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 1.378 by sf-exg, Thu Jul 7 16:30:49 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	// break 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
138	either it is interrupted or the given time interval has passed. Basically	184	until either it is interrupted or the given time interval has
		185	passed (approximately - it might return a bit earlier even if not
		186	interrupted). Returns immediately if C<< interval <= 0 >>.
		187
139	this is a subsecond-resolution C<sleep ()>.	188	Basically this is a sub-second-resolution C<sleep ()>.
		189
		190	The range of the C<interval> is limited - libev only guarantees to work
		191	with sleep times of up to one day (C<< interval <= 86400 >>).
140		192
141	=item int ev_version_major ()	193	=item int ev_version_major ()
142		194
143	=item int ev_version_minor ()	195	=item int ev_version_minor ()
144		196
…		…
155	as this indicates an incompatible change. Minor versions are usually	207	as this indicates an incompatible change. Minor versions are usually
156	compatible to older versions, so a larger minor version alone is usually	208	compatible to older versions, so a larger minor version alone is usually
157	not a problem.	209	not a problem.
158		210
159	Example: Make sure we haven't accidentally been linked against the wrong	211	Example: Make sure we haven't accidentally been linked against the wrong
160	version.	212	version (note, however, that this will not detect other ABI mismatches,
		213	such as LFS or reentrancy).
161		214
162	assert (("libev version mismatch",	215	assert (("libev version mismatch",
163	ev_version_major () == EV_VERSION_MAJOR	216	ev_version_major () == EV_VERSION_MAJOR
164	&& ev_version_minor () >= EV_VERSION_MINOR));	217	&& ev_version_minor () >= EV_VERSION_MINOR));
165		218
166	=item unsigned int ev_supported_backends ()	219	=item unsigned int ev_supported_backends ()
167		220
168	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>	221	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
169	value) compiled into this binary of libev (independent of their	222	value) compiled into this binary of libev (independent of their
…		…
171	a description of the set values.	224	a description of the set values.
172		225
173	Example: make sure we have the epoll method, because yeah this is cool and	226	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	227	a must have and can we have a torrent of it please!!!11
175		228
176	assert (("sorry, no epoll, no sex",	229	assert (("sorry, no epoll, no sex",
177	ev_supported_backends () & EVBACKEND_EPOLL));	230	ev_supported_backends () & EVBACKEND_EPOLL));
178		231
179	=item unsigned int ev_recommended_backends ()	232	=item unsigned int ev_recommended_backends ()
180		233
181	Return the set of all backends compiled into this binary of libev and also	234	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	235	also recommended for this platform, meaning it will work for most file
		236	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	237	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	238	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	239	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.	240	probe for if you specify no backends explicitly.
187		241
188	=item unsigned int ev_embeddable_backends ()	242	=item unsigned int ev_embeddable_backends ()
189		243
190	Returns the set of backends that are embeddable in other event loops. This	244	Returns the set of backends that are embeddable in other event loops. This
191	is the theoretical, all-platform, value. To find which backends	245	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	246	current system. To find which embeddable backends might be supported on
193	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	247	the current system, you would need to look at C<ev_embeddable_backends ()
194	recommended ones.	248	& ev_supported_backends ()>, likewise for recommended ones.
195		249
196	See the description of C<ev_embed> watchers for more info.	250	See the description of C<ev_embed> watchers for more info.
197		251
198	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	252	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
199		253
200	Sets the allocation function to use (the prototype is similar - the	254	Sets the allocation function to use (the prototype is similar - the
201	semantics is identical - to the realloc C function). It is used to	255	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	256	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	257	when memory needs to be allocated (C<size != 0>), the library might abort
204	potentially destructive action. The default is your system realloc	258	or take some potentially destructive action.
205	function.	259
		260	Since some systems (at least OpenBSD and Darwin) fail to implement
		261	correct C<realloc> semantics, libev will use a wrapper around the system
		262	C<realloc> and C<free> functions by default.
206		263
207	You could override this function in high-availability programs to, say,	264	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,	265	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.	266	or even to sleep a while and retry until some memory is available.
210		267
211	Example: Replace the libev allocator with one that waits a bit and then	268	Example: Replace the libev allocator with one that waits a bit and then
212	retries).	269	retries (example requires a standards-compliant C<realloc>).
