Starts (activates) the given watcher. Only active watchers will receive
935	events. If the watcher is already active nothing will happen.	1251	events. If the watcher is already active nothing will happen.
936		1252
		1253	Example: Start the C<ev_io> watcher that is being abused as example in this
		1254	whole section.
		1255
		1256	ev_io_start (EV_DEFAULT_UC, &w);
		1257
937	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	1258	=item C<ev_TYPE_stop> (loop, ev_TYPE *watcher)
938		1259
939	Stops the given watcher again (if active) and clears the pending	1260	Stops the given watcher if active, and clears the pending status (whether
		1261	the watcher was active or not).
		1262
940	status. It is possible that stopped watchers are pending (for example,	1263	It is possible that stopped watchers are pending - for example,
941	non-repeating timers are being stopped when they become pending), but	1264	non-repeating timers are being stopped when they become pending - but
942	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	1265	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
943	you want to free or reuse the memory used by the watcher it is therefore a	1266	pending. If you want to free or reuse the memory used by the watcher it is
944	good idea to always call its C<ev_TYPE_stop> function.	1267	therefore a good idea to always call its C<ev_TYPE_stop> function.
945		1268
946	=item bool ev_is_active (ev_TYPE *watcher)	1269	=item bool ev_is_active (ev_TYPE *watcher)
947		1270
948	Returns a true value iff the watcher is active (i.e. it has been started	1271	Returns a true value iff the watcher is active (i.e. it has been started
949	and not yet been stopped). As long as a watcher is active you must not modify	1272	and not yet been stopped). As long as a watcher is active you must not modify
…		…
965	=item ev_cb_set (ev_TYPE *watcher, callback)	1288	=item ev_cb_set (ev_TYPE *watcher, callback)
966		1289
967	Change the callback. You can change the callback at virtually any time	1290	Change the callback. You can change the callback at virtually any time
968	(modulo threads).	1291	(modulo threads).
969		1292
970	=item ev_set_priority (ev_TYPE *watcher, priority)	1293	=item ev_set_priority (ev_TYPE *watcher, int priority)
971		1294
972	=item int ev_priority (ev_TYPE *watcher)	1295	=item int ev_priority (ev_TYPE *watcher)
973		1296
974	Set and query the priority of the watcher. The priority is a small	1297	Set and query the priority of the watcher. The priority is a small
975	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>	1298	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
976	(default: C<-2>). Pending watchers with higher priority will be invoked	1299	(default: C<-2>). Pending watchers with higher priority will be invoked
977	before watchers with lower priority, but priority will not keep watchers	1300	before watchers with lower priority, but priority will not keep watchers
978	from being executed (except for C<ev_idle> watchers).	1301	from being executed (except for C<ev_idle> watchers).
979		1302
980	This means that priorities are I<only> used for ordering callback
981	invocation after new events have been received. This is useful, for
982	example, to reduce latency after idling, or more often, to bind two
983	watchers on the same event and make sure one is called first.
984
985	If you need to suppress invocation when higher priority events are pending	1303	If you need to suppress invocation when higher priority events are pending
986	you need to look at C<ev_idle> watchers, which provide this functionality.	1304	you need to look at C<ev_idle> watchers, which provide this functionality.
987		1305
988	You I<must not> change the priority of a watcher as long as it is active or	1306	You I<must not> change the priority of a watcher as long as it is active or
989	pending.	1307	pending.
990		1308
		1309	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1310	fine, as long as you do not mind that the priority value you query might
		1311	or might not have been clamped to the valid range.
		1312
991	The default priority used by watchers when no priority has been set is	1313	The default priority used by watchers when no priority has been set is
992	always C<0>, which is supposed to not be too high and not be too low :).	1314	always C<0>, which is supposed to not be too high and not be too low :).
993		1315
994	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is	1316	See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
995	fine, as long as you do not mind that the priority value you query might	1317	priorities.
996	or might not have been adjusted to be within valid range.
997		1318
998	=item ev_invoke (loop, ev_TYPE *watcher, int revents)	1319	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
999		1320
1000	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither	1321	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
1001	C<loop> nor C<revents> need to be valid as long as the watcher callback	1322	C<loop> nor C<revents> need to be valid as long as the watcher callback
1002	can deal with that fact.	1323	can deal with that fact, as both are simply passed through to the
		1324	callback.
1003		1325
1004	=item int ev_clear_pending (loop, ev_TYPE *watcher)	1326	=item int ev_clear_pending (loop, ev_TYPE *watcher)
1005		1327
1006	If the watcher is pending, this function returns clears its pending status	1328	If the watcher is pending, this function clears its pending status and
1007	and returns its C<revents> bitset (as if its callback was invoked). If the	1329	returns its C<revents> bitset (as if its callback was invoked). If the
1008	watcher isn't pending it does nothing and returns C<0>.	1330	watcher isn't pending it does nothing and returns C<0>.
1009		1331
		1332	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1333	callback to be invoked, which can be accomplished with this function.
		1334
		1335	=item ev_feed_event (loop, ev_TYPE *watcher, int revents)
		1336
		1337	Feeds the given event set into the event loop, as if the specified event
		1338	had happened for the specified watcher (which must be a pointer to an
		1339	initialised but not necessarily started event watcher). Obviously you must
		1340	not free the watcher as long as it has pending events.
		1341
		1342	Stopping the watcher, letting libev invoke it, or calling
		1343	C<ev_clear_pending> will clear the pending event, even if the watcher was
		1344	not started in the first place.
		1345
		1346	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
		1347	functions that do not need a watcher.
		1348
1010	=back	1349	=back
1011		1350
1012		1351
1013	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1352	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
1014		1353
1015	Each watcher has, by default, a member C<void *data> that you can change	1354	Each watcher has, by default, a member C<void *data> that you can change
1016	and read at any time, libev will completely ignore it. This can be used	1355	and read at any time: libev will completely ignore it. This can be used
1017	to associate arbitrary data with your watcher. If you need more data and	1356	to associate arbitrary data with your watcher. If you need more data and
1018	don't want to allocate memory and store a pointer to it in that data	1357	don't want to allocate memory and store a pointer to it in that data
1019	member, you can also "subclass" the watcher type and provide your own	1358	member, you can also "subclass" the watcher type and provide your own
1020	data:	1359	data:
1021		1360
1022	struct my_io	1361	struct my_io
1023	{	1362	{
1024	struct ev_io io;	1363	ev_io io;
1025	int otherfd;	1364	int otherfd;
1026	void *somedata;	1365	void *somedata;
1027	struct whatever *mostinteresting;	1366	struct whatever *mostinteresting;
1028	};	1367	};
1029		1368
…		…
1032	ev_io_init (&w.io, my_cb, fd, EV_READ);	1371	ev_io_init (&w.io, my_cb, fd, EV_READ);
1033		1372
1034	And since your callback will be called with a pointer to the watcher, you	1373	And since your callback will be called with a pointer to the watcher, you
1035	can cast it back to your own type:	1374	can cast it back to your own type:
1036		1375
1037	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1376	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
1038	{	1377	{
1039	struct my_io *w = (struct my_io *)w_;	1378	struct my_io *w = (struct my_io *)w_;
1040	...	1379	...
1041	}	1380	}
1042		1381
…		…
1053	ev_timer t2;	1392	ev_timer t2;
1054	}	1393	}
1055		1394
1056	In this case getting the pointer to C<my_biggy> is a bit more	1395	In this case getting the pointer to C<my_biggy> is a bit more
1057	complicated: Either you store the address of your C<my_biggy> struct	1396	complicated: Either you store the address of your C<my_biggy> struct
1058	in the C<data> member of the watcher, or you need to use some pointer	1397	in the C<data> member of the watcher (for woozies), or you need to use
1059	arithmetic using C<offsetof> inside your watchers:	1398	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1399	programmers):
1060		1400
1061	#include <stddef.h>	1401	#include <stddef.h>
1062		1402
1063	static void	1403	static void
1064	t1_cb (EV_P_ struct ev_timer *w, int revents)	1404	t1_cb (EV_P_ ev_timer *w, int revents)
1065	{	1405	{
1066	struct my_biggy big = (struct my_biggy *	1406	struct my_biggy big = (struct my_biggy *)
1067	(((char *)w) - offsetof (struct my_biggy, t1));	1407	(((char *)w) - offsetof (struct my_biggy, t1));
1068	}	1408	}
1069		1409
1070	static void	1410	static void
1071	t2_cb (EV_P_ struct ev_timer *w, int revents)	1411	t2_cb (EV_P_ ev_timer *w, int revents)
1072	{	1412	{
1073	struct my_biggy big = (struct my_biggy *	1413	struct my_biggy big = (struct my_biggy *)
1074	(((char *)w) - offsetof (struct my_biggy, t2));	1414	(((char *)w) - offsetof (struct my_biggy, t2));
1075	}	1415	}
		1416
		1417	=head2 WATCHER PRIORITY MODELS
		1418
		1419	Many event loops support I<watcher priorities>, which are usually small
		1420	integers that influence the ordering of event callback invocation
		1421	between watchers in some way, all else being equal.
		1422
		1423	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1424	description for the more technical details such as the actual priority
		1425	range.
		1426
		1427	There are two common ways how these these priorities are being interpreted
		1428	by event loops:
		1429
		1430	In the more common lock-out model, higher priorities "lock out" invocation
		1431	of lower priority watchers, which means as long as higher priority
		1432	watchers receive events, lower priority watchers are not being invoked.
		1433
		1434	The less common only-for-ordering model uses priorities solely to order
		1435	callback invocation within a single event loop iteration: Higher priority
		1436	watchers are invoked before lower priority ones, but they all get invoked
		1437	before polling for new events.
		1438
		1439	Libev uses the second (only-for-ordering) model for all its watchers
		1440	except for idle watchers (which use the lock-out model).
		1441
		1442	The rationale behind this is that implementing the lock-out model for
		1443	watchers is not well supported by most kernel interfaces, and most event
		1444	libraries will just poll for the same events again and again as long as
		1445	their callbacks have not been executed, which is very inefficient in the
		1446	common case of one high-priority watcher locking out a mass of lower
		1447	priority ones.
		1448
		1449	Static (ordering) priorities are most useful when you have two or more
		1450	watchers handling the same resource: a typical usage example is having an
		1451	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1452	timeouts. Under load, data might be received while the program handles
		1453	other jobs, but since timers normally get invoked first, the timeout
		1454	handler will be executed before checking for data. In that case, giving
		1455	the timer a lower priority than the I/O watcher ensures that I/O will be
		1456	handled first even under adverse conditions (which is usually, but not
		1457	always, what you want).
		1458
		1459	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1460	will only be executed when no same or higher priority watchers have
		1461	received events, they can be used to implement the "lock-out" model when
		1462	required.
		1463
		1464	For example, to emulate how many other event libraries handle priorities,
		1465	you can associate an C<ev_idle> watcher to each such watcher, and in
		1466	the normal watcher callback, you just start the idle watcher. The real
		1467	processing is done in the idle watcher callback. This causes libev to
		1468	continuously poll and process kernel event data for the watcher, but when
		1469	the lock-out case is known to be rare (which in turn is rare :), this is
		1470	workable.
		1471
		1472	Usually, however, the lock-out model implemented that way will perform
		1473	miserably under the type of load it was designed to handle. In that case,
		1474	it might be preferable to stop the real watcher before starting the
		1475	idle watcher, so the kernel will not have to process the event in case
		1476	the actual processing will be delayed for considerable time.
		1477
		1478	Here is an example of an I/O watcher that should run at a strictly lower
		1479	priority than the default, and which should only process data when no
		1480	other events are pending:
		1481
		1482	ev_idle idle; // actual processing watcher
		1483	ev_io io; // actual event watcher
		1484
		1485	static void
		1486	io_cb (EV_P_ ev_io *w, int revents)
		1487	{
		1488	// stop the I/O watcher, we received the event, but
		1489	// are not yet ready to handle it.
		1490	ev_io_stop (EV_A_ w);
		1491
		1492	// start the idle watcher to handle the actual event.
		1493	// it will not be executed as long as other watchers
		1494	// with the default priority are receiving events.
		1495	ev_idle_start (EV_A_ &idle);
		1496	}
		1497
		1498	static void
		1499	idle_cb (EV_P_ ev_idle *w, int revents)
		1500	{
		1501	// actual processing
		1502	read (STDIN_FILENO, ...);
		1503
		1504	// have to start the I/O watcher again, as
		1505	// we have handled the event
		1506	ev_io_start (EV_P_ &io);
		1507	}
		1508
		1509	// initialisation
		1510	ev_idle_init (&idle, idle_cb);
		1511	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1512	ev_io_start (EV_DEFAULT_ &io);
		1513
		1514	In the "real" world, it might also be beneficial to start a timer, so that
		1515	low-priority connections can not be locked out forever under load. This
		1516	enables your program to keep a lower latency for important connections
		1517	during short periods of high load, while not completely locking out less
		1518	important ones.
1076		1519
1077		1520
1078	=head1 WATCHER TYPES	1521	=head1 WATCHER TYPES
1079		1522
1080	This section describes each watcher in detail, but will not repeat	1523	This section describes each watcher in detail, but will not repeat
…		…
1104	In general you can register as many read and/or write event watchers per	1547	In general you can register as many read and/or write event watchers per
1105	fd as you want (as long as you don't confuse yourself). Setting all file	1548	fd as you want (as long as you don't confuse yourself). Setting all file
1106	descriptors to non-blocking mode is also usually a good idea (but not	1549	descriptors to non-blocking mode is also usually a good idea (but not
1107	required if you know what you are doing).	1550	required if you know what you are doing).
1108		1551
1109	If you must do this, then force the use of a known-to-be-good backend	1552	If you cannot use non-blocking mode, then force the use of a
1110	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1553	known-to-be-good backend (at the time of this writing, this includes only
1111	C<EVBACKEND_POLL>).	1554	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
		1555	descriptors for which non-blocking operation makes no sense (such as
		1556	files) - libev doesn't guarantee any specific behaviour in that case.
1112		1557
1113	Another thing you have to watch out for is that it is quite easy to	1558	Another thing you have to watch out for is that it is quite easy to
1114	receive "spurious" readiness notifications, that is your callback might	1559	receive "spurious" readiness notifications, that is your callback might
1115	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1560	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
1116	because there is no data. Not only are some backends known to create a	1561	because there is no data. Not only are some backends known to create a
1117	lot of those (for example Solaris ports), it is very easy to get into	1562	lot of those (for example Solaris ports), it is very easy to get into
1118	this situation even with a relatively standard program structure. Thus	1563	this situation even with a relatively standard program structure. Thus
1119	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1564	it is best to always use non-blocking I/O: An extra C<read>(2) returning
1120	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1565	C<EAGAIN> is far preferable to a program hanging until some data arrives.
1121		1566
1122	If you cannot run the fd in non-blocking mode (for example you should not	1567	If you cannot run the fd in non-blocking mode (for example you should
1123	play around with an Xlib connection), then you have to separately re-test	1568	not play around with an Xlib connection), then you have to separately
1124	whether a file descriptor is really ready with a known-to-be good interface	1569	re-test whether a file descriptor is really ready with a known-to-be good
1125	such as poll (fortunately in our Xlib example, Xlib already does this on	1570	interface such as poll (fortunately in our Xlib example, Xlib already
1126	its own, so its quite safe to use).	1571	does this on its own, so its quite safe to use). Some people additionally
		1572	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1573	indefinitely.
		1574
		1575	But really, best use non-blocking mode.
1127		1576
1128	=head3 The special problem of disappearing file descriptors	1577	=head3 The special problem of disappearing file descriptors
1129		1578
1130	Some backends (e.g. kqueue, epoll) need to be told about closing a file	1579	Some backends (e.g. kqueue, epoll) need to be told about closing a file
1131	descriptor (either by calling C<close> explicitly or by any other means,	1580	descriptor (either due to calling C<close> explicitly or any other means,
1132	such as C<dup>). The reason is that you register interest in some file	1581	such as C<dup2>). The reason is that you register interest in some file
1133	descriptor, but when it goes away, the operating system will silently drop	1582	descriptor, but when it goes away, the operating system will silently drop
1134	this interest. If another file descriptor with the same number then is	1583	this interest. If another file descriptor with the same number then is
1135	registered with libev, there is no efficient way to see that this is, in	1584	registered with libev, there is no efficient way to see that this is, in
1136	fact, a different file descriptor.	1585	fact, a different file descriptor.
1137		1586
…		…
1168	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1617	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
1169	C<EVBACKEND_POLL>.	1618	C<EVBACKEND_POLL>.
1170		1619
1171	=head3 The special problem of SIGPIPE	1620	=head3 The special problem of SIGPIPE
1172		1621
1173	While not really specific to libev, it is easy to forget about SIGPIPE:	1622	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
1174	when writing to a pipe whose other end has been closed, your program gets	1623	when writing to a pipe whose other end has been closed, your program gets
1175	send a SIGPIPE, which, by default, aborts your program. For most programs	1624	sent a SIGPIPE, which, by default, aborts your program. For most programs
1176	this is sensible behaviour, for daemons, this is usually undesirable.	1625	this is sensible behaviour, for daemons, this is usually undesirable.
1177		1626
1178	So when you encounter spurious, unexplained daemon exits, make sure you	1627	So when you encounter spurious, unexplained daemon exits, make sure you
1179	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon	1628	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
1180	somewhere, as that would have given you a big clue).	1629	somewhere, as that would have given you a big clue).
1181		1630
		1631	=head3 The special problem of accept()ing when you can't
		1632
		1633	Many implementations of the POSIX C<accept> function (for example,
		1634	found in post-2004 Linux) have the peculiar behaviour of not removing a
		1635	connection from the pending queue in all error cases.
		1636
		1637	For example, larger servers often run out of file descriptors (because
		1638	of resource limits), causing C<accept> to fail with C<ENFILE> but not
		1639	rejecting the connection, leading to libev signalling readiness on
		1640	the next iteration again (the connection still exists after all), and
		1641	typically causing the program to loop at 100% CPU usage.
		1642
		1643	Unfortunately, the set of errors that cause this issue differs between
		1644	operating systems, there is usually little the app can do to remedy the
		1645	situation, and no known thread-safe method of removing the connection to
		1646	cope with overload is known (to me).
		1647
		1648	One of the easiest ways to handle this situation is to just ignore it
		1649	- when the program encounters an overload, it will just loop until the
		1650	situation is over. While this is a form of busy waiting, no OS offers an
		1651	event-based way to handle this situation, so it's the best one can do.
		1652
		1653	A better way to handle the situation is to log any errors other than
		1654	C<EAGAIN> and C<EWOULDBLOCK>, making sure not to flood the log with such
		1655	messages, and continue as usual, which at least gives the user an idea of
		1656	what could be wrong ("raise the ulimit!"). For extra points one could stop
		1657	the C<ev_io> watcher on the listening fd "for a while", which reduces CPU
		1658	usage.
		1659
		1660	If your program is single-threaded, then you could also keep a dummy file
		1661	descriptor for overload situations (e.g. by opening F</dev/null>), and
		1662	when you run into C<ENFILE> or C<EMFILE>, close it, run C<accept>,
		1663	close that fd, and create a new dummy fd. This will gracefully refuse
		1664	clients under typical overload conditions.
		1665
		1666	The last way to handle it is to simply log the error and C<exit>, as
		1667	is often done with C<malloc> failures, but this results in an easy
		1668	opportunity for a DoS attack.
1182		1669
1183	=head3 Watcher-Specific Functions	1670	=head3 Watcher-Specific Functions
1184		1671
1185	=over 4	1672	=over 4
1186		1673
1187	=item ev_io_init (ev_io *, callback, int fd, int events)	1674	=item ev_io_init (ev_io *, callback, int fd, int events)
1188		1675
1189	=item ev_io_set (ev_io *, int fd, int events)	1676	=item ev_io_set (ev_io *, int fd, int events)
1190		1677
1191	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1678	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
1192	receive events for and events is either C<EV_READ>, C<EV_WRITE> or	1679	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
1193	C<EV_READ | EV_WRITE> to receive the given events.	1680	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
1194		1681
1195	=item int fd [read-only]	1682	=item int fd [read-only]
1196		1683
1197	The file descriptor being watched.	1684	The file descriptor being watched.
1198		1685
…		…
1207	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1694	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
1208	readable, but only once. Since it is likely line-buffered, you could	1695	readable, but only once. Since it is likely line-buffered, you could
1209	attempt to read a whole line in the callback.	1696	attempt to read a whole line in the callback.
1210		1697
1211	static void	1698	static void
1212	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1699	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
1213	{	1700	{
1214	ev_io_stop (loop, w);	1701	ev_io_stop (loop, w);
1215	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1702	.. read from stdin here (or from w->fd) and handle any I/O errors
1216	}	1703	}
1217		1704
1218	...	1705	...
