ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.136 by root, Thu Mar 13 13:06:16 2008 UTC vs.
Revision 1.349 by root, Mon Jan 10 01:58:55 2011 UTC

…		…
2		2
3	libev - a high performance full-featured event loop written in C	3	libev - a high performance full-featured event loop written in C
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
8		8
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	{
45	// use the default event loop unless you have special needs	47	// use the default event loop unless you have special needs
46	struct ev_loop *loop = ev_default_loop (0);	48	struct ev_loop *loop = EV_DEFAULT;
47		49
48	// initialise an io watcher, then start it	50	// initialise an io watcher, then start it
49	// this one will watch for stdin to become readable	51	// this one will watch for stdin to become readable
50	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);	52	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
51	ev_io_start (loop, &stdin_watcher);	53	ev_io_start (loop, &stdin_watcher);
52		54
53	// initialise a timer watcher, then start it	55	// initialise a timer watcher, then start it
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://cvs.schmorp.de/libev/ev.html>.	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 WHAT TO READ WHEN IN A HURRY
		84
		85	This manual tries to be very detailed, but unfortunately, this also makes
		86	it very long. If you just want to know the basics of libev, I suggest
		87	reading L<ANATOMY OF A WATCHER>, then the L<EXAMPLE PROGRAM> above and
		88	look up the missing functions in L<GLOBAL FUNCTIONS> and the C<ev_io> and
		89	C<ev_timer> sections in L<WATCHER TYPES>.
		90
		91	=head1 ABOUT LIBEV
70		92
71	Libev is an event loop: you register interest in certain events (such as a	93	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	94	file descriptor being readable or a timeout occurring), and it will manage
73	these event sources and provide your program with events.	95	these event sources and provide your program with events.
74		96
…		…
84	=head2 FEATURES	106	=head2 FEATURES
85		107
86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the	108	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms	109	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface	110	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers	111	(for C<ev_stat>), Linux eventfd/signalfd (for faster and cleaner
90	with customised rescheduling (C<ev_periodic>), synchronous signals	112	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	113	timers (C<ev_timer>), absolute timers with customised rescheduling
92	watchers dealing with the event loop mechanism itself (C<ev_idle>,	114	(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	115	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	116	loop mechanism itself (C<ev_idle>, C<ev_embed>, C<ev_prepare> and
95	(C<ev_fork>).	117	C<ev_check> watchers) as well as file watchers (C<ev_stat>) and even
		118	limited support for fork events (C<ev_fork>).
96		119
97	It also is quite fast (see this	120	It also is quite fast (see this
98	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent	121	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
99	for example).	122	for example).
100		123
…		…
108	name C<loop> (which is always of type C<struct ev_loop *>) will not have	131	name C<loop> (which is always of type C<struct ev_loop *>) will not have
109	this argument.	132	this argument.
110		133
111	=head2 TIME REPRESENTATION	134	=head2 TIME REPRESENTATION
112		135
113	Libev represents time as a single floating point number, representing the	136	Libev represents time as a single floating point number, representing
114	(fractional) number of seconds since the (POSIX) epoch (somewhere near	137	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	138	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	139	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	140	too. It usually aliases to the C<double> type in C. When you need to do
118	it, you should treat it as some floatingpoint value. Unlike the name	141	any calculations on it, you should treat it as some floating point value.
		142
119	component C<stamp> might indicate, it is also used for time differences	143	Unlike the name component C<stamp> might indicate, it is also used for
120	throughout libev.	144	time differences (e.g. delays) throughout libev.
		145
		146	=head1 ERROR HANDLING
		147
		148	Libev knows three classes of errors: operating system errors, usage errors
		149	and internal errors (bugs).
		150
		151	When libev catches an operating system error it cannot handle (for example
		152	a system call indicating a condition libev cannot fix), it calls the callback
		153	set via C<ev_set_syserr_cb>, which is supposed to fix the problem or
		154	abort. The default is to print a diagnostic message and to call C<abort
		155	()>.
		156
		157	When libev detects a usage error such as a negative timer interval, then
		158	it will print a diagnostic message and abort (via the C<assert> mechanism,
		159	so C<NDEBUG> will disable this checking): these are programming errors in
		160	the libev caller and need to be fixed there.
		161
		162	Libev also has a few internal error-checking C<assert>ions, and also has
		163	extensive consistency checking code. These do not trigger under normal
		164	circumstances, as they indicate either a bug in libev or worse.
		165
121		166
122	=head1 GLOBAL FUNCTIONS	167	=head1 GLOBAL FUNCTIONS
123		168
124	These functions can be called anytime, even before initialising the	169	These functions can be called anytime, even before initialising the
125	library in any way.	170	library in any way.
…		…
128		173
129	=item ev_tstamp ev_time ()	174	=item ev_tstamp ev_time ()
130		175
131	Returns the current time as libev would use it. Please note that the	176	Returns the current time as libev would use it. Please note that the
132	C<ev_now> function is usually faster and also often returns the timestamp	177	C<ev_now> function is usually faster and also often returns the timestamp
133	you actually want to know.	178	you actually want to know. Also interesting is the combination of
		179	C<ev_update_now> and C<ev_now>.
134		180
135	=item ev_sleep (ev_tstamp interval)	181	=item ev_sleep (ev_tstamp interval)
136		182
137	Sleep for the given interval: The current thread will be blocked until	183	Sleep for the given interval: The current thread will be blocked until
138	either it is interrupted or the given time interval has passed. Basically	184	either it is interrupted or the given time interval has passed. Basically
139	this is a subsecond-resolution C<sleep ()>.	185	this is a sub-second-resolution C<sleep ()>.
140		186
141	=item int ev_version_major ()	187	=item int ev_version_major ()
142		188
143	=item int ev_version_minor ()	189	=item int ev_version_minor ()
144		190
…		…
155	as this indicates an incompatible change. Minor versions are usually	201	as this indicates an incompatible change. Minor versions are usually
156	compatible to older versions, so a larger minor version alone is usually	202	compatible to older versions, so a larger minor version alone is usually
157	not a problem.	203	not a problem.
158		204
159	Example: Make sure we haven't accidentally been linked against the wrong	205	Example: Make sure we haven't accidentally been linked against the wrong
160	version.	206	version (note, however, that this will not detect other ABI mismatches,
		207	such as LFS or reentrancy).
161		208
162	assert (("libev version mismatch",	209	assert (("libev version mismatch",
163	ev_version_major () == EV_VERSION_MAJOR	210	ev_version_major () == EV_VERSION_MAJOR
164	&& ev_version_minor () >= EV_VERSION_MINOR));	211	&& ev_version_minor () >= EV_VERSION_MINOR));
165		212
166	=item unsigned int ev_supported_backends ()	213	=item unsigned int ev_supported_backends ()
167		214
168	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>	215	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
169	value) compiled into this binary of libev (independent of their	216	value) compiled into this binary of libev (independent of their
…		…
171	a description of the set values.	218	a description of the set values.
172		219
173	Example: make sure we have the epoll method, because yeah this is cool and	220	Example: make sure we have the epoll method, because yeah this is cool and
174	a must have and can we have a torrent of it please!!!11	221	a must have and can we have a torrent of it please!!!11
175		222
176	assert (("sorry, no epoll, no sex",	223	assert (("sorry, no epoll, no sex",
177	ev_supported_backends () & EVBACKEND_EPOLL));	224	ev_supported_backends () & EVBACKEND_EPOLL));
178		225
179	=item unsigned int ev_recommended_backends ()	226	=item unsigned int ev_recommended_backends ()
180		227
181	Return the set of all backends compiled into this binary of libev and also	228	Return the set of all backends compiled into this binary of libev and
182	recommended for this platform. This set is often smaller than the one	229	also recommended for this platform, meaning it will work for most file
		230	descriptor types. This set is often smaller than the one returned by
183	returned by C<ev_supported_backends>, as for example kqueue is broken on	231	C<ev_supported_backends>, as for example kqueue is broken on most BSDs
184	most BSDs and will not be autodetected unless you explicitly request it	232	and will not be auto-detected unless you explicitly request it (assuming
185	(assuming you know what you are doing). This is the set of backends that	233	you know what you are doing). This is the set of backends that libev will
186	libev will probe for if you specify no backends explicitly.	234	probe for if you specify no backends explicitly.
187		235
188	=item unsigned int ev_embeddable_backends ()	236	=item unsigned int ev_embeddable_backends ()
189		237
190	Returns the set of backends that are embeddable in other event loops. This	238	Returns the set of backends that are embeddable in other event loops. This
191	is the theoretical, all-platform, value. To find which backends	239	value is platform-specific but can include backends not available on the
192	might be supported on the current system, you would need to look at	240	current system. To find which embeddable backends might be supported on
193	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	241	the current system, you would need to look at C<ev_embeddable_backends ()
194	recommended ones.	242	& ev_supported_backends ()>, likewise for recommended ones.
195		243
196	See the description of C<ev_embed> watchers for more info.	244	See the description of C<ev_embed> watchers for more info.
197		245
198	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	246	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
199		247
200	Sets the allocation function to use (the prototype is similar - the	248	Sets the allocation function to use (the prototype is similar - the
201	semantics is identical - to the realloc C function). It is used to	249	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
202	allocate and free memory (no surprises here). If it returns zero when	250	used to allocate and free memory (no surprises here). If it returns zero
203	memory needs to be allocated, the library might abort or take some	251	when memory needs to be allocated (C<size != 0>), the library might abort
204	potentially destructive action. The default is your system realloc	252	or take some potentially destructive action.
205	function.	253
		254	Since some systems (at least OpenBSD and Darwin) fail to implement
		255	correct C<realloc> semantics, libev will use a wrapper around the system
		256	C<realloc> and C<free> functions by default.
206		257
207	You could override this function in high-availability programs to, say,	258	You could override this function in high-availability programs to, say,
208	free some memory if it cannot allocate memory, to use a special allocator,	259	free some memory if it cannot allocate memory, to use a special allocator,
209	or even to sleep a while and retry until some memory is available.	260	or even to sleep a while and retry until some memory is available.
210		261
211	Example: Replace the libev allocator with one that waits a bit and then	262	Example: Replace the libev allocator with one that waits a bit and then
212	retries).	263	retries (example requires a standards-compliant C<realloc>).
213		264
214	static void *	265	static void *
215	persistent_realloc (void *ptr, size_t size)	266	persistent_realloc (void *ptr, size_t size)
216	{	267	{
217	for (;;)	268	for (;;)
…		…
226	}	277	}
227		278
228	...	279	...
229	ev_set_allocator (persistent_realloc);	280	ev_set_allocator (persistent_realloc);
230		281
231	=item ev_set_syserr_cb (void (*cb)(const char *msg));	282	=item ev_set_syserr_cb (void (*cb)(const char *msg))
232		283
233	Set the callback function to call on a retryable syscall error (such	284	Set the callback function to call on a retryable system call error (such
234	as failed select, poll, epoll_wait). The message is a printable string	285	as failed select, poll, epoll_wait). The message is a printable string
235	indicating the system call or subsystem causing the problem. If this	286	indicating the system call or subsystem causing the problem. If this
236	callback is set, then libev will expect it to remedy the sitution, no	287	callback is set, then libev will expect it to remedy the situation, no
237	matter what, when it returns. That is, libev will generally retry the	288	matter what, when it returns. That is, libev will generally retry the
238	requested operation, or, if the condition doesn't go away, do bad stuff	289	requested operation, or, if the condition doesn't go away, do bad stuff
239	(such as abort).	290	(such as abort).
240		291
241	Example: This is basically the same thing that libev does internally, too.	292	Example: This is basically the same thing that libev does internally, too.
…		…
248	}	299	}
249		300
250	...	301	...
251	ev_set_syserr_cb (fatal_error);	302	ev_set_syserr_cb (fatal_error);
252		303
		304	=item ev_feed_signal (int signum)
		305
		306	This function can be used to "simulate" a signal receive. It is completely
		307	safe to call this function at any time, from any context, including signal
		308	handlers or random threads.
		309
		310	It's main use is to customise signal handling in your process, especially
		311	in the presence of threads. For example, you could block signals
		312	by default in all threads (and specifying C<EVFLAG_NOSIGMASK> when
		313	creating any loops), and in one thread, use C<sigwait> or any other
		314	mechanism to wait for signals, then "deliver" them to libev by calling
		315	C<ev_feed_signal>.
		316
253	=back	317	=back
254		318
255	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	319	=head1 FUNCTIONS CONTROLLING EVENT LOOPS
256		320
257	An event loop is described by a C<struct ev_loop *>. The library knows two	321	An event loop is described by a C<struct ev_loop *> (the C<struct> is
258	types of such loops, the I<default> loop, which supports signals and child	322	I<not> optional in this case unless libev 3 compatibility is disabled, as
259	events, and dynamically created loops which do not.	323	libev 3 had an C<ev_loop> function colliding with the struct name).
260		324
261	If you use threads, a common model is to run the default event loop	325	The library knows two types of such loops, the I<default> loop, which
262	in your main thread (or in a separate thread) and for each thread you	326	supports child process events, and dynamically created event loops which
263	create, you also create another event loop. Libev itself does no locking	327	do not.
264	whatsoever, so if you mix calls to the same event loop in different
265	threads, make sure you lock (this is usually a bad idea, though, even if
266	done correctly, because it's hideous and inefficient).
267		328
268	=over 4	329	=over 4
269		330
270	=item struct ev_loop *ev_default_loop (unsigned int flags)	331	=item struct ev_loop *ev_default_loop (unsigned int flags)
271		332
272	This will initialise the default event loop if it hasn't been initialised	333	This returns the "default" event loop object, which is what you should
273	yet and return it. If the default loop could not be initialised, returns	334	normally use when you just need "the event loop". Event loop objects and
274	false. If it already was initialised it simply returns it (and ignores the	335	the C<flags> parameter are described in more detail in the entry for
275	flags. If that is troubling you, check C<ev_backend ()> afterwards).	336	C<ev_loop_new>.
		337
		338	If the default loop is already initialised then this function simply
		339	returns it (and ignores the flags. If that is troubling you, check
		340	C<ev_backend ()> afterwards). Otherwise it will create it with the given
		341	flags, which should almost always be C<0>, unless the caller is also the
		342	one calling C<ev_run> or otherwise qualifies as "the main program".
276		343
277	If you don't know what event loop to use, use the one returned from this	344	If you don't know what event loop to use, use the one returned from this
278	function.	345	function (or via the C<EV_DEFAULT> macro).
279		346
		347	Note that this function is I<not> thread-safe, so if you want to use it
		348	from multiple threads, you have to employ some kind of mutex (note also
		349	that this case is unlikely, as loops cannot be shared easily between
		350	threads anyway).
		351
280	The default loop is the only loop that can handle C<ev_signal> and	352	The default loop is the only loop that can handle C<ev_child> watchers,
281	C<ev_child> watchers, and to do this, it always registers a handler	353	and to do this, it always registers a handler for C<SIGCHLD>. If this is
282	for C<SIGCHLD>. If this is a problem for your app you can either	354	a problem for your application you can either create a dynamic loop with
283	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you	355	C<ev_loop_new> which doesn't do that, or you can simply overwrite the
284	can simply overwrite the C<SIGCHLD> signal handler I<after> calling	356	C<SIGCHLD> signal handler I<after> calling C<ev_default_init>.
285	C<ev_default_init>.	357
		358	Example: This is the most typical usage.
		359
		360	if (!ev_default_loop (0))
		361	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
		362
		363	Example: Restrict libev to the select and poll backends, and do not allow
		364	environment settings to be taken into account:
		365
		366	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
		367
		368	=item struct ev_loop *ev_loop_new (unsigned int flags)
		369
		370	This will create and initialise a new event loop object. If the loop
		371	could not be initialised, returns false.
		372
		373	This function is thread-safe, and one common way to use libev with
		374	threads is indeed to create one loop per thread, and using the default
		375	loop in the "main" or "initial" thread.
286		376
287	The flags argument can be used to specify special behaviour or specific	377	The flags argument can be used to specify special behaviour or specific
288	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).	378	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
289		379
290	The following flags are supported:	380	The following flags are supported:
…		…
296	The default flags value. Use this if you have no clue (it's the right	386	The default flags value. Use this if you have no clue (it's the right
297	thing, believe me).	387	thing, believe me).
298		388
299	=item C<EVFLAG_NOENV>	389	=item C<EVFLAG_NOENV>
300		390
301	If this flag bit is ored into the flag value (or the program runs setuid	391	If this flag bit is or'ed into the flag value (or the program runs setuid
302	or setgid) then libev will I<not> look at the environment variable	392	or setgid) then libev will I<not> look at the environment variable
303	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	393	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
304	override the flags completely if it is found in the environment. This is	394	override the flags completely if it is found in the environment. This is
305	useful to try out specific backends to test their performance, or to work	395	useful to try out specific backends to test their performance, or to work
306	around bugs.	396	around bugs.
307		397
308	=item C<EVFLAG_FORKCHECK>	398	=item C<EVFLAG_FORKCHECK>
309		399
310	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after	400	Instead of calling C<ev_loop_fork> manually after a fork, you can also
311	a fork, you can also make libev check for a fork in each iteration by	401	make libev check for a fork in each iteration by enabling this flag.
312	enabling this flag.
313		402
314	This works by calling C<getpid ()> on every iteration of the loop,	403	This works by calling C<getpid ()> on every iteration of the loop,
315	and thus this might slow down your event loop if you do a lot of loop	404	and thus this might slow down your event loop if you do a lot of loop
316	iterations and little real work, but is usually not noticeable (on my	405	iterations and little real work, but is usually not noticeable (on my
317	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence	406	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
318	without a syscall and thus I<very> fast, but my GNU/Linux system also has	407	without a system call and thus I<very> fast, but my GNU/Linux system also has
319	C<pthread_atfork> which is even faster).	408	C<pthread_atfork> which is even faster).
320		409
321	The big advantage of this flag is that you can forget about fork (and	410	The big advantage of this flag is that you can forget about fork (and
322	forget about forgetting to tell libev about forking) when you use this	411	forget about forgetting to tell libev about forking) when you use this
323	flag.	412	flag.
324		413
325	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>	414	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
326	environment variable.	415	environment variable.
		416
		417	=item C<EVFLAG_NOINOTIFY>
		418
		419	When this flag is specified, then libev will not attempt to use the
		420	I<inotify> API for its C<ev_stat> watchers. Apart from debugging and
		421	testing, this flag can be useful to conserve inotify file descriptors, as
		422	otherwise each loop using C<ev_stat> watchers consumes one inotify handle.
		423
		424	=item C<EVFLAG_SIGNALFD>
		425
		426	When this flag is specified, then libev will attempt to use the
		427	I<signalfd> API for its C<ev_signal> (and C<ev_child>) watchers. This API
		428	delivers signals synchronously, which makes it both faster and might make
		429	it possible to get the queued signal data. It can also simplify signal
		430	handling with threads, as long as you properly block signals in your
		431	threads that are not interested in handling them.
		432
		433	Signalfd will not be used by default as this changes your signal mask, and
		434	there are a lot of shoddy libraries and programs (glib's threadpool for
		435	example) that can't properly initialise their signal masks.
		436
		437	=item C<EVFLAG_NOSIGMASK>
		438
		439	When this flag is specified, then libev will avoid to modify the signal
		440	mask. Specifically, this means you ahve to make sure signals are unblocked
		441	when you want to receive them.
		442
		443	This behaviour is useful when you want to do your own signal handling, or
		444	want to handle signals only in specific threads and want to avoid libev
		445	unblocking the signals.
		446
		447	This flag's behaviour will become the default in future versions of libev.
327		448
328	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	449	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
329		450
330	This is your standard select(2) backend. Not I<completely> standard, as	451	This is your standard select(2) backend. Not I<completely> standard, as
331	libev tries to roll its own fd_set with no limits on the number of fds,	452	libev tries to roll its own fd_set with no limits on the number of fds,
332	but if that fails, expect a fairly low limit on the number of fds when	453	but if that fails, expect a fairly low limit on the number of fds when
333	using this backend. It doesn't scale too well (O(highest_fd)), but its	454	using this backend. It doesn't scale too well (O(highest_fd)), but its
334	usually the fastest backend for a low number of (low-numbered :) fds.	455	usually the fastest backend for a low number of (low-numbered :) fds.
335		456
336	To get good performance out of this backend you need a high amount of	457	To get good performance out of this backend you need a high amount of
337	parallelity (most of the file descriptors should be busy). If you are	458	parallelism (most of the file descriptors should be busy). If you are
338	writing a server, you should C<accept ()> in a loop to accept as many	459	writing a server, you should C<accept ()> in a loop to accept as many
339	connections as possible during one iteration. You might also want to have	460	connections as possible during one iteration. You might also want to have
340	a look at C<ev_set_io_collect_interval ()> to increase the amount of	461	a look at C<ev_set_io_collect_interval ()> to increase the amount of
341	readyness notifications you get per iteration.	462	readiness notifications you get per iteration.
		463
		464	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		465	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		466	C<exceptfds> set on that platform).
342		467
343	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	468	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
344		469
345	And this is your standard poll(2) backend. It's more complicated	470	And this is your standard poll(2) backend. It's more complicated
346	than select, but handles sparse fds better and has no artificial	471	than select, but handles sparse fds better and has no artificial
347	limit on the number of fds you can use (except it will slow down	472	limit on the number of fds you can use (except it will slow down
348	considerably with a lot of inactive fds). It scales similarly to select,	473	considerably with a lot of inactive fds). It scales similarly to select,
349	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for	474	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
350	performance tips.	475	performance tips.
351		476
		477	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		478	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
		479
352	=item C<EVBACKEND_EPOLL> (value 4, Linux)	480	=item C<EVBACKEND_EPOLL> (value 4, Linux)
		481
		482	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
		483	kernels).
353		484
354	For few fds, this backend is a bit little slower than poll and select,	485	For few fds, this backend is a bit little slower than poll and select,
355	but it scales phenomenally better. While poll and select usually scale	486	but it scales phenomenally better. While poll and select usually scale
356	like O(total_fds) where n is the total number of fds (or the highest fd),	487	like O(total_fds) where n is the total number of fds (or the highest fd),
357	epoll scales either O(1) or O(active_fds). The epoll design has a number	488	epoll scales either O(1) or O(active_fds).
358	of shortcomings, such as silently dropping events in some hard-to-detect	489
359	cases and rewiring a syscall per fd change, no fork support and bad	490	The epoll mechanism deserves honorable mention as the most misdesigned
360	support for dup.	491	of the more advanced event mechanisms: mere annoyances include silently
		492	dropping file descriptors, requiring a system call per change per file
		493	descriptor (and unnecessary guessing of parameters), problems with dup,
		494	returning before the timeout value, resulting in additional iterations
		495	(and only giving 5ms accuracy while select on the same platform gives
		496	0.1ms) and so on. The biggest issue is fork races, however - if a program
		497	forks then I<both> parent and child process have to recreate the epoll
		498	set, which can take considerable time (one syscall per file descriptor)
		499	and is of course hard to detect.
		500
		501	Epoll is also notoriously buggy - embedding epoll fds I<should> work, but
		502	of course I<doesn't>, and epoll just loves to report events for totally
		503	I<different> file descriptors (even already closed ones, so one cannot
		504	even remove them from the set) than registered in the set (especially
		505	on SMP systems). Libev tries to counter these spurious notifications by
		506	employing an additional generation counter and comparing that against the
		507	events to filter out spurious ones, recreating the set when required. Last
		508	not least, it also refuses to work with some file descriptors which work
		509	perfectly fine with C<select> (files, many character devices...).
		510
		511	Epoll is truly the train wreck analog among event poll mechanisms.
361		512
362	While stopping, setting and starting an I/O watcher in the same iteration	513	While stopping, setting and starting an I/O watcher in the same iteration
363	will result in some caching, there is still a syscall per such incident	514	will result in some caching, there is still a system call per such
364	(because the fd could point to a different file description now), so its	515	incident (because the same I<file descriptor> could point to a different
365	best to avoid that. Also, C<dup ()>'ed file descriptors might not work	516	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
366	very well if you register events for both fds.	517	file descriptors might not work very well if you register events for both
367		518	file descriptors.
368	Please note that epoll sometimes generates spurious notifications, so you
369	need to use non-blocking I/O or other means to avoid blocking when no data
370	(or space) is available.
371		519
372	Best performance from this backend is achieved by not unregistering all	520	Best performance from this backend is achieved by not unregistering all
373	watchers for a file descriptor until it has been closed, if possible, i.e.	521	watchers for a file descriptor until it has been closed, if possible,
374	keep at least one watcher active per fd at all times.	522	i.e. keep at least one watcher active per fd at all times. Stopping and
		523	starting a watcher (without re-setting it) also usually doesn't cause
		524	extra overhead. A fork can both result in spurious notifications as well
		525	as in libev having to destroy and recreate the epoll object, which can
		526	take considerable time and thus should be avoided.
375		527
		528	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		529	faster than epoll for maybe up to a hundred file descriptors, depending on
		530	the usage. So sad.
		531
376	While nominally embeddeble in other event loops, this feature is broken in	532	While nominally embeddable in other event loops, this feature is broken in
377	all kernel versions tested so far.	533	all kernel versions tested so far.
		534
		535	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		536	C<EVBACKEND_POLL>.
378		537
379	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	538	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
380		539
381	Kqueue deserves special mention, as at the time of this writing, it	540	Kqueue deserves special mention, as at the time of this writing, it
382	was broken on all BSDs except NetBSD (usually it doesn't work reliably	541	was broken on all BSDs except NetBSD (usually it doesn't work reliably
383	with anything but sockets and pipes, except on Darwin, where of course	542	with anything but sockets and pipes, except on Darwin, where of course
384	it's completely useless). For this reason it's not being "autodetected"	543	it's completely useless). Unlike epoll, however, whose brokenness
		544	is by design, these kqueue bugs can (and eventually will) be fixed
		545	without API changes to existing programs. For this reason it's not being
385	unless you explicitly specify it explicitly in the flags (i.e. using	546	"auto-detected" unless you explicitly specify it in the flags (i.e. using
386	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)	547	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
387	system like NetBSD.	548	system like NetBSD.
388		549
389	You still can embed kqueue into a normal poll or select backend and use it	550	You still can embed kqueue into a normal poll or select backend and use it
390	only for sockets (after having made sure that sockets work with kqueue on	551	only for sockets (after having made sure that sockets work with kqueue on
391	the target platform). See C<ev_embed> watchers for more info.	552	the target platform). See C<ev_embed> watchers for more info.
392		553
393	It scales in the same way as the epoll backend, but the interface to the	554	It scales in the same way as the epoll backend, but the interface to the
394	kernel is more efficient (which says nothing about its actual speed, of	555	kernel is more efficient (which says nothing about its actual speed, of
395	course). While stopping, setting and starting an I/O watcher does never	556	course). While stopping, setting and starting an I/O watcher does never
396	cause an extra syscall as with C<EVBACKEND_EPOLL>, it still adds up to	557	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
397	two event changes per incident, support for C<fork ()> is very bad and it	558	two event changes per incident. Support for C<fork ()> is very bad (but
398	drops fds silently in similarly hard-to-detect cases.	559	sane, unlike epoll) and it drops fds silently in similarly hard-to-detect
		560	cases
399		561
400	This backend usually performs well under most conditions.	562	This backend usually performs well under most conditions.
401		563
402	While nominally embeddable in other event loops, this doesn't work	564	While nominally embeddable in other event loops, this doesn't work
403	everywhere, so you might need to test for this. And since it is broken	565	everywhere, so you might need to test for this. And since it is broken
404	almost everywhere, you should only use it when you have a lot of sockets	566	almost everywhere, you should only use it when you have a lot of sockets
405	(for which it usually works), by embedding it into another event loop	567	(for which it usually works), by embedding it into another event loop
406	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for	568	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
407	sockets.	569	also broken on OS X)) and, did I mention it, using it only for sockets.
		570
		571	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		572	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		573	C<NOTE_EOF>.
408		574
409	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	575	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
410		576
411	This is not implemented yet (and might never be, unless you send me an	577	This is not implemented yet (and might never be, unless you send me an
412	implementation). According to reports, C</dev/poll> only supports sockets	578	implementation). According to reports, C</dev/poll> only supports sockets
…		…
416	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	582	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
417		583
418	This uses the Solaris 10 event port mechanism. As with everything on Solaris,	584	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
419	it's really slow, but it still scales very well (O(active_fds)).	585	it's really slow, but it still scales very well (O(active_fds)).
420		586
421	Please note that solaris event ports can deliver a lot of spurious	587	Please note that Solaris event ports can deliver a lot of spurious
422	notifications, so you need to use non-blocking I/O or other means to avoid	588	notifications, so you need to use non-blocking I/O or other means to avoid
423	blocking when no data (or space) is available.	589	blocking when no data (or space) is available.
424		590
425	While this backend scales well, it requires one system call per active	591	While this backend scales well, it requires one system call per active
426	file descriptor per loop iteration. For small and medium numbers of file	592	file descriptor per loop iteration. For small and medium numbers of file
427	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	593	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
428	might perform better.	594	might perform better.
429		595
430	On the positive side, ignoring the spurious readyness notifications, this	596	On the positive side, with the exception of the spurious readiness
431	backend actually performed to specification in all tests and is fully	597	notifications, this backend actually performed fully to specification
432	embeddable, which is a rare feat among the OS-specific backends.	598	in all tests and is fully embeddable, which is a rare feat among the
		599	OS-specific backends (I vastly prefer correctness over speed hacks).
		600
		601	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		602	C<EVBACKEND_POLL>.
433		603
434	=item C<EVBACKEND_ALL>	604	=item C<EVBACKEND_ALL>
435		605
436	Try all backends (even potentially broken ones that wouldn't be tried	606	Try all backends (even potentially broken ones that wouldn't be tried
437	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	607	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
438	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	608	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
439		609
440	It is definitely not recommended to use this flag.	610	It is definitely not recommended to use this flag, use whatever
		611	C<ev_recommended_backends ()> returns, or simply do not specify a backend
		612	at all.
		613
		614	=item C<EVBACKEND_MASK>
		615
		616	Not a backend at all, but a mask to select all backend bits from a
		617	C<flags> value, in case you want to mask out any backends from a flags
		618	value (e.g. when modifying the C<LIBEV_FLAGS> environment variable).
441		619
442	=back	620	=back
443		621
444	If one or more of these are ored into the flags value, then only these	622	If one or more of the backend flags are or'ed into the flags value,
445	backends will be tried (in the reverse order as listed here). If none are	623	then only these backends will be tried (in the reverse order as listed
446	specified, all backends in C<ev_recommended_backends ()> will be tried.	624	here). If none are specified, all backends in C<ev_recommended_backends
447		625	()> will be tried.
448	The most typical usage is like this:
449
450	if (!ev_default_loop (0))
451	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
452
453	Restrict libev to the select and poll backends, and do not allow
454	environment settings to be taken into account:
455
456	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
457
458	Use whatever libev has to offer, but make sure that kqueue is used if
459	available (warning, breaks stuff, best use only with your own private
460	event loop and only if you know the OS supports your types of fds):
461
462	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
463
464	=item struct ev_loop *ev_loop_new (unsigned int flags)
465
466	Similar to C<ev_default_loop>, but always creates a new event loop that is
467	always distinct from the default loop. Unlike the default loop, it cannot
468	handle signal and child watchers, and attempts to do so will be greeted by
469	undefined behaviour (or a failed assertion if assertions are enabled).
470		626
471	Example: Try to create a event loop that uses epoll and nothing else.	627	Example: Try to create a event loop that uses epoll and nothing else.
472		628
473	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	629	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
474	if (!epoller)	630	if (!epoller)
475	fatal ("no epoll found here, maybe it hides under your chair");	631	fatal ("no epoll found here, maybe it hides under your chair");
476		632
		633	Example: Use whatever libev has to offer, but make sure that kqueue is
		634	used if available.
		635
		636	struct ev_loop *loop = ev_loop_new (ev_recommended_backends () | EVBACKEND_KQUEUE);
		637
477	=item ev_default_destroy ()	638	=item ev_loop_destroy (loop)
478		639
479	Destroys the default loop again (frees all memory and kernel state	640	Destroys an event loop object (frees all memory and kernel state
480	etc.). None of the active event watchers will be stopped in the normal	641	etc.). None of the active event watchers will be stopped in the normal
481	sense, so e.g. C<ev_is_active> might still return true. It is your	642	sense, so e.g. C<ev_is_active> might still return true. It is your
482	responsibility to either stop all watchers cleanly yoursef I<before>	643	responsibility to either stop all watchers cleanly yourself I<before>
483	calling this function, or cope with the fact afterwards (which is usually	644	calling this function, or cope with the fact afterwards (which is usually
484	the easiest thing, you can just ignore the watchers and/or C<free ()> them	645	the easiest thing, you can just ignore the watchers and/or C<free ()> them
485	for example).	646	for example).
486		647
487	Note that certain global state, such as signal state, will not be freed by	648	Note that certain global state, such as signal state (and installed signal
488	this function, and related watchers (such as signal and child watchers)	649	handlers), will not be freed by this function, and related watchers (such
489	would need to be stopped manually.	650	as signal and child watchers) would need to be stopped manually.