213		270
214	static void *	271	static void *
215	persistent_realloc (void *ptr, size_t size)	272	persistent_realloc (void *ptr, size_t size)
216	{	273	{
217	for (;;)	274	for (;;)
…		…
226	}	283	}
227		284
228	...	285	...
229	ev_set_allocator (persistent_realloc);	286	ev_set_allocator (persistent_realloc);
230		287
231	=item ev_set_syserr_cb (void (*cb)(const char *msg));	288	=item ev_set_syserr_cb (void (*cb)(const char *msg))
232		289
233	Set the callback function to call on a retryable syscall error (such	290	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	291	as failed select, poll, epoll_wait). The message is a printable string
235	indicating the system call or subsystem causing the problem. If this	292	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	293	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	294	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	295	requested operation, or, if the condition doesn't go away, do bad stuff
239	(such as abort).	296	(such as abort).
240		297
241	Example: This is basically the same thing that libev does internally, too.	298	Example: This is basically the same thing that libev does internally, too.
…		…
248	}	305	}
249		306
250	...	307	...
251	ev_set_syserr_cb (fatal_error);	308	ev_set_syserr_cb (fatal_error);
252		309
		310	=item ev_feed_signal (int signum)
		311
		312	This function can be used to "simulate" a signal receive. It is completely
		313	safe to call this function at any time, from any context, including signal
		314	handlers or random threads.
		315
		316	Its main use is to customise signal handling in your process, especially
		317	in the presence of threads. For example, you could block signals
		318	by default in all threads (and specifying C<EVFLAG_NOSIGMASK> when
		319	creating any loops), and in one thread, use C<sigwait> or any other
		320	mechanism to wait for signals, then "deliver" them to libev by calling
		321	C<ev_feed_signal>.
		322
253	=back	323	=back
254		324
255	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	325	=head1 FUNCTIONS CONTROLLING EVENT LOOPS
256		326
257	An event loop is described by a C<struct ev_loop *>. The library knows two	327	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	328	I<not> optional in this case unless libev 3 compatibility is disabled, as
259	events, and dynamically created loops which do not.	329	libev 3 had an C<ev_loop> function colliding with the struct name).
260		330
261	If you use threads, a common model is to run the default event loop	331	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	332	supports child process events, and dynamically created event loops which
263	create, you also create another event loop. Libev itself does no locking	333	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		334
268	=over 4	335	=over 4
269		336
270	=item struct ev_loop *ev_default_loop (unsigned int flags)	337	=item struct ev_loop *ev_default_loop (unsigned int flags)
271		338
272	This will initialise the default event loop if it hasn't been initialised	339	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	340	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	341	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).	342	C<ev_loop_new>.
		343
		344	If the default loop is already initialised then this function simply
		345	returns it (and ignores the flags. If that is troubling you, check
		346	C<ev_backend ()> afterwards). Otherwise it will create it with the given
		347	flags, which should almost always be C<0>, unless the caller is also the
		348	one calling C<ev_run> or otherwise qualifies as "the main program".
276		349
277	If you don't know what event loop to use, use the one returned from this	350	If you don't know what event loop to use, use the one returned from this
278	function.	351	function (or via the C<EV_DEFAULT> macro).
279		352
		353	Note that this function is I<not> thread-safe, so if you want to use it
		354	from multiple threads, you have to employ some kind of mutex (note also
		355	that this case is unlikely, as loops cannot be shared easily between
		356	threads anyway).
		357
280	The default loop is the only loop that can handle C<ev_signal> and	358	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	359	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	360	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	361	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	362	C<SIGCHLD> signal handler I<after> calling C<ev_default_init>.
285	C<ev_default_init>.	363
		364	Example: This is the most typical usage.
		365
		366	if (!ev_default_loop (0))
		367	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
		368
		369	Example: Restrict libev to the select and poll backends, and do not allow
		370	environment settings to be taken into account:
		371
		372	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
		373
		374	=item struct ev_loop *ev_loop_new (unsigned int flags)
		375
		376	This will create and initialise a new event loop object. If the loop
		377	could not be initialised, returns false.
		378
		379	This function is thread-safe, and one common way to use libev with
		380	threads is indeed to create one loop per thread, and using the default
		381	loop in the "main" or "initial" thread.
286		382
287	The flags argument can be used to specify special behaviour or specific	383	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>).	384	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
289		385
290	The following flags are supported:	386	The following flags are supported:
…		…
296	The default flags value. Use this if you have no clue (it's the right	392	The default flags value. Use this if you have no clue (it's the right
297	thing, believe me).	393	thing, believe me).