1219	struct ev_loop *loop = ev_default_init (0);	1706	struct ev_loop *loop = ev_default_init (0);
1220	struct ev_io stdin_readable;	1707	ev_io stdin_readable;
1221	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1708	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1222	ev_io_start (loop, &stdin_readable);	1709	ev_io_start (loop, &stdin_readable);
1223	ev_loop (loop, 0);	1710	ev_run (loop, 0);
1224		1711
1225		1712
1226	=head2 C<ev_timer> - relative and optionally repeating timeouts	1713	=head2 C<ev_timer> - relative and optionally repeating timeouts
1227		1714
1228	Timer watchers are simple relative timers that generate an event after a	1715	Timer watchers are simple relative timers that generate an event after a
1229	given time, and optionally repeating in regular intervals after that.	1716	given time, and optionally repeating in regular intervals after that.
1230		1717
1231	The timers are based on real time, that is, if you register an event that	1718	The timers are based on real time, that is, if you register an event that
1232	times out after an hour and you reset your system clock to January last	1719	times out after an hour and you reset your system clock to January last
1233	year, it will still time out after (roughly) and hour. "Roughly" because	1720	year, it will still time out after (roughly) one hour. "Roughly" because
1234	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1721	detecting time jumps is hard, and some inaccuracies are unavoidable (the
1235	monotonic clock option helps a lot here).	1722	monotonic clock option helps a lot here).
1236		1723
1237	The callback is guaranteed to be invoked only after its timeout has passed,	1724	The callback is guaranteed to be invoked only I<after> its timeout has
1238	but if multiple timers become ready during the same loop iteration then	1725	passed (not I<at>, so on systems with very low-resolution clocks this
1239	order of execution is undefined.	1726	might introduce a small delay). If multiple timers become ready during the
		1727	same loop iteration then the ones with earlier time-out values are invoked
		1728	before ones of the same priority with later time-out values (but this is
		1729	no longer true when a callback calls C<ev_run> recursively).
		1730
		1731	=head3 Be smart about timeouts
		1732
		1733	Many real-world problems involve some kind of timeout, usually for error
		1734	recovery. A typical example is an HTTP request - if the other side hangs,
		1735	you want to raise some error after a while.
		1736
		1737	What follows are some ways to handle this problem, from obvious and
		1738	inefficient to smart and efficient.
		1739
		1740	In the following, a 60 second activity timeout is assumed - a timeout that
		1741	gets reset to 60 seconds each time there is activity (e.g. each time some
		1742	data or other life sign was received).
		1743
		1744	=over 4
		1745
		1746	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1747
		1748	This is the most obvious, but not the most simple way: In the beginning,
		1749	start the watcher:
		1750
		1751	ev_timer_init (timer, callback, 60., 0.);
		1752	ev_timer_start (loop, timer);
		1753
		1754	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1755	and start it again:
		1756
		1757	ev_timer_stop (loop, timer);
		1758	ev_timer_set (timer, 60., 0.);
		1759	ev_timer_start (loop, timer);
		1760
		1761	This is relatively simple to implement, but means that each time there is
		1762	some activity, libev will first have to remove the timer from its internal
		1763	data structure and then add it again. Libev tries to be fast, but it's
		1764	still not a constant-time operation.
		1765
		1766	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1767
		1768	This is the easiest way, and involves using C<ev_timer_again> instead of
		1769	C<ev_timer_start>.
		1770
		1771	To implement this, configure an C<ev_timer> with a C<repeat> value
		1772	of C<60> and then call C<ev_timer_again> at start and each time you
		1773	successfully read or write some data. If you go into an idle state where
		1774	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1775	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1776
		1777	That means you can ignore both the C<ev_timer_start> function and the
		1778	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1779	member and C<ev_timer_again>.
		1780
		1781	At start:
		1782
		1783	ev_init (timer, callback);
		1784	timer->repeat = 60.;
		1785	ev_timer_again (loop, timer);
		1786
		1787	Each time there is some activity:
		1788
		1789	ev_timer_again (loop, timer);
		1790
		1791	It is even possible to change the time-out on the fly, regardless of
		1792	whether the watcher is active or not:
		1793
		1794	timer->repeat = 30.;
		1795	ev_timer_again (loop, timer);
		1796
		1797	This is slightly more efficient then stopping/starting the timer each time
		1798	you want to modify its timeout value, as libev does not have to completely
		1799	remove and re-insert the timer from/into its internal data structure.
		1800
		1801	It is, however, even simpler than the "obvious" way to do it.
		1802
		1803	=item 3. Let the timer time out, but then re-arm it as required.
		1804
		1805	This method is more tricky, but usually most efficient: Most timeouts are
		1806	relatively long compared to the intervals between other activity - in
		1807	our example, within 60 seconds, there are usually many I/O events with
		1808	associated activity resets.
		1809
		1810	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1811	but remember the time of last activity, and check for a real timeout only
		1812	within the callback:
		1813
		1814	ev_tstamp last_activity; // time of last activity
		1815
		1816	static void
		1817	callback (EV_P_ ev_timer *w, int revents)
		1818	{
		1819	ev_tstamp now = ev_now (EV_A);
		1820	ev_tstamp timeout = last_activity + 60.;
		1821
		1822	// if last_activity + 60. is older than now, we did time out
		1823	if (timeout < now)
		1824	{
		1825	// timeout occurred, take action
		1826	}
		1827	else
		1828	{
		1829	// callback was invoked, but there was some activity, re-arm
		1830	// the watcher to fire in last_activity + 60, which is
		1831	// guaranteed to be in the future, so "again" is positive:
		1832	w->repeat = timeout - now;
		1833	ev_timer_again (EV_A_ w);
		1834	}
		1835	}
		1836
		1837	To summarise the callback: first calculate the real timeout (defined
		1838	as "60 seconds after the last activity"), then check if that time has
		1839	been reached, which means something I<did>, in fact, time out. Otherwise
		1840	the callback was invoked too early (C<timeout> is in the future), so
		1841	re-schedule the timer to fire at that future time, to see if maybe we have
		1842	a timeout then.
		1843
		1844	Note how C<ev_timer_again> is used, taking advantage of the
		1845	C<ev_timer_again> optimisation when the timer is already running.
		1846
		1847	This scheme causes more callback invocations (about one every 60 seconds
		1848	minus half the average time between activity), but virtually no calls to
		1849	libev to change the timeout.
		1850
		1851	To start the timer, simply initialise the watcher and set C<last_activity>
		1852	to the current time (meaning we just have some activity :), then call the
		1853	callback, which will "do the right thing" and start the timer:
		1854
		1855	ev_init (timer, callback);
		1856	last_activity = ev_now (loop);
		1857	callback (loop, timer, EV_TIMER);
		1858
		1859	And when there is some activity, simply store the current time in
		1860	C<last_activity>, no libev calls at all:
		1861
		1862	last_activity = ev_now (loop);
		1863
		1864	This technique is slightly more complex, but in most cases where the
		1865	time-out is unlikely to be triggered, much more efficient.
		1866
		1867	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1868	callback :) - just change the timeout and invoke the callback, which will
		1869	fix things for you.
		1870
		1871	=item 4. Wee, just use a double-linked list for your timeouts.
		1872
		1873	If there is not one request, but many thousands (millions...), all
		1874	employing some kind of timeout with the same timeout value, then one can
		1875	do even better:
		1876
		1877	When starting the timeout, calculate the timeout value and put the timeout
		1878	at the I<end> of the list.
		1879
		1880	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1881	the list is expected to fire (for example, using the technique #3).
		1882
		1883	When there is some activity, remove the timer from the list, recalculate
		1884	the timeout, append it to the end of the list again, and make sure to
		1885	update the C<ev_timer> if it was taken from the beginning of the list.
		1886
		1887	This way, one can manage an unlimited number of timeouts in O(1) time for
		1888	starting, stopping and updating the timers, at the expense of a major
		1889	complication, and having to use a constant timeout. The constant timeout
		1890	ensures that the list stays sorted.
		1891
		1892	=back
		1893
		1894	So which method the best?
		1895
		1896	Method #2 is a simple no-brain-required solution that is adequate in most
		1897	situations. Method #3 requires a bit more thinking, but handles many cases
		1898	better, and isn't very complicated either. In most case, choosing either
		1899	one is fine, with #3 being better in typical situations.
		1900
		1901	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1902	rather complicated, but extremely efficient, something that really pays
		1903	off after the first million or so of active timers, i.e. it's usually
		1904	overkill :)
1240		1905
1241	=head3 The special problem of time updates	1906	=head3 The special problem of time updates
1242		1907
1243	Establishing the current time is a costly operation (it usually takes at	1908	Establishing the current time is a costly operation (it usually takes at
1244	least two system calls): EV therefore updates its idea of the current	1909	least two system calls): EV therefore updates its idea of the current
1245	time only before and after C<ev_loop> polls for new events, which causes	1910	time only before and after C<ev_run> collects new events, which causes a
1246	a growing difference between C<ev_now ()> and C<ev_time ()> when handling	1911	growing difference between C<ev_now ()> and C<ev_time ()> when handling
1247	lots of events.	1912	lots of events in one iteration.
1248		1913
1249	The relative timeouts are calculated relative to the C<ev_now ()>	1914	The relative timeouts are calculated relative to the C<ev_now ()>
1250	time. This is usually the right thing as this timestamp refers to the time	1915	time. This is usually the right thing as this timestamp refers to the time
1251	of the event triggering whatever timeout you are modifying/starting. If	1916	of the event triggering whatever timeout you are modifying/starting. If
1252	you suspect event processing to be delayed and you I<need> to base the	1917	you suspect event processing to be delayed and you I<need> to base the
…		…
1256		1921
1257	If the event loop is suspended for a long time, you can also force an	1922	If the event loop is suspended for a long time, you can also force an
1258	update of the time returned by C<ev_now ()> by calling C<ev_now_update	1923	update of the time returned by C<ev_now ()> by calling C<ev_now_update
1259	()>.	1924	()>.
1260		1925
		1926	=head3 The special problems of suspended animation
		1927
		1928	When you leave the server world it is quite customary to hit machines that
		1929	can suspend/hibernate - what happens to the clocks during such a suspend?
		1930
		1931	Some quick tests made with a Linux 2.6.28 indicate that a suspend freezes
		1932	all processes, while the clocks (C<times>, C<CLOCK_MONOTONIC>) continue
		1933	to run until the system is suspended, but they will not advance while the
		1934	system is suspended. That means, on resume, it will be as if the program
		1935	was frozen for a few seconds, but the suspend time will not be counted
		1936	towards C<ev_timer> when a monotonic clock source is used. The real time
		1937	clock advanced as expected, but if it is used as sole clocksource, then a
		1938	long suspend would be detected as a time jump by libev, and timers would
		1939	be adjusted accordingly.
		1940
		1941	I would not be surprised to see different behaviour in different between
		1942	operating systems, OS versions or even different hardware.
		1943
		1944	The other form of suspend (job control, or sending a SIGSTOP) will see a
		1945	time jump in the monotonic clocks and the realtime clock. If the program
		1946	is suspended for a very long time, and monotonic clock sources are in use,
		1947	then you can expect C<ev_timer>s to expire as the full suspension time
		1948	will be counted towards the timers. When no monotonic clock source is in
		1949	use, then libev will again assume a timejump and adjust accordingly.
		1950
		1951	It might be beneficial for this latter case to call C<ev_suspend>
		1952	and C<ev_resume> in code that handles C<SIGTSTP>, to at least get
		1953	deterministic behaviour in this case (you can do nothing against
		1954	C<SIGSTOP>).
		1955
1261	=head3 Watcher-Specific Functions and Data Members	1956	=head3 Watcher-Specific Functions and Data Members
1262		1957
1263	=over 4	1958	=over 4
1264		1959
1265	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1960	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
…		…
1288	If the timer is started but non-repeating, stop it (as if it timed out).	1983	If the timer is started but non-repeating, stop it (as if it timed out).
1289		1984
1290	If the timer is repeating, either start it if necessary (with the	1985	If the timer is repeating, either start it if necessary (with the
1291	C<repeat> value), or reset the running timer to the C<repeat> value.	1986	C<repeat> value), or reset the running timer to the C<repeat> value.
1292		1987
1293	This sounds a bit complicated, but here is a useful and typical	1988	This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
1294	example: Imagine you have a TCP connection and you want a so-called idle	1989	usage example.
1295	timeout, that is, you want to be called when there have been, say, 60
1296	seconds of inactivity on the socket. The easiest way to do this is to
1297	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
1298	C<ev_timer_again> each time you successfully read or write some data. If
1299	you go into an idle state where you do not expect data to travel on the
1300	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
1301	automatically restart it if need be.
1302		1990
1303	That means you can ignore the C<after> value and C<ev_timer_start>	1991	=item ev_tstamp ev_timer_remaining (loop, ev_timer *)
1304	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
1305		1992
1306	ev_timer_init (timer, callback, 0., 5.);	1993	Returns the remaining time until a timer fires. If the timer is active,
1307	ev_timer_again (loop, timer);	1994	then this time is relative to the current event loop time, otherwise it's
1308	...	1995	the timeout value currently configured.
1309	timer->again = 17.;
1310	ev_timer_again (loop, timer);
1311	...
1312	timer->again = 10.;
1313	ev_timer_again (loop, timer);
1314		1996
1315	This is more slightly efficient then stopping/starting the timer each time	1997	That is, after an C<ev_timer_set (w, 5, 7)>, C<ev_timer_remaining> returns
1316	you want to modify its timeout value.	1998	C<5>. When the timer is started and one second passes, C<ev_timer_remaining>
		1999	will return C<4>. When the timer expires and is restarted, it will return
		2000	roughly C<7> (likely slightly less as callback invocation takes some time,
		2001	too), and so on.
1317		2002
1318	=item ev_tstamp repeat [read-write]	2003	=item ev_tstamp repeat [read-write]
1319		2004
1320	The current C<repeat> value. Will be used each time the watcher times out	2005	The current C<repeat> value. Will be used each time the watcher times out
1321	or C<ev_timer_again> is called and determines the next timeout (if any),	2006	or C<ev_timer_again> is called, and determines the next timeout (if any),
1322	which is also when any modifications are taken into account.	2007	which is also when any modifications are taken into account.
1323		2008
1324	=back	2009	=back
1325		2010
1326	=head3 Examples	2011	=head3 Examples
1327		2012
1328	Example: Create a timer that fires after 60 seconds.	2013	Example: Create a timer that fires after 60 seconds.
1329		2014
1330	static void	2015	static void
1331	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2016	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
1332	{	2017	{
1333	.. one minute over, w is actually stopped right here	2018	.. one minute over, w is actually stopped right here
1334	}	2019	}
1335		2020
1336	struct ev_timer mytimer;	2021	ev_timer mytimer;
1337	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	2022	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1338	ev_timer_start (loop, &mytimer);	2023	ev_timer_start (loop, &mytimer);
1339		2024
1340	Example: Create a timeout timer that times out after 10 seconds of	2025	Example: Create a timeout timer that times out after 10 seconds of
1341	inactivity.	2026	inactivity.
1342		2027
1343	static void	2028	static void
1344	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2029	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
1345	{	2030	{
1346	.. ten seconds without any activity	2031	.. ten seconds without any activity
1347	}	2032	}
1348		2033
1349	struct ev_timer mytimer;	2034	ev_timer mytimer;
1350	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	2035	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1351	ev_timer_again (&mytimer); /* start timer */	2036	ev_timer_again (&mytimer); /* start timer */
1352	ev_loop (loop, 0);	2037	ev_run (loop, 0);
1353		2038
1354	// and in some piece of code that gets executed on any "activity":	2039	// and in some piece of code that gets executed on any "activity":
1355	// reset the timeout to start ticking again at 10 seconds	2040	// reset the timeout to start ticking again at 10 seconds
1356	ev_timer_again (&mytimer);	2041	ev_timer_again (&mytimer);
1357		2042
…		…
1359	=head2 C<ev_periodic> - to cron or not to cron?	2044	=head2 C<ev_periodic> - to cron or not to cron?
1360		2045
1361	Periodic watchers are also timers of a kind, but they are very versatile	2046	Periodic watchers are also timers of a kind, but they are very versatile
1362	(and unfortunately a bit complex).	2047	(and unfortunately a bit complex).
1363		2048
1364	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	2049	Unlike C<ev_timer>, periodic watchers are not based on real time (or
1365	but on wall clock time (absolute time). You can tell a periodic watcher	2050	relative time, the physical time that passes) but on wall clock time
1366	to trigger after some specific point in time. For example, if you tell a	2051	(absolute time, the thing you can read on your calender or clock). The
1367	periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()	2052	difference is that wall clock time can run faster or slower than real
1368	+ 10.>, that is, an absolute time not a delay) and then reset your system	2053	time, and time jumps are not uncommon (e.g. when you adjust your
1369	clock to January of the previous year, then it will take more than year	2054	wrist-watch).
1370	to trigger the event (unlike an C<ev_timer>, which would still trigger
1371	roughly 10 seconds later as it uses a relative timeout).
1372		2055
		2056	You can tell a periodic watcher to trigger after some specific point
		2057	in time: for example, if you tell a periodic watcher to trigger "in 10
		2058	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		2059	not a delay) and then reset your system clock to January of the previous
		2060	year, then it will take a year or more to trigger the event (unlike an
		2061	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		2062	it, as it uses a relative timeout).
		2063
1373	C<ev_periodic>s can also be used to implement vastly more complex timers,	2064	C<ev_periodic> watchers can also be used to implement vastly more complex
1374	such as triggering an event on each "midnight, local time", or other	2065	timers, such as triggering an event on each "midnight, local time", or
1375	complicated, rules.	2066	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		2067	those cannot react to time jumps.
1376		2068
1377	As with timers, the callback is guaranteed to be invoked only when the	2069	As with timers, the callback is guaranteed to be invoked only when the
1378	time (C<at>) has passed, but if multiple periodic timers become ready	2070	point in time where it is supposed to trigger has passed. If multiple
1379	during the same loop iteration then order of execution is undefined.	2071	timers become ready during the same loop iteration then the ones with
		2072	earlier time-out values are invoked before ones with later time-out values
		2073	(but this is no longer true when a callback calls C<ev_run> recursively).
1380		2074
1381	=head3 Watcher-Specific Functions and Data Members	2075	=head3 Watcher-Specific Functions and Data Members
1382		2076
1383	=over 4	2077	=over 4
1384		2078
1385	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	2079	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1386		2080
1387	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	2081	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1388		2082
1389	Lots of arguments, lets sort it out... There are basically three modes of	2083	Lots of arguments, let's sort it out... There are basically three modes of
1390	operation, and we will explain them from simplest to complex:	2084	operation, and we will explain them from simplest to most complex:
1391		2085
1392	=over 4	2086	=over 4
1393		2087
1394	=item * absolute timer (at = time, interval = reschedule_cb = 0)	2088	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
1395		2089
1396	In this configuration the watcher triggers an event after the wall clock	2090	In this configuration the watcher triggers an event after the wall clock
1397	time C<at> has passed and doesn't repeat. It will not adjust when a time	2091	time C<offset> has passed. It will not repeat and will not adjust when a
1398	jump occurs, that is, if it is to be run at January 1st 2011 then it will	2092	time jump occurs, that is, if it is to be run at January 1st 2011 then it
1399	run when the system time reaches or surpasses this time.	2093	will be stopped and invoked when the system clock reaches or surpasses
		2094	this point in time.
1400		2095
1401	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	2096	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
1402		2097
1403	In this mode the watcher will always be scheduled to time out at the next	2098	In this mode the watcher will always be scheduled to time out at the next
1404	C<at + N * interval> time (for some integer N, which can also be negative)	2099	C<offset + N * interval> time (for some integer N, which can also be
1405	and then repeat, regardless of any time jumps.	2100	negative) and then repeat, regardless of any time jumps. The C<offset>
		2101	argument is merely an offset into the C<interval> periods.
1406		2102
1407	This can be used to create timers that do not drift with respect to system	2103	This can be used to create timers that do not drift with respect to the
1408	time, for example, here is a C<ev_periodic> that triggers each hour, on	2104	system clock, for example, here is an C<ev_periodic> that triggers each
1409	the hour:	2105	hour, on the hour (with respect to UTC):
1410		2106
1411	ev_periodic_set (&periodic, 0., 3600., 0);	2107	ev_periodic_set (&periodic, 0., 3600., 0);
1412		2108
1413	This doesn't mean there will always be 3600 seconds in between triggers,	2109	This doesn't mean there will always be 3600 seconds in between triggers,
1414	but only that the callback will be called when the system time shows a	2110	but only that the callback will be called when the system time shows a
1415	full hour (UTC), or more correctly, when the system time is evenly divisible	2111	full hour (UTC), or more correctly, when the system time is evenly divisible
1416	by 3600.	2112	by 3600.
1417		2113
1418	Another way to think about it (for the mathematically inclined) is that	2114	Another way to think about it (for the mathematically inclined) is that
1419	C<ev_periodic> will try to run the callback in this mode at the next possible	2115	C<ev_periodic> will try to run the callback in this mode at the next possible
1420	time where C<time = at (mod interval)>, regardless of any time jumps.	2116	time where C<time = offset (mod interval)>, regardless of any time jumps.