490		651
491	In general it is not advisable to call this function except in the	652	This function is normally used on loop objects allocated by
492	rare occasion where you really need to free e.g. the signal handling	653	C<ev_loop_new>, but it can also be used on the default loop returned by
		654	C<ev_default_loop>, in which case it is not thread-safe.
		655
		656	Note that it is not advisable to call this function on the default loop
		657	except in the rare occasion where you really need to free its resources.
493	pipe fds. If you need dynamically allocated loops it is better to use	658	If you need dynamically allocated loops it is better to use C<ev_loop_new>
494	C<ev_loop_new> and C<ev_loop_destroy>).	659	and C<ev_loop_destroy>.
495		660
496	=item ev_loop_destroy (loop)	661	=item ev_loop_fork (loop)
497		662
498	Like C<ev_default_destroy>, but destroys an event loop created by an
499	earlier call to C<ev_loop_new>.
500
501	=item ev_default_fork ()
502
503	This function sets a flag that causes subsequent C<ev_loop> iterations	663	This function sets a flag that causes subsequent C<ev_run> iterations to
504	to reinitialise the kernel state for backends that have one. Despite the	664	reinitialise the kernel state for backends that have one. Despite the
505	name, you can call it anytime, but it makes most sense after forking, in	665	name, you can call it anytime, but it makes most sense after forking, in
506	the child process (or both child and parent, but that again makes little	666	the child process. You I<must> call it (or use C<EVFLAG_FORKCHECK>) in the
507	sense). You I<must> call it in the child before using any of the libev	667	child before resuming or calling C<ev_run>.
508	functions, and it will only take effect at the next C<ev_loop> iteration.	668
		669	Again, you I<have> to call it on I<any> loop that you want to re-use after
		670	a fork, I<even if you do not plan to use the loop in the parent>. This is
		671	because some kernel interfaces *cough* I<kqueue> *cough* do funny things
		672	during fork.
509		673
510	On the other hand, you only need to call this function in the child	674	On the other hand, you only need to call this function in the child
511	process if and only if you want to use the event library in the child. If	675	process if and only if you want to use the event loop in the child. If
512	you just fork+exec, you don't have to call it at all.	676	you just fork+exec or create a new loop in the child, you don't have to
		677	call it at all (in fact, C<epoll> is so badly broken that it makes a
		678	difference, but libev will usually detect this case on its own and do a
		679	costly reset of the backend).
513		680
514	The function itself is quite fast and it's usually not a problem to call	681	The function itself is quite fast and it's usually not a problem to call
515	it just in case after a fork. To make this easy, the function will fit in	682	it just in case after a fork.
516	quite nicely into a call to C<pthread_atfork>:
517		683
		684	Example: Automate calling C<ev_loop_fork> on the default loop when
		685	using pthreads.
		686
		687	static void
		688	post_fork_child (void)
		689	{
		690	ev_loop_fork (EV_DEFAULT);
		691	}
		692
		693	...
518	pthread_atfork (0, 0, ev_default_fork);	694	pthread_atfork (0, 0, post_fork_child);
519
520	=item ev_loop_fork (loop)
521
522	Like C<ev_default_fork>, but acts on an event loop created by
523	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
524	after fork, and how you do this is entirely your own problem.
525		695
526	=item int ev_is_default_loop (loop)	696	=item int ev_is_default_loop (loop)
527		697
528	Returns true when the given loop actually is the default loop, false otherwise.	698	Returns true when the given loop is, in fact, the default loop, and false
		699	otherwise.
529		700
530	=item unsigned int ev_loop_count (loop)	701	=item unsigned int ev_iteration (loop)
531		702
532	Returns the count of loop iterations for the loop, which is identical to	703	Returns the current iteration count for the event loop, which is identical
533	the number of times libev did poll for new events. It starts at C<0> and	704	to the number of times libev did poll for new events. It starts at C<0>
534	happily wraps around with enough iterations.	705	and happily wraps around with enough iterations.
535		706
536	This value can sometimes be useful as a generation counter of sorts (it	707	This value can sometimes be useful as a generation counter of sorts (it
537	"ticks" the number of loop iterations), as it roughly corresponds with	708	"ticks" the number of loop iterations), as it roughly corresponds with
538	C<ev_prepare> and C<ev_check> calls.	709	C<ev_prepare> and C<ev_check> calls - and is incremented between the
		710	prepare and check phases.
		711
		712	=item unsigned int ev_depth (loop)
		713
		714	Returns the number of times C<ev_run> was entered minus the number of
		715	times C<ev_run> was exited normally, in other words, the recursion depth.
		716
		717	Outside C<ev_run>, this number is zero. In a callback, this number is
		718	C<1>, unless C<ev_run> was invoked recursively (or from another thread),
		719	in which case it is higher.
		720
		721	Leaving C<ev_run> abnormally (setjmp/longjmp, cancelling the thread,
		722	throwing an exception etc.), doesn't count as "exit" - consider this
		723	as a hint to avoid such ungentleman-like behaviour unless it's really
		724	convenient, in which case it is fully supported.
539		725
540	=item unsigned int ev_backend (loop)	726	=item unsigned int ev_backend (loop)
541		727
542	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	728	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
543	use.	729	use.
…		…
548	received events and started processing them. This timestamp does not	734	received events and started processing them. This timestamp does not
549	change as long as callbacks are being processed, and this is also the base	735	change as long as callbacks are being processed, and this is also the base
550	time used for relative timers. You can treat it as the timestamp of the	736	time used for relative timers. You can treat it as the timestamp of the
551	event occurring (or more correctly, libev finding out about it).	737	event occurring (or more correctly, libev finding out about it).
552		738
		739	=item ev_now_update (loop)
		740
		741	Establishes the current time by querying the kernel, updating the time
		742	returned by C<ev_now ()> in the progress. This is a costly operation and
		743	is usually done automatically within C<ev_run ()>.
		744
		745	This function is rarely useful, but when some event callback runs for a
		746	very long time without entering the event loop, updating libev's idea of
		747	the current time is a good idea.
		748
		749	See also L<The special problem of time updates> in the C<ev_timer> section.
		750
		751	=item ev_suspend (loop)
		752
		753	=item ev_resume (loop)
		754
		755	These two functions suspend and resume an event loop, for use when the
		756	loop is not used for a while and timeouts should not be processed.
		757
		758	A typical use case would be an interactive program such as a game: When
		759	the user presses C<^Z> to suspend the game and resumes it an hour later it
		760	would be best to handle timeouts as if no time had actually passed while
		761	the program was suspended. This can be achieved by calling C<ev_suspend>
		762	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		763	C<ev_resume> directly afterwards to resume timer processing.
		764
		765	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		766	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		767	will be rescheduled (that is, they will lose any events that would have
		768	occurred while suspended).
		769
		770	After calling C<ev_suspend> you B<must not> call I<any> function on the
		771	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		772	without a previous call to C<ev_suspend>.
		773
		774	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		775	event loop time (see C<ev_now_update>).
		776
553	=item ev_loop (loop, int flags)	777	=item ev_run (loop, int flags)
554		778
555	Finally, this is it, the event handler. This function usually is called	779	Finally, this is it, the event handler. This function usually is called
556	after you initialised all your watchers and you want to start handling	780	after you have initialised all your watchers and you want to start
557	events.	781	handling events. It will ask the operating system for any new events, call
		782	the watcher callbacks, an then repeat the whole process indefinitely: This
		783	is why event loops are called I<loops>.
558		784
559	If the flags argument is specified as C<0>, it will not return until	785	If the flags argument is specified as C<0>, it will keep handling events
560	either no event watchers are active anymore or C<ev_unloop> was called.	786	until either no event watchers are active anymore or C<ev_break> was
		787	called.
561		788
562	Please note that an explicit C<ev_unloop> is usually better than	789	Please note that an explicit C<ev_break> is usually better than
563	relying on all watchers to be stopped when deciding when a program has	790	relying on all watchers to be stopped when deciding when a program has
564	finished (especially in interactive programs), but having a program that	791	finished (especially in interactive programs), but having a program
565	automatically loops as long as it has to and no longer by virtue of	792	that automatically loops as long as it has to and no longer by virtue
566	relying on its watchers stopping correctly is a thing of beauty.	793	of relying on its watchers stopping correctly, that is truly a thing of
		794	beauty.
567		795
		796	This function is also I<mostly> exception-safe - you can break out of
		797	a C<ev_run> call by calling C<longjmp> in a callback, throwing a C++
		798	exception and so on. This does not decrement the C<ev_depth> value, nor
		799	will it clear any outstanding C<EVBREAK_ONE> breaks.
		800
568	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	801	A flags value of C<EVRUN_NOWAIT> will look for new events, will handle
569	those events and any outstanding ones, but will not block your process in	802	those events and any already outstanding ones, but will not wait and
570	case there are no events and will return after one iteration of the loop.	803	block your process in case there are no events and will return after one
		804	iteration of the loop. This is sometimes useful to poll and handle new
		805	events while doing lengthy calculations, to keep the program responsive.
571		806
572	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	807	A flags value of C<EVRUN_ONCE> will look for new events (waiting if
573	neccessary) and will handle those and any outstanding ones. It will block	808	necessary) and will handle those and any already outstanding ones. It
574	your process until at least one new event arrives, and will return after	809	will block your process until at least one new event arrives (which could
575	one iteration of the loop. This is useful if you are waiting for some	810	be an event internal to libev itself, so there is no guarantee that a
576	external event in conjunction with something not expressible using other	811	user-registered callback will be called), and will return after one
		812	iteration of the loop.
		813
		814	This is useful if you are waiting for some external event in conjunction
		815	with something not expressible using other libev watchers (i.e. "roll your
577	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	816	own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
578	usually a better approach for this kind of thing.	817	usually a better approach for this kind of thing.
579		818
580	Here are the gory details of what C<ev_loop> does:	819	Here are the gory details of what C<ev_run> does:
581		820
		821	- Increment loop depth.
		822	- Reset the ev_break status.
582	- Before the first iteration, call any pending watchers.	823	- Before the first iteration, call any pending watchers.
		824	LOOP:
583	* If EVFLAG_FORKCHECK was used, check for a fork.	825	- If EVFLAG_FORKCHECK was used, check for a fork.
584	- If a fork was detected, queue and call all fork watchers.	826	- If a fork was detected (by any means), queue and call all fork watchers.
585	- Queue and call all prepare watchers.	827	- Queue and call all prepare watchers.
		828	- If ev_break was called, goto FINISH.
586	- If we have been forked, recreate the kernel state.	829	- If we have been forked, detach and recreate the kernel state
		830	as to not disturb the other process.
587	- Update the kernel state with all outstanding changes.	831	- Update the kernel state with all outstanding changes.
588	- Update the "event loop time".	832	- Update the "event loop time" (ev_now ()).
589	- Calculate for how long to sleep or block, if at all	833	- Calculate for how long to sleep or block, if at all
590	(active idle watchers, EVLOOP_NONBLOCK or not having	834	(active idle watchers, EVRUN_NOWAIT or not having
591	any active watchers at all will result in not sleeping).	835	any active watchers at all will result in not sleeping).
592	- Sleep if the I/O and timer collect interval say so.	836	- Sleep if the I/O and timer collect interval say so.
		837	- Increment loop iteration counter.
593	- Block the process, waiting for any events.	838	- Block the process, waiting for any events.
594	- Queue all outstanding I/O (fd) events.	839	- Queue all outstanding I/O (fd) events.
595	- Update the "event loop time" and do time jump handling.	840	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
596	- Queue all outstanding timers.	841	- Queue all expired timers.
597	- Queue all outstanding periodics.	842	- Queue all expired periodics.
598	- If no events are pending now, queue all idle watchers.	843	- Queue all idle watchers with priority higher than that of pending events.
599	- Queue all check watchers.	844	- Queue all check watchers.
600	- Call all queued watchers in reverse order (i.e. check watchers first).	845	- Call all queued watchers in reverse order (i.e. check watchers first).
601	Signals and child watchers are implemented as I/O watchers, and will	846	Signals and child watchers are implemented as I/O watchers, and will
602	be handled here by queueing them when their watcher gets executed.	847	be handled here by queueing them when their watcher gets executed.
603	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	848	- If ev_break has been called, or EVRUN_ONCE or EVRUN_NOWAIT
604	were used, or there are no active watchers, return, otherwise	849	were used, or there are no active watchers, goto FINISH, otherwise
605	continue with step *.	850	continue with step LOOP.
		851	FINISH:
		852	- Reset the ev_break status iff it was EVBREAK_ONE.
		853	- Decrement the loop depth.
		854	- Return.
606		855
607	Example: Queue some jobs and then loop until no events are outstanding	856	Example: Queue some jobs and then loop until no events are outstanding
608	anymore.	857	anymore.
609		858
610	... queue jobs here, make sure they register event watchers as long	859	... queue jobs here, make sure they register event watchers as long
611	... as they still have work to do (even an idle watcher will do..)	860	... as they still have work to do (even an idle watcher will do..)
612	ev_loop (my_loop, 0);	861	ev_run (my_loop, 0);
613	... jobs done. yeah!	862	... jobs done or somebody called unloop. yeah!
614		863
615	=item ev_unloop (loop, how)	864	=item ev_break (loop, how)
616		865
617	Can be used to make a call to C<ev_loop> return early (but only after it	866	Can be used to make a call to C<ev_run> return early (but only after it
618	has processed all outstanding events). The C<how> argument must be either	867	has processed all outstanding events). The C<how> argument must be either
619	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	868	C<EVBREAK_ONE>, which will make the innermost C<ev_run> call return, or
620	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	869	C<EVBREAK_ALL>, which will make all nested C<ev_run> calls return.
621		870
622	This "unloop state" will be cleared when entering C<ev_loop> again.	871	This "break state" will be cleared on the next call to C<ev_run>.
		872
		873	It is safe to call C<ev_break> from outside any C<ev_run> calls, too, in
		874	which case it will have no effect.
623		875
624	=item ev_ref (loop)	876	=item ev_ref (loop)
625		877
626	=item ev_unref (loop)	878	=item ev_unref (loop)
627		879
628	Ref/unref can be used to add or remove a reference count on the event	880	Ref/unref can be used to add or remove a reference count on the event
629	loop: Every watcher keeps one reference, and as long as the reference	881	loop: Every watcher keeps one reference, and as long as the reference
630	count is nonzero, C<ev_loop> will not return on its own. If you have	882	count is nonzero, C<ev_run> will not return on its own.
631	a watcher you never unregister that should not keep C<ev_loop> from	883
632	returning, ev_unref() after starting, and ev_ref() before stopping it. For	884	This is useful when you have a watcher that you never intend to
		885	unregister, but that nevertheless should not keep C<ev_run> from
		886	returning. In such a case, call C<ev_unref> after starting, and C<ev_ref>
		887	before stopping it.
		888
633	example, libev itself uses this for its internal signal pipe: It is not	889	As an example, libev itself uses this for its internal signal pipe: It
634	visible to the libev user and should not keep C<ev_loop> from exiting if	890	is not visible to the libev user and should not keep C<ev_run> from
635	no event watchers registered by it are active. It is also an excellent	891	exiting if no event watchers registered by it are active. It is also an
636	way to do this for generic recurring timers or from within third-party	892	excellent way to do this for generic recurring timers or from within
637	libraries. Just remember to I<unref after start> and I<ref before stop>	893	third-party libraries. Just remember to I<unref after start> and I<ref
638	(but only if the watcher wasn't active before, or was active before,	894	before stop> (but only if the watcher wasn't active before, or was active
639	respectively).	895	before, respectively. Note also that libev might stop watchers itself
		896	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		897	in the callback).
640		898
641	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	899	Example: Create a signal watcher, but keep it from keeping C<ev_run>
642	running when nothing else is active.	900	running when nothing else is active.
643		901
644	struct ev_signal exitsig;	902	ev_signal exitsig;
645	ev_signal_init (&exitsig, sig_cb, SIGINT);	903	ev_signal_init (&exitsig, sig_cb, SIGINT);
646	ev_signal_start (loop, &exitsig);	904	ev_signal_start (loop, &exitsig);
647	evf_unref (loop);	905	ev_unref (loop);
648		906
649	Example: For some weird reason, unregister the above signal handler again.	907	Example: For some weird reason, unregister the above signal handler again.
650		908
651	ev_ref (loop);	909	ev_ref (loop);
652	ev_signal_stop (loop, &exitsig);	910	ev_signal_stop (loop, &exitsig);
653		911
654	=item ev_set_io_collect_interval (loop, ev_tstamp interval)	912	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
655		913
656	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)	914	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
657		915
658	These advanced functions influence the time that libev will spend waiting	916	These advanced functions influence the time that libev will spend waiting
659	for events. Both are by default C<0>, meaning that libev will try to	917	for events. Both time intervals are by default C<0>, meaning that libev
660	invoke timer/periodic callbacks and I/O callbacks with minimum latency.	918	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		919	latency.
661		920
662	Setting these to a higher value (the C<interval> I<must> be >= C<0>)	921	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
663	allows libev to delay invocation of I/O and timer/periodic callbacks to	922	allows libev to delay invocation of I/O and timer/periodic callbacks
664	increase efficiency of loop iterations.	923	to increase efficiency of loop iterations (or to increase power-saving
		924	opportunities).
665		925
666	The background is that sometimes your program runs just fast enough to	926	The idea is that sometimes your program runs just fast enough to handle
667	handle one (or very few) event(s) per loop iteration. While this makes	927	one (or very few) event(s) per loop iteration. While this makes the
668	the program responsive, it also wastes a lot of CPU time to poll for new	928	program responsive, it also wastes a lot of CPU time to poll for new
669	events, especially with backends like C<select ()> which have a high	929	events, especially with backends like C<select ()> which have a high
670	overhead for the actual polling but can deliver many events at once.	930	overhead for the actual polling but can deliver many events at once.
671		931
672	By setting a higher I<io collect interval> you allow libev to spend more	932	By setting a higher I<io collect interval> you allow libev to spend more
673	time collecting I/O events, so you can handle more events per iteration,	933	time collecting I/O events, so you can handle more events per iteration,
674	at the cost of increasing latency. Timeouts (both C<ev_periodic> and	934	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
675	C<ev_timer>) will be not affected. Setting this to a non-null value will	935	C<ev_timer>) will be not affected. Setting this to a non-null value will
676	introduce an additional C<ev_sleep ()> call into most loop iterations.	936	introduce an additional C<ev_sleep ()> call into most loop iterations. The
		937	sleep time ensures that libev will not poll for I/O events more often then
		938	once per this interval, on average.
677		939
678	Likewise, by setting a higher I<timeout collect interval> you allow libev	940	Likewise, by setting a higher I<timeout collect interval> you allow libev
679	to spend more time collecting timeouts, at the expense of increased	941	to spend more time collecting timeouts, at the expense of increased
680	latency (the watcher callback will be called later). C<ev_io> watchers	942	latency/jitter/inexactness (the watcher callback will be called
681	will not be affected. Setting this to a non-null value will not introduce	943	later). C<ev_io> watchers will not be affected. Setting this to a non-null
682	any overhead in libev.	944	value will not introduce any overhead in libev.
683		945
684	Many (busy) programs can usually benefit by setting the io collect	946	Many (busy) programs can usually benefit by setting the I/O collect
685	interval to a value near C<0.1> or so, which is often enough for	947	interval to a value near C<0.1> or so, which is often enough for
686	interactive servers (of course not for games), likewise for timeouts. It	948	interactive servers (of course not for games), likewise for timeouts. It
687	usually doesn't make much sense to set it to a lower value than C<0.01>,	949	usually doesn't make much sense to set it to a lower value than C<0.01>,
688	as this approsaches the timing granularity of most systems.	950	as this approaches the timing granularity of most systems. Note that if
		951	you do transactions with the outside world and you can't increase the
		952	parallelity, then this setting will limit your transaction rate (if you
		953	need to poll once per transaction and the I/O collect interval is 0.01,
		954	then you can't do more than 100 transactions per second).
		955
		956	Setting the I<timeout collect interval> can improve the opportunity for
		957	saving power, as the program will "bundle" timer callback invocations that
		958	are "near" in time together, by delaying some, thus reducing the number of
		959	times the process sleeps and wakes up again. Another useful technique to
		960	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		961	they fire on, say, one-second boundaries only.
		962
		963	Example: we only need 0.1s timeout granularity, and we wish not to poll
		964	more often than 100 times per second:
		965
		966	ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
		967	ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
		968
		969	=item ev_invoke_pending (loop)
		970
		971	This call will simply invoke all pending watchers while resetting their
		972	pending state. Normally, C<ev_run> does this automatically when required,
		973	but when overriding the invoke callback this call comes handy. This
		974	function can be invoked from a watcher - this can be useful for example
		975	when you want to do some lengthy calculation and want to pass further
		976	event handling to another thread (you still have to make sure only one
		977	thread executes within C<ev_invoke_pending> or C<ev_run> of course).
		978
		979	=item int ev_pending_count (loop)
		980
		981	Returns the number of pending watchers - zero indicates that no watchers
		982	are pending.
		983
		984	=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
		985
		986	This overrides the invoke pending functionality of the loop: Instead of
		987	invoking all pending watchers when there are any, C<ev_run> will call
		988	this callback instead. This is useful, for example, when you want to
		989	invoke the actual watchers inside another context (another thread etc.).
		990
		991	If you want to reset the callback, use C<ev_invoke_pending> as new
		992	callback.
		993
		994	=item ev_set_loop_release_cb (loop, void (*release)(EV_P), void (*acquire)(EV_P))
		995
		996	Sometimes you want to share the same loop between multiple threads. This
		997	can be done relatively simply by putting mutex_lock/unlock calls around
		998	each call to a libev function.
		999
		1000	However, C<ev_run> can run an indefinite time, so it is not feasible
		1001	to wait for it to return. One way around this is to wake up the event
		1002	loop via C<ev_break> and C<av_async_send>, another way is to set these
		1003	I<release> and I<acquire> callbacks on the loop.
		1004
		1005	When set, then C<release> will be called just before the thread is
		1006	suspended waiting for new events, and C<acquire> is called just
		1007	afterwards.
		1008
		1009	Ideally, C<release> will just call your mutex_unlock function, and
		1010	C<acquire> will just call the mutex_lock function again.
		1011
		1012	While event loop modifications are allowed between invocations of
		1013	C<release> and C<acquire> (that's their only purpose after all), no
		1014	modifications done will affect the event loop, i.e. adding watchers will
		1015	have no effect on the set of file descriptors being watched, or the time
		1016	waited. Use an C<ev_async> watcher to wake up C<ev_run> when you want it
		1017	to take note of any changes you made.
		1018
		1019	In theory, threads executing C<ev_run> will be async-cancel safe between
		1020	invocations of C<release> and C<acquire>.
		1021
		1022	See also the locking example in the C<THREADS> section later in this
		1023	document.
		1024
		1025	=item ev_set_userdata (loop, void *data)
		1026
		1027	=item void *ev_userdata (loop)
		1028
		1029	Set and retrieve a single C<void *> associated with a loop. When
		1030	C<ev_set_userdata> has never been called, then C<ev_userdata> returns
		1031	C<0>.
		1032
		1033	These two functions can be used to associate arbitrary data with a loop,
		1034	and are intended solely for the C<invoke_pending_cb>, C<release> and
		1035	C<acquire> callbacks described above, but of course can be (ab-)used for
		1036	any other purpose as well.
		1037
		1038	=item ev_verify (loop)
		1039
		1040	This function only does something when C<EV_VERIFY> support has been
		1041	compiled in, which is the default for non-minimal builds. It tries to go
		1042	through all internal structures and checks them for validity. If anything
		1043	is found to be inconsistent, it will print an error message to standard
		1044	error and call C<abort ()>.
		1045
		1046	This can be used to catch bugs inside libev itself: under normal
		1047	circumstances, this function will never abort as of course libev keeps its
		1048	data structures consistent.
689		1049
690	=back	1050	=back
691		1051
692		1052
693	=head1 ANATOMY OF A WATCHER	1053	=head1 ANATOMY OF A WATCHER
694		1054
		1055	In the following description, uppercase C<TYPE> in names stands for the
		1056	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		1057	watchers and C<ev_io_start> for I/O watchers.
		1058
695	A watcher is a structure that you create and register to record your	1059	A watcher is an opaque structure that you allocate and register to record
696	interest in some event. For instance, if you want to wait for STDIN to	1060	your interest in some event. To make a concrete example, imagine you want
697	become readable, you would create an C<ev_io> watcher for that:	1061	to wait for STDIN to become readable, you would create an C<ev_io> watcher
		1062	for that:
698		1063
699	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1064	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
700	{	1065	{
701	ev_io_stop (w);	1066	ev_io_stop (w);
702	ev_unloop (loop, EVUNLOOP_ALL);	1067	ev_break (loop, EVBREAK_ALL);
703	}	1068	}
704		1069
705	struct ev_loop *loop = ev_default_loop (0);	1070	struct ev_loop *loop = ev_default_loop (0);
		1071
706	struct ev_io stdin_watcher;	1072	ev_io stdin_watcher;
		1073
707	ev_init (&stdin_watcher, my_cb);	1074	ev_init (&stdin_watcher, my_cb);
708	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	1075	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
709	ev_io_start (loop, &stdin_watcher);	1076	ev_io_start (loop, &stdin_watcher);
		1077
710	ev_loop (loop, 0);	1078	ev_run (loop, 0);
711		1079
712	As you can see, you are responsible for allocating the memory for your	1080	As you can see, you are responsible for allocating the memory for your
713	watcher structures (and it is usually a bad idea to do this on the stack,	1081	watcher structures (and it is I<usually> a bad idea to do this on the
714	although this can sometimes be quite valid).	1082	stack).
715		1083
		1084	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		1085	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
		1086
716	Each watcher structure must be initialised by a call to C<ev_init	1087	Each watcher structure must be initialised by a call to C<ev_init (watcher
717	(watcher *, callback)>, which expects a callback to be provided. This	1088	*, callback)>, which expects a callback to be provided. This callback is
718	callback gets invoked each time the event occurs (or, in the case of io	1089	invoked each time the event occurs (or, in the case of I/O watchers, each
719	watchers, each time the event loop detects that the file descriptor given	1090	time the event loop detects that the file descriptor given is readable
720	is readable and/or writable).	1091	and/or writable).
721		1092
722	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	1093	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
723	with arguments specific to this watcher type. There is also a macro	1094	macro to configure it, with arguments specific to the watcher type. There
724	to combine initialisation and setting in one call: C<< ev_<type>_init	1095	is also a macro to combine initialisation and setting in one call: C<<
725	(watcher *, callback, ...) >>.	1096	ev_TYPE_init (watcher *, callback, ...) >>.
726		1097
727	To make the watcher actually watch out for events, you have to start it	1098	To make the watcher actually watch out for events, you have to start it
728	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	1099	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
729	*) >>), and you can stop watching for events at any time by calling the	1100	*) >>), and you can stop watching for events at any time by calling the
730	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	1101	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
731		1102
732	As long as your watcher is active (has been started but not stopped) you	1103	As long as your watcher is active (has been started but not stopped) you
733	must not touch the values stored in it. Most specifically you must never	1104	must not touch the values stored in it. Most specifically you must never
734	reinitialise it or call its C<set> macro.	1105	reinitialise it or call its C<ev_TYPE_set> macro.
735		1106
736	Each and every callback receives the event loop pointer as first, the	1107	Each and every callback receives the event loop pointer as first, the
737	registered watcher structure as second, and a bitset of received events as	1108	registered watcher structure as second, and a bitset of received events as
738	third argument.	1109	third argument.
739		1110
…		…
748	=item C<EV_WRITE>	1119	=item C<EV_WRITE>
749		1120
750	The file descriptor in the C<ev_io> watcher has become readable and/or	1121	The file descriptor in the C<ev_io> watcher has become readable and/or
751	writable.	1122	writable.
752		1123
753	=item C<EV_TIMEOUT>	1124	=item C<EV_TIMER>
754		1125
755	The C<ev_timer> watcher has timed out.	1126	The C<ev_timer> watcher has timed out.
756		1127
757	=item C<EV_PERIODIC>	1128	=item C<EV_PERIODIC>
758		1129
…		…
776		1147
777	=item C<EV_PREPARE>	1148	=item C<EV_PREPARE>
778		1149
779	=item C<EV_CHECK>	1150	=item C<EV_CHECK>
780		1151
781	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts	1152	All C<ev_prepare> watchers are invoked just I<before> C<ev_run> starts
782	to gather new events, and all C<ev_check> watchers are invoked just after	1153	to gather new events, and all C<ev_check> watchers are invoked just after
783	C<ev_loop> has gathered them, but before it invokes any callbacks for any	1154	C<ev_run> has gathered them, but before it invokes any callbacks for any
784	received events. Callbacks of both watcher types can start and stop as	1155	received events. Callbacks of both watcher types can start and stop as
785	many watchers as they want, and all of them will be taken into account	1156	many watchers as they want, and all of them will be taken into account
786	(for example, a C<ev_prepare> watcher might start an idle watcher to keep	1157	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
787	C<ev_loop> from blocking).	1158	C<ev_run> from blocking).
788		1159
789	=item C<EV_EMBED>	1160	=item C<EV_EMBED>
790		1161
791	The embedded event loop specified in the C<ev_embed> watcher needs attention.	1162	The embedded event loop specified in the C<ev_embed> watcher needs attention.
792		1163
793	=item C<EV_FORK>	1164	=item C<EV_FORK>
794		1165
795	The event loop has been resumed in the child process after fork (see	1166	The event loop has been resumed in the child process after fork (see
796	C<ev_fork>).	1167	C<ev_fork>).
797		1168
		1169	=item C<EV_CLEANUP>
		1170
		1171	The event loop is about to be destroyed (see C<ev_cleanup>).
		1172
798	=item C<EV_ASYNC>	1173	=item C<EV_ASYNC>
799		1174
800	The given async watcher has been asynchronously notified (see C<ev_async>).	1175	The given async watcher has been asynchronously notified (see C<ev_async>).
801		1176
		1177	=item C<EV_CUSTOM>
		1178
		1179	Not ever sent (or otherwise used) by libev itself, but can be freely used
		1180	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		1181
802	=item C<EV_ERROR>	1182	=item C<EV_ERROR>
803		1183
804	An unspecified error has occured, the watcher has been stopped. This might	1184	An unspecified error has occurred, the watcher has been stopped. This might
805	happen because the watcher could not be properly started because libev	1185	happen because the watcher could not be properly started because libev
806	ran out of memory, a file descriptor was found to be closed or any other	1186	ran out of memory, a file descriptor was found to be closed or any other
		1187	problem. Libev considers these application bugs.
		1188
807	problem. You best act on it by reporting the problem and somehow coping	1189	You best act on it by reporting the problem and somehow coping with the
808	with the watcher being stopped.	1190	watcher being stopped. Note that well-written programs should not receive
		1191	an error ever, so when your watcher receives it, this usually indicates a
		1192	bug in your program.
809		1193
810	Libev will usually signal a few "dummy" events together with an error,	1194	Libev will usually signal a few "dummy" events together with an error, for
811	for example it might indicate that a fd is readable or writable, and if	1195	example it might indicate that a fd is readable or writable, and if your
812	your callbacks is well-written it can just attempt the operation and cope	1196	callbacks is well-written it can just attempt the operation and cope with
813	with the error from read() or write(). This will not work in multithreaded	1197	the error from read() or write(). This will not work in multi-threaded
814	programs, though, so beware.	1198	programs, though, as the fd could already be closed and reused for another
		1199	thing, so beware.
815		1200
816	=back	1201	=back
817		1202
818	=head2 GENERIC WATCHER FUNCTIONS	1203	=head2 GENERIC WATCHER FUNCTIONS
819
820	In the following description, C<TYPE> stands for the watcher type,
821	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
822		1204
823	=over 4	1205	=over 4
824		1206
825	=item C<ev_init> (ev_TYPE *watcher, callback)	1207	=item C<ev_init> (ev_TYPE *watcher, callback)
826		1208
…		…
832	which rolls both calls into one.	1214	which rolls both calls into one.
833		1215
834	You can reinitialise a watcher at any time as long as it has been stopped	1216	You can reinitialise a watcher at any time as long as it has been stopped
835	(or never started) and there are no pending events outstanding.	1217	(or never started) and there are no pending events outstanding.
836		1218
837	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	1219	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
838	int revents)>.	1220	int revents)>.
839		1221
		1222	Example: Initialise an C<ev_io> watcher in two steps.
		1223
		1224	ev_io w;
		1225	ev_init (&w, my_cb);
		1226	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1227
840	=item C<ev_TYPE_set> (ev_TYPE *, [args])	1228	=item C<ev_TYPE_set> (ev_TYPE *watcher, [args])
841		1229
842	This macro initialises the type-specific parts of a watcher. You need to	1230	This macro initialises the type-specific parts of a watcher. You need to
843	call C<ev_init> at least once before you call this macro, but you can	1231	call C<ev_init> at least once before you call this macro, but you can
844	call C<ev_TYPE_set> any number of times. You must not, however, call this	1232	call C<ev_TYPE_set> any number of times. You must not, however, call this
845	macro on a watcher that is active (it can be pending, however, which is a	1233	macro on a watcher that is active (it can be pending, however, which is a
846	difference to the C<ev_init> macro).	1234	difference to the C<ev_init> macro).