298		394
299	=item C<EVFLAG_NOENV>	395	=item C<EVFLAG_NOENV>
300		396
301	If this flag bit is ored into the flag value (or the program runs setuid	397	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	398	or setgid) then libev will I<not> look at the environment variable
303	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	399	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
304	override the flags completely if it is found in the environment. This is	400	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	401	useful to try out specific backends to test their performance, or to work
306	around bugs.	402	around bugs.
307		403
308	=item C<EVFLAG_FORKCHECK>	404	=item C<EVFLAG_FORKCHECK>
309		405
310	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after	406	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	407	make libev check for a fork in each iteration by enabling this flag.
312	enabling this flag.
313		408
314	This works by calling C<getpid ()> on every iteration of the loop,	409	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	410	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	411	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	412	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	413	without a system call and thus I<very> fast, but my GNU/Linux system also has
319	C<pthread_atfork> which is even faster).	414	C<pthread_atfork> which is even faster).
320		415
321	The big advantage of this flag is that you can forget about fork (and	416	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	417	forget about forgetting to tell libev about forking) when you use this
323	flag.	418	flag.
324		419
325	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>	420	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
326	environment variable.	421	environment variable.
		422
		423	=item C<EVFLAG_NOINOTIFY>
		424
		425	When this flag is specified, then libev will not attempt to use the
		426	I<inotify> API for its C<ev_stat> watchers. Apart from debugging and
		427	testing, this flag can be useful to conserve inotify file descriptors, as
		428	otherwise each loop using C<ev_stat> watchers consumes one inotify handle.
		429
		430	=item C<EVFLAG_SIGNALFD>
		431
		432	When this flag is specified, then libev will attempt to use the
		433	I<signalfd> API for its C<ev_signal> (and C<ev_child>) watchers. This API
		434	delivers signals synchronously, which makes it both faster and might make
		435	it possible to get the queued signal data. It can also simplify signal
		436	handling with threads, as long as you properly block signals in your
		437	threads that are not interested in handling them.
		438
		439	Signalfd will not be used by default as this changes your signal mask, and
		440	there are a lot of shoddy libraries and programs (glib's threadpool for
		441	example) that can't properly initialise their signal masks.
		442
		443	=item C<EVFLAG_NOSIGMASK>
		444
		445	When this flag is specified, then libev will avoid to modify the signal
		446	mask. Specifically, this means you have to make sure signals are unblocked
		447	when you want to receive them.
		448
		449	This behaviour is useful when you want to do your own signal handling, or
		450	want to handle signals only in specific threads and want to avoid libev
		451	unblocking the signals.
		452
		453	It's also required by POSIX in a threaded program, as libev calls
		454	C<sigprocmask>, whose behaviour is officially unspecified.
		455
		456	This flag's behaviour will become the default in future versions of libev.
327		457
328	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	458	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
329		459
330	This is your standard select(2) backend. Not I<completely> standard, as	460	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,	461	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	462	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	463	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.	464	usually the fastest backend for a low number of (low-numbered :) fds.
335		465
336	To get good performance out of this backend you need a high amount of	466	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	467	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	468	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	469	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	470	a look at C<ev_set_io_collect_interval ()> to increase the amount of
341	readyness notifications you get per iteration.	471	readiness notifications you get per iteration.
		472
		473	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		474	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		475	C<exceptfds> set on that platform).
342		476
343	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	477	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
344		478
345	And this is your standard poll(2) backend. It's more complicated	479	And this is your standard poll(2) backend. It's more complicated
346	than select, but handles sparse fds better and has no artificial	480	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	481	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,	482	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	483	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
350	performance tips.	484	performance tips.
351		485
		486	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		487	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
		488
352	=item C<EVBACKEND_EPOLL> (value 4, Linux)	489	=item C<EVBACKEND_EPOLL> (value 4, Linux)
353		490
		491	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
		492	kernels).
		493
354	For few fds, this backend is a bit little slower than poll and select,	494	For few fds, this backend is a bit little slower than poll and select, but
355	but it scales phenomenally better. While poll and select usually scale	495	it scales phenomenally better. While poll and select usually scale like
356	like O(total_fds) where n is the total number of fds (or the highest fd),	496	O(total_fds) where total_fds is the total number of fds (or the highest
357	epoll scales either O(1) or O(active_fds). The epoll design has a number	497	fd), epoll scales either O(1) or O(active_fds).