1421		2117
1422	For numerical stability it is preferable that the C<at> value is near	2118	For numerical stability it is preferable that the C<offset> value is near
1423	C<ev_now ()> (the current time), but there is no range requirement for	2119	C<ev_now ()> (the current time), but there is no range requirement for
1424	this value, and in fact is often specified as zero.	2120	this value, and in fact is often specified as zero.
1425		2121
1426	Note also that there is an upper limit to how often a timer can fire (CPU	2122	Note also that there is an upper limit to how often a timer can fire (CPU
1427	speed for example), so if C<interval> is very small then timing stability	2123	speed for example), so if C<interval> is very small then timing stability
1428	will of course deteriorate. Libev itself tries to be exact to be about one	2124	will of course deteriorate. Libev itself tries to be exact to be about one
1429	millisecond (if the OS supports it and the machine is fast enough).	2125	millisecond (if the OS supports it and the machine is fast enough).
1430		2126
1431	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)	2127	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
1432		2128
1433	In this mode the values for C<interval> and C<at> are both being	2129	In this mode the values for C<interval> and C<offset> are both being
1434	ignored. Instead, each time the periodic watcher gets scheduled, the	2130	ignored. Instead, each time the periodic watcher gets scheduled, the
1435	reschedule callback will be called with the watcher as first, and the	2131	reschedule callback will be called with the watcher as first, and the
1436	current time as second argument.	2132	current time as second argument.
1437		2133
1438	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	2134	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
1439	ever, or make ANY event loop modifications whatsoever>.	2135	or make ANY other event loop modifications whatsoever, unless explicitly
		2136	allowed by documentation here>.
1440		2137
1441	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop	2138	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
1442	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the	2139	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
1443	only event loop modification you are allowed to do).	2140	only event loop modification you are allowed to do).
1444		2141
1445	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic	2142	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
1446	*w, ev_tstamp now)>, e.g.:	2143	*w, ev_tstamp now)>, e.g.:
1447		2144
		2145	static ev_tstamp
1448	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	2146	my_rescheduler (ev_periodic *w, ev_tstamp now)
1449	{	2147	{
1450	return now + 60.;	2148	return now + 60.;
1451	}	2149	}
1452		2150
1453	It must return the next time to trigger, based on the passed time value	2151	It must return the next time to trigger, based on the passed time value
…		…
1473	a different time than the last time it was called (e.g. in a crond like	2171	a different time than the last time it was called (e.g. in a crond like
1474	program when the crontabs have changed).	2172	program when the crontabs have changed).
1475		2173
1476	=item ev_tstamp ev_periodic_at (ev_periodic *)	2174	=item ev_tstamp ev_periodic_at (ev_periodic *)
1477		2175
1478	When active, returns the absolute time that the watcher is supposed to	2176	When active, returns the absolute time that the watcher is supposed
1479	trigger next.	2177	to trigger next. This is not the same as the C<offset> argument to
		2178	C<ev_periodic_set>, but indeed works even in interval and manual
		2179	rescheduling modes.
1480		2180
1481	=item ev_tstamp offset [read-write]	2181	=item ev_tstamp offset [read-write]
1482		2182
1483	When repeating, this contains the offset value, otherwise this is the	2183	When repeating, this contains the offset value, otherwise this is the
1484	absolute point in time (the C<at> value passed to C<ev_periodic_set>).	2184	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		2185	although libev might modify this value for better numerical stability).
1485		2186
1486	Can be modified any time, but changes only take effect when the periodic	2187	Can be modified any time, but changes only take effect when the periodic
1487	timer fires or C<ev_periodic_again> is being called.	2188	timer fires or C<ev_periodic_again> is being called.
1488		2189
1489	=item ev_tstamp interval [read-write]	2190	=item ev_tstamp interval [read-write]
1490		2191
1491	The current interval value. Can be modified any time, but changes only	2192	The current interval value. Can be modified any time, but changes only
1492	take effect when the periodic timer fires or C<ev_periodic_again> is being	2193	take effect when the periodic timer fires or C<ev_periodic_again> is being
1493	called.	2194	called.
1494		2195
1495	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]	2196	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
1496		2197
1497	The current reschedule callback, or C<0>, if this functionality is	2198	The current reschedule callback, or C<0>, if this functionality is
1498	switched off. Can be changed any time, but changes only take effect when	2199	switched off. Can be changed any time, but changes only take effect when
1499	the periodic timer fires or C<ev_periodic_again> is being called.	2200	the periodic timer fires or C<ev_periodic_again> is being called.
1500		2201
1501	=back	2202	=back
1502		2203
1503	=head3 Examples	2204	=head3 Examples
1504		2205
1505	Example: Call a callback every hour, or, more precisely, whenever the	2206	Example: Call a callback every hour, or, more precisely, whenever the
1506	system clock is divisible by 3600. The callback invocation times have	2207	system time is divisible by 3600. The callback invocation times have
1507	potentially a lot of jitter, but good long-term stability.	2208	potentially a lot of jitter, but good long-term stability.
1508		2209
1509	static void	2210	static void
1510	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	2211	clock_cb (struct ev_loop *loop, ev_periodic *w, int revents)
1511	{	2212	{
1512	... its now a full hour (UTC, or TAI or whatever your clock follows)	2213	... its now a full hour (UTC, or TAI or whatever your clock follows)
1513	}	2214	}
1514		2215
1515	struct ev_periodic hourly_tick;	2216	ev_periodic hourly_tick;
1516	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	2217	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1517	ev_periodic_start (loop, &hourly_tick);	2218	ev_periodic_start (loop, &hourly_tick);
1518		2219
1519	Example: The same as above, but use a reschedule callback to do it:	2220	Example: The same as above, but use a reschedule callback to do it:
1520		2221
1521	#include <math.h>	2222	#include <math.h>
1522		2223
1523	static ev_tstamp	2224	static ev_tstamp
1524	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	2225	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
1525	{	2226	{
1526	return fmod (now, 3600.) + 3600.;	2227	return now + (3600. - fmod (now, 3600.));
1527	}	2228	}
1528		2229
1529	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	2230	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1530		2231
1531	Example: Call a callback every hour, starting now:	2232	Example: Call a callback every hour, starting now:
1532		2233
1533	struct ev_periodic hourly_tick;	2234	ev_periodic hourly_tick;
1534	ev_periodic_init (&hourly_tick, clock_cb,	2235	ev_periodic_init (&hourly_tick, clock_cb,
1535	fmod (ev_now (loop), 3600.), 3600., 0);	2236	fmod (ev_now (loop), 3600.), 3600., 0);
1536	ev_periodic_start (loop, &hourly_tick);	2237	ev_periodic_start (loop, &hourly_tick);
1537		2238
1538		2239
…		…
1541	Signal watchers will trigger an event when the process receives a specific	2242	Signal watchers will trigger an event when the process receives a specific
1542	signal one or more times. Even though signals are very asynchronous, libev	2243	signal one or more times. Even though signals are very asynchronous, libev
1543	will try it's best to deliver signals synchronously, i.e. as part of the	2244	will try it's best to deliver signals synchronously, i.e. as part of the
1544	normal event processing, like any other event.	2245	normal event processing, like any other event.
1545		2246
		2247	If you want signals to be delivered truly asynchronously, just use
		2248	C<sigaction> as you would do without libev and forget about sharing
		2249	the signal. You can even use C<ev_async> from a signal handler to
		2250	synchronously wake up an event loop.
		2251
1546	You can configure as many watchers as you like per signal. Only when the	2252	You can configure as many watchers as you like for the same signal, but
		2253	only within the same loop, i.e. you can watch for C<SIGINT> in your
		2254	default loop and for C<SIGIO> in another loop, but you cannot watch for
		2255	C<SIGINT> in both the default loop and another loop at the same time. At
		2256	the moment, C<SIGCHLD> is permanently tied to the default loop.
		2257
1547	first watcher gets started will libev actually register a signal watcher	2258	When the first watcher gets started will libev actually register something
1548	with the kernel (thus it coexists with your own signal handlers as long	2259	with the kernel (thus it coexists with your own signal handlers as long as
1549	as you don't register any with libev). Similarly, when the last signal	2260	you don't register any with libev for the same signal).
1550	watcher for a signal is stopped libev will reset the signal handler to
1551	SIG_DFL (regardless of what it was set to before).
1552		2261
1553	If possible and supported, libev will install its handlers with	2262	If possible and supported, libev will install its handlers with
1554	C<SA_RESTART> behaviour enabled, so system calls should not be unduly	2263	C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should
1555	interrupted. If you have a problem with system calls getting interrupted by	2264	not be unduly interrupted. If you have a problem with system calls getting
1556	signals you can block all signals in an C<ev_check> watcher and unblock	2265	interrupted by signals you can block all signals in an C<ev_check> watcher
1557	them in an C<ev_prepare> watcher.	2266	and unblock them in an C<ev_prepare> watcher.
		2267
		2268	=head3 The special problem of inheritance over fork/execve/pthread_create
		2269
		2270	Both the signal mask (C<sigprocmask>) and the signal disposition
		2271	(C<sigaction>) are unspecified after starting a signal watcher (and after
		2272	stopping it again), that is, libev might or might not block the signal,
		2273	and might or might not set or restore the installed signal handler.
		2274
		2275	While this does not matter for the signal disposition (libev never
		2276	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
		2277	C<execve>), this matters for the signal mask: many programs do not expect
		2278	certain signals to be blocked.
		2279
		2280	This means that before calling C<exec> (from the child) you should reset
		2281	the signal mask to whatever "default" you expect (all clear is a good
		2282	choice usually).
		2283
		2284	The simplest way to ensure that the signal mask is reset in the child is
		2285	to install a fork handler with C<pthread_atfork> that resets it. That will
		2286	catch fork calls done by libraries (such as the libc) as well.
		2287
		2288	In current versions of libev, the signal will not be blocked indefinitely
		2289	unless you use the C<signalfd> API (C<EV_SIGNALFD>). While this reduces
		2290	the window of opportunity for problems, it will not go away, as libev
		2291	I<has> to modify the signal mask, at least temporarily.
		2292
		2293	So I can't stress this enough: I<If you do not reset your signal mask when
		2294	you expect it to be empty, you have a race condition in your code>. This
		2295	is not a libev-specific thing, this is true for most event libraries.
1558		2296
1559	=head3 Watcher-Specific Functions and Data Members	2297	=head3 Watcher-Specific Functions and Data Members
1560		2298
1561	=over 4	2299	=over 4
1562		2300
…		…
1573		2311
1574	=back	2312	=back
1575		2313
1576	=head3 Examples	2314	=head3 Examples
1577		2315
1578	Example: Try to exit cleanly on SIGINT and SIGTERM.	2316	Example: Try to exit cleanly on SIGINT.
1579		2317
1580	static void	2318	static void
1581	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	2319	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
1582	{	2320	{
1583	ev_unloop (loop, EVUNLOOP_ALL);	2321	ev_break (loop, EVBREAK_ALL);
1584	}	2322	}
1585		2323
1586	struct ev_signal signal_watcher;	2324	ev_signal signal_watcher;
1587	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	2325	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1588	ev_signal_start (loop, &sigint_cb);	2326	ev_signal_start (loop, &signal_watcher);
1589		2327
1590		2328
1591	=head2 C<ev_child> - watch out for process status changes	2329	=head2 C<ev_child> - watch out for process status changes
1592		2330
1593	Child watchers trigger when your process receives a SIGCHLD in response to	2331	Child watchers trigger when your process receives a SIGCHLD in response to
1594	some child status changes (most typically when a child of yours dies). It	2332	some child status changes (most typically when a child of yours dies or
1595	is permissible to install a child watcher I<after> the child has been	2333	exits). It is permissible to install a child watcher I<after> the child
1596	forked (which implies it might have already exited), as long as the event	2334	has been forked (which implies it might have already exited), as long
1597	loop isn't entered (or is continued from a watcher).	2335	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2336	forking and then immediately registering a watcher for the child is fine,
		2337	but forking and registering a watcher a few event loop iterations later or
		2338	in the next callback invocation is not.
1598		2339
1599	Only the default event loop is capable of handling signals, and therefore	2340	Only the default event loop is capable of handling signals, and therefore
1600	you can only register child watchers in the default event loop.	2341	you can only register child watchers in the default event loop.
1601		2342
		2343	Due to some design glitches inside libev, child watchers will always be
		2344	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2345	libev)
		2346
1602	=head3 Process Interaction	2347	=head3 Process Interaction
1603		2348
1604	Libev grabs C<SIGCHLD> as soon as the default event loop is	2349	Libev grabs C<SIGCHLD> as soon as the default event loop is
1605	initialised. This is necessary to guarantee proper behaviour even if	2350	initialised. This is necessary to guarantee proper behaviour even if the
1606	the first child watcher is started after the child exits. The occurrence	2351	first child watcher is started after the child exits. The occurrence
1607	of C<SIGCHLD> is recorded asynchronously, but child reaping is done	2352	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
1608	synchronously as part of the event loop processing. Libev always reaps all	2353	synchronously as part of the event loop processing. Libev always reaps all
1609	children, even ones not watched.	2354	children, even ones not watched.
1610		2355
1611	=head3 Overriding the Built-In Processing	2356	=head3 Overriding the Built-In Processing
…		…
1621	=head3 Stopping the Child Watcher	2366	=head3 Stopping the Child Watcher
1622		2367
1623	Currently, the child watcher never gets stopped, even when the	2368	Currently, the child watcher never gets stopped, even when the
1624	child terminates, so normally one needs to stop the watcher in the	2369	child terminates, so normally one needs to stop the watcher in the
1625	callback. Future versions of libev might stop the watcher automatically	2370	callback. Future versions of libev might stop the watcher automatically
1626	when a child exit is detected.	2371	when a child exit is detected (calling C<ev_child_stop> twice is not a
		2372	problem).
1627		2373
1628	=head3 Watcher-Specific Functions and Data Members	2374	=head3 Watcher-Specific Functions and Data Members
1629		2375
1630	=over 4	2376	=over 4
1631		2377
…		…
1663	its completion.	2409	its completion.
1664		2410
1665	ev_child cw;	2411	ev_child cw;
1666		2412
1667	static void	2413	static void
1668	child_cb (EV_P_ struct ev_child *w, int revents)	2414	child_cb (EV_P_ ev_child *w, int revents)
1669	{	2415	{
1670	ev_child_stop (EV_A_ w);	2416	ev_child_stop (EV_A_ w);
1671	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);	2417	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1672	}	2418	}
1673		2419
…		…
1688		2434
1689		2435
1690	=head2 C<ev_stat> - did the file attributes just change?	2436	=head2 C<ev_stat> - did the file attributes just change?
1691		2437
1692	This watches a file system path for attribute changes. That is, it calls	2438	This watches a file system path for attribute changes. That is, it calls
1693	C<stat> regularly (or when the OS says it changed) and sees if it changed	2439	C<stat> on that path in regular intervals (or when the OS says it changed)
1694	compared to the last time, invoking the callback if it did.	2440	and sees if it changed compared to the last time, invoking the callback if
		2441	it did.
1695		2442
1696	The path does not need to exist: changing from "path exists" to "path does	2443	The path does not need to exist: changing from "path exists" to "path does
1697	not exist" is a status change like any other. The condition "path does	2444	not exist" is a status change like any other. The condition "path does not
1698	not exist" is signified by the C<st_nlink> field being zero (which is	2445	exist" (or more correctly "path cannot be stat'ed") is signified by the
1699	otherwise always forced to be at least one) and all the other fields of	2446	C<st_nlink> field being zero (which is otherwise always forced to be at
1700	the stat buffer having unspecified contents.	2447	least one) and all the other fields of the stat buffer having unspecified
		2448	contents.
1701		2449
1702	The path I<should> be absolute and I<must not> end in a slash. If it is	2450	The path I<must not> end in a slash or contain special components such as
		2451	C<.> or C<..>. The path I<should> be absolute: If it is relative and
1703	relative and your working directory changes, the behaviour is undefined.	2452	your working directory changes, then the behaviour is undefined.
1704		2453
1705	Since there is no standard to do this, the portable implementation simply	2454	Since there is no portable change notification interface available, the
1706	calls C<stat (2)> regularly on the path to see if it changed somehow. You	2455	portable implementation simply calls C<stat(2)> regularly on the path
1707	can specify a recommended polling interval for this case. If you specify	2456	to see if it changed somehow. You can specify a recommended polling
1708	a polling interval of C<0> (highly recommended!) then a I<suitable,	2457	interval for this case. If you specify a polling interval of C<0> (highly
1709	unspecified default> value will be used (which you can expect to be around	2458	recommended!) then a I<suitable, unspecified default> value will be used
1710	five seconds, although this might change dynamically). Libev will also	2459	(which you can expect to be around five seconds, although this might
1711	impose a minimum interval which is currently around C<0.1>, but thats	2460	change dynamically). Libev will also impose a minimum interval which is
1712	usually overkill.	2461	currently around C<0.1>, but that's usually overkill.
1713		2462
1714	This watcher type is not meant for massive numbers of stat watchers,	2463	This watcher type is not meant for massive numbers of stat watchers,
1715	as even with OS-supported change notifications, this can be	2464	as even with OS-supported change notifications, this can be
1716	resource-intensive.	2465	resource-intensive.
1717		2466
1718	At the time of this writing, only the Linux inotify interface is	2467	At the time of this writing, the only OS-specific interface implemented
1719	implemented (implementing kqueue support is left as an exercise for the	2468	is the Linux inotify interface (implementing kqueue support is left as an
1720	reader, note, however, that the author sees no way of implementing ev_stat	2469	exercise for the reader. Note, however, that the author sees no way of
1721	semantics with kqueue). Inotify will be used to give hints only and should	2470	implementing C<ev_stat> semantics with kqueue, except as a hint).
1722	not change the semantics of C<ev_stat> watchers, which means that libev
1723	sometimes needs to fall back to regular polling again even with inotify,
1724	but changes are usually detected immediately, and if the file exists there
1725	will be no polling.
1726		2471
1727	=head3 ABI Issues (Largefile Support)	2472	=head3 ABI Issues (Largefile Support)
1728		2473
1729	Libev by default (unless the user overrides this) uses the default	2474	Libev by default (unless the user overrides this) uses the default
1730	compilation environment, which means that on systems with large file	2475	compilation environment, which means that on systems with large file
1731	support disabled by default, you get the 32 bit version of the stat	2476	support disabled by default, you get the 32 bit version of the stat
1732	structure. When using the library from programs that change the ABI to	2477	structure. When using the library from programs that change the ABI to
1733	use 64 bit file offsets the programs will fail. In that case you have to	2478	use 64 bit file offsets the programs will fail. In that case you have to
1734	compile libev with the same flags to get binary compatibility. This is	2479	compile libev with the same flags to get binary compatibility. This is
1735	obviously the case with any flags that change the ABI, but the problem is	2480	obviously the case with any flags that change the ABI, but the problem is
1736	most noticeably disabled with ev_stat and large file support.	2481	most noticeably displayed with ev_stat and large file support.
1737		2482
1738	The solution for this is to lobby your distribution maker to make large	2483	The solution for this is to lobby your distribution maker to make large
1739	file interfaces available by default (as e.g. FreeBSD does) and not	2484	file interfaces available by default (as e.g. FreeBSD does) and not
1740	optional. Libev cannot simply switch on large file support because it has	2485	optional. Libev cannot simply switch on large file support because it has
1741	to exchange stat structures with application programs compiled using the	2486	to exchange stat structures with application programs compiled using the
1742	default compilation environment.	2487	default compilation environment.
1743		2488
1744	=head3 Inotify	2489	=head3 Inotify and Kqueue
1745		2490
1746	When C<inotify (7)> support has been compiled into libev (generally only	2491	When C<inotify (7)> support has been compiled into libev and present at
1747	available on Linux) and present at runtime, it will be used to speed up	2492	runtime, it will be used to speed up change detection where possible. The
1748	change detection where possible. The inotify descriptor will be created lazily	2493	inotify descriptor will be created lazily when the first C<ev_stat>
1749	when the first C<ev_stat> watcher is being started.	2494	watcher is being started.
1750		2495
1751	Inotify presence does not change the semantics of C<ev_stat> watchers	2496	Inotify presence does not change the semantics of C<ev_stat> watchers
1752	except that changes might be detected earlier, and in some cases, to avoid	2497	except that changes might be detected earlier, and in some cases, to avoid
1753	making regular C<stat> calls. Even in the presence of inotify support	2498	making regular C<stat> calls. Even in the presence of inotify support
1754	there are many cases where libev has to resort to regular C<stat> polling.	2499	there are many cases where libev has to resort to regular C<stat> polling,
		2500	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2501	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2502	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2503	xfs are fully working) libev usually gets away without polling.
1755		2504
1756	(There is no support for kqueue, as apparently it cannot be used to	2505	There is no support for kqueue, as apparently it cannot be used to
1757	implement this functionality, due to the requirement of having a file	2506	implement this functionality, due to the requirement of having a file
1758	descriptor open on the object at all times).	2507	descriptor open on the object at all times, and detecting renames, unlinks
		2508	etc. is difficult.