847		1235
848	Although some watcher types do not have type-specific arguments	1236	Although some watcher types do not have type-specific arguments
849	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	1237	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
850		1238
		1239	See C<ev_init>, above, for an example.
		1240
851	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	1241	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
852		1242
853	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	1243	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
854	calls into a single call. This is the most convinient method to initialise	1244	calls into a single call. This is the most convenient method to initialise
855	a watcher. The same limitations apply, of course.	1245	a watcher. The same limitations apply, of course.
856		1246
		1247	Example: Initialise and set an C<ev_io> watcher in one step.
		1248
		1249	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1250
857	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	1251	=item C<ev_TYPE_start> (loop, ev_TYPE *watcher)
858		1252
859	Starts (activates) the given watcher. Only active watchers will receive	1253	Starts (activates) the given watcher. Only active watchers will receive
860	events. If the watcher is already active nothing will happen.	1254	events. If the watcher is already active nothing will happen.
861		1255
		1256	Example: Start the C<ev_io> watcher that is being abused as example in this
		1257	whole section.
		1258
		1259	ev_io_start (EV_DEFAULT_UC, &w);
		1260
862	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	1261	=item C<ev_TYPE_stop> (loop, ev_TYPE *watcher)
863		1262
864	Stops the given watcher again (if active) and clears the pending	1263	Stops the given watcher if active, and clears the pending status (whether
		1264	the watcher was active or not).
		1265
865	status. It is possible that stopped watchers are pending (for example,	1266	It is possible that stopped watchers are pending - for example,
866	non-repeating timers are being stopped when they become pending), but	1267	non-repeating timers are being stopped when they become pending - but
867	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	1268	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
868	you want to free or reuse the memory used by the watcher it is therefore a	1269	pending. If you want to free or reuse the memory used by the watcher it is
869	good idea to always call its C<ev_TYPE_stop> function.	1270	therefore a good idea to always call its C<ev_TYPE_stop> function.
870		1271
871	=item bool ev_is_active (ev_TYPE *watcher)	1272	=item bool ev_is_active (ev_TYPE *watcher)
872		1273
873	Returns a true value iff the watcher is active (i.e. it has been started	1274	Returns a true value iff the watcher is active (i.e. it has been started
874	and not yet been stopped). As long as a watcher is active you must not modify	1275	and not yet been stopped). As long as a watcher is active you must not modify
…		…
890	=item ev_cb_set (ev_TYPE *watcher, callback)	1291	=item ev_cb_set (ev_TYPE *watcher, callback)
891		1292
892	Change the callback. You can change the callback at virtually any time	1293	Change the callback. You can change the callback at virtually any time
893	(modulo threads).	1294	(modulo threads).
894		1295
895	=item ev_set_priority (ev_TYPE *watcher, priority)	1296	=item ev_set_priority (ev_TYPE *watcher, int priority)
896		1297
897	=item int ev_priority (ev_TYPE *watcher)	1298	=item int ev_priority (ev_TYPE *watcher)
898		1299
899	Set and query the priority of the watcher. The priority is a small	1300	Set and query the priority of the watcher. The priority is a small
900	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>	1301	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
901	(default: C<-2>). Pending watchers with higher priority will be invoked	1302	(default: C<-2>). Pending watchers with higher priority will be invoked
902	before watchers with lower priority, but priority will not keep watchers	1303	before watchers with lower priority, but priority will not keep watchers
903	from being executed (except for C<ev_idle> watchers).	1304	from being executed (except for C<ev_idle> watchers).
904		1305
905	This means that priorities are I<only> used for ordering callback
906	invocation after new events have been received. This is useful, for
907	example, to reduce latency after idling, or more often, to bind two
908	watchers on the same event and make sure one is called first.
909
910	If you need to suppress invocation when higher priority events are pending	1306	If you need to suppress invocation when higher priority events are pending
911	you need to look at C<ev_idle> watchers, which provide this functionality.	1307	you need to look at C<ev_idle> watchers, which provide this functionality.
912		1308
913	You I<must not> change the priority of a watcher as long as it is active or	1309	You I<must not> change the priority of a watcher as long as it is active or
914	pending.	1310	pending.
915		1311
		1312	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1313	fine, as long as you do not mind that the priority value you query might
		1314	or might not have been clamped to the valid range.
		1315
916	The default priority used by watchers when no priority has been set is	1316	The default priority used by watchers when no priority has been set is
917	always C<0>, which is supposed to not be too high and not be too low :).	1317	always C<0>, which is supposed to not be too high and not be too low :).
918		1318
919	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is	1319	See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
920	fine, as long as you do not mind that the priority value you query might	1320	priorities.
921	or might not have been adjusted to be within valid range.
922		1321
923	=item ev_invoke (loop, ev_TYPE *watcher, int revents)	1322	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
924		1323
925	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither	1324	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
926	C<loop> nor C<revents> need to be valid as long as the watcher callback	1325	C<loop> nor C<revents> need to be valid as long as the watcher callback
927	can deal with that fact.	1326	can deal with that fact, as both are simply passed through to the
		1327	callback.
928		1328
929	=item int ev_clear_pending (loop, ev_TYPE *watcher)	1329	=item int ev_clear_pending (loop, ev_TYPE *watcher)
930		1330
931	If the watcher is pending, this function returns clears its pending status	1331	If the watcher is pending, this function clears its pending status and
932	and returns its C<revents> bitset (as if its callback was invoked). If the	1332	returns its C<revents> bitset (as if its callback was invoked). If the
933	watcher isn't pending it does nothing and returns C<0>.	1333	watcher isn't pending it does nothing and returns C<0>.
934		1334
		1335	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1336	callback to be invoked, which can be accomplished with this function.
		1337
		1338	=item ev_feed_event (loop, ev_TYPE *watcher, int revents)
		1339
		1340	Feeds the given event set into the event loop, as if the specified event
		1341	had happened for the specified watcher (which must be a pointer to an
		1342	initialised but not necessarily started event watcher). Obviously you must
		1343	not free the watcher as long as it has pending events.
		1344
		1345	Stopping the watcher, letting libev invoke it, or calling
		1346	C<ev_clear_pending> will clear the pending event, even if the watcher was
		1347	not started in the first place.
		1348
		1349	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
		1350	functions that do not need a watcher.
		1351
935	=back	1352	=back
936		1353
937
938	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1354	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
939		1355
940	Each watcher has, by default, a member C<void *data> that you can change	1356	Each watcher has, by default, a member C<void *data> that you can change
941	and read at any time, libev will completely ignore it. This can be used	1357	and read at any time: libev will completely ignore it. This can be used
942	to associate arbitrary data with your watcher. If you need more data and	1358	to associate arbitrary data with your watcher. If you need more data and
943	don't want to allocate memory and store a pointer to it in that data	1359	don't want to allocate memory and store a pointer to it in that data
944	member, you can also "subclass" the watcher type and provide your own	1360	member, you can also "subclass" the watcher type and provide your own
945	data:	1361	data:
946		1362
947	struct my_io	1363	struct my_io
948	{	1364	{
949	struct ev_io io;	1365	ev_io io;
950	int otherfd;	1366	int otherfd;
951	void *somedata;	1367	void *somedata;
952	struct whatever *mostinteresting;	1368	struct whatever *mostinteresting;
953	}	1369	};
		1370
		1371	...
		1372	struct my_io w;
		1373	ev_io_init (&w.io, my_cb, fd, EV_READ);
954		1374
955	And since your callback will be called with a pointer to the watcher, you	1375	And since your callback will be called with a pointer to the watcher, you
956	can cast it back to your own type:	1376	can cast it back to your own type:
957		1377
958	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1378	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
959	{	1379	{
960	struct my_io *w = (struct my_io *)w_;	1380	struct my_io *w = (struct my_io *)w_;
961	...	1381	...
962	}	1382	}
963		1383
964	More interesting and less C-conformant ways of casting your callback type	1384	More interesting and less C-conformant ways of casting your callback type
965	instead have been omitted.	1385	instead have been omitted.
966		1386
967	Another common scenario is having some data structure with multiple	1387	Another common scenario is to use some data structure with multiple
968	watchers:	1388	embedded watchers:
969		1389
970	struct my_biggy	1390	struct my_biggy
971	{	1391	{
972	int some_data;	1392	int some_data;
973	ev_timer t1;	1393	ev_timer t1;
974	ev_timer t2;	1394	ev_timer t2;
975	}	1395	}
976		1396
977	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1397	In this case getting the pointer to C<my_biggy> is a bit more
978	you need to use C<offsetof>:	1398	complicated: Either you store the address of your C<my_biggy> struct
		1399	in the C<data> member of the watcher (for woozies), or you need to use
		1400	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1401	programmers):
979		1402
980	#include <stddef.h>	1403	#include <stddef.h>
981		1404
982	static void	1405	static void
983	t1_cb (EV_P_ struct ev_timer *w, int revents)	1406	t1_cb (EV_P_ ev_timer *w, int revents)
984	{	1407	{
985	struct my_biggy big = (struct my_biggy *	1408	struct my_biggy big = (struct my_biggy *)
986	(((char *)w) - offsetof (struct my_biggy, t1));	1409	(((char *)w) - offsetof (struct my_biggy, t1));
987	}	1410	}
988		1411
989	static void	1412	static void
990	t2_cb (EV_P_ struct ev_timer *w, int revents)	1413	t2_cb (EV_P_ ev_timer *w, int revents)
991	{	1414	{
992	struct my_biggy big = (struct my_biggy *	1415	struct my_biggy big = (struct my_biggy *)
993	(((char *)w) - offsetof (struct my_biggy, t2));	1416	(((char *)w) - offsetof (struct my_biggy, t2));
994	}	1417	}
		1418
		1419	=head2 WATCHER STATES
		1420
		1421	There are various watcher states mentioned throughout this manual -
		1422	active, pending and so on. In this section these states and the rules to
		1423	transition between them will be described in more detail - and while these
		1424	rules might look complicated, they usually do "the right thing".
		1425
		1426	=over 4
		1427
		1428	=item initialiased
		1429
		1430	Before a watcher can be registered with the event looop it has to be
		1431	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to
		1432	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.
		1433
		1434	In this state it is simply some block of memory that is suitable for use
		1435	in an event loop. It can be moved around, freed, reused etc. at will.
		1436
		1437	=item started/running/active
		1438
		1439	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes
		1440	property of the event loop, and is actively waiting for events. While in
		1441	this state it cannot be accessed (except in a few documented ways), moved,
		1442	freed or anything else - the only legal thing is to keep a pointer to it,
		1443	and call libev functions on it that are documented to work on active watchers.
		1444
		1445	=item pending
		1446
		1447	If a watcher is active and libev determines that an event it is interested
		1448	in has occurred (such as a timer expiring), it will become pending. It will
		1449	stay in this pending state until either it is stopped or its callback is
		1450	about to be invoked, so it is not normally pending inside the watcher
		1451	callback.
		1452
		1453	The watcher might or might not be active while it is pending (for example,
		1454	an expired non-repeating timer can be pending but no longer active). If it
		1455	is stopped, it can be freely accessed (e.g. by calling C<ev_TYPE_set>),
		1456	but it is still property of the event loop at this time, so cannot be
		1457	moved, freed or reused. And if it is active the rules described in the
		1458	previous item still apply.
		1459
		1460	It is also possible to feed an event on a watcher that is not active (e.g.
		1461	via C<ev_feed_event>), in which case it becomes pending without being
		1462	active.
		1463
		1464	=item stopped
		1465
		1466	A watcher can be stopped implicitly by libev (in which case it might still
		1467	be pending), or explicitly by calling its C<ev_TYPE_stop> function. The
		1468	latter will clear any pending state the watcher might be in, regardless
		1469	of whether it was active or not, so stopping a watcher explicitly before
		1470	freeing it is often a good idea.
		1471
		1472	While stopped (and not pending) the watcher is essentially in the
		1473	initialised state, that is it can be reused, moved, modified in any way
		1474	you wish.
		1475
		1476	=back
		1477
		1478	=head2 WATCHER PRIORITY MODELS
		1479
		1480	Many event loops support I<watcher priorities>, which are usually small
		1481	integers that influence the ordering of event callback invocation
		1482	between watchers in some way, all else being equal.
		1483
		1484	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1485	description for the more technical details such as the actual priority
		1486	range.
		1487
		1488	There are two common ways how these these priorities are being interpreted
		1489	by event loops:
		1490
		1491	In the more common lock-out model, higher priorities "lock out" invocation
		1492	of lower priority watchers, which means as long as higher priority
		1493	watchers receive events, lower priority watchers are not being invoked.
		1494
		1495	The less common only-for-ordering model uses priorities solely to order
		1496	callback invocation within a single event loop iteration: Higher priority
		1497	watchers are invoked before lower priority ones, but they all get invoked
		1498	before polling for new events.
		1499
		1500	Libev uses the second (only-for-ordering) model for all its watchers
		1501	except for idle watchers (which use the lock-out model).
		1502
		1503	The rationale behind this is that implementing the lock-out model for
		1504	watchers is not well supported by most kernel interfaces, and most event
		1505	libraries will just poll for the same events again and again as long as
		1506	their callbacks have not been executed, which is very inefficient in the
		1507	common case of one high-priority watcher locking out a mass of lower
		1508	priority ones.
		1509
		1510	Static (ordering) priorities are most useful when you have two or more
		1511	watchers handling the same resource: a typical usage example is having an
		1512	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1513	timeouts. Under load, data might be received while the program handles
		1514	other jobs, but since timers normally get invoked first, the timeout
		1515	handler will be executed before checking for data. In that case, giving
		1516	the timer a lower priority than the I/O watcher ensures that I/O will be
		1517	handled first even under adverse conditions (which is usually, but not
		1518	always, what you want).
		1519
		1520	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1521	will only be executed when no same or higher priority watchers have
		1522	received events, they can be used to implement the "lock-out" model when
		1523	required.
		1524
		1525	For example, to emulate how many other event libraries handle priorities,
		1526	you can associate an C<ev_idle> watcher to each such watcher, and in
		1527	the normal watcher callback, you just start the idle watcher. The real
		1528	processing is done in the idle watcher callback. This causes libev to
		1529	continuously poll and process kernel event data for the watcher, but when
		1530	the lock-out case is known to be rare (which in turn is rare :), this is
		1531	workable.
		1532
		1533	Usually, however, the lock-out model implemented that way will perform
		1534	miserably under the type of load it was designed to handle. In that case,
		1535	it might be preferable to stop the real watcher before starting the
		1536	idle watcher, so the kernel will not have to process the event in case
		1537	the actual processing will be delayed for considerable time.
		1538
		1539	Here is an example of an I/O watcher that should run at a strictly lower
		1540	priority than the default, and which should only process data when no
		1541	other events are pending:
		1542
		1543	ev_idle idle; // actual processing watcher
		1544	ev_io io; // actual event watcher
		1545
		1546	static void
		1547	io_cb (EV_P_ ev_io *w, int revents)
		1548	{
		1549	// stop the I/O watcher, we received the event, but
		1550	// are not yet ready to handle it.
		1551	ev_io_stop (EV_A_ w);
		1552
		1553	// start the idle watcher to handle the actual event.
		1554	// it will not be executed as long as other watchers
		1555	// with the default priority are receiving events.
		1556	ev_idle_start (EV_A_ &idle);
		1557	}
		1558
		1559	static void
		1560	idle_cb (EV_P_ ev_idle *w, int revents)
		1561	{
		1562	// actual processing
		1563	read (STDIN_FILENO, ...);
		1564
		1565	// have to start the I/O watcher again, as
		1566	// we have handled the event
		1567	ev_io_start (EV_P_ &io);
		1568	}
		1569
		1570	// initialisation
		1571	ev_idle_init (&idle, idle_cb);
		1572	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1573	ev_io_start (EV_DEFAULT_ &io);
		1574
		1575	In the "real" world, it might also be beneficial to start a timer, so that
		1576	low-priority connections can not be locked out forever under load. This
		1577	enables your program to keep a lower latency for important connections
		1578	during short periods of high load, while not completely locking out less
		1579	important ones.
995		1580
996		1581
997	=head1 WATCHER TYPES	1582	=head1 WATCHER TYPES
998		1583
999	This section describes each watcher in detail, but will not repeat	1584	This section describes each watcher in detail, but will not repeat
…		…
1023	In general you can register as many read and/or write event watchers per	1608	In general you can register as many read and/or write event watchers per
1024	fd as you want (as long as you don't confuse yourself). Setting all file	1609	fd as you want (as long as you don't confuse yourself). Setting all file
1025	descriptors to non-blocking mode is also usually a good idea (but not	1610	descriptors to non-blocking mode is also usually a good idea (but not
1026	required if you know what you are doing).	1611	required if you know what you are doing).
1027		1612
1028	If you must do this, then force the use of a known-to-be-good backend	1613	If you cannot use non-blocking mode, then force the use of a
1029	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1614	known-to-be-good backend (at the time of this writing, this includes only
1030	C<EVBACKEND_POLL>).	1615	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
		1616	descriptors for which non-blocking operation makes no sense (such as
		1617	files) - libev doesn't guarantee any specific behaviour in that case.
1031		1618
1032	Another thing you have to watch out for is that it is quite easy to	1619	Another thing you have to watch out for is that it is quite easy to
1033	receive "spurious" readyness notifications, that is your callback might	1620	receive "spurious" readiness notifications, that is your callback might
1034	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1621	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
1035	because there is no data. Not only are some backends known to create a	1622	because there is no data. Not only are some backends known to create a
1036	lot of those (for example solaris ports), it is very easy to get into	1623	lot of those (for example Solaris ports), it is very easy to get into
1037	this situation even with a relatively standard program structure. Thus	1624	this situation even with a relatively standard program structure. Thus
1038	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1625	it is best to always use non-blocking I/O: An extra C<read>(2) returning
1039	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1626	C<EAGAIN> is far preferable to a program hanging until some data arrives.
1040		1627
1041	If you cannot run the fd in non-blocking mode (for example you should not	1628	If you cannot run the fd in non-blocking mode (for example you should
1042	play around with an Xlib connection), then you have to seperately re-test	1629	not play around with an Xlib connection), then you have to separately
1043	whether a file descriptor is really ready with a known-to-be good interface	1630	re-test whether a file descriptor is really ready with a known-to-be good
1044	such as poll (fortunately in our Xlib example, Xlib already does this on	1631	interface such as poll (fortunately in our Xlib example, Xlib already
1045	its own, so its quite safe to use).	1632	does this on its own, so its quite safe to use). Some people additionally
		1633	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1634	indefinitely.
		1635
		1636	But really, best use non-blocking mode.
1046		1637
1047	=head3 The special problem of disappearing file descriptors	1638	=head3 The special problem of disappearing file descriptors
1048		1639
1049	Some backends (e.g. kqueue, epoll) need to be told about closing a file	1640	Some backends (e.g. kqueue, epoll) need to be told about closing a file
1050	descriptor (either by calling C<close> explicitly or by any other means,	1641	descriptor (either due to calling C<close> explicitly or any other means,
1051	such as C<dup>). The reason is that you register interest in some file	1642	such as C<dup2>). The reason is that you register interest in some file
1052	descriptor, but when it goes away, the operating system will silently drop	1643	descriptor, but when it goes away, the operating system will silently drop
1053	this interest. If another file descriptor with the same number then is	1644	this interest. If another file descriptor with the same number then is
1054	registered with libev, there is no efficient way to see that this is, in	1645	registered with libev, there is no efficient way to see that this is, in
1055	fact, a different file descriptor.	1646	fact, a different file descriptor.
1056		1647
…		…
1085	To support fork in your programs, you either have to call	1676	To support fork in your programs, you either have to call
1086	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,	1677	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
1087	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1678	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
1088	C<EVBACKEND_POLL>.	1679	C<EVBACKEND_POLL>.
1089		1680
		1681	=head3 The special problem of SIGPIPE
		1682
		1683	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1684	when writing to a pipe whose other end has been closed, your program gets
		1685	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1686	this is sensible behaviour, for daemons, this is usually undesirable.
		1687
		1688	So when you encounter spurious, unexplained daemon exits, make sure you
		1689	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1690	somewhere, as that would have given you a big clue).
		1691
		1692	=head3 The special problem of accept()ing when you can't
		1693
		1694	Many implementations of the POSIX C<accept> function (for example,
		1695	found in post-2004 Linux) have the peculiar behaviour of not removing a
		1696	connection from the pending queue in all error cases.
		1697
		1698	For example, larger servers often run out of file descriptors (because
		1699	of resource limits), causing C<accept> to fail with C<ENFILE> but not
		1700	rejecting the connection, leading to libev signalling readiness on
		1701	the next iteration again (the connection still exists after all), and
		1702	typically causing the program to loop at 100% CPU usage.
		1703
		1704	Unfortunately, the set of errors that cause this issue differs between
		1705	operating systems, there is usually little the app can do to remedy the
		1706	situation, and no known thread-safe method of removing the connection to
		1707	cope with overload is known (to me).
		1708
		1709	One of the easiest ways to handle this situation is to just ignore it
		1710	- when the program encounters an overload, it will just loop until the
		1711	situation is over. While this is a form of busy waiting, no OS offers an
		1712	event-based way to handle this situation, so it's the best one can do.
		1713
		1714	A better way to handle the situation is to log any errors other than
		1715	C<EAGAIN> and C<EWOULDBLOCK>, making sure not to flood the log with such
		1716	messages, and continue as usual, which at least gives the user an idea of
		1717	what could be wrong ("raise the ulimit!"). For extra points one could stop
		1718	the C<ev_io> watcher on the listening fd "for a while", which reduces CPU
		1719	usage.
		1720
		1721	If your program is single-threaded, then you could also keep a dummy file
		1722	descriptor for overload situations (e.g. by opening F</dev/null>), and
		1723	when you run into C<ENFILE> or C<EMFILE>, close it, run C<accept>,
		1724	close that fd, and create a new dummy fd. This will gracefully refuse
		1725	clients under typical overload conditions.
		1726
		1727	The last way to handle it is to simply log the error and C<exit>, as
		1728	is often done with C<malloc> failures, but this results in an easy
		1729	opportunity for a DoS attack.
1090		1730
1091	=head3 Watcher-Specific Functions	1731	=head3 Watcher-Specific Functions
1092		1732
1093	=over 4	1733	=over 4
1094		1734
1095	=item ev_io_init (ev_io *, callback, int fd, int events)	1735	=item ev_io_init (ev_io *, callback, int fd, int events)
1096		1736
1097	=item ev_io_set (ev_io *, int fd, int events)	1737	=item ev_io_set (ev_io *, int fd, int events)
1098		1738
1099	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1739	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
1100	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or	1740	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
1101	C<EV_READ | EV_WRITE> to receive the given events.	1741	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
1102		1742
1103	=item int fd [read-only]	1743	=item int fd [read-only]
1104		1744
1105	The file descriptor being watched.	1745	The file descriptor being watched.
1106		1746
…		…
1114		1754
1115	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1755	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
1116	readable, but only once. Since it is likely line-buffered, you could	1756	readable, but only once. Since it is likely line-buffered, you could
1117	attempt to read a whole line in the callback.	1757	attempt to read a whole line in the callback.
1118		1758
1119	static void	1759	static void
1120	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1760	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
1121	{	1761	{
1122	ev_io_stop (loop, w);	1762	ev_io_stop (loop, w);
1123	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1763	.. read from stdin here (or from w->fd) and handle any I/O errors
1124	}	1764	}
1125		1765
1126	...	1766	...
1127	struct ev_loop *loop = ev_default_init (0);	1767	struct ev_loop *loop = ev_default_init (0);
1128	struct ev_io stdin_readable;	1768	ev_io stdin_readable;
1129	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1769	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1130	ev_io_start (loop, &stdin_readable);	1770	ev_io_start (loop, &stdin_readable);
1131	ev_loop (loop, 0);	1771	ev_run (loop, 0);
1132		1772
1133		1773
1134	=head2 C<ev_timer> - relative and optionally repeating timeouts	1774	=head2 C<ev_timer> - relative and optionally repeating timeouts
1135		1775
1136	Timer watchers are simple relative timers that generate an event after a	1776	Timer watchers are simple relative timers that generate an event after a
1137	given time, and optionally repeating in regular intervals after that.	1777	given time, and optionally repeating in regular intervals after that.
1138		1778
1139	The timers are based on real time, that is, if you register an event that	1779	The timers are based on real time, that is, if you register an event that
1140	times out after an hour and you reset your system clock to last years	1780	times out after an hour and you reset your system clock to January last
1141	time, it will still time out after (roughly) and hour. "Roughly" because	1781	year, it will still time out after (roughly) one hour. "Roughly" because
1142	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1782	detecting time jumps is hard, and some inaccuracies are unavoidable (the
1143	monotonic clock option helps a lot here).	1783	monotonic clock option helps a lot here).
		1784
		1785	The callback is guaranteed to be invoked only I<after> its timeout has
		1786	passed (not I<at>, so on systems with very low-resolution clocks this
		1787	might introduce a small delay). If multiple timers become ready during the
		1788	same loop iteration then the ones with earlier time-out values are invoked
		1789	before ones of the same priority with later time-out values (but this is
		1790	no longer true when a callback calls C<ev_run> recursively).
		1791
		1792	=head3 Be smart about timeouts
		1793
		1794	Many real-world problems involve some kind of timeout, usually for error
		1795	recovery. A typical example is an HTTP request - if the other side hangs,
		1796	you want to raise some error after a while.
		1797
		1798	What follows are some ways to handle this problem, from obvious and
		1799	inefficient to smart and efficient.
		1800
		1801	In the following, a 60 second activity timeout is assumed - a timeout that
		1802	gets reset to 60 seconds each time there is activity (e.g. each time some
		1803	data or other life sign was received).
		1804
		1805	=over 4
		1806
		1807	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1808
		1809	This is the most obvious, but not the most simple way: In the beginning,
		1810	start the watcher:
		1811
		1812	ev_timer_init (timer, callback, 60., 0.);
		1813	ev_timer_start (loop, timer);
		1814
		1815	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1816	and start it again:
		1817
		1818	ev_timer_stop (loop, timer);
		1819	ev_timer_set (timer, 60., 0.);
		1820	ev_timer_start (loop, timer);
		1821
		1822	This is relatively simple to implement, but means that each time there is
		1823	some activity, libev will first have to remove the timer from its internal
		1824	data structure and then add it again. Libev tries to be fast, but it's
		1825	still not a constant-time operation.
		1826
		1827	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1828
		1829	This is the easiest way, and involves using C<ev_timer_again> instead of
		1830	C<ev_timer_start>.
		1831
		1832	To implement this, configure an C<ev_timer> with a C<repeat> value
		1833	of C<60> and then call C<ev_timer_again> at start and each time you
		1834	successfully read or write some data. If you go into an idle state where
		1835	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1836	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1837
		1838	That means you can ignore both the C<ev_timer_start> function and the
		1839	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1840	member and C<ev_timer_again>.
		1841
		1842	At start:
		1843
		1844	ev_init (timer, callback);
		1845	timer->repeat = 60.;
		1846	ev_timer_again (loop, timer);
		1847
		1848	Each time there is some activity:
		1849
		1850	ev_timer_again (loop, timer);
		1851
		1852	It is even possible to change the time-out on the fly, regardless of
		1853	whether the watcher is active or not:
		1854
		1855	timer->repeat = 30.;
		1856	ev_timer_again (loop, timer);
		1857
		1858	This is slightly more efficient then stopping/starting the timer each time
		1859	you want to modify its timeout value, as libev does not have to completely
		1860	remove and re-insert the timer from/into its internal data structure.
		1861
		1862	It is, however, even simpler than the "obvious" way to do it.
		1863
		1864	=item 3. Let the timer time out, but then re-arm it as required.
		1865
		1866	This method is more tricky, but usually most efficient: Most timeouts are
		1867	relatively long compared to the intervals between other activity - in
		1868	our example, within 60 seconds, there are usually many I/O events with
		1869	associated activity resets.
		1870
		1871	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1872	but remember the time of last activity, and check for a real timeout only
		1873	within the callback:
		1874
		1875	ev_tstamp last_activity; // time of last activity
		1876
		1877	static void
		1878	callback (EV_P_ ev_timer *w, int revents)
		1879	{
		1880	ev_tstamp now = ev_now (EV_A);
		1881	ev_tstamp timeout = last_activity + 60.;
		1882
		1883	// if last_activity + 60. is older than now, we did time out
		1884	if (timeout < now)
		1885	{
		1886	// timeout occurred, take action
		1887	}
		1888	else
		1889	{
		1890	// callback was invoked, but there was some activity, re-arm
		1891	// the watcher to fire in last_activity + 60, which is
		1892	// guaranteed to be in the future, so "again" is positive:
		1893	w->repeat = timeout - now;
		1894	ev_timer_again (EV_A_ w);
		1895	}
		1896	}
		1897
		1898	To summarise the callback: first calculate the real timeout (defined
		1899	as "60 seconds after the last activity"), then check if that time has
		1900	been reached, which means something I<did>, in fact, time out. Otherwise
		1901	the callback was invoked too early (C<timeout> is in the future), so
		1902	re-schedule the timer to fire at that future time, to see if maybe we have
		1903	a timeout then.
		1904
		1905	Note how C<ev_timer_again> is used, taking advantage of the
		1906	C<ev_timer_again> optimisation when the timer is already running.
		1907
		1908	This scheme causes more callback invocations (about one every 60 seconds
		1909	minus half the average time between activity), but virtually no calls to
		1910	libev to change the timeout.
		1911
		1912	To start the timer, simply initialise the watcher and set C<last_activity>
		1913	to the current time (meaning we just have some activity :), then call the
		1914	callback, which will "do the right thing" and start the timer:
		1915
		1916	ev_init (timer, callback);
		1917	last_activity = ev_now (loop);
		1918	callback (loop, timer, EV_TIMER);
		1919
		1920	And when there is some activity, simply store the current time in
		1921	C<last_activity>, no libev calls at all:
		1922
		1923	last_activity = ev_now (loop);
		1924
		1925	This technique is slightly more complex, but in most cases where the
		1926	time-out is unlikely to be triggered, much more efficient.
		1927
		1928	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1929	callback :) - just change the timeout and invoke the callback, which will
		1930	fix things for you.
		1931
		1932	=item 4. Wee, just use a double-linked list for your timeouts.
		1933
		1934	If there is not one request, but many thousands (millions...), all
		1935	employing some kind of timeout with the same timeout value, then one can
		1936	do even better:
		1937
		1938	When starting the timeout, calculate the timeout value and put the timeout
		1939	at the I<end> of the list.
		1940
		1941	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1942	the list is expected to fire (for example, using the technique #3).
		1943
		1944	When there is some activity, remove the timer from the list, recalculate
		1945	the timeout, append it to the end of the list again, and make sure to
		1946	update the C<ev_timer> if it was taken from the beginning of the list.
		1947
		1948	This way, one can manage an unlimited number of timeouts in O(1) time for
		1949	starting, stopping and updating the timers, at the expense of a major
		1950	complication, and having to use a constant timeout. The constant timeout
		1951	ensures that the list stays sorted.
		1952
		1953	=back
		1954
		1955	So which method the best?
		1956
		1957	Method #2 is a simple no-brain-required solution that is adequate in most
		1958	situations. Method #3 requires a bit more thinking, but handles many cases
		1959	better, and isn't very complicated either. In most case, choosing either
		1960	one is fine, with #3 being better in typical situations.
		1961
		1962	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1963	rather complicated, but extremely efficient, something that really pays
		1964	off after the first million or so of active timers, i.e. it's usually
		1965	overkill :)
		1966
		1967	=head3 The special problem of time updates
		1968
		1969	Establishing the current time is a costly operation (it usually takes at
		1970	least two system calls): EV therefore updates its idea of the current
		1971	time only before and after C<ev_run> collects new events, which causes a
		1972	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1973	lots of events in one iteration.
1144		1974
1145	The relative timeouts are calculated relative to the C<ev_now ()>	1975	The relative timeouts are calculated relative to the C<ev_now ()>
1146	time. This is usually the right thing as this timestamp refers to the time	1976	time. This is usually the right thing as this timestamp refers to the time
1147	of the event triggering whatever timeout you are modifying/starting. If	1977	of the event triggering whatever timeout you are modifying/starting. If
1148	you suspect event processing to be delayed and you I<need> to base the timeout	1978	you suspect event processing to be delayed and you I<need> to base the
1149	on the current time, use something like this to adjust for this:	1979	timeout on the current time, use something like this to adjust for this:
1150		1980
1151	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1981	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
1152		1982
1153	The callback is guarenteed to be invoked only when its timeout has passed,	1983	If the event loop is suspended for a long time, you can also force an
1154	but if multiple timers become ready during the same loop iteration then	1984	update of the time returned by C<ev_now ()> by calling C<ev_now_update
1155	order of execution is undefined.	1985	()>.
		1986
		1987	=head3 The special problems of suspended animation
		1988
		1989	When you leave the server world it is quite customary to hit machines that
		1990	can suspend/hibernate - what happens to the clocks during such a suspend?
		1991
		1992	Some quick tests made with a Linux 2.6.28 indicate that a suspend freezes
		1993	all processes, while the clocks (C<times>, C<CLOCK_MONOTONIC>) continue
		1994	to run until the system is suspended, but they will not advance while the
		1995	system is suspended. That means, on resume, it will be as if the program
		1996	was frozen for a few seconds, but the suspend time will not be counted
		1997	towards C<ev_timer> when a monotonic clock source is used. The real time
		1998	clock advanced as expected, but if it is used as sole clocksource, then a
		1999	long suspend would be detected as a time jump by libev, and timers would
		2000	be adjusted accordingly.