358	of shortcomings, such as silently dropping events in some hard-to-detect	498
359	cases and rewiring a syscall per fd change, no fork support and bad	499	The epoll mechanism deserves honorable mention as the most misdesigned
360	support for dup.	500	of the more advanced event mechanisms: mere annoyances include silently
		501	dropping file descriptors, requiring a system call per change per file
		502	descriptor (and unnecessary guessing of parameters), problems with dup,
		503	returning before the timeout value, resulting in additional iterations
		504	(and only giving 5ms accuracy while select on the same platform gives
		505	0.1ms) and so on. The biggest issue is fork races, however - if a program
		506	forks then I<both> parent and child process have to recreate the epoll
		507	set, which can take considerable time (one syscall per file descriptor)
		508	and is of course hard to detect.
		509
		510	Epoll is also notoriously buggy - embedding epoll fds I<should> work,
		511	but of course I<doesn't>, and epoll just loves to report events for
		512	totally I<different> file descriptors (even already closed ones, so
		513	one cannot even remove them from the set) than registered in the set
		514	(especially on SMP systems). Libev tries to counter these spurious
		515	notifications by employing an additional generation counter and comparing
		516	that against the events to filter out spurious ones, recreating the set
		517	when required. Epoll also erroneously rounds down timeouts, but gives you
		518	no way to know when and by how much, so sometimes you have to busy-wait
		519	because epoll returns immediately despite a nonzero timeout. And last
		520	not least, it also refuses to work with some file descriptors which work
		521	perfectly fine with C<select> (files, many character devices...).
		522
		523	Epoll is truly the train wreck among event poll mechanisms, a frankenpoll,
		524	cobbled together in a hurry, no thought to design or interaction with
		525	others. Oh, the pain, will it ever stop...
361		526
362	While stopping, setting and starting an I/O watcher in the same iteration	527	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	528	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	529	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	530	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
366	very well if you register events for both fds.	531	file descriptors might not work very well if you register events for both
367		532	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		533
372	Best performance from this backend is achieved by not unregistering all	534	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.	535	watchers for a file descriptor until it has been closed, if possible,
374	keep at least one watcher active per fd at all times.	536	i.e. keep at least one watcher active per fd at all times. Stopping and
		537	starting a watcher (without re-setting it) also usually doesn't cause
		538	extra overhead. A fork can both result in spurious notifications as well
		539	as in libev having to destroy and recreate the epoll object, which can
		540	take considerable time and thus should be avoided.
375		541
		542	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		543	faster than epoll for maybe up to a hundred file descriptors, depending on
		544	the usage. So sad.
		545
376	While nominally embeddeble in other event loops, this feature is broken in	546	While nominally embeddable in other event loops, this feature is broken in
377	all kernel versions tested so far.	547	all kernel versions tested so far.
		548
		549	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		550	C<EVBACKEND_POLL>.
378		551
379	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	552	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
380		553
381	Kqueue deserves special mention, as at the time of this writing, it	554	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	555	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	556	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"	557	it's completely useless). Unlike epoll, however, whose brokenness
		558	is by design, these kqueue bugs can (and eventually will) be fixed
		559	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	560	"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)	561	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
387	system like NetBSD.	562	system like NetBSD.
388		563
389	You still can embed kqueue into a normal poll or select backend and use it	564	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	565	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.	566	the target platform). See C<ev_embed> watchers for more info.
392		567
393	It scales in the same way as the epoll backend, but the interface to the	568	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	569	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	570	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	571	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	572	two event changes per incident. Support for C<fork ()> is very bad (but
398	drops fds silently in similarly hard-to-detect cases.	573	sane, unlike epoll) and it drops fds silently in similarly hard-to-detect
		574	cases
399		575
400	This backend usually performs well under most conditions.	576	This backend usually performs well under most conditions.
401		577
402	While nominally embeddable in other event loops, this doesn't work	578	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	579	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	580	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	581	(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	582	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
407	sockets.	583	also broken on OS X)) and, did I mention it, using it only for sockets.
		584
		585	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		586	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		587	C<NOTE_EOF>.