		2509
		2510	=head3 C<stat ()> is a synchronous operation
		2511
		2512	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2513	the process. The exception are C<ev_stat> watchers - those call C<stat
		2514	()>, which is a synchronous operation.
		2515
		2516	For local paths, this usually doesn't matter: unless the system is very
		2517	busy or the intervals between stat's are large, a stat call will be fast,
		2518	as the path data is usually in memory already (except when starting the
		2519	watcher).
		2520
		2521	For networked file systems, calling C<stat ()> can block an indefinite
		2522	time due to network issues, and even under good conditions, a stat call
		2523	often takes multiple milliseconds.
		2524
		2525	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2526	paths, although this is fully supported by libev.
1759		2527
1760	=head3 The special problem of stat time resolution	2528	=head3 The special problem of stat time resolution
1761		2529
1762	The C<stat ()> system call only supports full-second resolution portably, and	2530	The C<stat ()> system call only supports full-second resolution portably,
1763	even on systems where the resolution is higher, many file systems still	2531	and even on systems where the resolution is higher, most file systems
1764	only support whole seconds.	2532	still only support whole seconds.
1765		2533
1766	That means that, if the time is the only thing that changes, you can	2534	That means that, if the time is the only thing that changes, you can
1767	easily miss updates: on the first update, C<ev_stat> detects a change and	2535	easily miss updates: on the first update, C<ev_stat> detects a change and
1768	calls your callback, which does something. When there is another update	2536	calls your callback, which does something. When there is another update
1769	within the same second, C<ev_stat> will be unable to detect it as the stat	2537	within the same second, C<ev_stat> will be unable to detect unless the
1770	data does not change.	2538	stat data does change in other ways (e.g. file size).
1771		2539
1772	The solution to this is to delay acting on a change for slightly more	2540	The solution to this is to delay acting on a change for slightly more
1773	than a second (or till slightly after the next full second boundary), using	2541	than a second (or till slightly after the next full second boundary), using
1774	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);	2542	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
1775	ev_timer_again (loop, w)>).	2543	ev_timer_again (loop, w)>).
…		…
1795	C<path>. The C<interval> is a hint on how quickly a change is expected to	2563	C<path>. The C<interval> is a hint on how quickly a change is expected to
1796	be detected and should normally be specified as C<0> to let libev choose	2564	be detected and should normally be specified as C<0> to let libev choose
1797	a suitable value. The memory pointed to by C<path> must point to the same	2565	a suitable value. The memory pointed to by C<path> must point to the same
1798	path for as long as the watcher is active.	2566	path for as long as the watcher is active.
1799		2567
1800	The callback will receive C<EV_STAT> when a change was detected, relative	2568	The callback will receive an C<EV_STAT> event when a change was detected,
1801	to the attributes at the time the watcher was started (or the last change	2569	relative to the attributes at the time the watcher was started (or the
1802	was detected).	2570	last change was detected).
1803		2571
1804	=item ev_stat_stat (loop, ev_stat *)	2572	=item ev_stat_stat (loop, ev_stat *)
1805		2573
1806	Updates the stat buffer immediately with new values. If you change the	2574	Updates the stat buffer immediately with new values. If you change the
1807	watched path in your callback, you could call this function to avoid	2575	watched path in your callback, you could call this function to avoid
…		…
1890		2658
1891		2659
1892	=head2 C<ev_idle> - when you've got nothing better to do...	2660	=head2 C<ev_idle> - when you've got nothing better to do...
1893		2661
1894	Idle watchers trigger events when no other events of the same or higher	2662	Idle watchers trigger events when no other events of the same or higher
1895	priority are pending (prepare, check and other idle watchers do not	2663	priority are pending (prepare, check and other idle watchers do not count
1896	count).	2664	as receiving "events").
1897		2665
1898	That is, as long as your process is busy handling sockets or timeouts	2666	That is, as long as your process is busy handling sockets or timeouts
1899	(or even signals, imagine) of the same or higher priority it will not be	2667	(or even signals, imagine) of the same or higher priority it will not be
1900	triggered. But when your process is idle (or only lower-priority watchers	2668	triggered. But when your process is idle (or only lower-priority watchers
1901	are pending), the idle watchers are being called once per event loop	2669	are pending), the idle watchers are being called once per event loop
…		…
1912		2680
1913	=head3 Watcher-Specific Functions and Data Members	2681	=head3 Watcher-Specific Functions and Data Members
1914		2682
1915	=over 4	2683	=over 4
1916		2684
1917	=item ev_idle_init (ev_signal *, callback)	2685	=item ev_idle_init (ev_idle *, callback)
1918		2686
1919	Initialises and configures the idle watcher - it has no parameters of any	2687	Initialises and configures the idle watcher - it has no parameters of any
1920	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2688	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1921	believe me.	2689	believe me.
1922		2690
…		…
1926		2694
1927	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	2695	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1928	callback, free it. Also, use no error checking, as usual.	2696	callback, free it. Also, use no error checking, as usual.
1929		2697
1930	static void	2698	static void
1931	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	2699	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
1932	{	2700	{
1933	free (w);	2701	free (w);
1934	// now do something you wanted to do when the program has	2702	// now do something you wanted to do when the program has
1935	// no longer anything immediate to do.	2703	// no longer anything immediate to do.
1936	}	2704	}
1937		2705
1938	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	2706	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
1939	ev_idle_init (idle_watcher, idle_cb);	2707	ev_idle_init (idle_watcher, idle_cb);
1940	ev_idle_start (loop, idle_cb);	2708	ev_idle_start (loop, idle_watcher);
1941		2709
1942		2710
1943	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2711	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1944		2712
1945	Prepare and check watchers are usually (but not always) used in tandem:	2713	Prepare and check watchers are usually (but not always) used in pairs:
1946	prepare watchers get invoked before the process blocks and check watchers	2714	prepare watchers get invoked before the process blocks and check watchers
1947	afterwards.	2715	afterwards.
1948		2716
1949	You I<must not> call C<ev_loop> or similar functions that enter	2717	You I<must not> call C<ev_run> or similar functions that enter
1950	the current event loop from either C<ev_prepare> or C<ev_check>	2718	the current event loop from either C<ev_prepare> or C<ev_check>
1951	watchers. Other loops than the current one are fine, however. The	2719	watchers. Other loops than the current one are fine, however. The
1952	rationale behind this is that you do not need to check for recursion in	2720	rationale behind this is that you do not need to check for recursion in
1953	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,	2721	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
1954	C<ev_check> so if you have one watcher of each kind they will always be	2722	C<ev_check> so if you have one watcher of each kind they will always be
1955	called in pairs bracketing the blocking call.	2723	called in pairs bracketing the blocking call.
1956		2724
1957	Their main purpose is to integrate other event mechanisms into libev and	2725	Their main purpose is to integrate other event mechanisms into libev and
1958	their use is somewhat advanced. This could be used, for example, to track	2726	their use is somewhat advanced. They could be used, for example, to track
1959	variable changes, implement your own watchers, integrate net-snmp or a	2727	variable changes, implement your own watchers, integrate net-snmp or a
1960	coroutine library and lots more. They are also occasionally useful if	2728	coroutine library and lots more. They are also occasionally useful if
1961	you cache some data and want to flush it before blocking (for example,	2729	you cache some data and want to flush it before blocking (for example,
1962	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>	2730	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1963	watcher).	2731	watcher).
1964		2732
1965	This is done by examining in each prepare call which file descriptors need	2733	This is done by examining in each prepare call which file descriptors
1966	to be watched by the other library, registering C<ev_io> watchers for	2734	need to be watched by the other library, registering C<ev_io> watchers
1967	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2735	for them and starting an C<ev_timer> watcher for any timeouts (many
1968	provide just this functionality). Then, in the check watcher you check for	2736	libraries provide exactly this functionality). Then, in the check watcher,
1969	any events that occurred (by checking the pending status of all watchers	2737	you check for any events that occurred (by checking the pending status
1970	and stopping them) and call back into the library. The I/O and timer	2738	of all watchers and stopping them) and call back into the library. The
1971	callbacks will never actually be called (but must be valid nevertheless,	2739	I/O and timer callbacks will never actually be called (but must be valid
1972	because you never know, you know?).	2740	nevertheless, because you never know, you know?).
1973		2741
1974	As another example, the Perl Coro module uses these hooks to integrate	2742	As another example, the Perl Coro module uses these hooks to integrate
1975	coroutines into libev programs, by yielding to other active coroutines	2743	coroutines into libev programs, by yielding to other active coroutines
1976	during each prepare and only letting the process block if no coroutines	2744	during each prepare and only letting the process block if no coroutines
1977	are ready to run (it's actually more complicated: it only runs coroutines	2745	are ready to run (it's actually more complicated: it only runs coroutines
…		…
1980	loop from blocking if lower-priority coroutines are active, thus mapping	2748	loop from blocking if lower-priority coroutines are active, thus mapping
1981	low-priority coroutines to idle/background tasks).	2749	low-priority coroutines to idle/background tasks).
1982		2750
1983	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	2751	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
1984	priority, to ensure that they are being run before any other watchers	2752	priority, to ensure that they are being run before any other watchers
		2753	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2754
1985	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	2755	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
1986	too) should not activate ("feed") events into libev. While libev fully	2756	activate ("feed") events into libev. While libev fully supports this, they
1987	supports this, they might get executed before other C<ev_check> watchers	2757	might get executed before other C<ev_check> watchers did their job. As
1988	did their job. As C<ev_check> watchers are often used to embed other	2758	C<ev_check> watchers are often used to embed other (non-libev) event
1989	(non-libev) event loops those other event loops might be in an unusable	2759	loops those other event loops might be in an unusable state until their
1990	state until their C<ev_check> watcher ran (always remind yourself to	2760	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
1991	coexist peacefully with others).	2761	others).
1992		2762
1993	=head3 Watcher-Specific Functions and Data Members	2763	=head3 Watcher-Specific Functions and Data Members
1994		2764
1995	=over 4	2765	=over 4
1996		2766
…		…
1998		2768
1999	=item ev_check_init (ev_check *, callback)	2769	=item ev_check_init (ev_check *, callback)
2000		2770
2001	Initialises and configures the prepare or check watcher - they have no	2771	Initialises and configures the prepare or check watcher - they have no
2002	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2772	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
2003	macros, but using them is utterly, utterly and completely pointless.	2773	macros, but using them is utterly, utterly, utterly and completely
		2774	pointless.
2004		2775
2005	=back	2776	=back
2006		2777
2007	=head3 Examples	2778	=head3 Examples
2008		2779
…		…
2021		2792
2022	static ev_io iow [nfd];	2793	static ev_io iow [nfd];
2023	static ev_timer tw;	2794	static ev_timer tw;
2024		2795
2025	static void	2796	static void
2026	io_cb (ev_loop *loop, ev_io *w, int revents)	2797	io_cb (struct ev_loop *loop, ev_io *w, int revents)
2027	{	2798	{
2028	}	2799	}
2029		2800
2030	// create io watchers for each fd and a timer before blocking	2801	// create io watchers for each fd and a timer before blocking
2031	static void	2802	static void
2032	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2803	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
2033	{	2804	{
2034	int timeout = 3600000;	2805	int timeout = 3600000;
2035	struct pollfd fds [nfd];	2806	struct pollfd fds [nfd];
2036	// actual code will need to loop here and realloc etc.	2807	// actual code will need to loop here and realloc etc.
2037	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2808	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
2038		2809
2039	/* the callback is illegal, but won't be called as we stop during check */	2810	/* the callback is illegal, but won't be called as we stop during check */
2040	ev_timer_init (&tw, 0, timeout * 1e-3);	2811	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
2041	ev_timer_start (loop, &tw);	2812	ev_timer_start (loop, &tw);
2042		2813
2043	// create one ev_io per pollfd	2814	// create one ev_io per pollfd
2044	for (int i = 0; i < nfd; ++i)	2815	for (int i = 0; i < nfd; ++i)
2045	{	2816	{
…		…
2052	}	2823	}
2053	}	2824	}
2054		2825
2055	// stop all watchers after blocking	2826	// stop all watchers after blocking
2056	static void	2827	static void
2057	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2828	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
2058	{	2829	{
2059	ev_timer_stop (loop, &tw);	2830	ev_timer_stop (loop, &tw);
2060		2831
2061	for (int i = 0; i < nfd; ++i)	2832	for (int i = 0; i < nfd; ++i)
2062	{	2833	{
…		…
2101	}	2872	}
2102		2873
2103	// do not ever call adns_afterpoll	2874	// do not ever call adns_afterpoll
2104		2875
2105	Method 4: Do not use a prepare or check watcher because the module you	2876	Method 4: Do not use a prepare or check watcher because the module you
2106	want to embed is too inflexible to support it. Instead, you can override	2877	want to embed is not flexible enough to support it. Instead, you can
2107	their poll function. The drawback with this solution is that the main	2878	override their poll function. The drawback with this solution is that the
2108	loop is now no longer controllable by EV. The C<Glib::EV> module does	2879	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
2109	this.	2880	this approach, effectively embedding EV as a client into the horrible
		2881	libglib event loop.
2110		2882
2111	static gint	2883	static gint
2112	event_poll_func (GPollFD *fds, guint nfds, gint timeout)	2884	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
2113	{	2885	{
2114	int got_events = 0;	2886	int got_events = 0;
…		…
2118		2890
2119	if (timeout >= 0)	2891	if (timeout >= 0)
2120	// create/start timer	2892	// create/start timer
2121		2893
2122	// poll	2894	// poll
2123	ev_loop (EV_A_ 0);	2895	ev_run (EV_A_ 0);
2124		2896
2125	// stop timer again	2897	// stop timer again
2126	if (timeout >= 0)	2898	if (timeout >= 0)
2127	ev_timer_stop (EV_A_ &to);	2899	ev_timer_stop (EV_A_ &to);
2128		2900
…		…
2145	prioritise I/O.	2917	prioritise I/O.
2146		2918
2147	As an example for a bug workaround, the kqueue backend might only support	2919	As an example for a bug workaround, the kqueue backend might only support
2148	sockets on some platform, so it is unusable as generic backend, but you	2920	sockets on some platform, so it is unusable as generic backend, but you
2149	still want to make use of it because you have many sockets and it scales	2921	still want to make use of it because you have many sockets and it scales
2150	so nicely. In this case, you would create a kqueue-based loop and embed it	2922	so nicely. In this case, you would create a kqueue-based loop and embed
2151	into your default loop (which might use e.g. poll). Overall operation will	2923	it into your default loop (which might use e.g. poll). Overall operation
2152	be a bit slower because first libev has to poll and then call kevent, but	2924	will be a bit slower because first libev has to call C<poll> and then
2153	at least you can use both at what they are best.	2925	C<kevent>, but at least you can use both mechanisms for what they are
		2926	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
2154		2927
2155	As for prioritising I/O: rarely you have the case where some fds have	2928	As for prioritising I/O: under rare circumstances you have the case where
2156	to be watched and handled very quickly (with low latency), and even	2929	some fds have to be watched and handled very quickly (with low latency),
2157	priorities and idle watchers might have too much overhead. In this case	2930	and even priorities and idle watchers might have too much overhead. In
2158	you would put all the high priority stuff in one loop and all the rest in	2931	this case you would put all the high priority stuff in one loop and all
2159	a second one, and embed the second one in the first.	2932	the rest in a second one, and embed the second one in the first.
2160		2933
2161	As long as the watcher is active, the callback will be invoked every time	2934	As long as the watcher is active, the callback will be invoked every
2162	there might be events pending in the embedded loop. The callback must then	2935	time there might be events pending in the embedded loop. The callback
2163	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke	2936	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
2164	their callbacks (you could also start an idle watcher to give the embedded	2937	sweep and invoke their callbacks (the callback doesn't need to invoke the
2165	loop strictly lower priority for example). You can also set the callback	2938	C<ev_embed_sweep> function directly, it could also start an idle watcher
2166	to C<0>, in which case the embed watcher will automatically execute the	2939	to give the embedded loop strictly lower priority for example).
2167	embedded loop sweep.
2168		2940
2169	As long as the watcher is started it will automatically handle events. The	2941	You can also set the callback to C<0>, in which case the embed watcher
2170	callback will be invoked whenever some events have been handled. You can	2942	will automatically execute the embedded loop sweep whenever necessary.
2171	set the callback to C<0> to avoid having to specify one if you are not
2172	interested in that.
2173		2943
2174	Also, there have not currently been made special provisions for forking:	2944	Fork detection will be handled transparently while the C<ev_embed> watcher
2175	when you fork, you not only have to call C<ev_loop_fork> on both loops,	2945	is active, i.e., the embedded loop will automatically be forked when the
2176	but you will also have to stop and restart any C<ev_embed> watchers	2946	embedding loop forks. In other cases, the user is responsible for calling
2177	yourself.	2947	C<ev_loop_fork> on the embedded loop.
2178		2948
2179	Unfortunately, not all backends are embeddable, only the ones returned by	2949	Unfortunately, not all backends are embeddable: only the ones returned by
2180	C<ev_embeddable_backends> are, which, unfortunately, does not include any	2950	C<ev_embeddable_backends> are, which, unfortunately, does not include any
2181	portable one.	2951	portable one.
2182		2952
2183	So when you want to use this feature you will always have to be prepared	2953	So when you want to use this feature you will always have to be prepared
2184	that you cannot get an embeddable loop. The recommended way to get around	2954	that you cannot get an embeddable loop. The recommended way to get around
2185	this is to have a separate variables for your embeddable loop, try to	2955	this is to have a separate variables for your embeddable loop, try to
2186	create it, and if that fails, use the normal loop for everything.	2956	create it, and if that fails, use the normal loop for everything.
		2957
		2958	=head3 C<ev_embed> and fork
		2959
		2960	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2961	automatically be applied to the embedded loop as well, so no special
		2962	fork handling is required in that case. When the watcher is not running,
		2963	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2964	as applicable.
2187		2965
2188	=head3 Watcher-Specific Functions and Data Members	2966	=head3 Watcher-Specific Functions and Data Members
2189		2967
2190	=over 4	2968	=over 4
2191		2969
…		…
2200	if you do not want that, you need to temporarily stop the embed watcher).	2978	if you do not want that, you need to temporarily stop the embed watcher).
2201		2979
2202	=item ev_embed_sweep (loop, ev_embed *)	2980	=item ev_embed_sweep (loop, ev_embed *)
2203		2981
2204	Make a single, non-blocking sweep over the embedded loop. This works	2982	Make a single, non-blocking sweep over the embedded loop. This works
2205	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	2983	similarly to C<ev_run (embedded_loop, EVRUN_NOWAIT)>, but in the most
2206	appropriate way for embedded loops.	2984	appropriate way for embedded loops.
2207		2985
2208	=item struct ev_loop *other [read-only]	2986	=item struct ev_loop *other [read-only]
2209		2987
2210	The embedded event loop.	2988	The embedded event loop.
…		…
2219	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be	2997	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
2220	used).	2998	used).
2221		2999
2222	struct ev_loop *loop_hi = ev_default_init (0);	3000	struct ev_loop *loop_hi = ev_default_init (0);
2223	struct ev_loop *loop_lo = 0;	3001	struct ev_loop *loop_lo = 0;
2224	struct ev_embed embed;	3002	ev_embed embed;
2225		3003
2226	// see if there is a chance of getting one that works	3004	// see if there is a chance of getting one that works
2227	// (remember that a flags value of 0 means autodetection)	3005	// (remember that a flags value of 0 means autodetection)
2228	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()	3006	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
2229	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())	3007	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
…		…
2243	kqueue implementation). Store the kqueue/socket-only event loop in	3021	kqueue implementation). Store the kqueue/socket-only event loop in
2244	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).	3022	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
2245		3023
2246	struct ev_loop *loop = ev_default_init (0);	3024	struct ev_loop *loop = ev_default_init (0);
2247	struct ev_loop *loop_socket = 0;	3025	struct ev_loop *loop_socket = 0;
2248	struct ev_embed embed;	3026	ev_embed embed;
2249		3027
2250	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)	3028	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
2251	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))	3029	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
2252	{	3030	{
2253	ev_embed_init (&embed, 0, loop_socket);	3031	ev_embed_init (&embed, 0, loop_socket);
…		…
2268	event loop blocks next and before C<ev_check> watchers are being called,	3046	event loop blocks next and before C<ev_check> watchers are being called,
2269	and only in the child after the fork. If whoever good citizen calling	3047	and only in the child after the fork. If whoever good citizen calling
2270	C<ev_default_fork> cheats and calls it in the wrong process, the fork	3048	C<ev_default_fork> cheats and calls it in the wrong process, the fork
2271	handlers will be invoked, too, of course.	3049	handlers will be invoked, too, of course.
2272		3050
		3051	=head3 The special problem of life after fork - how is it possible?
		3052
		3053	Most uses of C<fork()> consist of forking, then some simple calls to set
		3054	up/change the process environment, followed by a call to C<exec()>. This
		3055	sequence should be handled by libev without any problems.
		3056
		3057	This changes when the application actually wants to do event handling
		3058	in the child, or both parent in child, in effect "continuing" after the
		3059	fork.