		2001
		2002	I would not be surprised to see different behaviour in different between
		2003	operating systems, OS versions or even different hardware.
		2004
		2005	The other form of suspend (job control, or sending a SIGSTOP) will see a
		2006	time jump in the monotonic clocks and the realtime clock. If the program
		2007	is suspended for a very long time, and monotonic clock sources are in use,
		2008	then you can expect C<ev_timer>s to expire as the full suspension time
		2009	will be counted towards the timers. When no monotonic clock source is in
		2010	use, then libev will again assume a timejump and adjust accordingly.
		2011
		2012	It might be beneficial for this latter case to call C<ev_suspend>
		2013	and C<ev_resume> in code that handles C<SIGTSTP>, to at least get
		2014	deterministic behaviour in this case (you can do nothing against
		2015	C<SIGSTOP>).
1156		2016
1157	=head3 Watcher-Specific Functions and Data Members	2017	=head3 Watcher-Specific Functions and Data Members
1158		2018
1159	=over 4	2019	=over 4
1160		2020
1161	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	2021	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
1162		2022
1163	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	2023	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
1164		2024
1165	Configure the timer to trigger after C<after> seconds. If C<repeat> is	2025	Configure the timer to trigger after C<after> seconds. If C<repeat>
1166	C<0.>, then it will automatically be stopped. If it is positive, then the	2026	is C<0.>, then it will automatically be stopped once the timeout is
1167	timer will automatically be configured to trigger again C<repeat> seconds	2027	reached. If it is positive, then the timer will automatically be
1168	later, again, and again, until stopped manually.	2028	configured to trigger again C<repeat> seconds later, again, and again,
		2029	until stopped manually.
1169		2030
1170	The timer itself will do a best-effort at avoiding drift, that is, if you	2031	The timer itself will do a best-effort at avoiding drift, that is, if
1171	configure a timer to trigger every 10 seconds, then it will trigger at	2032	you configure a timer to trigger every 10 seconds, then it will normally
1172	exactly 10 second intervals. If, however, your program cannot keep up with	2033	trigger at exactly 10 second intervals. If, however, your program cannot
1173	the timer (because it takes longer than those 10 seconds to do stuff) the	2034	keep up with the timer (because it takes longer than those 10 seconds to
1174	timer will not fire more than once per event loop iteration.	2035	do stuff) the timer will not fire more than once per event loop iteration.
1175		2036
1176	=item ev_timer_again (loop, ev_timer *)	2037	=item ev_timer_again (loop, ev_timer *)
1177		2038
1178	This will act as if the timer timed out and restart it again if it is	2039	This will act as if the timer timed out and restart it again if it is
1179	repeating. The exact semantics are:	2040	repeating. The exact semantics are:
1180		2041
1181	If the timer is pending, its pending status is cleared.	2042	If the timer is pending, its pending status is cleared.
1182		2043
1183	If the timer is started but nonrepeating, stop it (as if it timed out).	2044	If the timer is started but non-repeating, stop it (as if it timed out).
1184		2045
1185	If the timer is repeating, either start it if necessary (with the	2046	If the timer is repeating, either start it if necessary (with the
1186	C<repeat> value), or reset the running timer to the C<repeat> value.	2047	C<repeat> value), or reset the running timer to the C<repeat> value.
1187		2048
1188	This sounds a bit complicated, but here is a useful and typical	2049	This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
1189	example: Imagine you have a tcp connection and you want a so-called idle	2050	usage example.
1190	timeout, that is, you want to be called when there have been, say, 60
1191	seconds of inactivity on the socket. The easiest way to do this is to
1192	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
1193	C<ev_timer_again> each time you successfully read or write some data. If
1194	you go into an idle state where you do not expect data to travel on the
1195	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
1196	automatically restart it if need be.
1197		2051
1198	That means you can ignore the C<after> value and C<ev_timer_start>	2052	=item ev_tstamp ev_timer_remaining (loop, ev_timer *)
1199	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
1200		2053
1201	ev_timer_init (timer, callback, 0., 5.);	2054	Returns the remaining time until a timer fires. If the timer is active,
1202	ev_timer_again (loop, timer);	2055	then this time is relative to the current event loop time, otherwise it's
1203	...	2056	the timeout value currently configured.
1204	timer->again = 17.;
1205	ev_timer_again (loop, timer);
1206	...
1207	timer->again = 10.;
1208	ev_timer_again (loop, timer);
1209		2057
1210	This is more slightly efficient then stopping/starting the timer each time	2058	That is, after an C<ev_timer_set (w, 5, 7)>, C<ev_timer_remaining> returns
1211	you want to modify its timeout value.	2059	C<5>. When the timer is started and one second passes, C<ev_timer_remaining>
		2060	will return C<4>. When the timer expires and is restarted, it will return
		2061	roughly C<7> (likely slightly less as callback invocation takes some time,
		2062	too), and so on.
1212		2063
1213	=item ev_tstamp repeat [read-write]	2064	=item ev_tstamp repeat [read-write]
1214		2065
1215	The current C<repeat> value. Will be used each time the watcher times out	2066	The current C<repeat> value. Will be used each time the watcher times out
1216	or C<ev_timer_again> is called and determines the next timeout (if any),	2067	or C<ev_timer_again> is called, and determines the next timeout (if any),
1217	which is also when any modifications are taken into account.	2068	which is also when any modifications are taken into account.
1218		2069
1219	=back	2070	=back
1220		2071
1221	=head3 Examples	2072	=head3 Examples
1222		2073
1223	Example: Create a timer that fires after 60 seconds.	2074	Example: Create a timer that fires after 60 seconds.
1224		2075
1225	static void	2076	static void
1226	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2077	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
1227	{	2078	{
1228	.. one minute over, w is actually stopped right here	2079	.. one minute over, w is actually stopped right here
1229	}	2080	}
1230		2081
1231	struct ev_timer mytimer;	2082	ev_timer mytimer;
1232	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	2083	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1233	ev_timer_start (loop, &mytimer);	2084	ev_timer_start (loop, &mytimer);
1234		2085
1235	Example: Create a timeout timer that times out after 10 seconds of	2086	Example: Create a timeout timer that times out after 10 seconds of
1236	inactivity.	2087	inactivity.
1237		2088
1238	static void	2089	static void
1239	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2090	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
1240	{	2091	{
1241	.. ten seconds without any activity	2092	.. ten seconds without any activity
1242	}	2093	}
1243		2094
1244	struct ev_timer mytimer;	2095	ev_timer mytimer;
1245	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	2096	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1246	ev_timer_again (&mytimer); /* start timer */	2097	ev_timer_again (&mytimer); /* start timer */
1247	ev_loop (loop, 0);	2098	ev_run (loop, 0);
1248		2099
1249	// and in some piece of code that gets executed on any "activity":	2100	// and in some piece of code that gets executed on any "activity":
1250	// reset the timeout to start ticking again at 10 seconds	2101	// reset the timeout to start ticking again at 10 seconds
1251	ev_timer_again (&mytimer);	2102	ev_timer_again (&mytimer);
1252		2103
1253		2104
1254	=head2 C<ev_periodic> - to cron or not to cron?	2105	=head2 C<ev_periodic> - to cron or not to cron?
1255		2106
1256	Periodic watchers are also timers of a kind, but they are very versatile	2107	Periodic watchers are also timers of a kind, but they are very versatile
1257	(and unfortunately a bit complex).	2108	(and unfortunately a bit complex).
1258		2109
1259	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	2110	Unlike C<ev_timer>, periodic watchers are not based on real time (or
1260	but on wallclock time (absolute time). You can tell a periodic watcher	2111	relative time, the physical time that passes) but on wall clock time
1261	to trigger "at" some specific point in time. For example, if you tell a	2112	(absolute time, the thing you can read on your calender or clock). The
1262	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	2113	difference is that wall clock time can run faster or slower than real
1263	+ 10.>) and then reset your system clock to the last year, then it will	2114	time, and time jumps are not uncommon (e.g. when you adjust your
1264	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	2115	wrist-watch).
1265	roughly 10 seconds later).
1266		2116
1267	They can also be used to implement vastly more complex timers, such as	2117	You can tell a periodic watcher to trigger after some specific point
1268	triggering an event on each midnight, local time or other, complicated,	2118	in time: for example, if you tell a periodic watcher to trigger "in 10
1269	rules.	2119	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		2120	not a delay) and then reset your system clock to January of the previous
		2121	year, then it will take a year or more to trigger the event (unlike an
		2122	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		2123	it, as it uses a relative timeout).
1270		2124
		2125	C<ev_periodic> watchers can also be used to implement vastly more complex
		2126	timers, such as triggering an event on each "midnight, local time", or
		2127	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		2128	those cannot react to time jumps.
		2129
1271	As with timers, the callback is guarenteed to be invoked only when the	2130	As with timers, the callback is guaranteed to be invoked only when the
1272	time (C<at>) has been passed, but if multiple periodic timers become ready	2131	point in time where it is supposed to trigger has passed. If multiple
1273	during the same loop iteration then order of execution is undefined.	2132	timers become ready during the same loop iteration then the ones with
		2133	earlier time-out values are invoked before ones with later time-out values
		2134	(but this is no longer true when a callback calls C<ev_run> recursively).
1274		2135
1275	=head3 Watcher-Specific Functions and Data Members	2136	=head3 Watcher-Specific Functions and Data Members
1276		2137
1277	=over 4	2138	=over 4
1278		2139
1279	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	2140	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1280		2141
1281	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	2142	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1282		2143
1283	Lots of arguments, lets sort it out... There are basically three modes of	2144	Lots of arguments, let's sort it out... There are basically three modes of
1284	operation, and we will explain them from simplest to complex:	2145	operation, and we will explain them from simplest to most complex:
1285		2146
1286	=over 4	2147	=over 4
1287		2148
1288	=item * absolute timer (at = time, interval = reschedule_cb = 0)	2149	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
1289		2150
1290	In this configuration the watcher triggers an event at the wallclock time	2151	In this configuration the watcher triggers an event after the wall clock
1291	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	2152	time C<offset> has passed. It will not repeat and will not adjust when a
1292	that is, if it is to be run at January 1st 2011 then it will run when the	2153	time jump occurs, that is, if it is to be run at January 1st 2011 then it
1293	system time reaches or surpasses this time.	2154	will be stopped and invoked when the system clock reaches or surpasses
		2155	this point in time.
1294		2156
1295	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	2157	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
1296		2158
1297	In this mode the watcher will always be scheduled to time out at the next	2159	In this mode the watcher will always be scheduled to time out at the next
1298	C<at + N * interval> time (for some integer N, which can also be negative)	2160	C<offset + N * interval> time (for some integer N, which can also be
1299	and then repeat, regardless of any time jumps.	2161	negative) and then repeat, regardless of any time jumps. The C<offset>
		2162	argument is merely an offset into the C<interval> periods.
1300		2163
1301	This can be used to create timers that do not drift with respect to system	2164	This can be used to create timers that do not drift with respect to the
1302	time:	2165	system clock, for example, here is an C<ev_periodic> that triggers each
		2166	hour, on the hour (with respect to UTC):
1303		2167
1304	ev_periodic_set (&periodic, 0., 3600., 0);	2168	ev_periodic_set (&periodic, 0., 3600., 0);
1305		2169
1306	This doesn't mean there will always be 3600 seconds in between triggers,	2170	This doesn't mean there will always be 3600 seconds in between triggers,
1307	but only that the the callback will be called when the system time shows a	2171	but only that the callback will be called when the system time shows a
1308	full hour (UTC), or more correctly, when the system time is evenly divisible	2172	full hour (UTC), or more correctly, when the system time is evenly divisible
1309	by 3600.	2173	by 3600.
1310		2174
1311	Another way to think about it (for the mathematically inclined) is that	2175	Another way to think about it (for the mathematically inclined) is that
1312	C<ev_periodic> will try to run the callback in this mode at the next possible	2176	C<ev_periodic> will try to run the callback in this mode at the next possible
1313	time where C<time = at (mod interval)>, regardless of any time jumps.	2177	time where C<time = offset (mod interval)>, regardless of any time jumps.
1314		2178
1315	For numerical stability it is preferable that the C<at> value is near	2179	For numerical stability it is preferable that the C<offset> value is near
1316	C<ev_now ()> (the current time), but there is no range requirement for	2180	C<ev_now ()> (the current time), but there is no range requirement for
1317	this value.	2181	this value, and in fact is often specified as zero.
1318		2182
		2183	Note also that there is an upper limit to how often a timer can fire (CPU
		2184	speed for example), so if C<interval> is very small then timing stability
		2185	will of course deteriorate. Libev itself tries to be exact to be about one
		2186	millisecond (if the OS supports it and the machine is fast enough).
		2187
1319	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)	2188	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
1320		2189
1321	In this mode the values for C<interval> and C<at> are both being	2190	In this mode the values for C<interval> and C<offset> are both being
1322	ignored. Instead, each time the periodic watcher gets scheduled, the	2191	ignored. Instead, each time the periodic watcher gets scheduled, the
1323	reschedule callback will be called with the watcher as first, and the	2192	reschedule callback will be called with the watcher as first, and the
1324	current time as second argument.	2193	current time as second argument.
1325		2194
1326	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	2195	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
1327	ever, or make any event loop modifications>. If you need to stop it,	2196	or make ANY other event loop modifications whatsoever, unless explicitly
1328	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by	2197	allowed by documentation here>.
1329	starting an C<ev_prepare> watcher, which is legal).
1330		2198
		2199	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		2200	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		2201	only event loop modification you are allowed to do).
		2202
1331	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	2203	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
1332	ev_tstamp now)>, e.g.:	2204	*w, ev_tstamp now)>, e.g.:
1333		2205
		2206	static ev_tstamp
1334	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	2207	my_rescheduler (ev_periodic *w, ev_tstamp now)
1335	{	2208	{
1336	return now + 60.;	2209	return now + 60.;
1337	}	2210	}
1338		2211
1339	It must return the next time to trigger, based on the passed time value	2212	It must return the next time to trigger, based on the passed time value
1340	(that is, the lowest time value larger than to the second argument). It	2213	(that is, the lowest time value larger than to the second argument). It
1341	will usually be called just before the callback will be triggered, but	2214	will usually be called just before the callback will be triggered, but
1342	might be called at other times, too.	2215	might be called at other times, too.
1343		2216
1344	NOTE: I<< This callback must always return a time that is later than the	2217	NOTE: I<< This callback must always return a time that is higher than or
1345	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	2218	equal to the passed C<now> value >>.
1346		2219
1347	This can be used to create very complex timers, such as a timer that	2220	This can be used to create very complex timers, such as a timer that
1348	triggers on each midnight, local time. To do this, you would calculate the	2221	triggers on "next midnight, local time". To do this, you would calculate the
1349	next midnight after C<now> and return the timestamp value for this. How	2222	next midnight after C<now> and return the timestamp value for this. How
1350	you do this is, again, up to you (but it is not trivial, which is the main	2223	you do this is, again, up to you (but it is not trivial, which is the main
1351	reason I omitted it as an example).	2224	reason I omitted it as an example).
1352		2225
1353	=back	2226	=back
…		…
1357	Simply stops and restarts the periodic watcher again. This is only useful	2230	Simply stops and restarts the periodic watcher again. This is only useful
1358	when you changed some parameters or the reschedule callback would return	2231	when you changed some parameters or the reschedule callback would return
1359	a different time than the last time it was called (e.g. in a crond like	2232	a different time than the last time it was called (e.g. in a crond like
1360	program when the crontabs have changed).	2233	program when the crontabs have changed).
1361		2234
		2235	=item ev_tstamp ev_periodic_at (ev_periodic *)
		2236
		2237	When active, returns the absolute time that the watcher is supposed
		2238	to trigger next. This is not the same as the C<offset> argument to
		2239	C<ev_periodic_set>, but indeed works even in interval and manual
		2240	rescheduling modes.
		2241
1362	=item ev_tstamp offset [read-write]	2242	=item ev_tstamp offset [read-write]
1363		2243
1364	When repeating, this contains the offset value, otherwise this is the	2244	When repeating, this contains the offset value, otherwise this is the
1365	absolute point in time (the C<at> value passed to C<ev_periodic_set>).	2245	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		2246	although libev might modify this value for better numerical stability).
1366		2247
1367	Can be modified any time, but changes only take effect when the periodic	2248	Can be modified any time, but changes only take effect when the periodic
1368	timer fires or C<ev_periodic_again> is being called.	2249	timer fires or C<ev_periodic_again> is being called.
1369		2250
1370	=item ev_tstamp interval [read-write]	2251	=item ev_tstamp interval [read-write]
1371		2252
1372	The current interval value. Can be modified any time, but changes only	2253	The current interval value. Can be modified any time, but changes only
1373	take effect when the periodic timer fires or C<ev_periodic_again> is being	2254	take effect when the periodic timer fires or C<ev_periodic_again> is being
1374	called.	2255	called.
1375		2256
1376	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]	2257	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
1377		2258
1378	The current reschedule callback, or C<0>, if this functionality is	2259	The current reschedule callback, or C<0>, if this functionality is
1379	switched off. Can be changed any time, but changes only take effect when	2260	switched off. Can be changed any time, but changes only take effect when
1380	the periodic timer fires or C<ev_periodic_again> is being called.	2261	the periodic timer fires or C<ev_periodic_again> is being called.
1381		2262
1382	=item ev_tstamp at [read-only]
1383
1384	When active, contains the absolute time that the watcher is supposed to
1385	trigger next.
1386
1387	=back	2263	=back
1388		2264
1389	=head3 Examples	2265	=head3 Examples
1390		2266
1391	Example: Call a callback every hour, or, more precisely, whenever the	2267	Example: Call a callback every hour, or, more precisely, whenever the
1392	system clock is divisible by 3600. The callback invocation times have	2268	system time is divisible by 3600. The callback invocation times have
1393	potentially a lot of jittering, but good long-term stability.	2269	potentially a lot of jitter, but good long-term stability.
1394		2270
1395	static void	2271	static void
1396	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	2272	clock_cb (struct ev_loop *loop, ev_periodic *w, int revents)
1397	{	2273	{
1398	... its now a full hour (UTC, or TAI or whatever your clock follows)	2274	... its now a full hour (UTC, or TAI or whatever your clock follows)
1399	}	2275	}
1400		2276
1401	struct ev_periodic hourly_tick;	2277	ev_periodic hourly_tick;
1402	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	2278	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1403	ev_periodic_start (loop, &hourly_tick);	2279	ev_periodic_start (loop, &hourly_tick);
1404		2280
1405	Example: The same as above, but use a reschedule callback to do it:	2281	Example: The same as above, but use a reschedule callback to do it:
1406		2282
1407	#include <math.h>	2283	#include <math.h>
1408		2284
1409	static ev_tstamp	2285	static ev_tstamp
1410	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	2286	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
1411	{	2287	{
1412	return fmod (now, 3600.) + 3600.;	2288	return now + (3600. - fmod (now, 3600.));
1413	}	2289	}
1414		2290
1415	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	2291	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1416		2292
1417	Example: Call a callback every hour, starting now:	2293	Example: Call a callback every hour, starting now:
1418		2294
1419	struct ev_periodic hourly_tick;	2295	ev_periodic hourly_tick;
1420	ev_periodic_init (&hourly_tick, clock_cb,	2296	ev_periodic_init (&hourly_tick, clock_cb,
1421	fmod (ev_now (loop), 3600.), 3600., 0);	2297	fmod (ev_now (loop), 3600.), 3600., 0);
1422	ev_periodic_start (loop, &hourly_tick);	2298	ev_periodic_start (loop, &hourly_tick);
1423		2299
1424		2300
1425	=head2 C<ev_signal> - signal me when a signal gets signalled!	2301	=head2 C<ev_signal> - signal me when a signal gets signalled!
1426		2302
1427	Signal watchers will trigger an event when the process receives a specific	2303	Signal watchers will trigger an event when the process receives a specific
1428	signal one or more times. Even though signals are very asynchronous, libev	2304	signal one or more times. Even though signals are very asynchronous, libev
1429	will try it's best to deliver signals synchronously, i.e. as part of the	2305	will try its best to deliver signals synchronously, i.e. as part of the
1430	normal event processing, like any other event.	2306	normal event processing, like any other event.
1431		2307
		2308	If you want signals to be delivered truly asynchronously, just use
		2309	C<sigaction> as you would do without libev and forget about sharing
		2310	the signal. You can even use C<ev_async> from a signal handler to
		2311	synchronously wake up an event loop.
		2312
1432	You can configure as many watchers as you like per signal. Only when the	2313	You can configure as many watchers as you like for the same signal, but
		2314	only within the same loop, i.e. you can watch for C<SIGINT> in your
		2315	default loop and for C<SIGIO> in another loop, but you cannot watch for
		2316	C<SIGINT> in both the default loop and another loop at the same time. At
		2317	the moment, C<SIGCHLD> is permanently tied to the default loop.
		2318
1433	first watcher gets started will libev actually register a signal watcher	2319	When the first watcher gets started will libev actually register something
1434	with the kernel (thus it coexists with your own signal handlers as long	2320	with the kernel (thus it coexists with your own signal handlers as long as
1435	as you don't register any with libev). Similarly, when the last signal	2321	you don't register any with libev for the same signal).
1436	watcher for a signal is stopped libev will reset the signal handler to
1437	SIG_DFL (regardless of what it was set to before).
1438		2322
1439	If possible and supported, libev will install its handlers with	2323	If possible and supported, libev will install its handlers with
1440	C<SA_RESTART> behaviour enabled, so syscalls should not be unduly	2324	C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should
1441	interrupted. If you have a problem with syscalls getting interrupted by	2325	not be unduly interrupted. If you have a problem with system calls getting
1442	signals you can block all signals in an C<ev_check> watcher and unblock	2326	interrupted by signals you can block all signals in an C<ev_check> watcher
1443	them in an C<ev_prepare> watcher.	2327	and unblock them in an C<ev_prepare> watcher.
		2328
		2329	=head3 The special problem of inheritance over fork/execve/pthread_create
		2330
		2331	Both the signal mask (C<sigprocmask>) and the signal disposition
		2332	(C<sigaction>) are unspecified after starting a signal watcher (and after
		2333	stopping it again), that is, libev might or might not block the signal,
		2334	and might or might not set or restore the installed signal handler.
		2335
		2336	While this does not matter for the signal disposition (libev never
		2337	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
		2338	C<execve>), this matters for the signal mask: many programs do not expect
		2339	certain signals to be blocked.
		2340
		2341	This means that before calling C<exec> (from the child) you should reset
		2342	the signal mask to whatever "default" you expect (all clear is a good
		2343	choice usually).
		2344
		2345	The simplest way to ensure that the signal mask is reset in the child is
		2346	to install a fork handler with C<pthread_atfork> that resets it. That will
		2347	catch fork calls done by libraries (such as the libc) as well.
		2348
		2349	In current versions of libev, the signal will not be blocked indefinitely
		2350	unless you use the C<signalfd> API (C<EV_SIGNALFD>). While this reduces
		2351	the window of opportunity for problems, it will not go away, as libev
		2352	I<has> to modify the signal mask, at least temporarily.
		2353
		2354	So I can't stress this enough: I<If you do not reset your signal mask when
		2355	you expect it to be empty, you have a race condition in your code>. This
		2356	is not a libev-specific thing, this is true for most event libraries.
		2357
		2358	=head3 The special problem of threads signal handling
		2359
		2360	POSIX threads has problematic signal handling semantics, specifically,
		2361	a lot of functionality (sigfd, sigwait etc.) only really works if all
		2362	threads in a process block signals, which is hard to achieve.
		2363
		2364	When you want to use sigwait (or mix libev signal handling with your own
		2365	for the same signals), you can tackle this problem by globally blocking
		2366	all signals before creating any threads (or creating them with a fully set
		2367	sigprocmask) and also specifying the C<EVFLAG_NOSIGMASK> when creating
		2368	loops. Then designate one thread as "signal receiver thread" which handles
		2369	these signals. You can pass on any signals that libev might be interested
		2370	in by calling C<ev_feed_signal>.
1444		2371
1445	=head3 Watcher-Specific Functions and Data Members	2372	=head3 Watcher-Specific Functions and Data Members
1446		2373
1447	=over 4	2374	=over 4
1448		2375
…		…
1459		2386
1460	=back	2387	=back
1461		2388
1462	=head3 Examples	2389	=head3 Examples
1463		2390
1464	Example: Try to exit cleanly on SIGINT and SIGTERM.	2391	Example: Try to exit cleanly on SIGINT.
1465		2392
1466	static void	2393	static void
1467	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	2394	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
1468	{	2395	{
1469	ev_unloop (loop, EVUNLOOP_ALL);	2396	ev_break (loop, EVBREAK_ALL);
1470	}	2397	}
1471		2398
1472	struct ev_signal signal_watcher;	2399	ev_signal signal_watcher;
1473	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	2400	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1474	ev_signal_start (loop, &sigint_cb);	2401	ev_signal_start (loop, &signal_watcher);
1475		2402
1476		2403
1477	=head2 C<ev_child> - watch out for process status changes	2404	=head2 C<ev_child> - watch out for process status changes
1478		2405
1479	Child watchers trigger when your process receives a SIGCHLD in response to	2406	Child watchers trigger when your process receives a SIGCHLD in response to
1480	some child status changes (most typically when a child of yours dies). It	2407	some child status changes (most typically when a child of yours dies or
1481	is permissible to install a child watcher I<after> the child has been	2408	exits). It is permissible to install a child watcher I<after> the child
1482	forked (which implies it might have already exited), as long as the event	2409	has been forked (which implies it might have already exited), as long
1483	loop isn't entered (or is continued from a watcher).	2410	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2411	forking and then immediately registering a watcher for the child is fine,
		2412	but forking and registering a watcher a few event loop iterations later or
		2413	in the next callback invocation is not.
1484		2414
1485	Only the default event loop is capable of handling signals, and therefore	2415	Only the default event loop is capable of handling signals, and therefore
1486	you can only rgeister child watchers in the default event loop.	2416	you can only register child watchers in the default event loop.
		2417
		2418	Due to some design glitches inside libev, child watchers will always be
		2419	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2420	libev)
1487		2421
1488	=head3 Process Interaction	2422	=head3 Process Interaction
1489		2423
1490	Libev grabs C<SIGCHLD> as soon as the default event loop is	2424	Libev grabs C<SIGCHLD> as soon as the default event loop is
1491	initialised. This is necessary to guarantee proper behaviour even if	2425	initialised. This is necessary to guarantee proper behaviour even if the
1492	the first child watcher is started after the child exits. The occurance	2426	first child watcher is started after the child exits. The occurrence
1493	of C<SIGCHLD> is recorded asynchronously, but child reaping is done	2427	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
1494	synchronously as part of the event loop processing. Libev always reaps all	2428	synchronously as part of the event loop processing. Libev always reaps all
1495	children, even ones not watched.	2429	children, even ones not watched.
1496		2430
1497	=head3 Overriding the Built-In Processing	2431	=head3 Overriding the Built-In Processing
…		…
1501	handler, you can override it easily by installing your own handler for	2435	handler, you can override it easily by installing your own handler for
1502	C<SIGCHLD> after initialising the default loop, and making sure the	2436	C<SIGCHLD> after initialising the default loop, and making sure the
1503	default loop never gets destroyed. You are encouraged, however, to use an	2437	default loop never gets destroyed. You are encouraged, however, to use an
1504	event-based approach to child reaping and thus use libev's support for	2438	event-based approach to child reaping and thus use libev's support for
1505	that, so other libev users can use C<ev_child> watchers freely.	2439	that, so other libev users can use C<ev_child> watchers freely.
		2440
		2441	=head3 Stopping the Child Watcher
		2442
		2443	Currently, the child watcher never gets stopped, even when the
		2444	child terminates, so normally one needs to stop the watcher in the
		2445	callback. Future versions of libev might stop the watcher automatically
		2446	when a child exit is detected (calling C<ev_child_stop> twice is not a
		2447	problem).
1506		2448
1507	=head3 Watcher-Specific Functions and Data Members	2449	=head3 Watcher-Specific Functions and Data Members
1508		2450
1509	=over 4	2451	=over 4
1510		2452
…		…
1539	=head3 Examples	2481	=head3 Examples
1540		2482
1541	Example: C<fork()> a new process and install a child handler to wait for	2483	Example: C<fork()> a new process and install a child handler to wait for
1542	its completion.	2484	its completion.
1543		2485
1544	ev_child cw;	2486	ev_child cw;
1545		2487
1546	static void	2488	static void
1547	child_cb (EV_P_ struct ev_child *w, int revents)	2489	child_cb (EV_P_ ev_child *w, int revents)
1548	{	2490	{
1549	ev_child_stop (EV_A_ w);	2491	ev_child_stop (EV_A_ w);
1550	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);	2492	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1551	}	2493	}
1552		2494
1553	pid_t pid = fork ();	2495	pid_t pid = fork ();
1554		2496
1555	if (pid < 0)	2497	if (pid < 0)
1556	// error	2498	// error
1557	else if (pid == 0)	2499	else if (pid == 0)
1558	{	2500	{
1559	// the forked child executes here	2501	// the forked child executes here
1560	exit (1);	2502	exit (1);
1561	}	2503	}
1562	else	2504	else
1563	{	2505	{
1564	ev_child_init (&cw, child_cb, pid, 0);	2506	ev_child_init (&cw, child_cb, pid, 0);
1565	ev_child_start (EV_DEFAULT_ &cw);	2507	ev_child_start (EV_DEFAULT_ &cw);
1566	}	2508	}
1567		2509
1568		2510
1569	=head2 C<ev_stat> - did the file attributes just change?	2511	=head2 C<ev_stat> - did the file attributes just change?
1570		2512
1571	This watches a filesystem path for attribute changes. That is, it calls	2513	This watches a file system path for attribute changes. That is, it calls
1572	C<stat> regularly (or when the OS says it changed) and sees if it changed	2514	C<stat> on that path in regular intervals (or when the OS says it changed)
1573	compared to the last time, invoking the callback if it did.	2515	and sees if it changed compared to the last time, invoking the callback if
		2516	it did.
1574		2517
1575	The path does not need to exist: changing from "path exists" to "path does	2518	The path does not need to exist: changing from "path exists" to "path does
1576	not exist" is a status change like any other. The condition "path does	2519	not exist" is a status change like any other. The condition "path does not
1577	not exist" is signified by the C<st_nlink> field being zero (which is	2520	exist" (or more correctly "path cannot be stat'ed") is signified by the
1578	otherwise always forced to be at least one) and all the other fields of	2521	C<st_nlink> field being zero (which is otherwise always forced to be at
1579	the stat buffer having unspecified contents.	2522	least one) and all the other fields of the stat buffer having unspecified
		2523	contents.
1580		2524
1581	The path I<should> be absolute and I<must not> end in a slash. If it is	2525	The path I<must not> end in a slash or contain special components such as
		2526	C<.> or C<..>. The path I<should> be absolute: If it is relative and
1582	relative and your working directory changes, the behaviour is undefined.	2527	your working directory changes, then the behaviour is undefined.
1583		2528
1584	Since there is no standard to do this, the portable implementation simply	2529	Since there is no portable change notification interface available, the
1585	calls C<stat (2)> regularly on the path to see if it changed somehow. You	2530	portable implementation simply calls C<stat(2)> regularly on the path
1586	can specify a recommended polling interval for this case. If you specify	2531	to see if it changed somehow. You can specify a recommended polling
1587	a polling interval of C<0> (highly recommended!) then a I<suitable,	2532	interval for this case. If you specify a polling interval of C<0> (highly
1588	unspecified default> value will be used (which you can expect to be around	2533	recommended!) then a I<suitable, unspecified default> value will be used
1589	five seconds, although this might change dynamically). Libev will also	2534	(which you can expect to be around five seconds, although this might
1590	impose a minimum interval which is currently around C<0.1>, but thats	2535	change dynamically). Libev will also impose a minimum interval which is
1591	usually overkill.	2536	currently around C<0.1>, but that's usually overkill.
1592		2537
1593	This watcher type is not meant for massive numbers of stat watchers,	2538	This watcher type is not meant for massive numbers of stat watchers,
1594	as even with OS-supported change notifications, this can be	2539	as even with OS-supported change notifications, this can be
1595	resource-intensive.	2540	resource-intensive.
1596		2541
1597	At the time of this writing, only the Linux inotify interface is	2542	At the time of this writing, the only OS-specific interface implemented
1598	implemented (implementing kqueue support is left as an exercise for the	2543	is the Linux inotify interface (implementing kqueue support is left as an
1599	reader). Inotify will be used to give hints only and should not change the	2544	exercise for the reader. Note, however, that the author sees no way of
1600	semantics of C<ev_stat> watchers, which means that libev sometimes needs	2545	implementing C<ev_stat> semantics with kqueue, except as a hint).
1601	to fall back to regular polling again even with inotify, but changes are
1602	usually detected immediately, and if the file exists there will be no
1603	polling.
1604		2546
1605	=head3 Inotify	2547	=head3 ABI Issues (Largefile Support)
1606		2548
		2549	Libev by default (unless the user overrides this) uses the default
		2550	compilation environment, which means that on systems with large file
		2551	support disabled by default, you get the 32 bit version of the stat
		2552	structure. When using the library from programs that change the ABI to
		2553	use 64 bit file offsets the programs will fail. In that case you have to
		2554	compile libev with the same flags to get binary compatibility. This is
		2555	obviously the case with any flags that change the ABI, but the problem is
		2556	most noticeably displayed with ev_stat and large file support.
		2557
		2558	The solution for this is to lobby your distribution maker to make large
		2559	file interfaces available by default (as e.g. FreeBSD does) and not
		2560	optional. Libev cannot simply switch on large file support because it has
		2561	to exchange stat structures with application programs compiled using the
		2562	default compilation environment.
		2563
		2564	=head3 Inotify and Kqueue
		2565
1607	When C<inotify (7)> support has been compiled into libev (generally only	2566	When C<inotify (7)> support has been compiled into libev and present at
1608	available on Linux) and present at runtime, it will be used to speed up	2567	runtime, it will be used to speed up change detection where possible. The
1609	change detection where possible. The inotify descriptor will be created lazily	2568	inotify descriptor will be created lazily when the first C<ev_stat>
1610	when the first C<ev_stat> watcher is being started.	2569	watcher is being started.
1611		2570
1612	Inotify presense does not change the semantics of C<ev_stat> watchers	2571	Inotify presence does not change the semantics of C<ev_stat> watchers
1613	except that changes might be detected earlier, and in some cases, to avoid	2572	except that changes might be detected earlier, and in some cases, to avoid
1614	making regular C<stat> calls. Even in the presense of inotify support	2573	making regular C<stat> calls. Even in the presence of inotify support
1615	there are many cases where libev has to resort to regular C<stat> polling.	2574	there are many cases where libev has to resort to regular C<stat> polling,
		2575	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2576	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2577	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2578	xfs are fully working) libev usually gets away without polling.
1616		2579
1617	(There is no support for kqueue, as apparently it cannot be used to	2580	There is no support for kqueue, as apparently it cannot be used to
1618	implement this functionality, due to the requirement of having a file	2581	implement this functionality, due to the requirement of having a file
1619	descriptor open on the object at all times).	2582	descriptor open on the object at all times, and detecting renames, unlinks
		2583	etc. is difficult.
		2584
		2585	=head3 C<stat ()> is a synchronous operation
		2586
		2587	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2588	the process. The exception are C<ev_stat> watchers - those call C<stat
		2589	()>, which is a synchronous operation.
		2590
		2591	For local paths, this usually doesn't matter: unless the system is very
		2592	busy or the intervals between stat's are large, a stat call will be fast,
		2593	as the path data is usually in memory already (except when starting the
		2594	watcher).
		2595
		2596	For networked file systems, calling C<stat ()> can block an indefinite
		2597	time due to network issues, and even under good conditions, a stat call
		2598	often takes multiple milliseconds.
		2599
		2600	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2601	paths, although this is fully supported by libev.
1620		2602
1621	=head3 The special problem of stat time resolution	2603	=head3 The special problem of stat time resolution
1622		2604
1623	The C<stat ()> syscall only supports full-second resolution portably, and	2605	The C<stat ()> system call only supports full-second resolution portably,
1624	even on systems where the resolution is higher, many filesystems still	2606	and even on systems where the resolution is higher, most file systems
1625	only support whole seconds.	2607	still only support whole seconds.
1626		2608
1627	That means that, if the time is the only thing that changes, you might	2609	That means that, if the time is the only thing that changes, you can
1628	miss updates: on the first update, C<ev_stat> detects a change and calls	2610	easily miss updates: on the first update, C<ev_stat> detects a change and
1629	your callback, which does something. When there is another update within	2611	calls your callback, which does something. When there is another update
1630	the same second, C<ev_stat> will be unable to detect it.	2612	within the same second, C<ev_stat> will be unable to detect unless the
		2613	stat data does change in other ways (e.g. file size).
1631		2614
1632	The solution to this is to delay acting on a change for a second (or till	2615	The solution to this is to delay acting on a change for slightly more
1633	the next second boundary), using a roughly one-second delay C<ev_timer>	2616	than a second (or till slightly after the next full second boundary), using
1634	(C<ev_timer_set (w, 0., 1.01); ev_timer_again (loop, w)>). The C<.01>	2617	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
1635	is added to work around small timing inconsistencies of some operating	2618	ev_timer_again (loop, w)>).
1636	systems.	2619
		2620	The C<.02> offset is added to work around small timing inconsistencies
		2621	of some operating systems (where the second counter of the current time
		2622	might be be delayed. One such system is the Linux kernel, where a call to
		2623	C<gettimeofday> might return a timestamp with a full second later than
		2624	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		2625	update file times then there will be a small window where the kernel uses
		2626	the previous second to update file times but libev might already execute
		2627	the timer callback).
1637		2628
1638	=head3 Watcher-Specific Functions and Data Members	2629	=head3 Watcher-Specific Functions and Data Members
1639		2630
1640	=over 4	2631	=over 4
1641		2632
…		…
1647	C<path>. The C<interval> is a hint on how quickly a change is expected to	2638	C<path>. The C<interval> is a hint on how quickly a change is expected to
1648	be detected and should normally be specified as C<0> to let libev choose	2639	be detected and should normally be specified as C<0> to let libev choose
1649	a suitable value. The memory pointed to by C<path> must point to the same	2640	a suitable value. The memory pointed to by C<path> must point to the same
1650	path for as long as the watcher is active.	2641	path for as long as the watcher is active.
1651		2642
1652	The callback will be receive C<EV_STAT> when a change was detected,	2643	The callback will receive an C<EV_STAT> event when a change was detected,
1653	relative to the attributes at the time the watcher was started (or the	2644	relative to the attributes at the time the watcher was started (or the
1654	last change was detected).	2645	last change was detected).
1655		2646
1656	=item ev_stat_stat (loop, ev_stat *)	2647	=item ev_stat_stat (loop, ev_stat *)
1657		2648
1658	Updates the stat buffer immediately with new values. If you change the	2649	Updates the stat buffer immediately with new values. If you change the
1659	watched path in your callback, you could call this fucntion to avoid	2650	watched path in your callback, you could call this function to avoid
1660	detecting this change (while introducing a race condition). Can also be	2651	detecting this change (while introducing a race condition if you are not
1661	useful simply to find out the new values.	2652	the only one changing the path). Can also be useful simply to find out the
		2653	new values.
1662		2654
1663	=item ev_statdata attr [read-only]	2655	=item ev_statdata attr [read-only]
1664		2656
1665	The most-recently detected attributes of the file. Although the type is of	2657	The most-recently detected attributes of the file. Although the type is
1666	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	2658	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1667	suitable for your system. If the C<st_nlink> member is C<0>, then there	2659	suitable for your system, but you can only rely on the POSIX-standardised
		2660	members to be present. If the C<st_nlink> member is C<0>, then there was
1668	was some error while C<stat>ing the file.	2661	some error while C<stat>ing the file.
1669		2662
1670	=item ev_statdata prev [read-only]	2663	=item ev_statdata prev [read-only]
1671		2664
1672	The previous attributes of the file. The callback gets invoked whenever	2665	The previous attributes of the file. The callback gets invoked whenever
1673	C<prev> != C<attr>.	2666	C<prev> != C<attr>, or, more precisely, one or more of these members
		2667	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		2668	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1674		2669
1675	=item ev_tstamp interval [read-only]	2670	=item ev_tstamp interval [read-only]
1676		2671
1677	The specified interval.	2672	The specified interval.
1678		2673
1679	=item const char *path [read-only]	2674	=item const char *path [read-only]
1680		2675
1681	The filesystem path that is being watched.	2676	The file system path that is being watched.
1682		2677
1683	=back	2678	=back
1684		2679
1685	=head3 Examples	2680	=head3 Examples
1686		2681
1687	Example: Watch C</etc/passwd> for attribute changes.	2682	Example: Watch C</etc/passwd> for attribute changes.
1688		2683
1689	static void	2684	static void
1690	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	2685	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1691	{	2686	{
1692	/* /etc/passwd changed in some way */	2687	/* /etc/passwd changed in some way */
1693	if (w->attr.st_nlink)	2688	if (w->attr.st_nlink)
1694	{	2689	{
1695	printf ("passwd current size %ld\n", (long)w->attr.st_size);	2690	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1696	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	2691	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1697	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	2692	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1698	}	2693	}
1699	else	2694	else
1700	/* you shalt not abuse printf for puts */	2695	/* you shalt not abuse printf for puts */
1701	puts ("wow, /etc/passwd is not there, expect problems. "	2696	puts ("wow, /etc/passwd is not there, expect problems. "
1702	"if this is windows, they already arrived\n");	2697	"if this is windows, they already arrived\n");
1703	}	2698	}
1704		2699
1705	...	2700	...
1706	ev_stat passwd;	2701	ev_stat passwd;
1707		2702
1708	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);	2703	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1709	ev_stat_start (loop, &passwd);	2704	ev_stat_start (loop, &passwd);
1710		2705
1711	Example: Like above, but additionally use a one-second delay so we do not	2706	Example: Like above, but additionally use a one-second delay so we do not
1712	miss updates (however, frequent updates will delay processing, too, so	2707	miss updates (however, frequent updates will delay processing, too, so
1713	one might do the work both on C<ev_stat> callback invocation I<and> on	2708	one might do the work both on C<ev_stat> callback invocation I<and> on
1714	C<ev_timer> callback invocation).	2709	C<ev_timer> callback invocation).
1715		2710
1716	static ev_stat passwd;	2711	static ev_stat passwd;
1717	static ev_timer timer;	2712	static ev_timer timer;
1718		2713
1719	static void	2714	static void
1720	timer_cb (EV_P_ ev_timer *w, int revents)	2715	timer_cb (EV_P_ ev_timer *w, int revents)
1721	{	2716	{
1722	ev_timer_stop (EV_A_ w);	2717	ev_timer_stop (EV_A_ w);
1723		2718
1724	/* now it's one second after the most recent passwd change */	2719	/* now it's one second after the most recent passwd change */
1725	}	2720	}
1726		2721
1727	static void	2722	static void
1728	stat_cb (EV_P_ ev_stat *w, int revents)	2723	stat_cb (EV_P_ ev_stat *w, int revents)
1729	{	2724	{
1730	/* reset the one-second timer */	2725	/* reset the one-second timer */
1731	ev_timer_again (EV_A_ &timer);	2726	ev_timer_again (EV_A_ &timer);
1732	}	2727	}
1733		2728
1734	...	2729	...
1735	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);	2730	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
1736	ev_stat_start (loop, &passwd);	2731	ev_stat_start (loop, &passwd);
1737	ev_timer_init (&timer, timer_cb, 0., 1.01);	2732	ev_timer_init (&timer, timer_cb, 0., 1.02);
1738		2733
1739		2734
1740	=head2 C<ev_idle> - when you've got nothing better to do...	2735	=head2 C<ev_idle> - when you've got nothing better to do...
1741		2736
1742	Idle watchers trigger events when no other events of the same or higher	2737	Idle watchers trigger events when no other events of the same or higher
1743	priority are pending (prepare, check and other idle watchers do not	2738	priority are pending (prepare, check and other idle watchers do not count
1744	count).	2739	as receiving "events").
1745		2740
1746	That is, as long as your process is busy handling sockets or timeouts	2741	That is, as long as your process is busy handling sockets or timeouts
1747	(or even signals, imagine) of the same or higher priority it will not be	2742	(or even signals, imagine) of the same or higher priority it will not be
1748	triggered. But when your process is idle (or only lower-priority watchers	2743	triggered. But when your process is idle (or only lower-priority watchers
1749	are pending), the idle watchers are being called once per event loop	2744	are pending), the idle watchers are being called once per event loop
…		…
1760		2755
1761	=head3 Watcher-Specific Functions and Data Members	2756	=head3 Watcher-Specific Functions and Data Members
1762		2757
1763	=over 4	2758	=over 4
1764		2759
1765	=item ev_idle_init (ev_signal *, callback)	2760	=item ev_idle_init (ev_idle *, callback)
1766		2761
1767	Initialises and configures the idle watcher - it has no parameters of any	2762	Initialises and configures the idle watcher - it has no parameters of any
1768	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2763	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1769	believe me.	2764	believe me.
1770		2765
…		…
1773	=head3 Examples	2768	=head3 Examples
1774		2769
1775	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	2770	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1776	callback, free it. Also, use no error checking, as usual.	2771	callback, free it. Also, use no error checking, as usual.
1777		2772
1778	static void	2773	static void
1779	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	2774	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
1780	{	2775	{
1781	free (w);	2776	free (w);
1782	// now do something you wanted to do when the program has	2777	// now do something you wanted to do when the program has
1783	// no longer anything immediate to do.	2778	// no longer anything immediate to do.
1784	}	2779	}
1785		2780
1786	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	2781	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
1787	ev_idle_init (idle_watcher, idle_cb);	2782	ev_idle_init (idle_watcher, idle_cb);
1788	ev_idle_start (loop, idle_cb);	2783	ev_idle_start (loop, idle_watcher);
1789		2784
1790		2785
1791	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2786	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1792		2787
1793	Prepare and check watchers are usually (but not always) used in tandem:	2788	Prepare and check watchers are usually (but not always) used in pairs:
1794	prepare watchers get invoked before the process blocks and check watchers	2789	prepare watchers get invoked before the process blocks and check watchers
1795	afterwards.	2790	afterwards.
1796		2791
1797	You I<must not> call C<ev_loop> or similar functions that enter	2792	You I<must not> call C<ev_run> or similar functions that enter
1798	the current event loop from either C<ev_prepare> or C<ev_check>	2793	the current event loop from either C<ev_prepare> or C<ev_check>
1799	watchers. Other loops than the current one are fine, however. The	2794	watchers. Other loops than the current one are fine, however. The
1800	rationale behind this is that you do not need to check for recursion in	2795	rationale behind this is that you do not need to check for recursion in
1801	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,	2796	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
1802	C<ev_check> so if you have one watcher of each kind they will always be	2797	C<ev_check> so if you have one watcher of each kind they will always be
1803	called in pairs bracketing the blocking call.	2798	called in pairs bracketing the blocking call.
1804		2799
1805	Their main purpose is to integrate other event mechanisms into libev and	2800	Their main purpose is to integrate other event mechanisms into libev and
1806	their use is somewhat advanced. This could be used, for example, to track	2801	their use is somewhat advanced. They could be used, for example, to track
1807	variable changes, implement your own watchers, integrate net-snmp or a	2802	variable changes, implement your own watchers, integrate net-snmp or a
1808	coroutine library and lots more. They are also occasionally useful if	2803	coroutine library and lots more. They are also occasionally useful if
1809	you cache some data and want to flush it before blocking (for example,	2804	you cache some data and want to flush it before blocking (for example,
1810	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>	2805	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1811	watcher).	2806	watcher).
1812		2807
1813	This is done by examining in each prepare call which file descriptors need	2808	This is done by examining in each prepare call which file descriptors
1814	to be watched by the other library, registering C<ev_io> watchers for	2809	need to be watched by the other library, registering C<ev_io> watchers
1815	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2810	for them and starting an C<ev_timer> watcher for any timeouts (many
1816	provide just this functionality). Then, in the check watcher you check for	2811	libraries provide exactly this functionality). Then, in the check watcher,
1817	any events that occured (by checking the pending status of all watchers	2812	you check for any events that occurred (by checking the pending status
1818	and stopping them) and call back into the library. The I/O and timer	2813	of all watchers and stopping them) and call back into the library. The
1819	callbacks will never actually be called (but must be valid nevertheless,	2814	I/O and timer callbacks will never actually be called (but must be valid
1820	because you never know, you know?).	2815	nevertheless, because you never know, you know?).
1821		2816
1822	As another example, the Perl Coro module uses these hooks to integrate	2817	As another example, the Perl Coro module uses these hooks to integrate
1823	coroutines into libev programs, by yielding to other active coroutines	2818	coroutines into libev programs, by yielding to other active coroutines
1824	during each prepare and only letting the process block if no coroutines	2819	during each prepare and only letting the process block if no coroutines
1825	are ready to run (it's actually more complicated: it only runs coroutines	2820	are ready to run (it's actually more complicated: it only runs coroutines
…		…
1828	loop from blocking if lower-priority coroutines are active, thus mapping	2823	loop from blocking if lower-priority coroutines are active, thus mapping
1829	low-priority coroutines to idle/background tasks).	2824	low-priority coroutines to idle/background tasks).
1830		2825
1831	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	2826	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
1832	priority, to ensure that they are being run before any other watchers	2827	priority, to ensure that they are being run before any other watchers
		2828	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2829
1833	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	2830	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
1834	too) should not activate ("feed") events into libev. While libev fully	2831	activate ("feed") events into libev. While libev fully supports this, they
1835	supports this, they will be called before other C<ev_check> watchers	2832	might get executed before other C<ev_check> watchers did their job. As
1836	did their job. As C<ev_check> watchers are often used to embed other	2833	C<ev_check> watchers are often used to embed other (non-libev) event
1837	(non-libev) event loops those other event loops might be in an unusable	2834	loops those other event loops might be in an unusable state until their
1838	state until their C<ev_check> watcher ran (always remind yourself to	2835	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
1839	coexist peacefully with others).	2836	others).
1840		2837
1841	=head3 Watcher-Specific Functions and Data Members	2838	=head3 Watcher-Specific Functions and Data Members
1842		2839
1843	=over 4	2840	=over 4
1844		2841
…		…
1846		2843
1847	=item ev_check_init (ev_check *, callback)	2844	=item ev_check_init (ev_check *, callback)
1848		2845
1849	Initialises and configures the prepare or check watcher - they have no	2846	Initialises and configures the prepare or check watcher - they have no
1850	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2847	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1851	macros, but using them is utterly, utterly and completely pointless.	2848	macros, but using them is utterly, utterly, utterly and completely
		2849	pointless.
1852		2850
1853	=back	2851	=back
1854		2852
1855	=head3 Examples	2853	=head3 Examples
1856		2854
1857	There are a number of principal ways to embed other event loops or modules	2855	There are a number of principal ways to embed other event loops or modules
1858	into libev. Here are some ideas on how to include libadns into libev	2856	into libev. Here are some ideas on how to include libadns into libev
1859	(there is a Perl module named C<EV::ADNS> that does this, which you could	2857	(there is a Perl module named C<EV::ADNS> that does this, which you could
1860	use for an actually working example. Another Perl module named C<EV::Glib>	2858	use as a working example. Another Perl module named C<EV::Glib> embeds a
1861	embeds a Glib main context into libev, and finally, C<Glib::EV> embeds EV	2859	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
1862	into the Glib event loop).	2860	Glib event loop).
1863		2861
1864	Method 1: Add IO watchers and a timeout watcher in a prepare handler,	2862	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1865	and in a check watcher, destroy them and call into libadns. What follows	2863	and in a check watcher, destroy them and call into libadns. What follows
1866	is pseudo-code only of course. This requires you to either use a low	2864	is pseudo-code only of course. This requires you to either use a low
1867	priority for the check watcher or use C<ev_clear_pending> explicitly, as	2865	priority for the check watcher or use C<ev_clear_pending> explicitly, as
1868	the callbacks for the IO/timeout watchers might not have been called yet.	2866	the callbacks for the IO/timeout watchers might not have been called yet.
1869		2867
1870	static ev_io iow [nfd];	2868	static ev_io iow [nfd];
1871	static ev_timer tw;	2869	static ev_timer tw;
1872		2870
1873	static void	2871	static void
1874	io_cb (ev_loop *loop, ev_io *w, int revents)	2872	io_cb (struct ev_loop *loop, ev_io *w, int revents)
1875	{	2873	{
1876	}	2874	}
1877		2875
1878	// create io watchers for each fd and a timer before blocking	2876	// create io watchers for each fd and a timer before blocking
1879	static void	2877	static void
1880	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2878	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
1881	{	2879	{
1882	int timeout = 3600000;	2880	int timeout = 3600000;
1883	struct pollfd fds [nfd];	2881	struct pollfd fds [nfd];
1884	// actual code will need to loop here and realloc etc.	2882	// actual code will need to loop here and realloc etc.
1885	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2883	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1886		2884
1887	/* the callback is illegal, but won't be called as we stop during check */	2885	/* the callback is illegal, but won't be called as we stop during check */
1888	ev_timer_init (&tw, 0, timeout * 1e-3);	2886	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
1889	ev_timer_start (loop, &tw);	2887	ev_timer_start (loop, &tw);
1890		2888
1891	// create one ev_io per pollfd	2889	// create one ev_io per pollfd
1892	for (int i = 0; i < nfd; ++i)	2890	for (int i = 0; i < nfd; ++i)
1893	{	2891	{
1894	ev_io_init (iow + i, io_cb, fds [i].fd,	2892	ev_io_init (iow + i, io_cb, fds [i].fd,
1895	((fds [i].events & POLLIN ? EV_READ : 0)	2893	((fds [i].events & POLLIN ? EV_READ : 0)
1896	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	2894	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1897		2895
1898	fds [i].revents = 0;	2896	fds [i].revents = 0;
1899	ev_io_start (loop, iow + i);	2897	ev_io_start (loop, iow + i);
1900	}	2898	}
1901	}	2899	}
1902		2900
1903	// stop all watchers after blocking	2901	// stop all watchers after blocking
1904	static void	2902	static void
1905	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2903	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
1906	{	2904	{
1907	ev_timer_stop (loop, &tw);	2905	ev_timer_stop (loop, &tw);
1908		2906
1909	for (int i = 0; i < nfd; ++i)	2907	for (int i = 0; i < nfd; ++i)
1910	{	2908	{
1911	// set the relevant poll flags	2909	// set the relevant poll flags
1912	// could also call adns_processreadable etc. here	2910	// could also call adns_processreadable etc. here
1913	struct pollfd *fd = fds + i;	2911	struct pollfd *fd = fds + i;
1914	int revents = ev_clear_pending (iow + i);	2912	int revents = ev_clear_pending (iow + i);
1915	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;	2913	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1916	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;	2914	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1917		2915
1918	// now stop the watcher	2916	// now stop the watcher
1919	ev_io_stop (loop, iow + i);	2917	ev_io_stop (loop, iow + i);
1920	}	2918	}
1921		2919
1922	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	2920	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1923	}	2921	}
1924		2922
1925	Method 2: This would be just like method 1, but you run C<adns_afterpoll>	2923	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
1926	in the prepare watcher and would dispose of the check watcher.	2924	in the prepare watcher and would dispose of the check watcher.
1927		2925
1928	Method 3: If the module to be embedded supports explicit event	2926	Method 3: If the module to be embedded supports explicit event
1929	notification (adns does), you can also make use of the actual watcher	2927	notification (libadns does), you can also make use of the actual watcher
1930	callbacks, and only destroy/create the watchers in the prepare watcher.	2928	callbacks, and only destroy/create the watchers in the prepare watcher.
1931		2929
1932	static void	2930	static void
1933	timer_cb (EV_P_ ev_timer *w, int revents)	2931	timer_cb (EV_P_ ev_timer *w, int revents)
1934	{	2932	{
1935	adns_state ads = (adns_state)w->data;	2933	adns_state ads = (adns_state)w->data;
1936	update_now (EV_A);	2934	update_now (EV_A);
1937		2935
1938	adns_processtimeouts (ads, &tv_now);	2936	adns_processtimeouts (ads, &tv_now);
1939	}	2937	}
1940		2938
1941	static void	2939	static void
1942	io_cb (EV_P_ ev_io *w, int revents)	2940	io_cb (EV_P_ ev_io *w, int revents)
1943	{	2941	{
1944	adns_state ads = (adns_state)w->data;	2942	adns_state ads = (adns_state)w->data;
1945	update_now (EV_A);	2943	update_now (EV_A);
1946		2944
1947	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);	2945	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
1948	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);	2946	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
1949	}	2947	}
1950		2948
1951	// do not ever call adns_afterpoll	2949	// do not ever call adns_afterpoll
1952		2950
1953	Method 4: Do not use a prepare or check watcher because the module you	2951	Method 4: Do not use a prepare or check watcher because the module you
1954	want to embed is too inflexible to support it. Instead, youc na override	2952	want to embed is not flexible enough to support it. Instead, you can
1955	their poll function. The drawback with this solution is that the main	2953	override their poll function. The drawback with this solution is that the
1956	loop is now no longer controllable by EV. The C<Glib::EV> module does	2954	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
1957	this.	2955	this approach, effectively embedding EV as a client into the horrible
		2956	libglib event loop.
1958		2957
1959	static gint	2958	static gint
1960	event_poll_func (GPollFD *fds, guint nfds, gint timeout)	2959	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
1961	{	2960	{
1962	int got_events = 0;	2961	int got_events = 0;
1963		2962
1964	for (n = 0; n < nfds; ++n)	2963	for (n = 0; n < nfds; ++n)
1965	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events	2964	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
1966		2965
1967	if (timeout >= 0)	2966	if (timeout >= 0)
1968	// create/start timer	2967	// create/start timer
1969		2968
1970	// poll	2969	// poll
1971	ev_loop (EV_A_ 0);	2970	ev_run (EV_A_ 0);
1972		2971
1973	// stop timer again	2972	// stop timer again
1974	if (timeout >= 0)	2973	if (timeout >= 0)
1975	ev_timer_stop (EV_A_ &to);	2974	ev_timer_stop (EV_A_ &to);
1976		2975
1977	// stop io watchers again - their callbacks should have set	2976	// stop io watchers again - their callbacks should have set
1978	for (n = 0; n < nfds; ++n)	2977	for (n = 0; n < nfds; ++n)
1979	ev_io_stop (EV_A_ iow [n]);	2978	ev_io_stop (EV_A_ iow [n]);
1980		2979
1981	return got_events;	2980	return got_events;
1982	}	2981	}
1983		2982
1984		2983
1985	=head2 C<ev_embed> - when one backend isn't enough...	2984	=head2 C<ev_embed> - when one backend isn't enough...
1986		2985
1987	This is a rather advanced watcher type that lets you embed one event loop	2986	This is a rather advanced watcher type that lets you embed one event loop
…		…
1993	prioritise I/O.	2992	prioritise I/O.
1994		2993
1995	As an example for a bug workaround, the kqueue backend might only support	2994	As an example for a bug workaround, the kqueue backend might only support
1996	sockets on some platform, so it is unusable as generic backend, but you	2995	sockets on some platform, so it is unusable as generic backend, but you
1997	still want to make use of it because you have many sockets and it scales	2996	still want to make use of it because you have many sockets and it scales
1998	so nicely. In this case, you would create a kqueue-based loop and embed it	2997	so nicely. In this case, you would create a kqueue-based loop and embed
1999	into your default loop (which might use e.g. poll). Overall operation will	2998	it into your default loop (which might use e.g. poll). Overall operation
2000	be a bit slower because first libev has to poll and then call kevent, but	2999	will be a bit slower because first libev has to call C<poll> and then
2001	at least you can use both at what they are best.	3000	C<kevent>, but at least you can use both mechanisms for what they are
		3001	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
2002		3002
2003	As for prioritising I/O: rarely you have the case where some fds have	3003	As for prioritising I/O: under rare circumstances you have the case where
2004	to be watched and handled very quickly (with low latency), and even	3004	some fds have to be watched and handled very quickly (with low latency),
2005	priorities and idle watchers might have too much overhead. In this case	3005	and even priorities and idle watchers might have too much overhead. In
2006	you would put all the high priority stuff in one loop and all the rest in	3006	this case you would put all the high priority stuff in one loop and all
2007	a second one, and embed the second one in the first.	3007	the rest in a second one, and embed the second one in the first.
2008		3008
2009	As long as the watcher is active, the callback will be invoked every time	3009	As long as the watcher is active, the callback will be invoked every
2010	there might be events pending in the embedded loop. The callback must then	3010	time there might be events pending in the embedded loop. The callback
2011	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke	3011	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
2012	their callbacks (you could also start an idle watcher to give the embedded	3012	sweep and invoke their callbacks (the callback doesn't need to invoke the
2013	loop strictly lower priority for example). You can also set the callback	3013	C<ev_embed_sweep> function directly, it could also start an idle watcher
2014	to C<0>, in which case the embed watcher will automatically execute the	3014	to give the embedded loop strictly lower priority for example).
2015	embedded loop sweep.
2016		3015
2017	As long as the watcher is started it will automatically handle events. The	3016	You can also set the callback to C<0>, in which case the embed watcher
2018	callback will be invoked whenever some events have been handled. You can	3017	will automatically execute the embedded loop sweep whenever necessary.
2019	set the callback to C<0> to avoid having to specify one if you are not
2020	interested in that.
2021		3018
2022	Also, there have not currently been made special provisions for forking:	3019	Fork detection will be handled transparently while the C<ev_embed> watcher
2023	when you fork, you not only have to call C<ev_loop_fork> on both loops,	3020	is active, i.e., the embedded loop will automatically be forked when the
2024	but you will also have to stop and restart any C<ev_embed> watchers	3021	embedding loop forks. In other cases, the user is responsible for calling
2025	yourself.	3022	C<ev_loop_fork> on the embedded loop.
2026		3023
2027	Unfortunately, not all backends are embeddable, only the ones returned by	3024	Unfortunately, not all backends are embeddable: only the ones returned by
2028	C<ev_embeddable_backends> are, which, unfortunately, does not include any	3025	C<ev_embeddable_backends> are, which, unfortunately, does not include any
2029	portable one.	3026	portable one.
2030		3027
2031	So when you want to use this feature you will always have to be prepared	3028	So when you want to use this feature you will always have to be prepared
2032	that you cannot get an embeddable loop. The recommended way to get around	3029	that you cannot get an embeddable loop. The recommended way to get around
2033	this is to have a separate variables for your embeddable loop, try to	3030	this is to have a separate variables for your embeddable loop, try to
2034	create it, and if that fails, use the normal loop for everything.	3031	create it, and if that fails, use the normal loop for everything.
2035		3032
		3033	=head3 C<ev_embed> and fork
		3034
		3035	While the C<ev_embed> watcher is running, forks in the embedding loop will
		3036	automatically be applied to the embedded loop as well, so no special
		3037	fork handling is required in that case. When the watcher is not running,
		3038	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		3039	as applicable.
		3040
2036	=head3 Watcher-Specific Functions and Data Members	3041	=head3 Watcher-Specific Functions and Data Members
2037		3042
2038	=over 4	3043	=over 4
2039		3044
2040	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	3045	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
…		…
2043		3048
2044	Configures the watcher to embed the given loop, which must be	3049	Configures the watcher to embed the given loop, which must be
2045	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be	3050	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
2046	invoked automatically, otherwise it is the responsibility of the callback	3051	invoked automatically, otherwise it is the responsibility of the callback
2047	to invoke it (it will continue to be called until the sweep has been done,	3052	to invoke it (it will continue to be called until the sweep has been done,
2048	if you do not want thta, you need to temporarily stop the embed watcher).	3053	if you do not want that, you need to temporarily stop the embed watcher).
2049		3054
2050	=item ev_embed_sweep (loop, ev_embed *)	3055	=item ev_embed_sweep (loop, ev_embed *)
2051		3056
2052	Make a single, non-blocking sweep over the embedded loop. This works	3057	Make a single, non-blocking sweep over the embedded loop. This works
2053	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	3058	similarly to C<ev_run (embedded_loop, EVRUN_NOWAIT)>, but in the most
2054	apropriate way for embedded loops.	3059	appropriate way for embedded loops.
2055		3060
2056	=item struct ev_loop *other [read-only]	3061	=item struct ev_loop *other [read-only]
2057		3062
2058	The embedded event loop.	3063	The embedded event loop.
2059		3064
…		…
2061		3066
2062	=head3 Examples	3067	=head3 Examples
2063		3068
2064	Example: Try to get an embeddable event loop and embed it into the default	3069	Example: Try to get an embeddable event loop and embed it into the default
2065	event loop. If that is not possible, use the default loop. The default	3070	event loop. If that is not possible, use the default loop. The default
2066	loop is stored in C<loop_hi>, while the mebeddable loop is stored in	3071	loop is stored in C<loop_hi>, while the embeddable loop is stored in
2067	C<loop_lo> (which is C<loop_hi> in the acse no embeddable loop can be	3072	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
2068	used).	3073	used).