		3060
		3061	The default mode of operation (for libev, with application help to detect
		3062	forks) is to duplicate all the state in the child, as would be expected
		3063	when I<either> the parent I<or> the child process continues.
		3064
		3065	When both processes want to continue using libev, then this is usually the
		3066	wrong result. In that case, usually one process (typically the parent) is
		3067	supposed to continue with all watchers in place as before, while the other
		3068	process typically wants to start fresh, i.e. without any active watchers.
		3069
		3070	The cleanest and most efficient way to achieve that with libev is to
		3071	simply create a new event loop, which of course will be "empty", and
		3072	use that for new watchers. This has the advantage of not touching more
		3073	memory than necessary, and thus avoiding the copy-on-write, and the
		3074	disadvantage of having to use multiple event loops (which do not support
		3075	signal watchers).
		3076
		3077	When this is not possible, or you want to use the default loop for
		3078	other reasons, then in the process that wants to start "fresh", call
		3079	C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying
		3080	the default loop will "orphan" (not stop) all registered watchers, so you
		3081	have to be careful not to execute code that modifies those watchers. Note
		3082	also that in that case, you have to re-register any signal watchers.
		3083
2273	=head3 Watcher-Specific Functions and Data Members	3084	=head3 Watcher-Specific Functions and Data Members
2274		3085
2275	=over 4	3086	=over 4
2276		3087
2277	=item ev_fork_init (ev_signal *, callback)	3088	=item ev_fork_init (ev_signal *, callback)
…		…
2281	believe me.	3092	believe me.
2282		3093
2283	=back	3094	=back
2284		3095
2285		3096
2286	=head2 C<ev_async> - how to wake up another event loop	3097	=head2 C<ev_async> - how to wake up an event loop
2287		3098
2288	In general, you cannot use an C<ev_loop> from multiple threads or other	3099	In general, you cannot use an C<ev_run> from multiple threads or other
2289	asynchronous sources such as signal handlers (as opposed to multiple event	3100	asynchronous sources such as signal handlers (as opposed to multiple event
2290	loops - those are of course safe to use in different threads).	3101	loops - those are of course safe to use in different threads).
2291		3102
2292	Sometimes, however, you need to wake up another event loop you do not	3103	Sometimes, however, you need to wake up an event loop you do not control,
2293	control, for example because it belongs to another thread. This is what	3104	for example because it belongs to another thread. This is what C<ev_async>
2294	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you	3105	watchers do: as long as the C<ev_async> watcher is active, you can signal
2295	can signal it by calling C<ev_async_send>, which is thread- and signal	3106	it by calling C<ev_async_send>, which is thread- and signal safe.
2296	safe.
2297		3107
2298	This functionality is very similar to C<ev_signal> watchers, as signals,	3108	This functionality is very similar to C<ev_signal> watchers, as signals,
2299	too, are asynchronous in nature, and signals, too, will be compressed	3109	too, are asynchronous in nature, and signals, too, will be compressed
2300	(i.e. the number of callback invocations may be less than the number of	3110	(i.e. the number of callback invocations may be less than the number of
2301	C<ev_async_sent> calls).	3111	C<ev_async_sent> calls).
…		…
2306	=head3 Queueing	3116	=head3 Queueing
2307		3117
2308	C<ev_async> does not support queueing of data in any way. The reason	3118	C<ev_async> does not support queueing of data in any way. The reason
2309	is that the author does not know of a simple (or any) algorithm for a	3119	is that the author does not know of a simple (or any) algorithm for a
2310	multiple-writer-single-reader queue that works in all cases and doesn't	3120	multiple-writer-single-reader queue that works in all cases and doesn't
2311	need elaborate support such as pthreads.	3121	need elaborate support such as pthreads or unportable memory access
		3122	semantics.
2312		3123
2313	That means that if you want to queue data, you have to provide your own	3124	That means that if you want to queue data, you have to provide your own
2314	queue. But at least I can tell you would implement locking around your	3125	queue. But at least I can tell you how to implement locking around your
2315	queue:	3126	queue:
2316		3127
2317	=over 4	3128	=over 4
2318		3129
2319	=item queueing from a signal handler context	3130	=item queueing from a signal handler context
2320		3131
2321	To implement race-free queueing, you simply add to the queue in the signal	3132	To implement race-free queueing, you simply add to the queue in the signal
2322	handler but you block the signal handler in the watcher callback. Here is an example that does that for	3133	handler but you block the signal handler in the watcher callback. Here is
2323	some fictitious SIGUSR1 handler:	3134	an example that does that for some fictitious SIGUSR1 handler:
2324		3135
2325	static ev_async mysig;	3136	static ev_async mysig;
2326		3137
2327	static void	3138	static void
2328	sigusr1_handler (void)	3139	sigusr1_handler (void)
…		…
2394	=over 4	3205	=over 4
2395		3206
2396	=item ev_async_init (ev_async *, callback)	3207	=item ev_async_init (ev_async *, callback)
2397		3208
2398	Initialises and configures the async watcher - it has no parameters of any	3209	Initialises and configures the async watcher - it has no parameters of any
2399	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,	3210	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
2400	believe me.	3211	trust me.
2401		3212
2402	=item ev_async_send (loop, ev_async *)	3213	=item ev_async_send (loop, ev_async *)
2403		3214
2404	Sends/signals/activates the given C<ev_async> watcher, that is, feeds	3215	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
2405	an C<EV_ASYNC> event on the watcher into the event loop. Unlike	3216	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
2406	C<ev_feed_event>, this call is safe to do in other threads, signal or	3217	C<ev_feed_event>, this call is safe to do from other threads, signal or
2407	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding	3218	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
2408	section below on what exactly this means).	3219	section below on what exactly this means).
2409		3220
		3221	Note that, as with other watchers in libev, multiple events might get
		3222	compressed into a single callback invocation (another way to look at this
		3223	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
		3224	reset when the event loop detects that).
		3225
2410	This call incurs the overhead of a system call only once per loop iteration,	3226	This call incurs the overhead of a system call only once per event loop
2411	so while the overhead might be noticeable, it doesn't apply to repeated	3227	iteration, so while the overhead might be noticeable, it doesn't apply to
2412	calls to C<ev_async_send>.	3228	repeated calls to C<ev_async_send> for the same event loop.
2413		3229
2414	=item bool = ev_async_pending (ev_async *)	3230	=item bool = ev_async_pending (ev_async *)
2415		3231
2416	Returns a non-zero value when C<ev_async_send> has been called on the	3232	Returns a non-zero value when C<ev_async_send> has been called on the
2417	watcher but the event has not yet been processed (or even noted) by the	3233	watcher but the event has not yet been processed (or even noted) by the
…		…
2420	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When	3236	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
2421	the loop iterates next and checks for the watcher to have become active,	3237	the loop iterates next and checks for the watcher to have become active,
2422	it will reset the flag again. C<ev_async_pending> can be used to very	3238	it will reset the flag again. C<ev_async_pending> can be used to very
2423	quickly check whether invoking the loop might be a good idea.	3239	quickly check whether invoking the loop might be a good idea.
2424		3240
2425	Not that this does I<not> check whether the watcher itself is pending, only	3241	Not that this does I<not> check whether the watcher itself is pending,
2426	whether it has been requested to make this watcher pending.	3242	only whether it has been requested to make this watcher pending: there
		3243	is a time window between the event loop checking and resetting the async
		3244	notification, and the callback being invoked.
2427		3245
2428	=back	3246	=back
2429		3247
2430		3248
2431	=head1 OTHER FUNCTIONS	3249	=head1 OTHER FUNCTIONS
…		…
2435	=over 4	3253	=over 4
2436		3254
2437	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	3255	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2438		3256
2439	This function combines a simple timer and an I/O watcher, calls your	3257	This function combines a simple timer and an I/O watcher, calls your
2440	callback on whichever event happens first and automatically stop both	3258	callback on whichever event happens first and automatically stops both
2441	watchers. This is useful if you want to wait for a single event on an fd	3259	watchers. This is useful if you want to wait for a single event on an fd
2442	or timeout without having to allocate/configure/start/stop/free one or	3260	or timeout without having to allocate/configure/start/stop/free one or
2443	more watchers yourself.	3261	more watchers yourself.
2444		3262
2445	If C<fd> is less than 0, then no I/O watcher will be started and events	3263	If C<fd> is less than 0, then no I/O watcher will be started and the
2446	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	3264	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2447	C<events> set will be created and started.	3265	the given C<fd> and C<events> set will be created and started.
2448		3266
2449	If C<timeout> is less than 0, then no timeout watcher will be	3267	If C<timeout> is less than 0, then no timeout watcher will be
2450	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	3268	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
2451	repeat = 0) will be started. While C<0> is a valid timeout, it is of	3269	repeat = 0) will be started. C<0> is a valid timeout.
2452	dubious value.
2453		3270
2454	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	3271	The callback has the type C<void (*cb)(int revents, void *arg)> and is
2455	passed an C<revents> set like normal event callbacks (a combination of	3272	passed an C<revents> set like normal event callbacks (a combination of
2456	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	3273	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMER>) and the C<arg>
2457	value passed to C<ev_once>:	3274	value passed to C<ev_once>. Note that it is possible to receive I<both>
		3275	a timeout and an io event at the same time - you probably should give io
		3276	events precedence.
		3277
		3278	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
2458		3279
2459	static void stdin_ready (int revents, void *arg)	3280	static void stdin_ready (int revents, void *arg)
2460	{	3281	{
		3282	if (revents & EV_READ)
		3283	/* stdin might have data for us, joy! */;
2461	if (revents & EV_TIMEOUT)	3284	else if (revents & EV_TIMER)
2462	/* doh, nothing entered */;	3285	/* doh, nothing entered */;
2463	else if (revents & EV_READ)
2464	/* stdin might have data for us, joy! */;
2465	}	3286	}
2466		3287
2467	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	3288	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2468		3289
2469	=item ev_feed_event (ev_loop *, watcher *, int revents)
2470
2471	Feeds the given event set into the event loop, as if the specified event
2472	had happened for the specified watcher (which must be a pointer to an
2473	initialised but not necessarily started event watcher).
2474
2475	=item ev_feed_fd_event (ev_loop *, int fd, int revents)	3290	=item ev_feed_fd_event (loop, int fd, int revents)
2476		3291
2477	Feed an event on the given fd, as if a file descriptor backend detected	3292	Feed an event on the given fd, as if a file descriptor backend detected
2478	the given events it.	3293	the given events it.
2479		3294
2480	=item ev_feed_signal_event (ev_loop *loop, int signum)	3295	=item ev_feed_signal_event (loop, int signum)
2481		3296
2482	Feed an event as if the given signal occurred (C<loop> must be the default	3297	Feed an event as if the given signal occurred (C<loop> must be the default
2483	loop!).	3298	loop!).
2484		3299
2485	=back	3300	=back
…		…
2565		3380
2566	=over 4	3381	=over 4
2567		3382
2568	=item ev::TYPE::TYPE ()	3383	=item ev::TYPE::TYPE ()
2569		3384
2570	=item ev::TYPE::TYPE (struct ev_loop *)	3385	=item ev::TYPE::TYPE (loop)
2571		3386
2572	=item ev::TYPE::~TYPE	3387	=item ev::TYPE::~TYPE
2573		3388
2574	The constructor (optionally) takes an event loop to associate the watcher	3389	The constructor (optionally) takes an event loop to associate the watcher
2575	with. If it is omitted, it will use C<EV_DEFAULT>.	3390	with. If it is omitted, it will use C<EV_DEFAULT>.
…		…
2607		3422
2608	myclass obj;	3423	myclass obj;
2609	ev::io iow;	3424	ev::io iow;
2610	iow.set <myclass, &myclass::io_cb> (&obj);	3425	iow.set <myclass, &myclass::io_cb> (&obj);
2611		3426
		3427	=item w->set (object *)
		3428
		3429	This is a variation of a method callback - leaving out the method to call
		3430	will default the method to C<operator ()>, which makes it possible to use
		3431	functor objects without having to manually specify the C<operator ()> all
		3432	the time. Incidentally, you can then also leave out the template argument
		3433	list.
		3434
		3435	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3436	int revents)>.
		3437
		3438	See the method-C<set> above for more details.
		3439
		3440	Example: use a functor object as callback.
		3441
		3442	struct myfunctor
		3443	{
		3444	void operator() (ev::io &w, int revents)
		3445	{
		3446	...
		3447	}
		3448	}
		3449
		3450	myfunctor f;
		3451
		3452	ev::io w;
		3453	w.set (&f);
		3454
2612	=item w->set<function> (void *data = 0)	3455	=item w->set<function> (void *data = 0)
2613		3456
2614	Also sets a callback, but uses a static method or plain function as	3457	Also sets a callback, but uses a static method or plain function as
2615	callback. The optional C<data> argument will be stored in the watcher's	3458	callback. The optional C<data> argument will be stored in the watcher's
2616	C<data> member and is free for you to use.	3459	C<data> member and is free for you to use.
2617		3460
2618	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.	3461	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
2619		3462
2620	See the method-C<set> above for more details.	3463	See the method-C<set> above for more details.
2621		3464
2622	Example:	3465	Example: Use a plain function as callback.
2623		3466
2624	static void io_cb (ev::io &w, int revents) { }	3467	static void io_cb (ev::io &w, int revents) { }
2625	iow.set <io_cb> ();	3468	iow.set <io_cb> ();
2626		3469
2627	=item w->set (struct ev_loop *)	3470	=item w->set (loop)
2628		3471
2629	Associates a different C<struct ev_loop> with this watcher. You can only	3472	Associates a different C<struct ev_loop> with this watcher. You can only
2630	do this when the watcher is inactive (and not pending either).	3473	do this when the watcher is inactive (and not pending either).
2631		3474
2632	=item w->set ([arguments])	3475	=item w->set ([arguments])
2633		3476
2634	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be	3477	Basically the same as C<ev_TYPE_set>, with the same arguments. Either this
2635	called at least once. Unlike the C counterpart, an active watcher gets	3478	method or a suitable start method must be called at least once. Unlike the
2636	automatically stopped and restarted when reconfiguring it with this	3479	C counterpart, an active watcher gets automatically stopped and restarted
2637	method.	3480	when reconfiguring it with this method.
2638		3481
2639	=item w->start ()	3482	=item w->start ()
2640		3483
2641	Starts the watcher. Note that there is no C<loop> argument, as the	3484	Starts the watcher. Note that there is no C<loop> argument, as the
2642	constructor already stores the event loop.	3485	constructor already stores the event loop.
2643		3486
		3487	=item w->start ([arguments])
		3488
		3489	Instead of calling C<set> and C<start> methods separately, it is often
		3490	convenient to wrap them in one call. Uses the same type of arguments as
		3491	the configure C<set> method of the watcher.
		3492
2644	=item w->stop ()	3493	=item w->stop ()
2645		3494
2646	Stops the watcher if it is active. Again, no C<loop> argument.	3495	Stops the watcher if it is active. Again, no C<loop> argument.
2647		3496
2648	=item w->again () (C<ev::timer>, C<ev::periodic> only)	3497	=item w->again () (C<ev::timer>, C<ev::periodic> only)
…		…
2660		3509
2661	=back	3510	=back
2662		3511
2663	=back	3512	=back
2664		3513
2665	Example: Define a class with an IO and idle watcher, start one of them in	3514	Example: Define a class with two I/O and idle watchers, start the I/O
2666	the constructor.	3515	watchers in the constructor.
2667		3516
2668	class myclass	3517	class myclass
2669	{	3518	{
2670	ev::io io; void io_cb (ev::io &w, int revents);	3519	ev::io io ; void io_cb (ev::io &w, int revents);
		3520	ev::io2 io2 ; void io2_cb (ev::io &w, int revents);
2671	ev:idle idle void idle_cb (ev::idle &w, int revents);	3521	ev::idle idle; void idle_cb (ev::idle &w, int revents);
2672		3522
2673	myclass (int fd)	3523	myclass (int fd)
2674	{	3524	{
2675	io .set <myclass, &myclass::io_cb > (this);	3525	io .set <myclass, &myclass::io_cb > (this);
		3526	io2 .set <myclass, &myclass::io2_cb > (this);
2676	idle.set <myclass, &myclass::idle_cb> (this);	3527	idle.set <myclass, &myclass::idle_cb> (this);
2677		3528
2678	io.start (fd, ev::READ);	3529	io.set (fd, ev::WRITE); // configure the watcher
		3530	io.start (); // start it whenever convenient
		3531
		3532	io2.start (fd, ev::READ); // set + start in one call
2679	}	3533	}
2680	};	3534	};
2681		3535
2682		3536
2683	=head1 OTHER LANGUAGE BINDINGS	3537	=head1 OTHER LANGUAGE BINDINGS
…		…
2692	=item Perl	3546	=item Perl
2693		3547
2694	The EV module implements the full libev API and is actually used to test	3548	The EV module implements the full libev API and is actually used to test
2695	libev. EV is developed together with libev. Apart from the EV core module,	3549	libev. EV is developed together with libev. Apart from the EV core module,
2696	there are additional modules that implement libev-compatible interfaces	3550	there are additional modules that implement libev-compatible interfaces
2697	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the	3551	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
2698	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).	3552	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3553	and C<EV::Glib>).
2699		3554
2700	It can be found and installed via CPAN, its homepage is at	3555	It can be found and installed via CPAN, its homepage is at
2701	L<http://software.schmorp.de/pkg/EV>.	3556	L<http://software.schmorp.de/pkg/EV>.
2702		3557
2703	=item Python	3558	=item Python
2704		3559
2705	Python bindings can be found at L<http://code.google.com/p/pyev/>. It	3560	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
2706	seems to be quite complete and well-documented. Note, however, that the	3561	seems to be quite complete and well-documented.
2707	patch they require for libev is outright dangerous as it breaks the ABI
2708	for everybody else, and therefore, should never be applied in an installed
2709	libev (if python requires an incompatible ABI then it needs to embed
2710	libev).
2711		3562
2712	=item Ruby	3563	=item Ruby
2713		3564
2714	Tony Arcieri has written a ruby extension that offers access to a subset	3565	Tony Arcieri has written a ruby extension that offers access to a subset
2715	of the libev API and adds file handle abstractions, asynchronous DNS and	3566	of the libev API and adds file handle abstractions, asynchronous DNS and
2716	more on top of it. It can be found via gem servers. Its homepage is at	3567	more on top of it. It can be found via gem servers. Its homepage is at
2717	L<http://rev.rubyforge.org/>.	3568	L<http://rev.rubyforge.org/>.
2718		3569
		3570	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		3571	makes rev work even on mingw.
		3572
		3573	=item Haskell
		3574
		3575	A haskell binding to libev is available at
		3576	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		3577
2719	=item D	3578	=item D
2720		3579
2721	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to	3580	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
2722	be found at L<http://proj.llucax.com.ar/wiki/evd>.	3581	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		3582
		3583	=item Ocaml
		3584
		3585	Erkki Seppala has written Ocaml bindings for libev, to be found at
		3586	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		3587
		3588	=item Lua
		3589
		3590	Brian Maher has written a partial interface to libev for lua (at the
		3591	time of this writing, only C<ev_io> and C<ev_timer>), to be found at
		3592	L<http://github.com/brimworks/lua-ev>.
2723		3593
2724	=back	3594	=back
2725		3595
2726		3596
2727	=head1 MACRO MAGIC	3597	=head1 MACRO MAGIC
…		…
2741	loop argument"). The C<EV_A> form is used when this is the sole argument,	3611	loop argument"). The C<EV_A> form is used when this is the sole argument,
2742	C<EV_A_> is used when other arguments are following. Example:	3612	C<EV_A_> is used when other arguments are following. Example:
2743		3613
2744	ev_unref (EV_A);	3614	ev_unref (EV_A);
2745	ev_timer_add (EV_A_ watcher);	3615	ev_timer_add (EV_A_ watcher);
2746	ev_loop (EV_A_ 0);	3616	ev_run (EV_A_ 0);
2747		3617
2748	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	3618	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
2749	which is often provided by the following macro.	3619	which is often provided by the following macro.
2750		3620
2751	=item C<EV_P>, C<EV_P_>	3621	=item C<EV_P>, C<EV_P_>
…		…
2791	}	3661	}
2792		3662
2793	ev_check check;	3663	ev_check check;
2794	ev_check_init (&check, check_cb);	3664	ev_check_init (&check, check_cb);
2795	ev_check_start (EV_DEFAULT_ &check);	3665	ev_check_start (EV_DEFAULT_ &check);
2796	ev_loop (EV_DEFAULT_ 0);	3666	ev_run (EV_DEFAULT_ 0);
2797		3667
2798	=head1 EMBEDDING	3668	=head1 EMBEDDING
2799		3669
2800	Libev can (and often is) directly embedded into host	3670	Libev can (and often is) directly embedded into host
2801	applications. Examples of applications that embed it include the Deliantra	3671	applications. Examples of applications that embed it include the Deliantra
…		…
2828		3698
2829	#define EV_STANDALONE 1	3699	#define EV_STANDALONE 1
2830	#include "ev.h"	3700	#include "ev.h"
2831		3701
2832	Both header files and implementation files can be compiled with a C++	3702	Both header files and implementation files can be compiled with a C++
2833	compiler (at least, thats a stated goal, and breakage will be treated	3703	compiler (at least, that's a stated goal, and breakage will be treated
2834	as a bug).	3704	as a bug).