2069		3074
2070	struct ev_loop *loop_hi = ev_default_init (0);	3075	struct ev_loop *loop_hi = ev_default_init (0);
2071	struct ev_loop *loop_lo = 0;	3076	struct ev_loop *loop_lo = 0;
2072	struct ev_embed embed;	3077	ev_embed embed;
2073		3078
2074	// see if there is a chance of getting one that works	3079	// see if there is a chance of getting one that works
2075	// (remember that a flags value of 0 means autodetection)	3080	// (remember that a flags value of 0 means autodetection)
2076	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()	3081	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
2077	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())	3082	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
2078	: 0;	3083	: 0;
2079		3084
2080	// if we got one, then embed it, otherwise default to loop_hi	3085	// if we got one, then embed it, otherwise default to loop_hi
2081	if (loop_lo)	3086	if (loop_lo)
2082	{	3087	{
2083	ev_embed_init (&embed, 0, loop_lo);	3088	ev_embed_init (&embed, 0, loop_lo);
2084	ev_embed_start (loop_hi, &embed);	3089	ev_embed_start (loop_hi, &embed);
2085	}	3090	}
2086	else	3091	else
2087	loop_lo = loop_hi;	3092	loop_lo = loop_hi;
2088		3093
2089	Example: Check if kqueue is available but not recommended and create	3094	Example: Check if kqueue is available but not recommended and create
2090	a kqueue backend for use with sockets (which usually work with any	3095	a kqueue backend for use with sockets (which usually work with any
2091	kqueue implementation). Store the kqueue/socket-only event loop in	3096	kqueue implementation). Store the kqueue/socket-only event loop in
2092	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).	3097	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
2093		3098
2094	struct ev_loop *loop = ev_default_init (0);	3099	struct ev_loop *loop = ev_default_init (0);
2095	struct ev_loop *loop_socket = 0;	3100	struct ev_loop *loop_socket = 0;
2096	struct ev_embed embed;	3101	ev_embed embed;
2097		3102
2098	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)	3103	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
2099	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))	3104	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
2100	{	3105	{
2101	ev_embed_init (&embed, 0, loop_socket);	3106	ev_embed_init (&embed, 0, loop_socket);
2102	ev_embed_start (loop, &embed);	3107	ev_embed_start (loop, &embed);
2103	}	3108	}
2104		3109
2105	if (!loop_socket)	3110	if (!loop_socket)
2106	loop_socket = loop;	3111	loop_socket = loop;
2107		3112
2108	// now use loop_socket for all sockets, and loop for everything else	3113	// now use loop_socket for all sockets, and loop for everything else
2109		3114
2110		3115
2111	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	3116	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
2112		3117
2113	Fork watchers are called when a C<fork ()> was detected (usually because	3118	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
2116	event loop blocks next and before C<ev_check> watchers are being called,	3121	event loop blocks next and before C<ev_check> watchers are being called,
2117	and only in the child after the fork. If whoever good citizen calling	3122	and only in the child after the fork. If whoever good citizen calling
2118	C<ev_default_fork> cheats and calls it in the wrong process, the fork	3123	C<ev_default_fork> cheats and calls it in the wrong process, the fork
2119	handlers will be invoked, too, of course.	3124	handlers will be invoked, too, of course.
2120		3125
		3126	=head3 The special problem of life after fork - how is it possible?
		3127
		3128	Most uses of C<fork()> consist of forking, then some simple calls to set
		3129	up/change the process environment, followed by a call to C<exec()>. This
		3130	sequence should be handled by libev without any problems.
		3131
		3132	This changes when the application actually wants to do event handling
		3133	in the child, or both parent in child, in effect "continuing" after the
		3134	fork.
		3135
		3136	The default mode of operation (for libev, with application help to detect
		3137	forks) is to duplicate all the state in the child, as would be expected
		3138	when I<either> the parent I<or> the child process continues.
		3139
		3140	When both processes want to continue using libev, then this is usually the
		3141	wrong result. In that case, usually one process (typically the parent) is
		3142	supposed to continue with all watchers in place as before, while the other
		3143	process typically wants to start fresh, i.e. without any active watchers.
		3144
		3145	The cleanest and most efficient way to achieve that with libev is to
		3146	simply create a new event loop, which of course will be "empty", and
		3147	use that for new watchers. This has the advantage of not touching more
		3148	memory than necessary, and thus avoiding the copy-on-write, and the
		3149	disadvantage of having to use multiple event loops (which do not support
		3150	signal watchers).
		3151
		3152	When this is not possible, or you want to use the default loop for
		3153	other reasons, then in the process that wants to start "fresh", call
		3154	C<ev_loop_destroy (EV_DEFAULT)> followed by C<ev_default_loop (...)>.
		3155	Destroying the default loop will "orphan" (not stop) all registered
		3156	watchers, so you have to be careful not to execute code that modifies
		3157	those watchers. Note also that in that case, you have to re-register any
		3158	signal watchers.
		3159
2121	=head3 Watcher-Specific Functions and Data Members	3160	=head3 Watcher-Specific Functions and Data Members
2122		3161
2123	=over 4	3162	=over 4
2124		3163
2125	=item ev_fork_init (ev_signal *, callback)	3164	=item ev_fork_init (ev_fork *, callback)
2126		3165
2127	Initialises and configures the fork watcher - it has no parameters of any	3166	Initialises and configures the fork watcher - it has no parameters of any
2128	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,	3167	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
2129	believe me.	3168	really.
2130		3169
2131	=back	3170	=back
2132		3171
2133		3172
		3173	=head2 C<ev_cleanup> - even the best things end
		3174
		3175	Cleanup watchers are called just before the event loop is being destroyed
		3176	by a call to C<ev_loop_destroy>.
		3177
		3178	While there is no guarantee that the event loop gets destroyed, cleanup
		3179	watchers provide a convenient method to install cleanup hooks for your
		3180	program, worker threads and so on - you just to make sure to destroy the
		3181	loop when you want them to be invoked.
		3182
		3183	Cleanup watchers are invoked in the same way as any other watcher. Unlike
		3184	all other watchers, they do not keep a reference to the event loop (which
		3185	makes a lot of sense if you think about it). Like all other watchers, you
		3186	can call libev functions in the callback, except C<ev_cleanup_start>.
		3187
		3188	=head3 Watcher-Specific Functions and Data Members
		3189
		3190	=over 4
		3191
		3192	=item ev_cleanup_init (ev_cleanup *, callback)
		3193
		3194	Initialises and configures the cleanup watcher - it has no parameters of
		3195	any kind. There is a C<ev_cleanup_set> macro, but using it is utterly
		3196	pointless, I assure you.
		3197
		3198	=back
		3199
		3200	Example: Register an atexit handler to destroy the default loop, so any
		3201	cleanup functions are called.
		3202
		3203	static void
		3204	program_exits (void)
		3205	{
		3206	ev_loop_destroy (EV_DEFAULT_UC);
		3207	}
		3208
		3209	...
		3210	atexit (program_exits);
		3211
		3212
2134	=head2 C<ev_async> - how to wake up another event loop	3213	=head2 C<ev_async> - how to wake up an event loop
2135		3214
2136	In general, you cannot use an C<ev_loop> from multiple threads or other	3215	In general, you cannot use an C<ev_run> from multiple threads or other
2137	asynchronous sources such as signal handlers (as opposed to multiple event	3216	asynchronous sources such as signal handlers (as opposed to multiple event
2138	loops - those are of course safe to use in different threads).	3217	loops - those are of course safe to use in different threads).
2139		3218
2140	Sometimes, however, you need to wake up another event loop you do not	3219	Sometimes, however, you need to wake up an event loop you do not control,
2141	control, for example because it belongs to another thread. This is what	3220	for example because it belongs to another thread. This is what C<ev_async>
2142	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you	3221	watchers do: as long as the C<ev_async> watcher is active, you can signal
2143	can signal it by calling C<ev_async_send>, which is thread- and signal	3222	it by calling C<ev_async_send>, which is thread- and signal safe.
2144	safe.
2145		3223
2146	This functionality is very similar to C<ev_signal> watchers, as signals,	3224	This functionality is very similar to C<ev_signal> watchers, as signals,
2147	too, are asynchronous in nature, and signals, too, will be compressed	3225	too, are asynchronous in nature, and signals, too, will be compressed
2148	(i.e. the number of callback invocations may be less than the number of	3226	(i.e. the number of callback invocations may be less than the number of
2149	C<ev_async_sent> calls).	3227	C<ev_async_sent> calls). In fact, you could use signal watchers as a kind
		3228	of "global async watchers" by using a watcher on an otherwise unused
		3229	signal, and C<ev_feed_signal> to signal this watcher from another thread,
		3230	even without knowing which loop owns the signal.
2150		3231
2151	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not	3232	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
2152	just the default loop.	3233	just the default loop.
2153		3234
2154	=head3 Queueing	3235	=head3 Queueing
2155		3236
2156	C<ev_async> does not support queueing of data in any way. The reason	3237	C<ev_async> does not support queueing of data in any way. The reason
2157	is that the author does not know of a simple (or any) algorithm for a	3238	is that the author does not know of a simple (or any) algorithm for a
2158	multiple-writer-single-reader queue that works in all cases and doesn't	3239	multiple-writer-single-reader queue that works in all cases and doesn't
2159	need elaborate support such as pthreads.	3240	need elaborate support such as pthreads or unportable memory access
		3241	semantics.
2160		3242
2161	That means that if you want to queue data, you have to provide your own	3243	That means that if you want to queue data, you have to provide your own
2162	queue. But at least I can tell you would implement locking around your	3244	queue. But at least I can tell you how to implement locking around your
2163	queue:	3245	queue:
2164		3246
2165	=over 4	3247	=over 4
2166		3248
2167	=item queueing from a signal handler context	3249	=item queueing from a signal handler context
2168		3250
2169	To implement race-free queueing, you simply add to the queue in the signal	3251	To implement race-free queueing, you simply add to the queue in the signal
2170	handler but you block the signal handler in the watcher callback. Here is an example that does that for	3252	handler but you block the signal handler in the watcher callback. Here is
2171	some fictitiuous SIGUSR1 handler:	3253	an example that does that for some fictitious SIGUSR1 handler:
2172		3254
2173	static ev_async mysig;	3255	static ev_async mysig;
2174		3256
2175	static void	3257	static void
2176	sigusr1_handler (void)	3258	sigusr1_handler (void)
…		…
2242	=over 4	3324	=over 4
2243		3325
2244	=item ev_async_init (ev_async *, callback)	3326	=item ev_async_init (ev_async *, callback)
2245		3327
2246	Initialises and configures the async watcher - it has no parameters of any	3328	Initialises and configures the async watcher - it has no parameters of any
2247	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,	3329	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
2248	believe me.	3330	trust me.
2249		3331
2250	=item ev_async_send (loop, ev_async *)	3332	=item ev_async_send (loop, ev_async *)
2251		3333
2252	Sends/signals/activates the given C<ev_async> watcher, that is, feeds	3334	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
2253	an C<EV_ASYNC> event on the watcher into the event loop. Unlike	3335	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
2254	C<ev_feed_event>, this call is safe to do in other threads, signal or	3336	C<ev_feed_event>, this call is safe to do from other threads, signal or
2255	similar contexts (see the dicusssion of C<EV_ATOMIC_T> in the embedding	3337	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
2256	section below on what exactly this means).	3338	section below on what exactly this means).
2257		3339
		3340	Note that, as with other watchers in libev, multiple events might get
		3341	compressed into a single callback invocation (another way to look at this
		3342	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
		3343	reset when the event loop detects that).
		3344
2258	This call incurs the overhead of a syscall only once per loop iteration,	3345	This call incurs the overhead of a system call only once per event loop
2259	so while the overhead might be noticable, it doesn't apply to repeated	3346	iteration, so while the overhead might be noticeable, it doesn't apply to
2260	calls to C<ev_async_send>.	3347	repeated calls to C<ev_async_send> for the same event loop.
		3348
		3349	=item bool = ev_async_pending (ev_async *)
		3350
		3351	Returns a non-zero value when C<ev_async_send> has been called on the
		3352	watcher but the event has not yet been processed (or even noted) by the
		3353	event loop.
		3354
		3355	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		3356	the loop iterates next and checks for the watcher to have become active,
		3357	it will reset the flag again. C<ev_async_pending> can be used to very
		3358	quickly check whether invoking the loop might be a good idea.
		3359
		3360	Not that this does I<not> check whether the watcher itself is pending,
		3361	only whether it has been requested to make this watcher pending: there
		3362	is a time window between the event loop checking and resetting the async
		3363	notification, and the callback being invoked.
2261		3364
2262	=back	3365	=back
2263		3366
2264		3367
2265	=head1 OTHER FUNCTIONS	3368	=head1 OTHER FUNCTIONS
…		…
2269	=over 4	3372	=over 4
2270		3373
2271	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	3374	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2272		3375
2273	This function combines a simple timer and an I/O watcher, calls your	3376	This function combines a simple timer and an I/O watcher, calls your
2274	callback on whichever event happens first and automatically stop both	3377	callback on whichever event happens first and automatically stops both
2275	watchers. This is useful if you want to wait for a single event on an fd	3378	watchers. This is useful if you want to wait for a single event on an fd
2276	or timeout without having to allocate/configure/start/stop/free one or	3379	or timeout without having to allocate/configure/start/stop/free one or
2277	more watchers yourself.	3380	more watchers yourself.
2278		3381
2279	If C<fd> is less than 0, then no I/O watcher will be started and events	3382	If C<fd> is less than 0, then no I/O watcher will be started and the
2280	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	3383	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2281	C<events> set will be craeted and started.	3384	the given C<fd> and C<events> set will be created and started.
2282		3385
2283	If C<timeout> is less than 0, then no timeout watcher will be	3386	If C<timeout> is less than 0, then no timeout watcher will be
2284	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	3387	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
2285	repeat = 0) will be started. While C<0> is a valid timeout, it is of	3388	repeat = 0) will be started. C<0> is a valid timeout.
2286	dubious value.
2287		3389
2288	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	3390	The callback has the type C<void (*cb)(int revents, void *arg)> and is
2289	passed an C<revents> set like normal event callbacks (a combination of	3391	passed an C<revents> set like normal event callbacks (a combination of
2290	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	3392	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMER>) and the C<arg>
2291	value passed to C<ev_once>:	3393	value passed to C<ev_once>. Note that it is possible to receive I<both>
		3394	a timeout and an io event at the same time - you probably should give io
		3395	events precedence.
2292		3396
		3397	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		3398
2293	static void stdin_ready (int revents, void *arg)	3399	static void stdin_ready (int revents, void *arg)
2294	{	3400	{
2295	if (revents & EV_TIMEOUT)
2296	/* doh, nothing entered */;
2297	else if (revents & EV_READ)	3401	if (revents & EV_READ)
2298	/* stdin might have data for us, joy! */;	3402	/* stdin might have data for us, joy! */;
		3403	else if (revents & EV_TIMER)
		3404	/* doh, nothing entered */;
2299	}	3405	}
2300		3406
2301	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	3407	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2302		3408
2303	=item ev_feed_event (ev_loop *, watcher *, int revents)
2304
2305	Feeds the given event set into the event loop, as if the specified event
2306	had happened for the specified watcher (which must be a pointer to an
2307	initialised but not necessarily started event watcher).
2308
2309	=item ev_feed_fd_event (ev_loop *, int fd, int revents)	3409	=item ev_feed_fd_event (loop, int fd, int revents)
2310		3410
2311	Feed an event on the given fd, as if a file descriptor backend detected	3411	Feed an event on the given fd, as if a file descriptor backend detected
2312	the given events it.	3412	the given events it.
2313		3413
2314	=item ev_feed_signal_event (ev_loop *loop, int signum)	3414	=item ev_feed_signal_event (loop, int signum)
2315		3415
2316	Feed an event as if the given signal occured (C<loop> must be the default	3416	Feed an event as if the given signal occurred. See also C<ev_feed_signal>,
2317	loop!).	3417	which is async-safe.
		3418
		3419	=back
		3420
		3421
		3422	=head1 COMMON OR USEFUL IDIOMS (OR BOTH)
		3423
		3424	This section explains some common idioms that are not immediately
		3425	obvious. Note that examples are sprinkled over the whole manual, and this
		3426	section only contains stuff that wouldn't fit anywhere else.
		3427
		3428	=over 4
		3429
		3430	=item Model/nested event loop invocations and exit conditions.
		3431
		3432	Often (especially in GUI toolkits) there are places where you have
		3433	I<modal> interaction, which is most easily implemented by recursively
		3434	invoking C<ev_run>.
		3435
		3436	This brings the problem of exiting - a callback might want to finish the
		3437	main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but
		3438	a modal "Are you sure?" dialog is still waiting), or just the nested one
		3439	and not the main one (e.g. user clocked "Ok" in a modal dialog), or some
		3440	other combination: In these cases, C<ev_break> will not work alone.
		3441
		3442	The solution is to maintain "break this loop" variable for each C<ev_run>
		3443	invocation, and use a loop around C<ev_run> until the condition is
		3444	triggered, using C<EVRUN_ONCE>:
		3445
		3446	// main loop
		3447	int exit_main_loop = 0;
		3448
		3449	while (!exit_main_loop)
		3450	ev_run (EV_DEFAULT_ EVRUN_ONCE);
		3451
		3452	// in a model watcher
		3453	int exit_nested_loop = 0;
		3454
		3455	while (!exit_nested_loop)
		3456	ev_run (EV_A_ EVRUN_ONCE);
		3457
		3458	To exit from any of these loops, just set the corresponding exit variable:
		3459
		3460	// exit modal loop
		3461	exit_nested_loop = 1;
		3462
		3463	// exit main program, after modal loop is finished
		3464	exit_main_loop = 1;
		3465
		3466	// exit both
		3467	exit_main_loop = exit_nested_loop = 1;
2318		3468
2319	=back	3469	=back
2320		3470
2321		3471
2322	=head1 LIBEVENT EMULATION	3472	=head1 LIBEVENT EMULATION
2323		3473
2324	Libev offers a compatibility emulation layer for libevent. It cannot	3474	Libev offers a compatibility emulation layer for libevent. It cannot
2325	emulate the internals of libevent, so here are some usage hints:	3475	emulate the internals of libevent, so here are some usage hints:
2326		3476
2327	=over 4	3477	=over 4
		3478
		3479	=item * Only the libevent-1.4.1-beta API is being emulated.
		3480
		3481	This was the newest libevent version available when libev was implemented,
		3482	and is still mostly unchanged in 2010.
2328		3483
2329	=item * Use it by including <event.h>, as usual.	3484	=item * Use it by including <event.h>, as usual.
2330		3485
2331	=item * The following members are fully supported: ev_base, ev_callback,	3486	=item * The following members are fully supported: ev_base, ev_callback,
2332	ev_arg, ev_fd, ev_res, ev_events.	3487	ev_arg, ev_fd, ev_res, ev_events.
…		…
2337		3492
2338	=item * Priorities are not currently supported. Initialising priorities	3493	=item * Priorities are not currently supported. Initialising priorities
2339	will fail and all watchers will have the same priority, even though there	3494	will fail and all watchers will have the same priority, even though there
2340	is an ev_pri field.	3495	is an ev_pri field.
2341		3496
		3497	=item * In libevent, the last base created gets the signals, in libev, the
		3498	base that registered the signal gets the signals.
		3499
2342	=item * Other members are not supported.	3500	=item * Other members are not supported.
2343		3501
2344	=item * The libev emulation is I<not> ABI compatible to libevent, you need	3502	=item * The libev emulation is I<not> ABI compatible to libevent, you need
2345	to use the libev header file and library.	3503	to use the libev header file and library.
2346		3504
2347	=back	3505	=back
2348		3506
2349	=head1 C++ SUPPORT	3507	=head1 C++ SUPPORT
2350		3508
2351	Libev comes with some simplistic wrapper classes for C++ that mainly allow	3509	Libev comes with some simplistic wrapper classes for C++ that mainly allow
2352	you to use some convinience methods to start/stop watchers and also change	3510	you to use some convenience methods to start/stop watchers and also change
2353	the callback model to a model using method callbacks on objects.	3511	the callback model to a model using method callbacks on objects.
2354		3512
2355	To use it,	3513	To use it,
2356		3514
2357	#include <ev++.h>	3515	#include <ev++.h>
2358		3516
2359	This automatically includes F<ev.h> and puts all of its definitions (many	3517	This automatically includes F<ev.h> and puts all of its definitions (many
2360	of them macros) into the global namespace. All C++ specific things are	3518	of them macros) into the global namespace. All C++ specific things are
2361	put into the C<ev> namespace. It should support all the same embedding	3519	put into the C<ev> namespace. It should support all the same embedding
2362	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.	3520	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
…		…
2364	Care has been taken to keep the overhead low. The only data member the C++	3522	Care has been taken to keep the overhead low. The only data member the C++
2365	classes add (compared to plain C-style watchers) is the event loop pointer	3523	classes add (compared to plain C-style watchers) is the event loop pointer
2366	that the watcher is associated with (or no additional members at all if	3524	that the watcher is associated with (or no additional members at all if
2367	you disable C<EV_MULTIPLICITY> when embedding libev).	3525	you disable C<EV_MULTIPLICITY> when embedding libev).
2368		3526
2369	Currently, functions, and static and non-static member functions can be	3527	Currently, functions, static and non-static member functions and classes
2370	used as callbacks. Other types should be easy to add as long as they only	3528	with C<operator ()> can be used as callbacks. Other types should be easy
2371	need one additional pointer for context. If you need support for other	3529	to add as long as they only need one additional pointer for context. If
2372	types of functors please contact the author (preferably after implementing	3530	you need support for other types of functors please contact the author
2373	it).	3531	(preferably after implementing it).
2374		3532
2375	Here is a list of things available in the C<ev> namespace:	3533	Here is a list of things available in the C<ev> namespace:
2376		3534
2377	=over 4	3535	=over 4
2378		3536
…		…
2396		3554
2397	=over 4	3555	=over 4
2398		3556
2399	=item ev::TYPE::TYPE ()	3557	=item ev::TYPE::TYPE ()
2400		3558
2401	=item ev::TYPE::TYPE (struct ev_loop *)	3559	=item ev::TYPE::TYPE (loop)
2402		3560
2403	=item ev::TYPE::~TYPE	3561	=item ev::TYPE::~TYPE
2404		3562
2405	The constructor (optionally) takes an event loop to associate the watcher	3563	The constructor (optionally) takes an event loop to associate the watcher
2406	with. If it is omitted, it will use C<EV_DEFAULT>.	3564	with. If it is omitted, it will use C<EV_DEFAULT>.
…		…
2429	your compiler is good :), then the method will be fully inlined into the	3587	your compiler is good :), then the method will be fully inlined into the
2430	thunking function, making it as fast as a direct C callback.	3588	thunking function, making it as fast as a direct C callback.
2431		3589
2432	Example: simple class declaration and watcher initialisation	3590	Example: simple class declaration and watcher initialisation
2433		3591
2434	struct myclass	3592	struct myclass
2435	{	3593	{
2436	void io_cb (ev::io &w, int revents) { }	3594	void io_cb (ev::io &w, int revents) { }
2437	}	3595	}
2438		3596
2439	myclass obj;	3597	myclass obj;
2440	ev::io iow;	3598	ev::io iow;
2441	iow.set <myclass, &myclass::io_cb> (&obj);	3599	iow.set <myclass, &myclass::io_cb> (&obj);
		3600
		3601	=item w->set (object *)
		3602
		3603	This is a variation of a method callback - leaving out the method to call
		3604	will default the method to C<operator ()>, which makes it possible to use
		3605	functor objects without having to manually specify the C<operator ()> all
		3606	the time. Incidentally, you can then also leave out the template argument
		3607	list.
		3608
		3609	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3610	int revents)>.
		3611
		3612	See the method-C<set> above for more details.
		3613
		3614	Example: use a functor object as callback.
		3615
		3616	struct myfunctor
		3617	{
		3618	void operator() (ev::io &w, int revents)
		3619	{
		3620	...
		3621	}
		3622	}
		3623
		3624	myfunctor f;
		3625
		3626	ev::io w;
		3627	w.set (&f);
2442		3628
2443	=item w->set<function> (void *data = 0)	3629	=item w->set<function> (void *data = 0)
2444		3630
2445	Also sets a callback, but uses a static method or plain function as	3631	Also sets a callback, but uses a static method or plain function as
2446	callback. The optional C<data> argument will be stored in the watcher's	3632	callback. The optional C<data> argument will be stored in the watcher's
…		…
2448		3634
2449	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.	3635	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
2450		3636
2451	See the method-C<set> above for more details.	3637	See the method-C<set> above for more details.
2452		3638
2453	Example:	3639	Example: Use a plain function as callback.
2454		3640
2455	static void io_cb (ev::io &w, int revents) { }	3641	static void io_cb (ev::io &w, int revents) { }
2456	iow.set <io_cb> ();	3642	iow.set <io_cb> ();
2457		3643
2458	=item w->set (struct ev_loop *)	3644	=item w->set (loop)
2459		3645
2460	Associates a different C<struct ev_loop> with this watcher. You can only	3646	Associates a different C<struct ev_loop> with this watcher. You can only
2461	do this when the watcher is inactive (and not pending either).	3647	do this when the watcher is inactive (and not pending either).
2462		3648
2463	=item w->set ([args])	3649	=item w->set ([arguments])
2464		3650
2465	Basically the same as C<ev_TYPE_set>, with the same args. Must be	3651	Basically the same as C<ev_TYPE_set>, with the same arguments. Either this
2466	called at least once. Unlike the C counterpart, an active watcher gets	3652	method or a suitable start method must be called at least once. Unlike the
2467	automatically stopped and restarted when reconfiguring it with this	3653	C counterpart, an active watcher gets automatically stopped and restarted
2468	method.	3654	when reconfiguring it with this method.
2469		3655
2470	=item w->start ()	3656	=item w->start ()
2471		3657
2472	Starts the watcher. Note that there is no C<loop> argument, as the	3658	Starts the watcher. Note that there is no C<loop> argument, as the
2473	constructor already stores the event loop.	3659	constructor already stores the event loop.
2474		3660
		3661	=item w->start ([arguments])
		3662
		3663	Instead of calling C<set> and C<start> methods separately, it is often
		3664	convenient to wrap them in one call. Uses the same type of arguments as
		3665	the configure C<set> method of the watcher.
		3666
2475	=item w->stop ()	3667	=item w->stop ()
2476		3668
2477	Stops the watcher if it is active. Again, no C<loop> argument.	3669	Stops the watcher if it is active. Again, no C<loop> argument.
2478		3670
2479	=item w->again () (C<ev::timer>, C<ev::periodic> only)	3671	=item w->again () (C<ev::timer>, C<ev::periodic> only)
…		…
2491		3683
2492	=back	3684	=back
2493		3685
2494	=back	3686	=back
2495		3687
2496	Example: Define a class with an IO and idle watcher, start one of them in	3688	Example: Define a class with two I/O and idle watchers, start the I/O
2497	the constructor.	3689	watchers in the constructor.
2498		3690
2499	class myclass	3691	class myclass
2500	{	3692	{
2501	ev::io io; void io_cb (ev::io &w, int revents);	3693	ev::io io ; void io_cb (ev::io &w, int revents);
		3694	ev::io2 io2 ; void io2_cb (ev::io &w, int revents);
2502	ev:idle idle void idle_cb (ev::idle &w, int revents);	3695	ev::idle idle; void idle_cb (ev::idle &w, int revents);
2503		3696
2504	myclass (int fd)	3697	myclass (int fd)
2505	{	3698	{
2506	io .set <myclass, &myclass::io_cb > (this);	3699	io .set <myclass, &myclass::io_cb > (this);
		3700	io2 .set <myclass, &myclass::io2_cb > (this);
2507	idle.set <myclass, &myclass::idle_cb> (this);	3701	idle.set <myclass, &myclass::idle_cb> (this);
2508		3702
2509	io.start (fd, ev::READ);	3703	io.set (fd, ev::WRITE); // configure the watcher
		3704	io.start (); // start it whenever convenient
		3705
		3706	io2.start (fd, ev::READ); // set + start in one call
2510	}	3707	}
2511	};	3708	};
2512		3709
2513		3710
2514	=head1 OTHER LANGUAGE BINDINGS	3711	=head1 OTHER LANGUAGE BINDINGS
2515		3712
2516	Libev does not offer other language bindings itself, but bindings for a	3713	Libev does not offer other language bindings itself, but bindings for a
2517	numbe rof languages exist in the form of third-party packages. If you know	3714	number of languages exist in the form of third-party packages. If you know
2518	any interesting language binding in addition to the ones listed here, drop	3715	any interesting language binding in addition to the ones listed here, drop
2519	me a note.	3716	me a note.
2520		3717
2521	=over 4	3718	=over 4
2522		3719
2523	=item Perl	3720	=item Perl
2524		3721
2525	The EV module implements the full libev API and is actually used to test	3722	The EV module implements the full libev API and is actually used to test
2526	libev. EV is developed together with libev. Apart from the EV core module,	3723	libev. EV is developed together with libev. Apart from the EV core module,
2527	there are additional modules that implement libev-compatible interfaces	3724	there are additional modules that implement libev-compatible interfaces
2528	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the	3725	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
2529	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).	3726	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3727	and C<EV::Glib>).
2530		3728
2531	It can be found and installed via CPAN, its homepage is found at	3729	It can be found and installed via CPAN, its homepage is at
2532	L<http://software.schmorp.de/pkg/EV>.	3730	L<http://software.schmorp.de/pkg/EV>.
2533		3731
		3732	=item Python
		3733
		3734	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		3735	seems to be quite complete and well-documented.
		3736
2534	=item Ruby	3737	=item Ruby
2535		3738
2536	Tony Arcieri has written a ruby extension that offers access to a subset	3739	Tony Arcieri has written a ruby extension that offers access to a subset
2537	of the libev API and adds filehandle abstractions, asynchronous DNS and	3740	of the libev API and adds file handle abstractions, asynchronous DNS and
2538	more on top of it. It can be found via gem servers. Its homepage is at	3741	more on top of it. It can be found via gem servers. Its homepage is at
2539	L<http://rev.rubyforge.org/>.	3742	L<http://rev.rubyforge.org/>.
2540		3743
		3744	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		3745	makes rev work even on mingw.
		3746
		3747	=item Haskell
		3748
		3749	A haskell binding to libev is available at
		3750	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		3751
2541	=item D	3752	=item D
2542		3753
2543	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to	3754	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
2544	be found at L<http://git.llucax.com.ar/?p=software/ev.d.git;a=summary>.	3755	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		3756
		3757	=item Ocaml
		3758
		3759	Erkki Seppala has written Ocaml bindings for libev, to be found at
		3760	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		3761
		3762	=item Lua
		3763
		3764	Brian Maher has written a partial interface to libev for lua (at the
		3765	time of this writing, only C<ev_io> and C<ev_timer>), to be found at
		3766	L<http://github.com/brimworks/lua-ev>.
2545		3767
2546	=back	3768	=back
2547		3769
2548		3770
2549	=head1 MACRO MAGIC	3771	=head1 MACRO MAGIC
2550		3772
2551	Libev can be compiled with a variety of options, the most fundamantal	3773	Libev can be compiled with a variety of options, the most fundamental
2552	of which is C<EV_MULTIPLICITY>. This option determines whether (most)	3774	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
2553	functions and callbacks have an initial C<struct ev_loop *> argument.	3775	functions and callbacks have an initial C<struct ev_loop *> argument.
2554		3776
2555	To make it easier to write programs that cope with either variant, the	3777	To make it easier to write programs that cope with either variant, the
2556	following macros are defined:	3778	following macros are defined:
…		…
2561		3783
2562	This provides the loop I<argument> for functions, if one is required ("ev	3784	This provides the loop I<argument> for functions, if one is required ("ev
2563	loop argument"). The C<EV_A> form is used when this is the sole argument,	3785	loop argument"). The C<EV_A> form is used when this is the sole argument,
2564	C<EV_A_> is used when other arguments are following. Example:	3786	C<EV_A_> is used when other arguments are following. Example:
2565		3787
2566	ev_unref (EV_A);	3788	ev_unref (EV_A);
2567	ev_timer_add (EV_A_ watcher);	3789	ev_timer_add (EV_A_ watcher);
2568	ev_loop (EV_A_ 0);	3790	ev_run (EV_A_ 0);
2569		3791
2570	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	3792	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
2571	which is often provided by the following macro.	3793	which is often provided by the following macro.
2572		3794
2573	=item C<EV_P>, C<EV_P_>	3795	=item C<EV_P>, C<EV_P_>
2574		3796
2575	This provides the loop I<parameter> for functions, if one is required ("ev	3797	This provides the loop I<parameter> for functions, if one is required ("ev
2576	loop parameter"). The C<EV_P> form is used when this is the sole parameter,	3798	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
2577	C<EV_P_> is used when other parameters are following. Example:	3799	C<EV_P_> is used when other parameters are following. Example:
2578		3800
2579	// this is how ev_unref is being declared	3801	// this is how ev_unref is being declared
2580	static void ev_unref (EV_P);	3802	static void ev_unref (EV_P);
2581		3803
2582	// this is how you can declare your typical callback	3804	// this is how you can declare your typical callback
2583	static void cb (EV_P_ ev_timer *w, int revents)	3805	static void cb (EV_P_ ev_timer *w, int revents)
2584		3806
2585	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	3807	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
2586	suitable for use with C<EV_A>.	3808	suitable for use with C<EV_A>.
2587		3809
2588	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	3810	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
2589		3811
2590	Similar to the other two macros, this gives you the value of the default	3812	Similar to the other two macros, this gives you the value of the default
2591	loop, if multiple loops are supported ("ev loop default").	3813	loop, if multiple loops are supported ("ev loop default").
		3814
		3815	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		3816
		3817	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		3818	default loop has been initialised (C<UC> == unchecked). Their behaviour
		3819	is undefined when the default loop has not been initialised by a previous
		3820	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		3821
		3822	It is often prudent to use C<EV_DEFAULT> when initialising the first
		3823	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
2592		3824
2593	=back	3825	=back
2594		3826
2595	Example: Declare and initialise a check watcher, utilising the above	3827	Example: Declare and initialise a check watcher, utilising the above
2596	macros so it will work regardless of whether multiple loops are supported	3828	macros so it will work regardless of whether multiple loops are supported
2597	or not.	3829	or not.
2598		3830
2599	static void	3831	static void
2600	check_cb (EV_P_ ev_timer *w, int revents)	3832	check_cb (EV_P_ ev_timer *w, int revents)
2601	{	3833	{
2602	ev_check_stop (EV_A_ w);	3834	ev_check_stop (EV_A_ w);
2603	}	3835	}
2604		3836
2605	ev_check check;	3837	ev_check check;
2606	ev_check_init (&check, check_cb);	3838	ev_check_init (&check, check_cb);
2607	ev_check_start (EV_DEFAULT_ &check);	3839	ev_check_start (EV_DEFAULT_ &check);
2608	ev_loop (EV_DEFAULT_ 0);	3840	ev_run (EV_DEFAULT_ 0);
2609		3841
2610	=head1 EMBEDDING	3842	=head1 EMBEDDING
2611		3843
2612	Libev can (and often is) directly embedded into host	3844	Libev can (and often is) directly embedded into host
2613	applications. Examples of applications that embed it include the Deliantra	3845	applications. Examples of applications that embed it include the Deliantra
…		…
2620	libev somewhere in your source tree).	3852	libev somewhere in your source tree).
2621		3853
2622	=head2 FILESETS	3854	=head2 FILESETS
2623		3855
2624	Depending on what features you need you need to include one or more sets of files	3856	Depending on what features you need you need to include one or more sets of files
2625	in your app.	3857	in your application.
2626		3858
2627	=head3 CORE EVENT LOOP	3859	=head3 CORE EVENT LOOP
2628		3860
2629	To include only the libev core (all the C<ev_*> functions), with manual	3861	To include only the libev core (all the C<ev_*> functions), with manual
2630	configuration (no autoconf):	3862	configuration (no autoconf):
2631		3863
2632	#define EV_STANDALONE 1	3864	#define EV_STANDALONE 1
2633	#include "ev.c"	3865	#include "ev.c"
2634		3866
2635	This will automatically include F<ev.h>, too, and should be done in a	3867	This will automatically include F<ev.h>, too, and should be done in a
2636	single C source file only to provide the function implementations. To use	3868	single C source file only to provide the function implementations. To use
2637	it, do the same for F<ev.h> in all files wishing to use this API (best	3869	it, do the same for F<ev.h> in all files wishing to use this API (best
2638	done by writing a wrapper around F<ev.h> that you can include instead and	3870	done by writing a wrapper around F<ev.h> that you can include instead and
2639	where you can put other configuration options):	3871	where you can put other configuration options):
2640		3872
2641	#define EV_STANDALONE 1	3873	#define EV_STANDALONE 1
2642	#include "ev.h"	3874	#include "ev.h"
2643		3875
2644	Both header files and implementation files can be compiled with a C++	3876	Both header files and implementation files can be compiled with a C++
2645	compiler (at least, thats a stated goal, and breakage will be treated	3877	compiler (at least, that's a stated goal, and breakage will be treated
2646	as a bug).	3878	as a bug).
2647		3879
2648	You need the following files in your source tree, or in a directory	3880	You need the following files in your source tree, or in a directory
2649	in your include path (e.g. in libev/ when using -Ilibev):	3881	in your include path (e.g. in libev/ when using -Ilibev):
2650		3882
2651	ev.h	3883	ev.h
2652	ev.c	3884	ev.c
2653	ev_vars.h	3885	ev_vars.h
2654	ev_wrap.h	3886	ev_wrap.h
2655		3887
2656	ev_win32.c required on win32 platforms only	3888	ev_win32.c required on win32 platforms only
2657		3889
2658	ev_select.c only when select backend is enabled (which is enabled by default)	3890	ev_select.c only when select backend is enabled (which is enabled by default)
2659	ev_poll.c only when poll backend is enabled (disabled by default)	3891	ev_poll.c only when poll backend is enabled (disabled by default)
2660	ev_epoll.c only when the epoll backend is enabled (disabled by default)	3892	ev_epoll.c only when the epoll backend is enabled (disabled by default)
2661	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	3893	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
2662	ev_port.c only when the solaris port backend is enabled (disabled by default)	3894	ev_port.c only when the solaris port backend is enabled (disabled by default)
2663		3895
2664	F<ev.c> includes the backend files directly when enabled, so you only need	3896	F<ev.c> includes the backend files directly when enabled, so you only need
2665	to compile this single file.	3897	to compile this single file.
2666		3898
2667	=head3 LIBEVENT COMPATIBILITY API	3899	=head3 LIBEVENT COMPATIBILITY API
2668		3900
2669	To include the libevent compatibility API, also include:	3901	To include the libevent compatibility API, also include:
2670		3902
2671	#include "event.c"	3903	#include "event.c"
2672		3904
2673	in the file including F<ev.c>, and:	3905	in the file including F<ev.c>, and:
2674		3906
2675	#include "event.h"	3907	#include "event.h"
2676		3908
2677	in the files that want to use the libevent API. This also includes F<ev.h>.	3909	in the files that want to use the libevent API. This also includes F<ev.h>.
2678		3910
2679	You need the following additional files for this:	3911	You need the following additional files for this:
2680		3912
2681	event.h	3913	event.h
2682	event.c	3914	event.c
2683		3915
2684	=head3 AUTOCONF SUPPORT	3916	=head3 AUTOCONF SUPPORT
2685		3917
2686	Instead of using C<EV_STANDALONE=1> and providing your config in	3918	Instead of using C<EV_STANDALONE=1> and providing your configuration in
2687	whatever way you want, you can also C<m4_include([libev.m4])> in your	3919	whatever way you want, you can also C<m4_include([libev.m4])> in your
2688	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then	3920	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
2689	include F<config.h> and configure itself accordingly.	3921	include F<config.h> and configure itself accordingly.
2690		3922
2691	For this of course you need the m4 file:	3923	For this of course you need the m4 file:
2692		3924
2693	libev.m4	3925	libev.m4
2694		3926
2695	=head2 PREPROCESSOR SYMBOLS/MACROS	3927	=head2 PREPROCESSOR SYMBOLS/MACROS
2696		3928
2697	Libev can be configured via a variety of preprocessor symbols you have to define	3929	Libev can be configured via a variety of preprocessor symbols you have to
2698	before including any of its files. The default is not to build for multiplicity	3930	define before including (or compiling) any of its files. The default in
2699	and only include the select backend.	3931	the absence of autoconf is documented for every option.
		3932
		3933	Symbols marked with "(h)" do not change the ABI, and can have different
		3934	values when compiling libev vs. including F<ev.h>, so it is permissible
		3935	to redefine them before including F<ev.h> without breaking compatibility
		3936	to a compiled library. All other symbols change the ABI, which means all
		3937	users of libev and the libev code itself must be compiled with compatible
		3938	settings.
2700		3939
2701	=over 4	3940	=over 4
2702		3941
		3942	=item EV_COMPAT3 (h)
		3943
		3944	Backwards compatibility is a major concern for libev. This is why this
		3945	release of libev comes with wrappers for the functions and symbols that
		3946	have been renamed between libev version 3 and 4.
		3947
		3948	You can disable these wrappers (to test compatibility with future
		3949	versions) by defining C<EV_COMPAT3> to C<0> when compiling your
		3950	sources. This has the additional advantage that you can drop the C<struct>
		3951	from C<struct ev_loop> declarations, as libev will provide an C<ev_loop>
		3952	typedef in that case.
		3953
		3954	In some future version, the default for C<EV_COMPAT3> will become C<0>,
		3955	and in some even more future version the compatibility code will be
		3956	removed completely.
		3957
2703	=item EV_STANDALONE	3958	=item EV_STANDALONE (h)
2704		3959
2705	Must always be C<1> if you do not use autoconf configuration, which	3960	Must always be C<1> if you do not use autoconf configuration, which
2706	keeps libev from including F<config.h>, and it also defines dummy	3961	keeps libev from including F<config.h>, and it also defines dummy
2707	implementations for some libevent functions (such as logging, which is not	3962	implementations for some libevent functions (such as logging, which is not
2708	supported). It will also not define any of the structs usually found in	3963	supported). It will also not define any of the structs usually found in
2709	F<event.h> that are not directly supported by the libev core alone.	3964	F<event.h> that are not directly supported by the libev core alone.
2710		3965
		3966	In standalone mode, libev will still try to automatically deduce the
		3967	configuration, but has to be more conservative.
		3968
2711	=item EV_USE_MONOTONIC	3969	=item EV_USE_MONOTONIC
2712		3970
2713	If defined to be C<1>, libev will try to detect the availability of the	3971	If defined to be C<1>, libev will try to detect the availability of the
2714	monotonic clock option at both compiletime and runtime. Otherwise no use	3972	monotonic clock option at both compile time and runtime. Otherwise no
2715	of the monotonic clock option will be attempted. If you enable this, you	3973	use of the monotonic clock option will be attempted. If you enable this,
2716	usually have to link against librt or something similar. Enabling it when	3974	you usually have to link against librt or something similar. Enabling it
2717	the functionality isn't available is safe, though, although you have	3975	when the functionality isn't available is safe, though, although you have
2718	to make sure you link against any libraries where the C<clock_gettime>	3976	to make sure you link against any libraries where the C<clock_gettime>
2719	function is hiding in (often F<-lrt>).	3977	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
2720		3978
2721	=item EV_USE_REALTIME	3979	=item EV_USE_REALTIME
2722		3980
2723	If defined to be C<1>, libev will try to detect the availability of the	3981	If defined to be C<1>, libev will try to detect the availability of the
2724	realtime clock option at compiletime (and assume its availability at	3982	real-time clock option at compile time (and assume its availability
2725	runtime if successful). Otherwise no use of the realtime clock option will	3983	at runtime if successful). Otherwise no use of the real-time clock
2726	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	3984	option will be attempted. This effectively replaces C<gettimeofday>
2727	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the	3985	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
2728	note about libraries in the description of C<EV_USE_MONOTONIC>, though.	3986	correctness. See the note about libraries in the description of
		3987	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		3988	C<EV_USE_CLOCK_SYSCALL>.
		3989
		3990	=item EV_USE_CLOCK_SYSCALL
		3991
		3992	If defined to be C<1>, libev will try to use a direct syscall instead
		3993	of calling the system-provided C<clock_gettime> function. This option
		3994	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		3995	unconditionally pulls in C<libpthread>, slowing down single-threaded
		3996	programs needlessly. Using a direct syscall is slightly slower (in
		3997	theory), because no optimised vdso implementation can be used, but avoids
		3998	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		3999	higher, as it simplifies linking (no need for C<-lrt>).
2729		4000
2730	=item EV_USE_NANOSLEEP	4001	=item EV_USE_NANOSLEEP
2731		4002
2732	If defined to be C<1>, libev will assume that C<nanosleep ()> is available	4003	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
2733	and will use it for delays. Otherwise it will use C<select ()>.	4004	and will use it for delays. Otherwise it will use C<select ()>.
2734		4005
		4006	=item EV_USE_EVENTFD
		4007
		4008	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		4009	available and will probe for kernel support at runtime. This will improve
		4010	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		4011	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		4012	2.7 or newer, otherwise disabled.
		4013
2735	=item EV_USE_SELECT	4014	=item EV_USE_SELECT
2736		4015
2737	If undefined or defined to be C<1>, libev will compile in support for the	4016	If undefined or defined to be C<1>, libev will compile in support for the
2738	C<select>(2) backend. No attempt at autodetection will be done: if no	4017	C<select>(2) backend. No attempt at auto-detection will be done: if no
2739	other method takes over, select will be it. Otherwise the select backend	4018	other method takes over, select will be it. Otherwise the select backend
2740	will not be compiled in.	4019	will not be compiled in.
2741		4020
2742	=item EV_SELECT_USE_FD_SET	4021	=item EV_SELECT_USE_FD_SET
2743		4022
2744	If defined to C<1>, then the select backend will use the system C<fd_set>	4023	If defined to C<1>, then the select backend will use the system C<fd_set>
2745	structure. This is useful if libev doesn't compile due to a missing	4024	structure. This is useful if libev doesn't compile due to a missing
2746	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on	4025	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
2747	exotic systems. This usually limits the range of file descriptors to some	4026	on exotic systems. This usually limits the range of file descriptors to
2748	low limit such as 1024 or might have other limitations (winsocket only	4027	some low limit such as 1024 or might have other limitations (winsocket
2749	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	4028	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
2750	influence the size of the C<fd_set> used.	4029	configures the maximum size of the C<fd_set>.
2751		4030
2752	=item EV_SELECT_IS_WINSOCKET	4031	=item EV_SELECT_IS_WINSOCKET
2753		4032
2754	When defined to C<1>, the select backend will assume that	4033	When defined to C<1>, the select backend will assume that
2755	select/socket/connect etc. don't understand file descriptors but	4034	select/socket/connect etc. don't understand file descriptors but
…		…
2757	be used is the winsock select). This means that it will call	4036	be used is the winsock select). This means that it will call
2758	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	4037	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
2759	it is assumed that all these functions actually work on fds, even	4038	it is assumed that all these functions actually work on fds, even
2760	on win32. Should not be defined on non-win32 platforms.	4039	on win32. Should not be defined on non-win32 platforms.
2761		4040
2762	=item EV_FD_TO_WIN32_HANDLE	4041	=item EV_FD_TO_WIN32_HANDLE(fd)
2763		4042
2764	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map	4043	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
2765	file descriptors to socket handles. When not defining this symbol (the	4044	file descriptors to socket handles. When not defining this symbol (the
2766	default), then libev will call C<_get_osfhandle>, which is usually	4045	default), then libev will call C<_get_osfhandle>, which is usually
2767	correct. In some cases, programs use their own file descriptor management,	4046	correct. In some cases, programs use their own file descriptor management,
2768	in which case they can provide this function to map fds to socket handles.	4047	in which case they can provide this function to map fds to socket handles.
2769		4048
		4049	=item EV_WIN32_HANDLE_TO_FD(handle)
		4050
		4051	If C<EV_SELECT_IS_WINSOCKET> then libev maps handles to file descriptors
		4052	using the standard C<_open_osfhandle> function. For programs implementing
		4053	their own fd to handle mapping, overwriting this function makes it easier
		4054	to do so. This can be done by defining this macro to an appropriate value.
		4055
		4056	=item EV_WIN32_CLOSE_FD(fd)
		4057
		4058	If programs implement their own fd to handle mapping on win32, then this
		4059	macro can be used to override the C<close> function, useful to unregister
		4060	file descriptors again. Note that the replacement function has to close
		4061	the underlying OS handle.
		4062
2770	=item EV_USE_POLL	4063	=item EV_USE_POLL
2771		4064
2772	If defined to be C<1>, libev will compile in support for the C<poll>(2)	4065	If defined to be C<1>, libev will compile in support for the C<poll>(2)
2773	backend. Otherwise it will be enabled on non-win32 platforms. It	4066	backend. Otherwise it will be enabled on non-win32 platforms. It
2774	takes precedence over select.	4067	takes precedence over select.
2775		4068
2776	=item EV_USE_EPOLL	4069	=item EV_USE_EPOLL
2777		4070
2778	If defined to be C<1>, libev will compile in support for the Linux	4071	If defined to be C<1>, libev will compile in support for the Linux
2779	C<epoll>(7) backend. Its availability will be detected at runtime,	4072	C<epoll>(7) backend. Its availability will be detected at runtime,
2780	otherwise another method will be used as fallback. This is the	4073	otherwise another method will be used as fallback. This is the preferred
2781	preferred backend for GNU/Linux systems.	4074	backend for GNU/Linux systems. If undefined, it will be enabled if the
		4075	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2782		4076
2783	=item EV_USE_KQUEUE	4077	=item EV_USE_KQUEUE
2784		4078
2785	If defined to be C<1>, libev will compile in support for the BSD style	4079	If defined to be C<1>, libev will compile in support for the BSD style
2786	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	4080	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
2799	otherwise another method will be used as fallback. This is the preferred	4093	otherwise another method will be used as fallback. This is the preferred
2800	backend for Solaris 10 systems.	4094	backend for Solaris 10 systems.
2801		4095
2802	=item EV_USE_DEVPOLL	4096	=item EV_USE_DEVPOLL
2803		4097
2804	reserved for future expansion, works like the USE symbols above.	4098	Reserved for future expansion, works like the USE symbols above.
2805		4099
2806	=item EV_USE_INOTIFY	4100	=item EV_USE_INOTIFY
2807		4101
2808	If defined to be C<1>, libev will compile in support for the Linux inotify	4102	If defined to be C<1>, libev will compile in support for the Linux inotify
2809	interface to speed up C<ev_stat> watchers. Its actual availability will	4103	interface to speed up C<ev_stat> watchers. Its actual availability will
2810	be detected at runtime.	4104	be detected at runtime. If undefined, it will be enabled if the headers
		4105	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2811		4106
2812	=item EV_ATOMIC_T	4107	=item EV_ATOMIC_T
2813		4108
2814	Libev requires an integer type (suitable for storing C<0> or C<1>) whose	4109	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
2815	access is atomic with respect to other threads or signal contexts. No such	4110	access is atomic with respect to other threads or signal contexts. No such
2816	type is easily found in the C language, so you can provide your own type	4111	type is easily found in the C language, so you can provide your own type
2817	that you know is safe for your purposes. It is used both for signal handler "locking"	4112	that you know is safe for your purposes. It is used both for signal handler "locking"
2818	as well as for signal and thread safety in C<ev_async> watchers.	4113	as well as for signal and thread safety in C<ev_async> watchers.
2819		4114
2820	In the absense of this define, libev will use C<sig_atomic_t volatile>	4115	In the absence of this define, libev will use C<sig_atomic_t volatile>
2821	(from F<signal.h>), which is usually good enough on most platforms.	4116	(from F<signal.h>), which is usually good enough on most platforms.
2822		4117
2823	=item EV_H	4118	=item EV_H (h)
2824		4119
2825	The name of the F<ev.h> header file used to include it. The default if	4120	The name of the F<ev.h> header file used to include it. The default if
2826	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be	4121	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2827	used to virtually rename the F<ev.h> header file in case of conflicts.	4122	used to virtually rename the F<ev.h> header file in case of conflicts.
2828		4123
2829	=item EV_CONFIG_H	4124	=item EV_CONFIG_H (h)
2830		4125
2831	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	4126	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2832	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	4127	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2833	C<EV_H>, above.	4128	C<EV_H>, above.
2834		4129
2835	=item EV_EVENT_H	4130	=item EV_EVENT_H (h)
2836		4131
2837	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	4132	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2838	of how the F<event.h> header can be found, the default is C<"event.h">.	4133	of how the F<event.h> header can be found, the default is C<"event.h">.
2839		4134
2840	=item EV_PROTOTYPES	4135	=item EV_PROTOTYPES (h)
2841		4136
2842	If defined to be C<0>, then F<ev.h> will not define any function	4137	If defined to be C<0>, then F<ev.h> will not define any function
2843	prototypes, but still define all the structs and other symbols. This is	4138	prototypes, but still define all the structs and other symbols. This is
2844	occasionally useful if you want to provide your own wrapper functions	4139	occasionally useful if you want to provide your own wrapper functions
2845	around libev functions.	4140	around libev functions.
…		…
2864	When doing priority-based operations, libev usually has to linearly search	4159	When doing priority-based operations, libev usually has to linearly search
2865	all the priorities, so having many of them (hundreds) uses a lot of space	4160	all the priorities, so having many of them (hundreds) uses a lot of space
2866	and time, so using the defaults of five priorities (-2 .. +2) is usually	4161	and time, so using the defaults of five priorities (-2 .. +2) is usually
2867	fine.	4162	fine.
2868		4163
2869	If your embedding app does not need any priorities, defining these both to	4164	If your embedding application does not need any priorities, defining these
2870	C<0> will save some memory and cpu.	4165	both to C<0> will save some memory and CPU.
2871		4166
2872	=item EV_PERIODIC_ENABLE	4167	=item EV_PERIODIC_ENABLE, EV_IDLE_ENABLE, EV_EMBED_ENABLE, EV_STAT_ENABLE,
		4168	EV_PREPARE_ENABLE, EV_CHECK_ENABLE, EV_FORK_ENABLE, EV_SIGNAL_ENABLE,
		4169	EV_ASYNC_ENABLE, EV_CHILD_ENABLE.
2873		4170
2874	If undefined or defined to be C<1>, then periodic timers are supported. If	4171	If undefined or defined to be C<1> (and the platform supports it), then
2875	defined to be C<0>, then they are not. Disabling them saves a few kB of	4172	the respective watcher type is supported. If defined to be C<0>, then it
2876	code.	4173	is not. Disabling watcher types mainly saves code size.
2877		4174
2878	=item EV_IDLE_ENABLE	4175	=item EV_FEATURES
2879
2880	If undefined or defined to be C<1>, then idle watchers are supported. If
2881	defined to be C<0>, then they are not. Disabling them saves a few kB of
2882	code.
2883
2884	=item EV_EMBED_ENABLE
2885
2886	If undefined or defined to be C<1>, then embed watchers are supported. If
2887	defined to be C<0>, then they are not.
2888
2889	=item EV_STAT_ENABLE
2890
2891	If undefined or defined to be C<1>, then stat watchers are supported. If
2892	defined to be C<0>, then they are not.
2893
2894	=item EV_FORK_ENABLE
2895
2896	If undefined or defined to be C<1>, then fork watchers are supported. If
2897	defined to be C<0>, then they are not.
2898
2899	=item EV_ASYNC_ENABLE
2900
2901	If undefined or defined to be C<1>, then async watchers are supported. If
2902	defined to be C<0>, then they are not.
2903
2904	=item EV_MINIMAL
2905		4176
2906	If you need to shave off some kilobytes of code at the expense of some	4177	If you need to shave off some kilobytes of code at the expense of some
2907	speed, define this symbol to C<1>. Currently only used for gcc to override	4178	speed (but with the full API), you can define this symbol to request
2908	some inlining decisions, saves roughly 30% codesize of amd64.	4179	certain subsets of functionality. The default is to enable all features
		4180	that can be enabled on the platform.
		4181
		4182	A typical way to use this symbol is to define it to C<0> (or to a bitset
		4183	with some broad features you want) and then selectively re-enable
		4184	additional parts you want, for example if you want everything minimal,
		4185	but multiple event loop support, async and child watchers and the poll
		4186	backend, use this:
		4187
		4188	#define EV_FEATURES 0
		4189	#define EV_MULTIPLICITY 1
		4190	#define EV_USE_POLL 1
		4191	#define EV_CHILD_ENABLE 1
		4192	#define EV_ASYNC_ENABLE 1
		4193
		4194	The actual value is a bitset, it can be a combination of the following
		4195	values:
		4196
		4197	=over 4
		4198
		4199	=item C<1> - faster/larger code
		4200
		4201	Use larger code to speed up some operations.
		4202
		4203	Currently this is used to override some inlining decisions (enlarging the
		4204	code size by roughly 30% on amd64).
		4205
		4206	When optimising for size, use of compiler flags such as C<-Os> with
		4207	gcc is recommended, as well as C<-DNDEBUG>, as libev contains a number of
		4208	assertions.
		4209
		4210	=item C<2> - faster/larger data structures
		4211
		4212	Replaces the small 2-heap for timer management by a faster 4-heap, larger
		4213	hash table sizes and so on. This will usually further increase code size
		4214	and can additionally have an effect on the size of data structures at
		4215	runtime.
		4216
		4217	=item C<4> - full API configuration
		4218
		4219	This enables priorities (sets C<EV_MAXPRI>=2 and C<EV_MINPRI>=-2), and
		4220	enables multiplicity (C<EV_MULTIPLICITY>=1).
		4221
		4222	=item C<8> - full API
		4223
		4224	This enables a lot of the "lesser used" API functions. See C<ev.h> for
		4225	details on which parts of the API are still available without this
		4226	feature, and do not complain if this subset changes over time.
		4227
		4228	=item C<16> - enable all optional watcher types
		4229
		4230	Enables all optional watcher types. If you want to selectively enable
		4231	only some watcher types other than I/O and timers (e.g. prepare,
		4232	embed, async, child...) you can enable them manually by defining
		4233	C<EV_watchertype_ENABLE> to C<1> instead.
		4234
		4235	=item C<32> - enable all backends
		4236
		4237	This enables all backends - without this feature, you need to enable at
		4238	least one backend manually (C<EV_USE_SELECT> is a good choice).
		4239
		4240	=item C<64> - enable OS-specific "helper" APIs
		4241
		4242	Enable inotify, eventfd, signalfd and similar OS-specific helper APIs by
		4243	default.
		4244
		4245	=back
		4246
		4247	Compiling with C<gcc -Os -DEV_STANDALONE -DEV_USE_EPOLL=1 -DEV_FEATURES=0>
		4248	reduces the compiled size of libev from 24.7Kb code/2.8Kb data to 6.5Kb
		4249	code/0.3Kb data on my GNU/Linux amd64 system, while still giving you I/O
		4250	watchers, timers and monotonic clock support.
		4251
		4252	With an intelligent-enough linker (gcc+binutils are intelligent enough
		4253	when you use C<-Wl,--gc-sections -ffunction-sections>) functions unused by
		4254	your program might be left out as well - a binary starting a timer and an
		4255	I/O watcher then might come out at only 5Kb.
		4256
		4257	=item EV_AVOID_STDIO
		4258
		4259	If this is set to C<1> at compiletime, then libev will avoid using stdio
		4260	functions (printf, scanf, perror etc.). This will increase the code size
		4261	somewhat, but if your program doesn't otherwise depend on stdio and your
		4262	libc allows it, this avoids linking in the stdio library which is quite
		4263	big.
		4264
		4265	Note that error messages might become less precise when this option is
		4266	enabled.
		4267
		4268	=item EV_NSIG
		4269
		4270	The highest supported signal number, +1 (or, the number of
		4271	signals): Normally, libev tries to deduce the maximum number of signals
		4272	automatically, but sometimes this fails, in which case it can be
		4273	specified. Also, using a lower number than detected (C<32> should be
		4274	good for about any system in existence) can save some memory, as libev
		4275	statically allocates some 12-24 bytes per signal number.
2909		4276
2910	=item EV_PID_HASHSIZE	4277	=item EV_PID_HASHSIZE
2911		4278
2912	C<ev_child> watchers use a small hash table to distribute workload by	4279	C<ev_child> watchers use a small hash table to distribute workload by
2913	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	4280	pid. The default size is C<16> (or C<1> with C<EV_FEATURES> disabled),
2914	than enough. If you need to manage thousands of children you might want to	4281	usually more than enough. If you need to manage thousands of children you
2915	increase this value (I<must> be a power of two).	4282	might want to increase this value (I<must> be a power of two).
2916		4283
2917	=item EV_INOTIFY_HASHSIZE	4284	=item EV_INOTIFY_HASHSIZE
2918		4285
2919	C<ev_stat> watchers use a small hash table to distribute workload by	4286	C<ev_stat> watchers use a small hash table to distribute workload by
2920	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	4287	inotify watch id. The default size is C<16> (or C<1> with C<EV_FEATURES>
2921	usually more than enough. If you need to manage thousands of C<ev_stat>	4288	disabled), usually more than enough. If you need to manage thousands of
2922	watchers you might want to increase this value (I<must> be a power of	4289	C<ev_stat> watchers you might want to increase this value (I<must> be a
2923	two).	4290	power of two).
		4291
		4292	=item EV_USE_4HEAP
		4293
		4294	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4295	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		4296	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		4297	faster performance with many (thousands) of watchers.
		4298
		4299	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4300	will be C<0>.
		4301
		4302	=item EV_HEAP_CACHE_AT
		4303
		4304	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4305	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		4306	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		4307	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		4308	but avoids random read accesses on heap changes. This improves performance
		4309	noticeably with many (hundreds) of watchers.
		4310
		4311	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4312	will be C<0>.
		4313
		4314	=item EV_VERIFY
		4315
		4316	Controls how much internal verification (see C<ev_verify ()>) will
		4317	be done: If set to C<0>, no internal verification code will be compiled
		4318	in. If set to C<1>, then verification code will be compiled in, but not
		4319	called. If set to C<2>, then the internal verification code will be
		4320	called once per loop, which can slow down libev. If set to C<3>, then the
		4321	verification code will be called very frequently, which will slow down
		4322	libev considerably.
		4323
		4324	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4325	will be C<0>.
2924		4326
2925	=item EV_COMMON	4327	=item EV_COMMON
2926		4328
2927	By default, all watchers have a C<void *data> member. By redefining	4329	By default, all watchers have a C<void *data> member. By redefining
2928	this macro to a something else you can include more and other types of	4330	this macro to something else you can include more and other types of
2929	members. You have to define it each time you include one of the files,	4331	members. You have to define it each time you include one of the files,
2930	though, and it must be identical each time.	4332	though, and it must be identical each time.
2931		4333
2932	For example, the perl EV module uses something like this:	4334	For example, the perl EV module uses something like this:
2933		4335
2934	#define EV_COMMON \	4336	#define EV_COMMON \
2935	SV *self; /* contains this struct */ \	4337	SV *self; /* contains this struct */ \
2936	SV *cb_sv, *fh /* note no trailing ";" */	4338	SV *cb_sv, *fh /* note no trailing ";" */
2937		4339
2938	=item EV_CB_DECLARE (type)	4340	=item EV_CB_DECLARE (type)
2939		4341
2940	=item EV_CB_INVOKE (watcher, revents)	4342	=item EV_CB_INVOKE (watcher, revents)
2941		4343
…		…
2946	definition and a statement, respectively. See the F<ev.h> header file for	4348	definition and a statement, respectively. See the F<ev.h> header file for
2947	their default definitions. One possible use for overriding these is to	4349	their default definitions. One possible use for overriding these is to
2948	avoid the C<struct ev_loop *> as first argument in all cases, or to use	4350	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2949	method calls instead of plain function calls in C++.	4351	method calls instead of plain function calls in C++.
2950		4352
		4353	=back
		4354
2951	=head2 EXPORTED API SYMBOLS	4355	=head2 EXPORTED API SYMBOLS
2952		4356
2953	If you need to re-export the API (e.g. via a dll) and you need a list of	4357	If you need to re-export the API (e.g. via a DLL) and you need a list of
2954	exported symbols, you can use the provided F<Symbol.*> files which list	4358	exported symbols, you can use the provided F<Symbol.*> files which list
2955	all public symbols, one per line:	4359	all public symbols, one per line:
2956		4360
2957	Symbols.ev for libev proper	4361	Symbols.ev for libev proper
2958	Symbols.event for the libevent emulation	4362	Symbols.event for the libevent emulation
2959		4363
2960	This can also be used to rename all public symbols to avoid clashes with	4364	This can also be used to rename all public symbols to avoid clashes with
2961	multiple versions of libev linked together (which is obviously bad in	4365	multiple versions of libev linked together (which is obviously bad in
2962	itself, but sometimes it is inconvinient to avoid this).	4366	itself, but sometimes it is inconvenient to avoid this).
2963		4367
2964	A sed command like this will create wrapper C<#define>'s that you need to	4368	A sed command like this will create wrapper C<#define>'s that you need to
2965	include before including F<ev.h>:	4369	include before including F<ev.h>:
2966		4370
2967	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h	4371	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
…		…
2984	file.	4388	file.
2985		4389
2986	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	4390	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2987	that everybody includes and which overrides some configure choices:	4391	that everybody includes and which overrides some configure choices:
2988		4392
2989	#define EV_MINIMAL 1	4393	#define EV_FEATURES 8
2990	#define EV_USE_POLL 0	4394	#define EV_USE_SELECT 1
2991	#define EV_MULTIPLICITY 0
2992	#define EV_PERIODIC_ENABLE 0	4395	#define EV_PREPARE_ENABLE 1
		4396	#define EV_IDLE_ENABLE 1
2993	#define EV_STAT_ENABLE 0	4397	#define EV_SIGNAL_ENABLE 1
2994	#define EV_FORK_ENABLE 0	4398	#define EV_CHILD_ENABLE 1
		4399	#define EV_USE_STDEXCEPT 0
2995	#define EV_CONFIG_H <config.h>	4400	#define EV_CONFIG_H <config.h>
2996	#define EV_MINPRI 0
2997	#define EV_MAXPRI 0
2998		4401
2999	#include "ev++.h"	4402	#include "ev++.h"
3000		4403
3001	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	4404	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
3002		4405
3003	#include "ev_cpp.h"	4406	#include "ev_cpp.h"
3004	#include "ev.c"	4407	#include "ev.c"
3005		4408
		4409	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
3006		4410
3007	=head1 COMPLEXITIES	4411	=head2 THREADS AND COROUTINES
3008		4412
3009	In this section the complexities of (many of) the algorithms used inside	4413	=head3 THREADS
3010	libev will be explained. For complexity discussions about backends see the
3011	documentation for C<ev_default_init>.
3012		4414
3013	All of the following are about amortised time: If an array needs to be	4415	All libev functions are reentrant and thread-safe unless explicitly
3014	extended, libev needs to realloc and move the whole array, but this	4416	documented otherwise, but libev implements no locking itself. This means
3015	happens asymptotically never with higher number of elements, so O(1) might	4417	that you can use as many loops as you want in parallel, as long as there
3016	mean it might do a lengthy realloc operation in rare cases, but on average	4418	are no concurrent calls into any libev function with the same loop
3017	it is much faster and asymptotically approaches constant time.	4419	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		4420	of course): libev guarantees that different event loops share no data
		4421	structures that need any locking.