2835		3705
2836	You need the following files in your source tree, or in a directory	3706	You need the following files in your source tree, or in a directory
2837	in your include path (e.g. in libev/ when using -Ilibev):	3707	in your include path (e.g. in libev/ when using -Ilibev):
2838		3708
…		…
2881	libev.m4	3751	libev.m4
2882		3752
2883	=head2 PREPROCESSOR SYMBOLS/MACROS	3753	=head2 PREPROCESSOR SYMBOLS/MACROS
2884		3754
2885	Libev can be configured via a variety of preprocessor symbols you have to	3755	Libev can be configured via a variety of preprocessor symbols you have to
2886	define before including any of its files. The default in the absence of	3756	define before including (or compiling) any of its files. The default in
2887	autoconf is noted for every option.	3757	the absence of autoconf is documented for every option.
		3758
		3759	Symbols marked with "(h)" do not change the ABI, and can have different
		3760	values when compiling libev vs. including F<ev.h>, so it is permissible
		3761	to redefine them before including F<ev.h> without breaking compatibility
		3762	to a compiled library. All other symbols change the ABI, which means all
		3763	users of libev and the libev code itself must be compiled with compatible
		3764	settings.
2888		3765
2889	=over 4	3766	=over 4
2890		3767
		3768	=item EV_COMPAT3 (h)
		3769
		3770	Backwards compatibility is a major concern for libev. This is why this
		3771	release of libev comes with wrappers for the functions and symbols that
		3772	have been renamed between libev version 3 and 4.
		3773
		3774	You can disable these wrappers (to test compatibility with future
		3775	versions) by defining C<EV_COMPAT3> to C<0> when compiling your
		3776	sources. This has the additional advantage that you can drop the C<struct>
		3777	from C<struct ev_loop> declarations, as libev will provide an C<ev_loop>
		3778	typedef in that case.
		3779
		3780	In some future version, the default for C<EV_COMPAT3> will become C<0>,
		3781	and in some even more future version the compatibility code will be
		3782	removed completely.
		3783
2891	=item EV_STANDALONE	3784	=item EV_STANDALONE (h)
2892		3785
2893	Must always be C<1> if you do not use autoconf configuration, which	3786	Must always be C<1> if you do not use autoconf configuration, which
2894	keeps libev from including F<config.h>, and it also defines dummy	3787	keeps libev from including F<config.h>, and it also defines dummy
2895	implementations for some libevent functions (such as logging, which is not	3788	implementations for some libevent functions (such as logging, which is not
2896	supported). It will also not define any of the structs usually found in	3789	supported). It will also not define any of the structs usually found in
2897	F<event.h> that are not directly supported by the libev core alone.	3790	F<event.h> that are not directly supported by the libev core alone.
2898		3791
		3792	In standalone mode, libev will still try to automatically deduce the
		3793	configuration, but has to be more conservative.
		3794
2899	=item EV_USE_MONOTONIC	3795	=item EV_USE_MONOTONIC
2900		3796
2901	If defined to be C<1>, libev will try to detect the availability of the	3797	If defined to be C<1>, libev will try to detect the availability of the
2902	monotonic clock option at both compile time and runtime. Otherwise no use	3798	monotonic clock option at both compile time and runtime. Otherwise no
2903	of the monotonic clock option will be attempted. If you enable this, you	3799	use of the monotonic clock option will be attempted. If you enable this,
2904	usually have to link against librt or something similar. Enabling it when	3800	you usually have to link against librt or something similar. Enabling it
2905	the functionality isn't available is safe, though, although you have	3801	when the functionality isn't available is safe, though, although you have
2906	to make sure you link against any libraries where the C<clock_gettime>	3802	to make sure you link against any libraries where the C<clock_gettime>
2907	function is hiding in (often F<-lrt>).	3803	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
2908		3804
2909	=item EV_USE_REALTIME	3805	=item EV_USE_REALTIME
2910		3806
2911	If defined to be C<1>, libev will try to detect the availability of the	3807	If defined to be C<1>, libev will try to detect the availability of the
2912	real-time clock option at compile time (and assume its availability at	3808	real-time clock option at compile time (and assume its availability
2913	runtime if successful). Otherwise no use of the real-time clock option will	3809	at runtime if successful). Otherwise no use of the real-time clock
2914	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	3810	option will be attempted. This effectively replaces C<gettimeofday>
2915	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the	3811	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
2916	note about libraries in the description of C<EV_USE_MONOTONIC>, though.	3812	correctness. See the note about libraries in the description of
		3813	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		3814	C<EV_USE_CLOCK_SYSCALL>.
		3815
		3816	=item EV_USE_CLOCK_SYSCALL
		3817
		3818	If defined to be C<1>, libev will try to use a direct syscall instead
		3819	of calling the system-provided C<clock_gettime> function. This option
		3820	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		3821	unconditionally pulls in C<libpthread>, slowing down single-threaded
		3822	programs needlessly. Using a direct syscall is slightly slower (in
		3823	theory), because no optimised vdso implementation can be used, but avoids
		3824	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		3825	higher, as it simplifies linking (no need for C<-lrt>).
2917		3826
2918	=item EV_USE_NANOSLEEP	3827	=item EV_USE_NANOSLEEP
2919		3828
2920	If defined to be C<1>, libev will assume that C<nanosleep ()> is available	3829	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
2921	and will use it for delays. Otherwise it will use C<select ()>.	3830	and will use it for delays. Otherwise it will use C<select ()>.
…		…
2937		3846
2938	=item EV_SELECT_USE_FD_SET	3847	=item EV_SELECT_USE_FD_SET
2939		3848
2940	If defined to C<1>, then the select backend will use the system C<fd_set>	3849	If defined to C<1>, then the select backend will use the system C<fd_set>
2941	structure. This is useful if libev doesn't compile due to a missing	3850	structure. This is useful if libev doesn't compile due to a missing
2942	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on	3851	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
2943	exotic systems. This usually limits the range of file descriptors to some	3852	on exotic systems. This usually limits the range of file descriptors to
2944	low limit such as 1024 or might have other limitations (winsocket only	3853	some low limit such as 1024 or might have other limitations (winsocket
2945	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	3854	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
2946	influence the size of the C<fd_set> used.	3855	configures the maximum size of the C<fd_set>.
2947		3856
2948	=item EV_SELECT_IS_WINSOCKET	3857	=item EV_SELECT_IS_WINSOCKET
2949		3858
2950	When defined to C<1>, the select backend will assume that	3859	When defined to C<1>, the select backend will assume that
2951	select/socket/connect etc. don't understand file descriptors but	3860	select/socket/connect etc. don't understand file descriptors but
…		…
2953	be used is the winsock select). This means that it will call	3862	be used is the winsock select). This means that it will call
2954	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	3863	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
2955	it is assumed that all these functions actually work on fds, even	3864	it is assumed that all these functions actually work on fds, even
2956	on win32. Should not be defined on non-win32 platforms.	3865	on win32. Should not be defined on non-win32 platforms.
2957		3866
2958	=item EV_FD_TO_WIN32_HANDLE	3867	=item EV_FD_TO_WIN32_HANDLE(fd)
2959		3868
2960	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map	3869	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
2961	file descriptors to socket handles. When not defining this symbol (the	3870	file descriptors to socket handles. When not defining this symbol (the
2962	default), then libev will call C<_get_osfhandle>, which is usually	3871	default), then libev will call C<_get_osfhandle>, which is usually
2963	correct. In some cases, programs use their own file descriptor management,	3872	correct. In some cases, programs use their own file descriptor management,
2964	in which case they can provide this function to map fds to socket handles.	3873	in which case they can provide this function to map fds to socket handles.
		3874
		3875	=item EV_WIN32_HANDLE_TO_FD(handle)
		3876
		3877	If C<EV_SELECT_IS_WINSOCKET> then libev maps handles to file descriptors
		3878	using the standard C<_open_osfhandle> function. For programs implementing
		3879	their own fd to handle mapping, overwriting this function makes it easier
		3880	to do so. This can be done by defining this macro to an appropriate value.
		3881
		3882	=item EV_WIN32_CLOSE_FD(fd)
		3883
		3884	If programs implement their own fd to handle mapping on win32, then this
		3885	macro can be used to override the C<close> function, useful to unregister
		3886	file descriptors again. Note that the replacement function has to close
		3887	the underlying OS handle.
2965		3888
2966	=item EV_USE_POLL	3889	=item EV_USE_POLL
2967		3890
2968	If defined to be C<1>, libev will compile in support for the C<poll>(2)	3891	If defined to be C<1>, libev will compile in support for the C<poll>(2)
2969	backend. Otherwise it will be enabled on non-win32 platforms. It	3892	backend. Otherwise it will be enabled on non-win32 platforms. It
…		…
3016	as well as for signal and thread safety in C<ev_async> watchers.	3939	as well as for signal and thread safety in C<ev_async> watchers.
3017		3940
3018	In the absence of this define, libev will use C<sig_atomic_t volatile>	3941	In the absence of this define, libev will use C<sig_atomic_t volatile>
3019	(from F<signal.h>), which is usually good enough on most platforms.	3942	(from F<signal.h>), which is usually good enough on most platforms.
3020		3943
3021	=item EV_H	3944	=item EV_H (h)
3022		3945
3023	The name of the F<ev.h> header file used to include it. The default if	3946	The name of the F<ev.h> header file used to include it. The default if
3024	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be	3947	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
3025	used to virtually rename the F<ev.h> header file in case of conflicts.	3948	used to virtually rename the F<ev.h> header file in case of conflicts.
3026		3949
3027	=item EV_CONFIG_H	3950	=item EV_CONFIG_H (h)
3028		3951
3029	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	3952	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
3030	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	3953	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
3031	C<EV_H>, above.	3954	C<EV_H>, above.
3032		3955
3033	=item EV_EVENT_H	3956	=item EV_EVENT_H (h)
3034		3957
3035	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	3958	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
3036	of how the F<event.h> header can be found, the default is C<"event.h">.	3959	of how the F<event.h> header can be found, the default is C<"event.h">.
3037		3960
3038	=item EV_PROTOTYPES	3961	=item EV_PROTOTYPES (h)
3039		3962
3040	If defined to be C<0>, then F<ev.h> will not define any function	3963	If defined to be C<0>, then F<ev.h> will not define any function
3041	prototypes, but still define all the structs and other symbols. This is	3964	prototypes, but still define all the structs and other symbols. This is
3042	occasionally useful if you want to provide your own wrapper functions	3965	occasionally useful if you want to provide your own wrapper functions
3043	around libev functions.	3966	around libev functions.
…		…
3062	When doing priority-based operations, libev usually has to linearly search	3985	When doing priority-based operations, libev usually has to linearly search
3063	all the priorities, so having many of them (hundreds) uses a lot of space	3986	all the priorities, so having many of them (hundreds) uses a lot of space
3064	and time, so using the defaults of five priorities (-2 .. +2) is usually	3987	and time, so using the defaults of five priorities (-2 .. +2) is usually
3065	fine.	3988	fine.
3066		3989
3067	If your embedding application does not need any priorities, defining these both to	3990	If your embedding application does not need any priorities, defining these
3068	C<0> will save some memory and CPU.	3991	both to C<0> will save some memory and CPU.
3069		3992
3070	=item EV_PERIODIC_ENABLE	3993	=item EV_PERIODIC_ENABLE, EV_IDLE_ENABLE, EV_EMBED_ENABLE, EV_STAT_ENABLE,
		3994	EV_PREPARE_ENABLE, EV_CHECK_ENABLE, EV_FORK_ENABLE, EV_SIGNAL_ENABLE,
		3995	EV_ASYNC_ENABLE, EV_CHILD_ENABLE.
3071		3996
3072	If undefined or defined to be C<1>, then periodic timers are supported. If	3997	If undefined or defined to be C<1> (and the platform supports it), then
3073	defined to be C<0>, then they are not. Disabling them saves a few kB of	3998	the respective watcher type is supported. If defined to be C<0>, then it
3074	code.	3999	is not. Disabling watcher types mainly saves code size.
3075		4000
3076	=item EV_IDLE_ENABLE	4001	=item EV_FEATURES
3077
3078	If undefined or defined to be C<1>, then idle watchers are supported. If
3079	defined to be C<0>, then they are not. Disabling them saves a few kB of
3080	code.
3081
3082	=item EV_EMBED_ENABLE
3083
3084	If undefined or defined to be C<1>, then embed watchers are supported. If
3085	defined to be C<0>, then they are not.
3086
3087	=item EV_STAT_ENABLE
3088
3089	If undefined or defined to be C<1>, then stat watchers are supported. If
3090	defined to be C<0>, then they are not.
3091
3092	=item EV_FORK_ENABLE
3093
3094	If undefined or defined to be C<1>, then fork watchers are supported. If
3095	defined to be C<0>, then they are not.
3096
3097	=item EV_ASYNC_ENABLE
3098
3099	If undefined or defined to be C<1>, then async watchers are supported. If
3100	defined to be C<0>, then they are not.
3101
3102	=item EV_MINIMAL
3103		4002
3104	If you need to shave off some kilobytes of code at the expense of some	4003	If you need to shave off some kilobytes of code at the expense of some
3105	speed, define this symbol to C<1>. Currently this is used to override some	4004	speed (but with the full API), you can define this symbol to request
3106	inlining decisions, saves roughly 30% code size on amd64. It also selects a	4005	certain subsets of functionality. The default is to enable all features
3107	much smaller 2-heap for timer management over the default 4-heap.	4006	that can be enabled on the platform.
		4007
		4008	A typical way to use this symbol is to define it to C<0> (or to a bitset
		4009	with some broad features you want) and then selectively re-enable
		4010	additional parts you want, for example if you want everything minimal,
		4011	but multiple event loop support, async and child watchers and the poll
		4012	backend, use this:
		4013
		4014	#define EV_FEATURES 0
		4015	#define EV_MULTIPLICITY 1
		4016	#define EV_USE_POLL 1
		4017	#define EV_CHILD_ENABLE 1
		4018	#define EV_ASYNC_ENABLE 1
		4019
		4020	The actual value is a bitset, it can be a combination of the following
		4021	values:
		4022
		4023	=over 4
		4024
		4025	=item C<1> - faster/larger code
		4026
		4027	Use larger code to speed up some operations.
		4028
		4029	Currently this is used to override some inlining decisions (enlarging the
		4030	code size by roughly 30% on amd64).
		4031
		4032	When optimising for size, use of compiler flags such as C<-Os> with
		4033	gcc is recommended, as well as C<-DNDEBUG>, as libev contains a number of
		4034	assertions.
		4035
		4036	=item C<2> - faster/larger data structures
		4037
		4038	Replaces the small 2-heap for timer management by a faster 4-heap, larger
		4039	hash table sizes and so on. This will usually further increase code size
		4040	and can additionally have an effect on the size of data structures at
		4041	runtime.
		4042
		4043	=item C<4> - full API configuration
		4044
		4045	This enables priorities (sets C<EV_MAXPRI>=2 and C<EV_MINPRI>=-2), and
		4046	enables multiplicity (C<EV_MULTIPLICITY>=1).
		4047
		4048	=item C<8> - full API
		4049
		4050	This enables a lot of the "lesser used" API functions. See C<ev.h> for
		4051	details on which parts of the API are still available without this
		4052	feature, and do not complain if this subset changes over time.
		4053
		4054	=item C<16> - enable all optional watcher types
		4055
		4056	Enables all optional watcher types. If you want to selectively enable
		4057	only some watcher types other than I/O and timers (e.g. prepare,
		4058	embed, async, child...) you can enable them manually by defining
		4059	C<EV_watchertype_ENABLE> to C<1> instead.
		4060
		4061	=item C<32> - enable all backends
		4062
		4063	This enables all backends - without this feature, you need to enable at
		4064	least one backend manually (C<EV_USE_SELECT> is a good choice).
		4065
		4066	=item C<64> - enable OS-specific "helper" APIs
		4067
		4068	Enable inotify, eventfd, signalfd and similar OS-specific helper APIs by
		4069	default.
		4070
		4071	=back
		4072
		4073	Compiling with C<gcc -Os -DEV_STANDALONE -DEV_USE_EPOLL=1 -DEV_FEATURES=0>
		4074	reduces the compiled size of libev from 24.7Kb code/2.8Kb data to 6.5Kb
		4075	code/0.3Kb data on my GNU/Linux amd64 system, while still giving you I/O
		4076	watchers, timers and monotonic clock support.
		4077
		4078	With an intelligent-enough linker (gcc+binutils are intelligent enough
		4079	when you use C<-Wl,--gc-sections -ffunction-sections>) functions unused by
		4080	your program might be left out as well - a binary starting a timer and an
		4081	I/O watcher then might come out at only 5Kb.
		4082
		4083	=item EV_AVOID_STDIO
		4084
		4085	If this is set to C<1> at compiletime, then libev will avoid using stdio
		4086	functions (printf, scanf, perror etc.). This will increase the code size
		4087	somewhat, but if your program doesn't otherwise depend on stdio and your
		4088	libc allows it, this avoids linking in the stdio library which is quite
		4089	big.
		4090
		4091	Note that error messages might become less precise when this option is
		4092	enabled.
		4093
		4094	=item EV_NSIG
		4095
		4096	The highest supported signal number, +1 (or, the number of
		4097	signals): Normally, libev tries to deduce the maximum number of signals
		4098	automatically, but sometimes this fails, in which case it can be
		4099	specified. Also, using a lower number than detected (C<32> should be
		4100	good for about any system in existence) can save some memory, as libev
		4101	statically allocates some 12-24 bytes per signal number.
3108		4102
3109	=item EV_PID_HASHSIZE	4103	=item EV_PID_HASHSIZE
3110		4104
3111	C<ev_child> watchers use a small hash table to distribute workload by	4105	C<ev_child> watchers use a small hash table to distribute workload by
3112	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	4106	pid. The default size is C<16> (or C<1> with C<EV_FEATURES> disabled),
3113	than enough. If you need to manage thousands of children you might want to	4107	usually more than enough. If you need to manage thousands of children you
3114	increase this value (I<must> be a power of two).	4108	might want to increase this value (I<must> be a power of two).
3115		4109
3116	=item EV_INOTIFY_HASHSIZE	4110	=item EV_INOTIFY_HASHSIZE
3117		4111
3118	C<ev_stat> watchers use a small hash table to distribute workload by	4112	C<ev_stat> watchers use a small hash table to distribute workload by
3119	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	4113	inotify watch id. The default size is C<16> (or C<1> with C<EV_FEATURES>
3120	usually more than enough. If you need to manage thousands of C<ev_stat>	4114	disabled), usually more than enough. If you need to manage thousands of
3121	watchers you might want to increase this value (I<must> be a power of	4115	C<ev_stat> watchers you might want to increase this value (I<must> be a
3122	two).	4116	power of two).
3123		4117
3124	=item EV_USE_4HEAP	4118	=item EV_USE_4HEAP
3125		4119
3126	Heaps are not very cache-efficient. To improve the cache-efficiency of the	4120	Heaps are not very cache-efficient. To improve the cache-efficiency of the
3127	timer and periodics heap, libev uses a 4-heap when this symbol is defined	4121	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
3128	to C<1>. The 4-heap uses more complicated (longer) code but has	4122	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
3129	noticeably faster performance with many (thousands) of watchers.	4123	faster performance with many (thousands) of watchers.
3130		4124
3131	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>	4125	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
3132	(disabled).	4126	will be C<0>.
3133		4127
3134	=item EV_HEAP_CACHE_AT	4128	=item EV_HEAP_CACHE_AT
3135		4129
3136	Heaps are not very cache-efficient. To improve the cache-efficiency of the	4130	Heaps are not very cache-efficient. To improve the cache-efficiency of the
3137	timer and periodics heap, libev can cache the timestamp (I<at>) within	4131	timer and periodics heaps, libev can cache the timestamp (I<at>) within
3138	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),	4132	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
3139	which uses 8-12 bytes more per watcher and a few hundred bytes more code,	4133	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
3140	but avoids random read accesses on heap changes. This improves performance	4134	but avoids random read accesses on heap changes. This improves performance
3141	noticeably with with many (hundreds) of watchers.	4135	noticeably with many (hundreds) of watchers.
3142		4136
3143	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>	4137	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
3144	(disabled).	4138	will be C<0>.