		4422
		4423	Or to put it differently: calls with different loop parameters can be done
		4424	concurrently from multiple threads, calls with the same loop parameter
		4425	must be done serially (but can be done from different threads, as long as
		4426	only one thread ever is inside a call at any point in time, e.g. by using
		4427	a mutex per loop).
		4428
		4429	Specifically to support threads (and signal handlers), libev implements
		4430	so-called C<ev_async> watchers, which allow some limited form of
		4431	concurrency on the same event loop, namely waking it up "from the
		4432	outside".
		4433
		4434	If you want to know which design (one loop, locking, or multiple loops
		4435	without or something else still) is best for your problem, then I cannot
		4436	help you, but here is some generic advice:
3018		4437
3019	=over 4	4438	=over 4
3020		4439
3021	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	4440	=item * most applications have a main thread: use the default libev loop
		4441	in that thread, or create a separate thread running only the default loop.
3022		4442
3023	This means that, when you have a watcher that triggers in one hour and	4443	This helps integrating other libraries or software modules that use libev
3024	there are 100 watchers that would trigger before that then inserting will	4444	themselves and don't care/know about threading.
3025	have to skip roughly seven (C<ld 100>) of these watchers.
3026		4445
3027	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	4446	=item * one loop per thread is usually a good model.
3028		4447
3029	That means that changing a timer costs less than removing/adding them	4448	Doing this is almost never wrong, sometimes a better-performance model
3030	as only the relative motion in the event queue has to be paid for.	4449	exists, but it is always a good start.
3031		4450
3032	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)	4451	=item * other models exist, such as the leader/follower pattern, where one
		4452	loop is handed through multiple threads in a kind of round-robin fashion.
3033		4453
3034	These just add the watcher into an array or at the head of a list.	4454	Choosing a model is hard - look around, learn, know that usually you can do
		4455	better than you currently do :-)
3035		4456
3036	=item Stopping check/prepare/idle/fork/async watchers: O(1)	4457	=item * often you need to talk to some other thread which blocks in the
		4458	event loop.
3037		4459
3038	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	4460	C<ev_async> watchers can be used to wake them up from other threads safely
		4461	(or from signal contexts...).
3039		4462
3040	These watchers are stored in lists then need to be walked to find the	4463	An example use would be to communicate signals or other events that only
3041	correct watcher to remove. The lists are usually short (you don't usually	4464	work in the default loop by registering the signal watcher with the
3042	have many watchers waiting for the same fd or signal).	4465	default loop and triggering an C<ev_async> watcher from the default loop
3043		4466	watcher callback into the event loop interested in the signal.
3044	=item Finding the next timer in each loop iteration: O(1)
3045
3046	By virtue of using a binary heap, the next timer is always found at the
3047	beginning of the storage array.
3048
3049	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
3050
3051	A change means an I/O watcher gets started or stopped, which requires
3052	libev to recalculate its status (and possibly tell the kernel, depending
3053	on backend and wether C<ev_io_set> was used).
3054
3055	=item Activating one watcher (putting it into the pending state): O(1)
3056
3057	=item Priority handling: O(number_of_priorities)
3058
3059	Priorities are implemented by allocating some space for each
3060	priority. When doing priority-based operations, libev usually has to
3061	linearly search all the priorities, but starting/stopping and activating
3062	watchers becomes O(1) w.r.t. priority handling.
3063
3064	=item Sending an ev_async: O(1)
3065
3066	=item Processing ev_async_send: O(number_of_async_watchers)
3067
3068	=item Processing signals: O(max_signal_number)
3069
3070	Sending involves a syscall I<iff> there were no other C<ev_async_send>
3071	calls in the current loop iteration. Checking for async and signal events
3072	involves iterating over all running async watchers or all signal numbers.
3073		4467
3074	=back	4468	=back
3075		4469
		4470	=head4 THREAD LOCKING EXAMPLE
3076		4471
3077	=head1 Win32 platform limitations and workarounds	4472	Here is a fictitious example of how to run an event loop in a different
		4473	thread than where callbacks are being invoked and watchers are
		4474	created/added/removed.
		4475
		4476	For a real-world example, see the C<EV::Loop::Async> perl module,
		4477	which uses exactly this technique (which is suited for many high-level
		4478	languages).
		4479
		4480	The example uses a pthread mutex to protect the loop data, a condition
		4481	variable to wait for callback invocations, an async watcher to notify the
		4482	event loop thread and an unspecified mechanism to wake up the main thread.
		4483
		4484	First, you need to associate some data with the event loop:
		4485
		4486	typedef struct {
		4487	mutex_t lock; /* global loop lock */
		4488	ev_async async_w;
		4489	thread_t tid;
		4490	cond_t invoke_cv;
		4491	} userdata;
		4492
		4493	void prepare_loop (EV_P)
		4494	{
		4495	// for simplicity, we use a static userdata struct.
		4496	static userdata u;
		4497
		4498	ev_async_init (&u->async_w, async_cb);
		4499	ev_async_start (EV_A_ &u->async_w);
		4500
		4501	pthread_mutex_init (&u->lock, 0);
		4502	pthread_cond_init (&u->invoke_cv, 0);
		4503
		4504	// now associate this with the loop
		4505	ev_set_userdata (EV_A_ u);
		4506	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		4507	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		4508
		4509	// then create the thread running ev_loop
		4510	pthread_create (&u->tid, 0, l_run, EV_A);
		4511	}
		4512
		4513	The callback for the C<ev_async> watcher does nothing: the watcher is used
		4514	solely to wake up the event loop so it takes notice of any new watchers
		4515	that might have been added:
		4516
		4517	static void
		4518	async_cb (EV_P_ ev_async *w, int revents)
		4519	{
		4520	// just used for the side effects
		4521	}
		4522
		4523	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		4524	protecting the loop data, respectively.
		4525
		4526	static void
		4527	l_release (EV_P)
		4528	{
		4529	userdata *u = ev_userdata (EV_A);
		4530	pthread_mutex_unlock (&u->lock);
		4531	}
		4532
		4533	static void
		4534	l_acquire (EV_P)
		4535	{
		4536	userdata *u = ev_userdata (EV_A);
		4537	pthread_mutex_lock (&u->lock);
		4538	}
		4539
		4540	The event loop thread first acquires the mutex, and then jumps straight
		4541	into C<ev_run>:
		4542
		4543	void *
		4544	l_run (void *thr_arg)
		4545	{
		4546	struct ev_loop *loop = (struct ev_loop *)thr_arg;
		4547
		4548	l_acquire (EV_A);
		4549	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		4550	ev_run (EV_A_ 0);
		4551	l_release (EV_A);
		4552
		4553	return 0;
		4554	}
		4555
		4556	Instead of invoking all pending watchers, the C<l_invoke> callback will
		4557	signal the main thread via some unspecified mechanism (signals? pipe
		4558	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		4559	have been called (in a while loop because a) spurious wakeups are possible
		4560	and b) skipping inter-thread-communication when there are no pending
		4561	watchers is very beneficial):
		4562
		4563	static void
		4564	l_invoke (EV_P)
		4565	{
		4566	userdata *u = ev_userdata (EV_A);
		4567
		4568	while (ev_pending_count (EV_A))
		4569	{
		4570	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		4571	pthread_cond_wait (&u->invoke_cv, &u->lock);
		4572	}
		4573	}
		4574
		4575	Now, whenever the main thread gets told to invoke pending watchers, it
		4576	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		4577	thread to continue:
		4578
		4579	static void
		4580	real_invoke_pending (EV_P)
		4581	{
		4582	userdata *u = ev_userdata (EV_A);
		4583
		4584	pthread_mutex_lock (&u->lock);
		4585	ev_invoke_pending (EV_A);
		4586	pthread_cond_signal (&u->invoke_cv);
		4587	pthread_mutex_unlock (&u->lock);
		4588	}
		4589
		4590	Whenever you want to start/stop a watcher or do other modifications to an
		4591	event loop, you will now have to lock:
		4592
		4593	ev_timer timeout_watcher;
		4594	userdata *u = ev_userdata (EV_A);
		4595
		4596	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		4597
		4598	pthread_mutex_lock (&u->lock);
		4599	ev_timer_start (EV_A_ &timeout_watcher);
		4600	ev_async_send (EV_A_ &u->async_w);
		4601	pthread_mutex_unlock (&u->lock);
		4602
		4603	Note that sending the C<ev_async> watcher is required because otherwise
		4604	an event loop currently blocking in the kernel will have no knowledge
		4605	about the newly added timer. By waking up the loop it will pick up any new
		4606	watchers in the next event loop iteration.
		4607
		4608	=head3 COROUTINES
		4609
		4610	Libev is very accommodating to coroutines ("cooperative threads"):
		4611	libev fully supports nesting calls to its functions from different
		4612	coroutines (e.g. you can call C<ev_run> on the same loop from two
		4613	different coroutines, and switch freely between both coroutines running
		4614	the loop, as long as you don't confuse yourself). The only exception is
		4615	that you must not do this from C<ev_periodic> reschedule callbacks.
		4616
		4617	Care has been taken to ensure that libev does not keep local state inside
		4618	C<ev_run>, and other calls do not usually allow for coroutine switches as
		4619	they do not call any callbacks.
		4620
		4621	=head2 COMPILER WARNINGS
		4622
		4623	Depending on your compiler and compiler settings, you might get no or a
		4624	lot of warnings when compiling libev code. Some people are apparently
		4625	scared by this.
		4626
		4627	However, these are unavoidable for many reasons. For one, each compiler
		4628	has different warnings, and each user has different tastes regarding
		4629	warning options. "Warn-free" code therefore cannot be a goal except when
		4630	targeting a specific compiler and compiler-version.
		4631
		4632	Another reason is that some compiler warnings require elaborate
		4633	workarounds, or other changes to the code that make it less clear and less
		4634	maintainable.
		4635
		4636	And of course, some compiler warnings are just plain stupid, or simply
		4637	wrong (because they don't actually warn about the condition their message
		4638	seems to warn about). For example, certain older gcc versions had some
		4639	warnings that resulted in an extreme number of false positives. These have
		4640	been fixed, but some people still insist on making code warn-free with
		4641	such buggy versions.
		4642
		4643	While libev is written to generate as few warnings as possible,
		4644	"warn-free" code is not a goal, and it is recommended not to build libev
		4645	with any compiler warnings enabled unless you are prepared to cope with
		4646	them (e.g. by ignoring them). Remember that warnings are just that:
		4647	warnings, not errors, or proof of bugs.
		4648
		4649
		4650	=head2 VALGRIND
		4651
		4652	Valgrind has a special section here because it is a popular tool that is
		4653	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		4654
		4655	If you think you found a bug (memory leak, uninitialised data access etc.)
		4656	in libev, then check twice: If valgrind reports something like:
		4657
		4658	==2274== definitely lost: 0 bytes in 0 blocks.
		4659	==2274== possibly lost: 0 bytes in 0 blocks.
		4660	==2274== still reachable: 256 bytes in 1 blocks.
		4661
		4662	Then there is no memory leak, just as memory accounted to global variables
		4663	is not a memleak - the memory is still being referenced, and didn't leak.
		4664
		4665	Similarly, under some circumstances, valgrind might report kernel bugs
		4666	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		4667	although an acceptable workaround has been found here), or it might be
		4668	confused.
		4669
		4670	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		4671	make it into some kind of religion.
		4672
		4673	If you are unsure about something, feel free to contact the mailing list
		4674	with the full valgrind report and an explanation on why you think this
		4675	is a bug in libev (best check the archives, too :). However, don't be
		4676	annoyed when you get a brisk "this is no bug" answer and take the chance
		4677	of learning how to interpret valgrind properly.
		4678
		4679	If you need, for some reason, empty reports from valgrind for your project
		4680	I suggest using suppression lists.
		4681
		4682
		4683	=head1 PORTABILITY NOTES
		4684
		4685	=head2 GNU/LINUX 32 BIT LIMITATIONS
		4686
		4687	GNU/Linux is the only common platform that supports 64 bit file/large file
		4688	interfaces but I<disables> them by default.
		4689
		4690	That means that libev compiled in the default environment doesn't support
		4691	files larger than 2GiB or so, which mainly affects C<ev_stat> watchers.
		4692
		4693	Unfortunately, many programs try to work around this GNU/Linux issue
		4694	by enabling the large file API, which makes them incompatible with the
		4695	standard libev compiled for their system.
		4696
		4697	Likewise, libev cannot enable the large file API itself as this would
		4698	suddenly make it incompatible to the default compile time environment,
		4699	i.e. all programs not using special compile switches.
		4700
		4701	=head2 OS/X AND DARWIN BUGS
		4702
		4703	The whole thing is a bug if you ask me - basically any system interface
		4704	you touch is broken, whether it is locales, poll, kqueue or even the
		4705	OpenGL drivers.
		4706
		4707	=head3 C<kqueue> is buggy
		4708
		4709	The kqueue syscall is broken in all known versions - most versions support
		4710	only sockets, many support pipes.
		4711
		4712	Libev tries to work around this by not using C<kqueue> by default on this
		4713	rotten platform, but of course you can still ask for it when creating a
		4714	loop - embedding a socket-only kqueue loop into a select-based one is
		4715	probably going to work well.
		4716
		4717	=head3 C<poll> is buggy
		4718
		4719	Instead of fixing C<kqueue>, Apple replaced their (working) C<poll>
		4720	implementation by something calling C<kqueue> internally around the 10.5.6
		4721	release, so now C<kqueue> I<and> C<poll> are broken.
		4722
		4723	Libev tries to work around this by not using C<poll> by default on
		4724	this rotten platform, but of course you can still ask for it when creating
		4725	a loop.
		4726
		4727	=head3 C<select> is buggy
		4728
		4729	All that's left is C<select>, and of course Apple found a way to fuck this
		4730	one up as well: On OS/X, C<select> actively limits the number of file
		4731	descriptors you can pass in to 1024 - your program suddenly crashes when
		4732	you use more.
		4733
		4734	There is an undocumented "workaround" for this - defining
		4735	C<_DARWIN_UNLIMITED_SELECT>, which libev tries to use, so select I<should>
		4736	work on OS/X.
		4737
		4738	=head2 SOLARIS PROBLEMS AND WORKAROUNDS
		4739
		4740	=head3 C<errno> reentrancy
		4741
		4742	The default compile environment on Solaris is unfortunately so
		4743	thread-unsafe that you can't even use components/libraries compiled
		4744	without C<-D_REENTRANT> in a threaded program, which, of course, isn't
		4745	defined by default. A valid, if stupid, implementation choice.
		4746
		4747	If you want to use libev in threaded environments you have to make sure
		4748	it's compiled with C<_REENTRANT> defined.
		4749
		4750	=head3 Event port backend
		4751
		4752	The scalable event interface for Solaris is called "event
		4753	ports". Unfortunately, this mechanism is very buggy in all major
		4754	releases. If you run into high CPU usage, your program freezes or you get
		4755	a large number of spurious wakeups, make sure you have all the relevant
		4756	and latest kernel patches applied. No, I don't know which ones, but there
		4757	are multiple ones to apply, and afterwards, event ports actually work
		4758	great.
		4759
		4760	If you can't get it to work, you can try running the program by setting
		4761	the environment variable C<LIBEV_FLAGS=3> to only allow C<poll> and
		4762	C<select> backends.
		4763
		4764	=head2 AIX POLL BUG
		4765
		4766	AIX unfortunately has a broken C<poll.h> header. Libev works around
		4767	this by trying to avoid the poll backend altogether (i.e. it's not even
		4768	compiled in), which normally isn't a big problem as C<select> works fine
		4769	with large bitsets on AIX, and AIX is dead anyway.
		4770
		4771	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		4772
		4773	=head3 General issues
3078		4774
3079	Win32 doesn't support any of the standards (e.g. POSIX) that libev	4775	Win32 doesn't support any of the standards (e.g. POSIX) that libev
3080	requires, and its I/O model is fundamentally incompatible with the POSIX	4776	requires, and its I/O model is fundamentally incompatible with the POSIX
3081	model. Libev still offers limited functionality on this platform in	4777	model. Libev still offers limited functionality on this platform in
3082	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	4778	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
3083	descriptors. This only applies when using Win32 natively, not when using	4779	descriptors. This only applies when using Win32 natively, not when using
3084	e.g. cygwin.	4780	e.g. cygwin. Actually, it only applies to the microsofts own compilers,
		4781	as every compielr comes with a slightly differently broken/incompatible
		4782	environment.
		4783
		4784	Lifting these limitations would basically require the full
		4785	re-implementation of the I/O system. If you are into this kind of thing,
		4786	then note that glib does exactly that for you in a very portable way (note
		4787	also that glib is the slowest event library known to man).
3085		4788
3086	There is no supported compilation method available on windows except	4789	There is no supported compilation method available on windows except
3087	embedding it into other applications.	4790	embedding it into other applications.
3088		4791
		4792	Sensible signal handling is officially unsupported by Microsoft - libev
		4793	tries its best, but under most conditions, signals will simply not work.
		4794
		4795	Not a libev limitation but worth mentioning: windows apparently doesn't
		4796	accept large writes: instead of resulting in a partial write, windows will
		4797	either accept everything or return C<ENOBUFS> if the buffer is too large,
		4798	so make sure you only write small amounts into your sockets (less than a
		4799	megabyte seems safe, but this apparently depends on the amount of memory
		4800	available).
		4801
3089	Due to the many, low, and arbitrary limits on the win32 platform and the	4802	Due to the many, low, and arbitrary limits on the win32 platform and
3090	abysmal performance of winsockets, using a large number of sockets is not	4803	the abysmal performance of winsockets, using a large number of sockets
3091	recommended (and not reasonable). If your program needs to use more than	4804	is not recommended (and not reasonable). If your program needs to use
3092	a hundred or so sockets, then likely it needs to use a totally different	4805	more than a hundred or so sockets, then likely it needs to use a totally
3093	implementation for windows, as libev offers the POSIX model, which cannot	4806	different implementation for windows, as libev offers the POSIX readiness
3094	be implemented efficiently on windows (microsoft monopoly games).	4807	notification model, which cannot be implemented efficiently on windows
		4808	(due to Microsoft monopoly games).
3095		4809
3096	=over 4	4810	A typical way to use libev under windows is to embed it (see the embedding
		4811	section for details) and use the following F<evwrap.h> header file instead
		4812	of F<ev.h>:
3097		4813
		4814	#define EV_STANDALONE /* keeps ev from requiring config.h */
		4815	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		4816
		4817	#include "ev.h"
		4818
		4819	And compile the following F<evwrap.c> file into your project (make sure
		4820	you do I<not> compile the F<ev.c> or any other embedded source files!):
		4821
		4822	#include "evwrap.h"
		4823	#include "ev.c"
		4824
3098	=item The winsocket select function	4825	=head3 The winsocket C<select> function
3099		4826
3100	The winsocket C<select> function doesn't follow POSIX in that it requires	4827	The winsocket C<select> function doesn't follow POSIX in that it
3101	socket I<handles> and not socket I<file descriptors>. This makes select	4828	requires socket I<handles> and not socket I<file descriptors> (it is
3102	very inefficient, and also requires a mapping from file descriptors	4829	also extremely buggy). This makes select very inefficient, and also
3103	to socket handles. See the discussion of the C<EV_SELECT_USE_FD_SET>,	4830	requires a mapping from file descriptors to socket handles (the Microsoft
3104	C<EV_SELECT_IS_WINSOCKET> and C<EV_FD_TO_WIN32_HANDLE> preprocessor	4831	C runtime provides the function C<_open_osfhandle> for this). See the
3105	symbols for more info.	4832	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		4833	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
3106		4834
3107	The configuration for a "naked" win32 using the microsoft runtime	4835	The configuration for a "naked" win32 using the Microsoft runtime
3108	libraries and raw winsocket select is:	4836	libraries and raw winsocket select is:
3109		4837
3110	#define EV_USE_SELECT 1	4838	#define EV_USE_SELECT 1
3111	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */	4839	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
3112		4840
3113	Note that winsockets handling of fd sets is O(n), so you can easily get a	4841	Note that winsockets handling of fd sets is O(n), so you can easily get a
3114	complexity in the O(n²) range when using win32.	4842	complexity in the O(n²) range when using win32.
3115		4843
3116	=item Limited number of file descriptors	4844	=head3 Limited number of file descriptors
3117		4845
3118	Windows has numerous arbitrary (and low) limits on things. Early versions	4846	Windows has numerous arbitrary (and low) limits on things.
3119	of winsocket's select only supported waiting for a max. of C<64> handles	4847
		4848	Early versions of winsocket's select only supported waiting for a maximum
3120	(probably owning to the fact that all windows kernels can only wait for	4849	of C<64> handles (probably owning to the fact that all windows kernels
3121	C<64> things at the same time internally; microsoft recommends spawning a	4850	can only wait for C<64> things at the same time internally; Microsoft
3122	chain of threads and wait for 63 handles and the previous thread in each).	4851	recommends spawning a chain of threads and wait for 63 handles and the
		4852	previous thread in each. Sounds great!).
3123		4853
3124	Newer versions support more handles, but you need to define C<FD_SETSIZE>	4854	Newer versions support more handles, but you need to define C<FD_SETSIZE>
3125	to some high number (e.g. C<2048>) before compiling the winsocket select	4855	to some high number (e.g. C<2048>) before compiling the winsocket select
3126	call (which might be in libev or elsewhere, for example, perl does its own	4856	call (which might be in libev or elsewhere, for example, perl and many
3127	select emulation on windows).	4857	other interpreters do their own select emulation on windows).
3128		4858
3129	Another limit is the number of file descriptors in the microsoft runtime	4859	Another limit is the number of file descriptors in the Microsoft runtime
3130	libraries, which by default is C<64> (there must be a hidden I<64> fetish	4860	libraries, which by default is C<64> (there must be a hidden I<64>
3131	or something like this inside microsoft). You can increase this by calling	4861	fetish or something like this inside Microsoft). You can increase this
3132	C<_setmaxstdio>, which can increase this limit to C<2048> (another	4862	by calling C<_setmaxstdio>, which can increase this limit to C<2048>
3133	arbitrary limit), but is broken in many versions of the microsoft runtime	4863	(another arbitrary limit), but is broken in many versions of the Microsoft
3134	libraries.
3135
3136	This might get you to about C<512> or C<2048> sockets (depending on	4864	runtime libraries. This might get you to about C<512> or C<2048> sockets
3137	windows version and/or the phase of the moon). To get more, you need to	4865	(depending on windows version and/or the phase of the moon). To get more,
3138	wrap all I/O functions and provide your own fd management, but the cost of	4866	you need to wrap all I/O functions and provide your own fd management, but
3139	calling select (O(n²)) will likely make this unworkable.	4867	the cost of calling select (O(n²)) will likely make this unworkable.
		4868
		4869	=head2 PORTABILITY REQUIREMENTS
		4870
		4871	In addition to a working ISO-C implementation and of course the
		4872	backend-specific APIs, libev relies on a few additional extensions:
		4873
		4874	=over 4
		4875
		4876	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		4877	calling conventions regardless of C<ev_watcher_type *>.
		4878
		4879	Libev assumes not only that all watcher pointers have the same internal
		4880	structure (guaranteed by POSIX but not by ISO C for example), but it also
		4881	assumes that the same (machine) code can be used to call any watcher
		4882	callback: The watcher callbacks have different type signatures, but libev
		4883	calls them using an C<ev_watcher *> internally.
		4884
		4885	=item pointer accesses must be thread-atomic
		4886
		4887	Accessing a pointer value must be atomic, it must both be readable and
		4888	writable in one piece - this is the case on all current architectures.
		4889
		4890	=item C<sig_atomic_t volatile> must be thread-atomic as well
		4891
		4892	The type C<sig_atomic_t volatile> (or whatever is defined as
		4893	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		4894	threads. This is not part of the specification for C<sig_atomic_t>, but is
		4895	believed to be sufficiently portable.
		4896
		4897	=item C<sigprocmask> must work in a threaded environment
		4898
		4899	Libev uses C<sigprocmask> to temporarily block signals. This is not
		4900	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		4901	pthread implementations will either allow C<sigprocmask> in the "main
		4902	thread" or will block signals process-wide, both behaviours would
		4903	be compatible with libev. Interaction between C<sigprocmask> and
		4904	C<pthread_sigmask> could complicate things, however.
		4905
		4906	The most portable way to handle signals is to block signals in all threads
		4907	except the initial one, and run the default loop in the initial thread as
		4908	well.
		4909
		4910	=item C<long> must be large enough for common memory allocation sizes
		4911
		4912	To improve portability and simplify its API, libev uses C<long> internally
		4913	instead of C<size_t> when allocating its data structures. On non-POSIX
		4914	systems (Microsoft...) this might be unexpectedly low, but is still at
		4915	least 31 bits everywhere, which is enough for hundreds of millions of
		4916	watchers.
		4917
		4918	=item C<double> must hold a time value in seconds with enough accuracy
		4919
		4920	The type C<double> is used to represent timestamps. It is required to
		4921	have at least 51 bits of mantissa (and 9 bits of exponent), which is
		4922	good enough for at least into the year 4000 with millisecond accuracy
		4923	(the design goal for libev). This requirement is overfulfilled by
		4924	implementations using IEEE 754, which is basically all existing ones. With
		4925	IEEE 754 doubles, you get microsecond accuracy until at least 2200.
3140		4926
3141	=back	4927	=back
3142		4928
		4929	If you know of other additional requirements drop me a note.
		4930
		4931
		4932	=head1 ALGORITHMIC COMPLEXITIES
		4933
		4934	In this section the complexities of (many of) the algorithms used inside
		4935	libev will be documented. For complexity discussions about backends see
		4936	the documentation for C<ev_default_init>.
		4937
		4938	All of the following are about amortised time: If an array needs to be
		4939	extended, libev needs to realloc and move the whole array, but this
		4940	happens asymptotically rarer with higher number of elements, so O(1) might
		4941	mean that libev does a lengthy realloc operation in rare cases, but on
		4942	average it is much faster and asymptotically approaches constant time.
		4943
		4944	=over 4
		4945
		4946	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		4947
		4948	This means that, when you have a watcher that triggers in one hour and
		4949	there are 100 watchers that would trigger before that, then inserting will
		4950	have to skip roughly seven (C<ld 100>) of these watchers.
		4951
		4952	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		4953
		4954	That means that changing a timer costs less than removing/adding them,
		4955	as only the relative motion in the event queue has to be paid for.
		4956
		4957	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
		4958
		4959	These just add the watcher into an array or at the head of a list.
		4960
		4961	=item Stopping check/prepare/idle/fork/async watchers: O(1)
		4962
		4963	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		4964
		4965	These watchers are stored in lists, so they need to be walked to find the
		4966	correct watcher to remove. The lists are usually short (you don't usually
		4967	have many watchers waiting for the same fd or signal: one is typical, two
		4968	is rare).
		4969
		4970	=item Finding the next timer in each loop iteration: O(1)
		4971
		4972	By virtue of using a binary or 4-heap, the next timer is always found at a
		4973	fixed position in the storage array.
		4974
		4975	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		4976
		4977	A change means an I/O watcher gets started or stopped, which requires
		4978	libev to recalculate its status (and possibly tell the kernel, depending
		4979	on backend and whether C<ev_io_set> was used).
		4980
		4981	=item Activating one watcher (putting it into the pending state): O(1)
		4982
		4983	=item Priority handling: O(number_of_priorities)
		4984
		4985	Priorities are implemented by allocating some space for each
		4986	priority. When doing priority-based operations, libev usually has to
		4987	linearly search all the priorities, but starting/stopping and activating
		4988	watchers becomes O(1) with respect to priority handling.
		4989
		4990	=item Sending an ev_async: O(1)
		4991
		4992	=item Processing ev_async_send: O(number_of_async_watchers)
		4993
		4994	=item Processing signals: O(max_signal_number)
		4995
		4996	Sending involves a system call I<iff> there were no other C<ev_async_send>
		4997	calls in the current loop iteration. Checking for async and signal events
		4998	involves iterating over all running async watchers or all signal numbers.
		4999
		5000	=back
		5001
		5002
		5003	=head1 PORTING FROM LIBEV 3.X TO 4.X
		5004
		5005	The major version 4 introduced some incompatible changes to the API.
		5006
		5007	At the moment, the C<ev.h> header file provides compatibility definitions
		5008	for all changes, so most programs should still compile. The compatibility
		5009	layer might be removed in later versions of libev, so better update to the
		5010	new API early than late.
		5011
		5012	=over 4
		5013
		5014	=item C<EV_COMPAT3> backwards compatibility mechanism
		5015
		5016	The backward compatibility mechanism can be controlled by
		5017	C<EV_COMPAT3>. See L<PREPROCESSOR SYMBOLS/MACROS> in the L<EMBEDDING>
		5018	section.
		5019
		5020	=item C<ev_default_destroy> and C<ev_default_fork> have been removed
		5021
		5022	These calls can be replaced easily by their C<ev_loop_xxx> counterparts:
		5023
		5024	ev_loop_destroy (EV_DEFAULT_UC);
		5025	ev_loop_fork (EV_DEFAULT);
		5026
		5027	=item function/symbol renames
		5028
		5029	A number of functions and symbols have been renamed:
		5030
		5031	ev_loop => ev_run
		5032	EVLOOP_NONBLOCK => EVRUN_NOWAIT
		5033	EVLOOP_ONESHOT => EVRUN_ONCE
		5034
		5035	ev_unloop => ev_break
		5036	EVUNLOOP_CANCEL => EVBREAK_CANCEL
		5037	EVUNLOOP_ONE => EVBREAK_ONE
		5038	EVUNLOOP_ALL => EVBREAK_ALL
		5039
		5040	EV_TIMEOUT => EV_TIMER
		5041
		5042	ev_loop_count => ev_iteration
		5043	ev_loop_depth => ev_depth
		5044	ev_loop_verify => ev_verify
		5045
		5046	Most functions working on C<struct ev_loop> objects don't have an
		5047	C<ev_loop_> prefix, so it was removed; C<ev_loop>, C<ev_unloop> and
		5048	associated constants have been renamed to not collide with the C<struct
		5049	ev_loop> anymore and C<EV_TIMER> now follows the same naming scheme
		5050	as all other watcher types. Note that C<ev_loop_fork> is still called
		5051	C<ev_loop_fork> because it would otherwise clash with the C<ev_fork>
		5052	typedef.
		5053
		5054	=item C<EV_MINIMAL> mechanism replaced by C<EV_FEATURES>
		5055
		5056	The preprocessor symbol C<EV_MINIMAL> has been replaced by a different
		5057	mechanism, C<EV_FEATURES>. Programs using C<EV_MINIMAL> usually compile
		5058	and work, but the library code will of course be larger.
		5059
		5060	=back
		5061
		5062
		5063	=head1 GLOSSARY
		5064
		5065	=over 4
		5066
		5067	=item active
		5068
		5069	A watcher is active as long as it has been started and not yet stopped.
		5070	See L<WATCHER STATES> for details.
		5071
		5072	=item application
		5073
		5074	In this document, an application is whatever is using libev.
		5075
		5076	=item backend
		5077
		5078	The part of the code dealing with the operating system interfaces.
		5079
		5080	=item callback
		5081
		5082	The address of a function that is called when some event has been
		5083	detected. Callbacks are being passed the event loop, the watcher that
		5084	received the event, and the actual event bitset.
		5085
		5086	=item callback/watcher invocation
		5087
		5088	The act of calling the callback associated with a watcher.
		5089
		5090	=item event
		5091
		5092	A change of state of some external event, such as data now being available
		5093	for reading on a file descriptor, time having passed or simply not having
		5094	any other events happening anymore.
		5095
		5096	In libev, events are represented as single bits (such as C<EV_READ> or
		5097	C<EV_TIMER>).
		5098
		5099	=item event library
		5100
		5101	A software package implementing an event model and loop.
		5102
		5103	=item event loop
		5104
		5105	An entity that handles and processes external events and converts them
		5106	into callback invocations.
		5107
		5108	=item event model
		5109
		5110	The model used to describe how an event loop handles and processes
		5111	watchers and events.
		5112
		5113	=item pending
		5114
		5115	A watcher is pending as soon as the corresponding event has been
		5116	detected. See L<WATCHER STATES> for details.
		5117
		5118	=item real time
		5119
		5120	The physical time that is observed. It is apparently strictly monotonic :)
		5121
		5122	=item wall-clock time
		5123
		5124	The time and date as shown on clocks. Unlike real time, it can actually
		5125	be wrong and jump forwards and backwards, e.g. when the you adjust your
		5126	clock.
		5127
		5128	=item watcher
		5129
		5130	A data structure that describes interest in certain events. Watchers need
		5131	to be started (attached to an event loop) before they can receive events.
		5132
		5133	=back
3143		5134
3144	=head1 AUTHOR	5135	=head1 AUTHOR
3145		5136
3146	Marc Lehmann <libev@schmorp.de>.	5137	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael
		5138	Magnusson and Emanuele Giaquinta.
3147		5139

Diff Legend

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