3145		4139
3146	=item EV_VERIFY	4140	=item EV_VERIFY
3147		4141
3148	Controls how much internal verification (see C<ev_loop_verify ()>) will	4142	Controls how much internal verification (see C<ev_verify ()>) will
3149	be done: If set to C<0>, no internal verification code will be compiled	4143	be done: If set to C<0>, no internal verification code will be compiled
3150	in. If set to C<1>, then verification code will be compiled in, but not	4144	in. If set to C<1>, then verification code will be compiled in, but not
3151	called. If set to C<2>, then the internal verification code will be	4145	called. If set to C<2>, then the internal verification code will be
3152	called once per loop, which can slow down libev. If set to C<3>, then the	4146	called once per loop, which can slow down libev. If set to C<3>, then the
3153	verification code will be called very frequently, which will slow down	4147	verification code will be called very frequently, which will slow down
3154	libev considerably.	4148	libev considerably.
3155		4149
3156	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be	4150	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
3157	C<0.>	4151	will be C<0>.
3158		4152
3159	=item EV_COMMON	4153	=item EV_COMMON
3160		4154
3161	By default, all watchers have a C<void *data> member. By redefining	4155	By default, all watchers have a C<void *data> member. By redefining
3162	this macro to a something else you can include more and other types of	4156	this macro to something else you can include more and other types of
3163	members. You have to define it each time you include one of the files,	4157	members. You have to define it each time you include one of the files,
3164	though, and it must be identical each time.	4158	though, and it must be identical each time.
3165		4159
3166	For example, the perl EV module uses something like this:	4160	For example, the perl EV module uses something like this:
3167		4161
…		…
3179	and the way callbacks are invoked and set. Must expand to a struct member	4173	and the way callbacks are invoked and set. Must expand to a struct member
3180	definition and a statement, respectively. See the F<ev.h> header file for	4174	definition and a statement, respectively. See the F<ev.h> header file for
3181	their default definitions. One possible use for overriding these is to	4175	their default definitions. One possible use for overriding these is to
3182	avoid the C<struct ev_loop *> as first argument in all cases, or to use	4176	avoid the C<struct ev_loop *> as first argument in all cases, or to use
3183	method calls instead of plain function calls in C++.	4177	method calls instead of plain function calls in C++.
		4178
		4179	=back
3184		4180
3185	=head2 EXPORTED API SYMBOLS	4181	=head2 EXPORTED API SYMBOLS
3186		4182
3187	If you need to re-export the API (e.g. via a DLL) and you need a list of	4183	If you need to re-export the API (e.g. via a DLL) and you need a list of
3188	exported symbols, you can use the provided F<Symbol.*> files which list	4184	exported symbols, you can use the provided F<Symbol.*> files which list
…		…
3218	file.	4214	file.
3219		4215
3220	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	4216	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
3221	that everybody includes and which overrides some configure choices:	4217	that everybody includes and which overrides some configure choices:
3222		4218
3223	#define EV_MINIMAL 1	4219	#define EV_FEATURES 8
3224	#define EV_USE_POLL 0	4220	#define EV_USE_SELECT 1
3225	#define EV_MULTIPLICITY 0
3226	#define EV_PERIODIC_ENABLE 0	4221	#define EV_PREPARE_ENABLE 1
		4222	#define EV_IDLE_ENABLE 1
3227	#define EV_STAT_ENABLE 0	4223	#define EV_SIGNAL_ENABLE 1
3228	#define EV_FORK_ENABLE 0	4224	#define EV_CHILD_ENABLE 1
		4225	#define EV_USE_STDEXCEPT 0
3229	#define EV_CONFIG_H <config.h>	4226	#define EV_CONFIG_H <config.h>
3230	#define EV_MINPRI 0
3231	#define EV_MAXPRI 0
3232		4227
3233	#include "ev++.h"	4228	#include "ev++.h"
3234		4229
3235	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	4230	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
3236		4231
3237	#include "ev_cpp.h"	4232	#include "ev_cpp.h"
3238	#include "ev.c"	4233	#include "ev.c"
3239		4234
		4235	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
3240		4236
3241	=head1 THREADS AND COROUTINES	4237	=head2 THREADS AND COROUTINES
3242		4238
3243	=head2 THREADS	4239	=head3 THREADS
3244		4240
3245	Libev itself is thread-safe (unless the opposite is specifically	4241	All libev functions are reentrant and thread-safe unless explicitly
3246	documented for a function), but it uses no locking itself. This means that	4242	documented otherwise, but libev implements no locking itself. This means
3247	you can use as many loops as you want in parallel, as long as only one	4243	that you can use as many loops as you want in parallel, as long as there
3248	thread ever calls into one libev function with the same loop parameter:	4244	are no concurrent calls into any libev function with the same loop
		4245	parameter (C<ev_default_*> calls have an implicit default loop parameter,
3249	libev guarentees that different event loops share no data structures that	4246	of course): libev guarantees that different event loops share no data
3250	need locking.	4247	structures that need any locking.
3251		4248
3252	Or to put it differently: calls with different loop parameters can be done	4249	Or to put it differently: calls with different loop parameters can be done
3253	concurrently from multiple threads, calls with the same loop parameter	4250	concurrently from multiple threads, calls with the same loop parameter
3254	must be done serially (but can be done from different threads, as long as	4251	must be done serially (but can be done from different threads, as long as
3255	only one thread ever is inside a call at any point in time, e.g. by using	4252	only one thread ever is inside a call at any point in time, e.g. by using
3256	a mutex per loop).	4253	a mutex per loop).
3257		4254
3258	Specifically to support threads (and signal handlers), libev implements	4255	Specifically to support threads (and signal handlers), libev implements
3259	so-called C<ev_async> watchers, which allow some limited form of	4256	so-called C<ev_async> watchers, which allow some limited form of
3260	concurrency on the same event loop.	4257	concurrency on the same event loop, namely waking it up "from the
		4258	outside".
3261		4259
3262	If you want to know which design (one loop, locking, or multiple loops	4260	If you want to know which design (one loop, locking, or multiple loops
3263	without or something else still) is best for your problem, then I cannot	4261	without or something else still) is best for your problem, then I cannot
3264	help you. I can give some generic advice however:	4262	help you, but here is some generic advice:
3265		4263
3266	=over 4	4264	=over 4
3267		4265
3268	=item * most applications have a main thread: use the default libev loop	4266	=item * most applications have a main thread: use the default libev loop
3269	in that thread, or create a separate thread running only the default loop.	4267	in that thread, or create a separate thread running only the default loop.
…		…
3281		4279
3282	Choosing a model is hard - look around, learn, know that usually you can do	4280	Choosing a model is hard - look around, learn, know that usually you can do
3283	better than you currently do :-)	4281	better than you currently do :-)
3284		4282
3285	=item * often you need to talk to some other thread which blocks in the	4283	=item * often you need to talk to some other thread which blocks in the
		4284	event loop.
		4285
3286	event loop - C<ev_async> watchers can be used to wake them up from other	4286	C<ev_async> watchers can be used to wake them up from other threads safely
3287	threads safely (or from signal contexts...).	4287	(or from signal contexts...).
3288		4288
3289	=item * some watcher types are only supported in the default loop - use	4289	An example use would be to communicate signals or other events that only
3290	C<ev_async> watchers to tell your other loops about any such events.	4290	work in the default loop by registering the signal watcher with the
		4291	default loop and triggering an C<ev_async> watcher from the default loop
		4292	watcher callback into the event loop interested in the signal.
3291		4293
3292	=back	4294	=back
3293		4295
		4296	=head4 THREAD LOCKING EXAMPLE
		4297
		4298	Here is a fictitious example of how to run an event loop in a different
		4299	thread than where callbacks are being invoked and watchers are
		4300	created/added/removed.
		4301
		4302	For a real-world example, see the C<EV::Loop::Async> perl module,
		4303	which uses exactly this technique (which is suited for many high-level
		4304	languages).
		4305
		4306	The example uses a pthread mutex to protect the loop data, a condition
		4307	variable to wait for callback invocations, an async watcher to notify the
		4308	event loop thread and an unspecified mechanism to wake up the main thread.
		4309
		4310	First, you need to associate some data with the event loop:
		4311
		4312	typedef struct {
		4313	mutex_t lock; /* global loop lock */
		4314	ev_async async_w;
		4315	thread_t tid;
		4316	cond_t invoke_cv;
		4317	} userdata;
		4318
		4319	void prepare_loop (EV_P)
		4320	{
		4321	// for simplicity, we use a static userdata struct.
		4322	static userdata u;
		4323
		4324	ev_async_init (&u->async_w, async_cb);
		4325	ev_async_start (EV_A_ &u->async_w);
		4326
		4327	pthread_mutex_init (&u->lock, 0);
		4328	pthread_cond_init (&u->invoke_cv, 0);
		4329
		4330	// now associate this with the loop
		4331	ev_set_userdata (EV_A_ u);
		4332	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		4333	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		4334
		4335	// then create the thread running ev_loop
		4336	pthread_create (&u->tid, 0, l_run, EV_A);
		4337	}
		4338
		4339	The callback for the C<ev_async> watcher does nothing: the watcher is used
		4340	solely to wake up the event loop so it takes notice of any new watchers
		4341	that might have been added:
		4342
		4343	static void
		4344	async_cb (EV_P_ ev_async *w, int revents)
		4345	{
		4346	// just used for the side effects
		4347	}
		4348
		4349	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		4350	protecting the loop data, respectively.
		4351
		4352	static void
		4353	l_release (EV_P)
		4354	{
		4355	userdata *u = ev_userdata (EV_A);
		4356	pthread_mutex_unlock (&u->lock);
		4357	}
		4358
		4359	static void
		4360	l_acquire (EV_P)
		4361	{
		4362	userdata *u = ev_userdata (EV_A);
		4363	pthread_mutex_lock (&u->lock);
		4364	}
		4365
		4366	The event loop thread first acquires the mutex, and then jumps straight
		4367	into C<ev_run>:
		4368
		4369	void *
		4370	l_run (void *thr_arg)
		4371	{
		4372	struct ev_loop *loop = (struct ev_loop *)thr_arg;
		4373
		4374	l_acquire (EV_A);
		4375	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		4376	ev_run (EV_A_ 0);
		4377	l_release (EV_A);
		4378
		4379	return 0;
		4380	}
		4381
		4382	Instead of invoking all pending watchers, the C<l_invoke> callback will
		4383	signal the main thread via some unspecified mechanism (signals? pipe
		4384	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		4385	have been called (in a while loop because a) spurious wakeups are possible
		4386	and b) skipping inter-thread-communication when there are no pending
		4387	watchers is very beneficial):
		4388
		4389	static void
		4390	l_invoke (EV_P)
		4391	{
		4392	userdata *u = ev_userdata (EV_A);
		4393
		4394	while (ev_pending_count (EV_A))
		4395	{
		4396	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		4397	pthread_cond_wait (&u->invoke_cv, &u->lock);
		4398	}
		4399	}
		4400
		4401	Now, whenever the main thread gets told to invoke pending watchers, it
		4402	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		4403	thread to continue:
		4404
		4405	static void
		4406	real_invoke_pending (EV_P)
		4407	{
		4408	userdata *u = ev_userdata (EV_A);
		4409
		4410	pthread_mutex_lock (&u->lock);
		4411	ev_invoke_pending (EV_A);
		4412	pthread_cond_signal (&u->invoke_cv);
		4413	pthread_mutex_unlock (&u->lock);
		4414	}
		4415
		4416	Whenever you want to start/stop a watcher or do other modifications to an
		4417	event loop, you will now have to lock:
		4418
		4419	ev_timer timeout_watcher;
		4420	userdata *u = ev_userdata (EV_A);
		4421
		4422	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		4423
		4424	pthread_mutex_lock (&u->lock);
		4425	ev_timer_start (EV_A_ &timeout_watcher);
		4426	ev_async_send (EV_A_ &u->async_w);
		4427	pthread_mutex_unlock (&u->lock);
		4428
		4429	Note that sending the C<ev_async> watcher is required because otherwise
		4430	an event loop currently blocking in the kernel will have no knowledge
		4431	about the newly added timer. By waking up the loop it will pick up any new
		4432	watchers in the next event loop iteration.
		4433
3294	=head2 COROUTINES	4434	=head3 COROUTINES
3295		4435
3296	Libev is much more accommodating to coroutines ("cooperative threads"):	4436	Libev is very accommodating to coroutines ("cooperative threads"):
3297	libev fully supports nesting calls to it's functions from different	4437	libev fully supports nesting calls to its functions from different
3298	coroutines (e.g. you can call C<ev_loop> on the same loop from two	4438	coroutines (e.g. you can call C<ev_run> on the same loop from two
3299	different coroutines and switch freely between both coroutines running the	4439	different coroutines, and switch freely between both coroutines running
3300	loop, as long as you don't confuse yourself). The only exception is that	4440	the loop, as long as you don't confuse yourself). The only exception is
3301	you must not do this from C<ev_periodic> reschedule callbacks.	4441	that you must not do this from C<ev_periodic> reschedule callbacks.
3302		4442
3303	Care has been taken to ensure that libev does not keep local state inside	4443	Care has been taken to ensure that libev does not keep local state inside
3304	C<ev_loop>, and other calls do not usually allow coroutine switches.	4444	C<ev_run>, and other calls do not usually allow for coroutine switches as
		4445	they do not call any callbacks.
3305		4446
		4447	=head2 COMPILER WARNINGS
3306		4448
3307	=head1 COMPLEXITIES	4449	Depending on your compiler and compiler settings, you might get no or a
		4450	lot of warnings when compiling libev code. Some people are apparently
		4451	scared by this.
3308		4452
3309	In this section the complexities of (many of) the algorithms used inside	4453	However, these are unavoidable for many reasons. For one, each compiler
3310	libev will be explained. For complexity discussions about backends see the	4454	has different warnings, and each user has different tastes regarding
3311	documentation for C<ev_default_init>.	4455	warning options. "Warn-free" code therefore cannot be a goal except when
		4456	targeting a specific compiler and compiler-version.
3312		4457
3313	All of the following are about amortised time: If an array needs to be	4458	Another reason is that some compiler warnings require elaborate
3314	extended, libev needs to realloc and move the whole array, but this	4459	workarounds, or other changes to the code that make it less clear and less
3315	happens asymptotically never with higher number of elements, so O(1) might	4460	maintainable.
3316	mean it might do a lengthy realloc operation in rare cases, but on average
3317	it is much faster and asymptotically approaches constant time.
3318		4461
3319	=over 4	4462	And of course, some compiler warnings are just plain stupid, or simply
		4463	wrong (because they don't actually warn about the condition their message
		4464	seems to warn about). For example, certain older gcc versions had some
		4465	warnings that resulted in an extreme number of false positives. These have
		4466	been fixed, but some people still insist on making code warn-free with
		4467	such buggy versions.
3320		4468
3321	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	4469	While libev is written to generate as few warnings as possible,
		4470	"warn-free" code is not a goal, and it is recommended not to build libev
		4471	with any compiler warnings enabled unless you are prepared to cope with
		4472	them (e.g. by ignoring them). Remember that warnings are just that:
		4473	warnings, not errors, or proof of bugs.
3322		4474
3323	This means that, when you have a watcher that triggers in one hour and
3324	there are 100 watchers that would trigger before that then inserting will
3325	have to skip roughly seven (C<ld 100>) of these watchers.
3326		4475
3327	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	4476	=head2 VALGRIND
3328		4477
3329	That means that changing a timer costs less than removing/adding them	4478	Valgrind has a special section here because it is a popular tool that is
3330	as only the relative motion in the event queue has to be paid for.	4479	highly useful. Unfortunately, valgrind reports are very hard to interpret.
3331		4480
3332	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)	4481	If you think you found a bug (memory leak, uninitialised data access etc.)
		4482	in libev, then check twice: If valgrind reports something like:
3333		4483
3334	These just add the watcher into an array or at the head of a list.	4484	==2274== definitely lost: 0 bytes in 0 blocks.
		4485	==2274== possibly lost: 0 bytes in 0 blocks.
		4486	==2274== still reachable: 256 bytes in 1 blocks.
3335		4487
3336	=item Stopping check/prepare/idle/fork/async watchers: O(1)	4488	Then there is no memory leak, just as memory accounted to global variables
		4489	is not a memleak - the memory is still being referenced, and didn't leak.
3337		4490
3338	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	4491	Similarly, under some circumstances, valgrind might report kernel bugs
		4492	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		4493	although an acceptable workaround has been found here), or it might be
		4494	confused.
3339		4495
3340	These watchers are stored in lists then need to be walked to find the	4496	Keep in mind that valgrind is a very good tool, but only a tool. Don't
3341	correct watcher to remove. The lists are usually short (you don't usually	4497	make it into some kind of religion.
3342	have many watchers waiting for the same fd or signal).
3343		4498
3344	=item Finding the next timer in each loop iteration: O(1)	4499	If you are unsure about something, feel free to contact the mailing list
		4500	with the full valgrind report and an explanation on why you think this
		4501	is a bug in libev (best check the archives, too :). However, don't be
		4502	annoyed when you get a brisk "this is no bug" answer and take the chance
		4503	of learning how to interpret valgrind properly.
3345		4504
3346	By virtue of using a binary or 4-heap, the next timer is always found at a	4505	If you need, for some reason, empty reports from valgrind for your project
3347	fixed position in the storage array.	4506	I suggest using suppression lists.
3348		4507
3349	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
3350		4508
3351	A change means an I/O watcher gets started or stopped, which requires	4509	=head1 PORTABILITY NOTES
3352	libev to recalculate its status (and possibly tell the kernel, depending
3353	on backend and whether C<ev_io_set> was used).
3354		4510
3355	=item Activating one watcher (putting it into the pending state): O(1)	4511	=head2 GNU/LINUX 32 BIT LIMITATIONS
3356		4512
3357	=item Priority handling: O(number_of_priorities)	4513	GNU/Linux is the only common platform that supports 64 bit file/large file
		4514	interfaces but I<disables> them by default.
3358		4515
3359	Priorities are implemented by allocating some space for each	4516	That means that libev compiled in the default environment doesn't support
3360	priority. When doing priority-based operations, libev usually has to	4517	files larger than 2GiB or so, which mainly affects C<ev_stat> watchers.
3361	linearly search all the priorities, but starting/stopping and activating
3362	watchers becomes O(1) w.r.t. priority handling.
3363		4518
3364	=item Sending an ev_async: O(1)	4519	Unfortunately, many programs try to work around this GNU/Linux issue
		4520	by enabling the large file API, which makes them incompatible with the
		4521	standard libev compiled for their system.
3365		4522
3366	=item Processing ev_async_send: O(number_of_async_watchers)	4523	Likewise, libev cannot enable the large file API itself as this would
		4524	suddenly make it incompatible to the default compile time environment,
		4525	i.e. all programs not using special compile switches.
3367		4526
3368	=item Processing signals: O(max_signal_number)	4527	=head2 OS/X AND DARWIN BUGS
3369		4528
3370	Sending involves a system call I<iff> there were no other C<ev_async_send>	4529	The whole thing is a bug if you ask me - basically any system interface
3371	calls in the current loop iteration. Checking for async and signal events	4530	you touch is broken, whether it is locales, poll, kqueue or even the
3372	involves iterating over all running async watchers or all signal numbers.	4531	OpenGL drivers.
3373		4532
3374	=back	4533	=head3 C<kqueue> is buggy
3375		4534
		4535	The kqueue syscall is broken in all known versions - most versions support
		4536	only sockets, many support pipes.
3376		4537
		4538	Libev tries to work around this by not using C<kqueue> by default on this
		4539	rotten platform, but of course you can still ask for it when creating a
		4540	loop - embedding a socket-only kqueue loop into a select-based one is
		4541	probably going to work well.
		4542
		4543	=head3 C<poll> is buggy
		4544
		4545	Instead of fixing C<kqueue>, Apple replaced their (working) C<poll>
		4546	implementation by something calling C<kqueue> internally around the 10.5.6
		4547	release, so now C<kqueue> I<and> C<poll> are broken.
		4548
		4549	Libev tries to work around this by not using C<poll> by default on
		4550	this rotten platform, but of course you can still ask for it when creating
		4551	a loop.
		4552
		4553	=head3 C<select> is buggy
		4554
		4555	All that's left is C<select>, and of course Apple found a way to fuck this
		4556	one up as well: On OS/X, C<select> actively limits the number of file
		4557	descriptors you can pass in to 1024 - your program suddenly crashes when
		4558	you use more.
		4559
		4560	There is an undocumented "workaround" for this - defining
		4561	C<_DARWIN_UNLIMITED_SELECT>, which libev tries to use, so select I<should>
		4562	work on OS/X.
		4563
		4564	=head2 SOLARIS PROBLEMS AND WORKAROUNDS
		4565
		4566	=head3 C<errno> reentrancy
		4567
		4568	The default compile environment on Solaris is unfortunately so
		4569	thread-unsafe that you can't even use components/libraries compiled
		4570	without C<-D_REENTRANT> in a threaded program, which, of course, isn't
		4571	defined by default. A valid, if stupid, implementation choice.
		4572
		4573	If you want to use libev in threaded environments you have to make sure
		4574	it's compiled with C<_REENTRANT> defined.
		4575
		4576	=head3 Event port backend
		4577
		4578	The scalable event interface for Solaris is called "event
		4579	ports". Unfortunately, this mechanism is very buggy in all major
		4580	releases. If you run into high CPU usage, your program freezes or you get
		4581	a large number of spurious wakeups, make sure you have all the relevant
		4582	and latest kernel patches applied. No, I don't know which ones, but there
		4583	are multiple ones to apply, and afterwards, event ports actually work
		4584	great.
		4585
		4586	If you can't get it to work, you can try running the program by setting
		4587	the environment variable C<LIBEV_FLAGS=3> to only allow C<poll> and
		4588	C<select> backends.
		4589
		4590	=head2 AIX POLL BUG
		4591
		4592	AIX unfortunately has a broken C<poll.h> header. Libev works around
		4593	this by trying to avoid the poll backend altogether (i.e. it's not even
		4594	compiled in), which normally isn't a big problem as C<select> works fine
		4595	with large bitsets on AIX, and AIX is dead anyway.
		4596
3377	=head1 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS	4597	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		4598
		4599	=head3 General issues
3378		4600
3379	Win32 doesn't support any of the standards (e.g. POSIX) that libev	4601	Win32 doesn't support any of the standards (e.g. POSIX) that libev
3380	requires, and its I/O model is fundamentally incompatible with the POSIX	4602	requires, and its I/O model is fundamentally incompatible with the POSIX
3381	model. Libev still offers limited functionality on this platform in	4603	model. Libev still offers limited functionality on this platform in
3382	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	4604	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
3383	descriptors. This only applies when using Win32 natively, not when using	4605	descriptors. This only applies when using Win32 natively, not when using
3384	e.g. cygwin.	4606	e.g. cygwin. Actually, it only applies to the microsofts own compilers,
		4607	as every compielr comes with a slightly differently broken/incompatible
		4608	environment.
3385		4609
3386	Lifting these limitations would basically require the full	4610	Lifting these limitations would basically require the full
3387	re-implementation of the I/O system. If you are into these kinds of	4611	re-implementation of the I/O system. If you are into this kind of thing,
3388	things, then note that glib does exactly that for you in a very portable	4612	then note that glib does exactly that for you in a very portable way (note
3389	way (note also that glib is the slowest event library known to man).	4613	also that glib is the slowest event library known to man).
3390		4614
3391	There is no supported compilation method available on windows except	4615	There is no supported compilation method available on windows except
3392	embedding it into other applications.	4616	embedding it into other applications.
		4617
		4618	Sensible signal handling is officially unsupported by Microsoft - libev
		4619	tries its best, but under most conditions, signals will simply not work.
3393		4620
3394	Not a libev limitation but worth mentioning: windows apparently doesn't	4621	Not a libev limitation but worth mentioning: windows apparently doesn't
3395	accept large writes: instead of resulting in a partial write, windows will	4622	accept large writes: instead of resulting in a partial write, windows will
3396	either accept everything or return C<ENOBUFS> if the buffer is too large,	4623	either accept everything or return C<ENOBUFS> if the buffer is too large,
3397	so make sure you only write small amounts into your sockets (less than a	4624	so make sure you only write small amounts into your sockets (less than a
3398	megabyte seems safe, but thsi apparently depends on the amount of memory	4625	megabyte seems safe, but this apparently depends on the amount of memory
3399	available).	4626	available).
3400		4627
3401	Due to the many, low, and arbitrary limits on the win32 platform and	4628	Due to the many, low, and arbitrary limits on the win32 platform and
3402	the abysmal performance of winsockets, using a large number of sockets	4629	the abysmal performance of winsockets, using a large number of sockets
3403	is not recommended (and not reasonable). If your program needs to use	4630	is not recommended (and not reasonable). If your program needs to use
3404	more than a hundred or so sockets, then likely it needs to use a totally	4631	more than a hundred or so sockets, then likely it needs to use a totally
3405	different implementation for windows, as libev offers the POSIX readiness	4632	different implementation for windows, as libev offers the POSIX readiness
3406	notification model, which cannot be implemented efficiently on windows	4633	notification model, which cannot be implemented efficiently on windows
3407	(Microsoft monopoly games).	4634	(due to Microsoft monopoly games).
3408		4635
3409	A typical way to use libev under windows is to embed it (see the embedding	4636	A typical way to use libev under windows is to embed it (see the embedding
3410	section for details) and use the following F<evwrap.h> header file instead	4637	section for details) and use the following F<evwrap.h> header file instead
3411	of F<ev.h>:	4638	of F<ev.h>:
3412		4639
…		…
3414	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */	4641	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
3415		4642
3416	#include "ev.h"	4643	#include "ev.h"
3417		4644
3418	And compile the following F<evwrap.c> file into your project (make sure	4645	And compile the following F<evwrap.c> file into your project (make sure
3419	you do I<not> compile the F<ev.c> or any other embedded soruce files!):	4646	you do I<not> compile the F<ev.c> or any other embedded source files!):
3420		4647
3421	#include "evwrap.h"	4648	#include "evwrap.h"
3422	#include "ev.c"	4649	#include "ev.c"
3423		4650
3424	=over 4
3425
3426	=item The winsocket select function	4651	=head3 The winsocket C<select> function
3427		4652
3428	The winsocket C<select> function doesn't follow POSIX in that it	4653	The winsocket C<select> function doesn't follow POSIX in that it
3429	requires socket I<handles> and not socket I<file descriptors> (it is	4654	requires socket I<handles> and not socket I<file descriptors> (it is
3430	also extremely buggy). This makes select very inefficient, and also	4655	also extremely buggy). This makes select very inefficient, and also
3431	requires a mapping from file descriptors to socket handles (the Microsoft	4656	requires a mapping from file descriptors to socket handles (the Microsoft
…		…
3440	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */	4665	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
3441		4666
3442	Note that winsockets handling of fd sets is O(n), so you can easily get a	4667	Note that winsockets handling of fd sets is O(n), so you can easily get a
3443	complexity in the O(n²) range when using win32.	4668	complexity in the O(n²) range when using win32.
3444		4669
3445	=item Limited number of file descriptors	4670	=head3 Limited number of file descriptors
3446		4671
3447	Windows has numerous arbitrary (and low) limits on things.	4672	Windows has numerous arbitrary (and low) limits on things.
3448		4673
3449	Early versions of winsocket's select only supported waiting for a maximum	4674	Early versions of winsocket's select only supported waiting for a maximum
3450	of C<64> handles (probably owning to the fact that all windows kernels	4675	of C<64> handles (probably owning to the fact that all windows kernels
3451	can only wait for C<64> things at the same time internally; Microsoft	4676	can only wait for C<64> things at the same time internally; Microsoft
3452	recommends spawning a chain of threads and wait for 63 handles and the	4677	recommends spawning a chain of threads and wait for 63 handles and the
3453	previous thread in each. Great).	4678	previous thread in each. Sounds great!).
3454		4679
3455	Newer versions support more handles, but you need to define C<FD_SETSIZE>	4680	Newer versions support more handles, but you need to define C<FD_SETSIZE>
3456	to some high number (e.g. C<2048>) before compiling the winsocket select	4681	to some high number (e.g. C<2048>) before compiling the winsocket select
3457	call (which might be in libev or elsewhere, for example, perl does its own	4682	call (which might be in libev or elsewhere, for example, perl and many
3458	select emulation on windows).	4683	other interpreters do their own select emulation on windows).
3459		4684
3460	Another limit is the number of file descriptors in the Microsoft runtime	4685	Another limit is the number of file descriptors in the Microsoft runtime
3461	libraries, which by default is C<64> (there must be a hidden I<64> fetish	4686	libraries, which by default is C<64> (there must be a hidden I<64>
3462	or something like this inside Microsoft). You can increase this by calling	4687	fetish or something like this inside Microsoft). You can increase this
3463	C<_setmaxstdio>, which can increase this limit to C<2048> (another	4688	by calling C<_setmaxstdio>, which can increase this limit to C<2048>
3464	arbitrary limit), but is broken in many versions of the Microsoft runtime	4689	(another arbitrary limit), but is broken in many versions of the Microsoft
3465	libraries.
3466
3467	This might get you to about C<512> or C<2048> sockets (depending on	4690	runtime libraries. This might get you to about C<512> or C<2048> sockets
3468	windows version and/or the phase of the moon). To get more, you need to	4691	(depending on windows version and/or the phase of the moon). To get more,
3469	wrap all I/O functions and provide your own fd management, but the cost of	4692	you need to wrap all I/O functions and provide your own fd management, but
3470	calling select (O(n²)) will likely make this unworkable.	4693	the cost of calling select (O(n²)) will likely make this unworkable.
3471		4694
3472	=back
3473
3474
3475	=head1 PORTABILITY REQUIREMENTS	4695	=head2 PORTABILITY REQUIREMENTS
3476		4696
3477	In addition to a working ISO-C implementation, libev relies on a few	4697	In addition to a working ISO-C implementation and of course the
3478	additional extensions:	4698	backend-specific APIs, libev relies on a few additional extensions:
3479		4699
3480	=over 4	4700	=over 4
3481		4701
3482	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible	4702	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
3483	calling conventions regardless of C<ev_watcher_type *>.	4703	calling conventions regardless of C<ev_watcher_type *>.
…		…
3489	calls them using an C<ev_watcher *> internally.	4709	calls them using an C<ev_watcher *> internally.
3490		4710
3491	=item C<sig_atomic_t volatile> must be thread-atomic as well	4711	=item C<sig_atomic_t volatile> must be thread-atomic as well
3492		4712
3493	The type C<sig_atomic_t volatile> (or whatever is defined as	4713	The type C<sig_atomic_t volatile> (or whatever is defined as
3494	C<EV_ATOMIC_T>) must be atomic w.r.t. accesses from different	4714	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
3495	threads. This is not part of the specification for C<sig_atomic_t>, but is	4715	threads. This is not part of the specification for C<sig_atomic_t>, but is
3496	believed to be sufficiently portable.	4716	believed to be sufficiently portable.
3497		4717
3498	=item C<sigprocmask> must work in a threaded environment	4718	=item C<sigprocmask> must work in a threaded environment
3499		4719
…		…
3508	except the initial one, and run the default loop in the initial thread as	4728	except the initial one, and run the default loop in the initial thread as
3509	well.	4729	well.
3510		4730
3511	=item C<long> must be large enough for common memory allocation sizes	4731	=item C<long> must be large enough for common memory allocation sizes
3512		4732
3513	To improve portability and simplify using libev, libev uses C<long>	4733	To improve portability and simplify its API, libev uses C<long> internally
3514	internally instead of C<size_t> when allocating its data structures. On	4734	instead of C<size_t> when allocating its data structures. On non-POSIX
3515	non-POSIX systems (Microsoft...) this might be unexpectedly low, but	4735	systems (Microsoft...) this might be unexpectedly low, but is still at
3516	is still at least 31 bits everywhere, which is enough for hundreds of	4736	least 31 bits everywhere, which is enough for hundreds of millions of
3517	millions of watchers.	4737	watchers.
3518		4738
3519	=item C<double> must hold a time value in seconds with enough accuracy	4739	=item C<double> must hold a time value in seconds with enough accuracy
3520		4740
3521	The type C<double> is used to represent timestamps. It is required to	4741	The type C<double> is used to represent timestamps. It is required to
3522	have at least 51 bits of mantissa (and 9 bits of exponent), which is good	4742	have at least 51 bits of mantissa (and 9 bits of exponent), which is
3523	enough for at least into the year 4000. This requirement is fulfilled by	4743	good enough for at least into the year 4000 with millisecond accuracy
		4744	(the design goal for libev). This requirement is overfulfilled by
3524	implementations implementing IEEE 754 (basically all existing ones).	4745	implementations using IEEE 754, which is basically all existing ones. With
		4746	IEEE 754 doubles, you get microsecond accuracy until at least 2200.
3525		4747
3526	=back	4748	=back
3527		4749
3528	If you know of other additional requirements drop me a note.	4750	If you know of other additional requirements drop me a note.
3529		4751
3530		4752
3531	=head1 COMPILER WARNINGS	4753	=head1 ALGORITHMIC COMPLEXITIES
3532		4754
3533	Depending on your compiler and compiler settings, you might get no or a	4755	In this section the complexities of (many of) the algorithms used inside
3534	lot of warnings when compiling libev code. Some people are apparently	4756	libev will be documented. For complexity discussions about backends see
3535	scared by this.	4757	the documentation for C<ev_default_init>.
3536		4758
3537	However, these are unavoidable for many reasons. For one, each compiler	4759	All of the following are about amortised time: If an array needs to be
3538	has different warnings, and each user has different tastes regarding	4760	extended, libev needs to realloc and move the whole array, but this
3539	warning options. "Warn-free" code therefore cannot be a goal except when	4761	happens asymptotically rarer with higher number of elements, so O(1) might
3540	targeting a specific compiler and compiler-version.	4762	mean that libev does a lengthy realloc operation in rare cases, but on
		4763	average it is much faster and asymptotically approaches constant time.
3541		4764
3542	Another reason is that some compiler warnings require elaborate	4765	=over 4
3543	workarounds, or other changes to the code that make it less clear and less
3544	maintainable.
3545		4766
3546	And of course, some compiler warnings are just plain stupid, or simply	4767	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
3547	wrong (because they don't actually warn about the condition their message
3548	seems to warn about).
3549		4768
3550	While libev is written to generate as few warnings as possible,	4769	This means that, when you have a watcher that triggers in one hour and
3551	"warn-free" code is not a goal, and it is recommended not to build libev	4770	there are 100 watchers that would trigger before that, then inserting will
3552	with any compiler warnings enabled unless you are prepared to cope with	4771	have to skip roughly seven (C<ld 100>) of these watchers.
3553	them (e.g. by ignoring them). Remember that warnings are just that:
3554	warnings, not errors, or proof of bugs.
3555		4772
		4773	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
3556		4774
3557	=head1 VALGRIND	4775	That means that changing a timer costs less than removing/adding them,
		4776	as only the relative motion in the event queue has to be paid for.
3558		4777
3559	Valgrind has a special section here because it is a popular tool that is	4778	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
3560	highly useful, but valgrind reports are very hard to interpret.
3561		4779
3562	If you think you found a bug (memory leak, uninitialised data access etc.)	4780	These just add the watcher into an array or at the head of a list.
3563	in libev, then check twice: If valgrind reports something like:
3564		4781
3565	==2274== definitely lost: 0 bytes in 0 blocks.	4782	=item Stopping check/prepare/idle/fork/async watchers: O(1)
3566	==2274== possibly lost: 0 bytes in 0 blocks.
3567	==2274== still reachable: 256 bytes in 1 blocks.
3568		4783
3569	Then there is no memory leak. Similarly, under some circumstances,	4784	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
3570	valgrind might report kernel bugs as if it were a bug in libev, or it
3571	might be confused (it is a very good tool, but only a tool).
3572		4785
3573	If you are unsure about something, feel free to contact the mailing list	4786	These watchers are stored in lists, so they need to be walked to find the
3574	with the full valgrind report and an explanation on why you think this is	4787	correct watcher to remove. The lists are usually short (you don't usually
3575	a bug in libev. However, don't be annoyed when you get a brisk "this is	4788	have many watchers waiting for the same fd or signal: one is typical, two
3576	no bug" answer and take the chance of learning how to interpret valgrind	4789	is rare).
3577	properly.
3578		4790
3579	If you need, for some reason, empty reports from valgrind for your project	4791	=item Finding the next timer in each loop iteration: O(1)
3580	I suggest using suppression lists.
3581		4792
		4793	By virtue of using a binary or 4-heap, the next timer is always found at a
		4794	fixed position in the storage array.
		4795
		4796	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		4797
		4798	A change means an I/O watcher gets started or stopped, which requires
		4799	libev to recalculate its status (and possibly tell the kernel, depending
		4800	on backend and whether C<ev_io_set> was used).
		4801
		4802	=item Activating one watcher (putting it into the pending state): O(1)
		4803
		4804	=item Priority handling: O(number_of_priorities)
		4805
		4806	Priorities are implemented by allocating some space for each
		4807	priority. When doing priority-based operations, libev usually has to
		4808	linearly search all the priorities, but starting/stopping and activating
		4809	watchers becomes O(1) with respect to priority handling.
		4810
		4811	=item Sending an ev_async: O(1)
		4812
		4813	=item Processing ev_async_send: O(number_of_async_watchers)
		4814
		4815	=item Processing signals: O(max_signal_number)
		4816
		4817	Sending involves a system call I<iff> there were no other C<ev_async_send>
		4818	calls in the current loop iteration. Checking for async and signal events
		4819	involves iterating over all running async watchers or all signal numbers.
		4820
		4821	=back
		4822
		4823
		4824	=head1 PORTING FROM LIBEV 3.X TO 4.X
		4825
		4826	The major version 4 introduced some minor incompatible changes to the API.
		4827
		4828	At the moment, the C<ev.h> header file tries to implement superficial
		4829	compatibility, so most programs should still compile. Those might be
		4830	removed in later versions of libev, so better update early than late.
		4831
		4832	=over 4
		4833
		4834	=item function/symbol renames
		4835
		4836	A number of functions and symbols have been renamed:
		4837
		4838	ev_loop => ev_run
		4839	EVLOOP_NONBLOCK => EVRUN_NOWAIT
		4840	EVLOOP_ONESHOT => EVRUN_ONCE
		4841
		4842	ev_unloop => ev_break
		4843	EVUNLOOP_CANCEL => EVBREAK_CANCEL
		4844	EVUNLOOP_ONE => EVBREAK_ONE
		4845	EVUNLOOP_ALL => EVBREAK_ALL
		4846
		4847	EV_TIMEOUT => EV_TIMER
		4848
		4849	ev_loop_count => ev_iteration
		4850	ev_loop_depth => ev_depth
		4851	ev_loop_verify => ev_verify
		4852
		4853	Most functions working on C<struct ev_loop> objects don't have an
		4854	C<ev_loop_> prefix, so it was removed; C<ev_loop>, C<ev_unloop> and
		4855	associated constants have been renamed to not collide with the C<struct
		4856	ev_loop> anymore and C<EV_TIMER> now follows the same naming scheme
		4857	as all other watcher types. Note that C<ev_loop_fork> is still called
		4858	C<ev_loop_fork> because it would otherwise clash with the C<ev_fork>
		4859	typedef.
		4860
		4861	=item C<EV_COMPAT3> backwards compatibility mechanism
		4862
		4863	The backward compatibility mechanism can be controlled by
		4864	C<EV_COMPAT3>. See L<PREPROCESSOR SYMBOLS/MACROS> in the L<EMBEDDING>
		4865	section.
		4866
		4867	=item C<EV_MINIMAL> mechanism replaced by C<EV_FEATURES>
		4868
		4869	The preprocessor symbol C<EV_MINIMAL> has been replaced by a different
		4870	mechanism, C<EV_FEATURES>. Programs using C<EV_MINIMAL> usually compile
		4871	and work, but the library code will of course be larger.
		4872
		4873	=back
		4874
		4875
		4876	=head1 GLOSSARY
		4877
		4878	=over 4
		4879
		4880	=item active
		4881
		4882	A watcher is active as long as it has been started and not yet stopped.
		4883	See L<WATCHER STATES> for details.
		4884
		4885	=item application
		4886
		4887	In this document, an application is whatever is using libev.
		4888
		4889	=item backend
		4890
		4891	The part of the code dealing with the operating system interfaces.
		4892
		4893	=item callback
		4894
		4895	The address of a function that is called when some event has been
		4896	detected. Callbacks are being passed the event loop, the watcher that
		4897	received the event, and the actual event bitset.
		4898
		4899	=item callback/watcher invocation
		4900
		4901	The act of calling the callback associated with a watcher.
		4902
		4903	=item event
		4904
		4905	A change of state of some external event, such as data now being available
		4906	for reading on a file descriptor, time having passed or simply not having
		4907	any other events happening anymore.
		4908
		4909	In libev, events are represented as single bits (such as C<EV_READ> or
		4910	C<EV_TIMER>).
		4911
		4912	=item event library
		4913
		4914	A software package implementing an event model and loop.
		4915
		4916	=item event loop
		4917
		4918	An entity that handles and processes external events and converts them
		4919	into callback invocations.
		4920
		4921	=item event model
		4922
		4923	The model used to describe how an event loop handles and processes
		4924	watchers and events.
		4925
		4926	=item pending
		4927
		4928	A watcher is pending as soon as the corresponding event has been
		4929	detected. See L<WATCHER STATES> for details.
		4930
		4931	=item real time
		4932
		4933	The physical time that is observed. It is apparently strictly monotonic :)
		4934
		4935	=item wall-clock time
		4936
		4937	The time and date as shown on clocks. Unlike real time, it can actually
		4938	be wrong and jump forwards and backwards, e.g. when the you adjust your
		4939	clock.
		4940
		4941	=item watcher
		4942
		4943	A data structure that describes interest in certain events. Watchers need
		4944	to be started (attached to an event loop) before they can receive events.
		4945
		4946	=back
3582		4947
3583	=head1 AUTHOR	4948	=head1 AUTHOR
3584		4949
3585	Marc Lehmann <libev@schmorp.de>.	4950	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.
3586		4951

Diff Legend

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