408		588
409	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	589	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
410		590
411	This is not implemented yet (and might never be, unless you send me an	591	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	592	implementation). According to reports, C</dev/poll> only supports sockets
…		…
416	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	596	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
417		597
418	This uses the Solaris 10 event port mechanism. As with everything on Solaris,	598	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)).	599	it's really slow, but it still scales very well (O(active_fds)).
420		600
421	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
423	blocking when no data (or space) is available.
424
425	While this backend scales well, it requires one system call per active	601	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	602	file descriptor per loop iteration. For small and medium numbers of file
427	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	603	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
428	might perform better.	604	might perform better.
429		605
430	On the positive side, ignoring the spurious readyness notifications, this	606	On the positive side, this backend actually performed fully to
431	backend actually performed to specification in all tests and is fully	607	specification in all tests and is fully embeddable, which is a rare feat
432	embeddable, which is a rare feat among the OS-specific backends.	608	among the OS-specific backends (I vastly prefer correctness over speed
		609	hacks).
		610
		611	On the negative side, the interface is I<bizarre> - so bizarre that
		612	even sun itself gets it wrong in their code examples: The event polling
		613	function sometimes returns events to the caller even though an error
		614	occurred, but with no indication whether it has done so or not (yes, it's
		615	even documented that way) - deadly for edge-triggered interfaces where you
		616	absolutely have to know whether an event occurred or not because you have
		617	to re-arm the watcher.
		618
		619	Fortunately libev seems to be able to work around these idiocies.
		620
		621	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		622	C<EVBACKEND_POLL>.
433		623
434	=item C<EVBACKEND_ALL>	624	=item C<EVBACKEND_ALL>
435		625
436	Try all backends (even potentially broken ones that wouldn't be tried	626	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	627	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
438	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	628	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
439		629
440	It is definitely not recommended to use this flag.	630	It is definitely not recommended to use this flag, use whatever
		631	C<ev_recommended_backends ()> returns, or simply do not specify a backend
		632	at all.
		633
		634	=item C<EVBACKEND_MASK>
		635
		636	Not a backend at all, but a mask to select all backend bits from a
		637	C<flags> value, in case you want to mask out any backends from a flags
		638	value (e.g. when modifying the C<LIBEV_FLAGS> environment variable).
441		639
442	=back	640	=back
443		641
444	If one or more of these are ored into the flags value, then only these	642	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	643	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.	644	here). If none are specified, all backends in C<ev_recommended_backends
447		645	()> 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		646
471	Example: Try to create a event loop that uses epoll and nothing else.	647	Example: Try to create a event loop that uses epoll and nothing else.
472		648
473	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	649	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
474	if (!epoller)	650	if (!epoller)
475	fatal ("no epoll found here, maybe it hides under your chair");	651	fatal ("no epoll found here, maybe it hides under your chair");
476		652
		653	Example: Use whatever libev has to offer, but make sure that kqueue is
		654	used if available.
		655
		656	struct ev_loop *loop = ev_loop_new (ev_recommended_backends () | EVBACKEND_KQUEUE);
		657
477	=item ev_default_destroy ()	658	=item ev_loop_destroy (loop)
478		659
479	Destroys the default loop again (frees all memory and kernel state	660	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	661	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	662	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>	663	responsibility to either stop all watchers cleanly yourself I<before>
483	calling this function, or cope with the fact afterwards (which is usually	664	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	665	the easiest thing, you can just ignore the watchers and/or C<free ()> them
485	for example).	666	for example).
486		667
487	Note that certain global state, such as signal state, will not be freed by	668	Note that certain global state, such as signal state (and installed signal
488	this function, and related watchers (such as signal and child watchers)	669	handlers), will not be freed by this function, and related watchers (such
489	would need to be stopped manually.	670	as signal and child watchers) would need to be stopped manually.
490		671
491	In general it is not advisable to call this function except in the	672	This function is normally used on loop objects allocated by
492	rare occasion where you really need to free e.g. the signal handling	673	C<ev_loop_new>, but it can also be used on the default loop returned by
		674	C<ev_default_loop>, in which case it is not thread-safe.
		675
		676	Note that it is not advisable to call this function on the default loop
		677	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	678	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>).	679	and C<ev_loop_destroy>.
495		680
496	=item ev_loop_destroy (loop)	681	=item ev_loop_fork (loop)
497		682
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	683	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	684	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	685	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	686	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	687	child before resuming or calling C<ev_run>.
508	functions, and it will only take effect at the next C<ev_loop> iteration.	688
		689	Again, you I<have> to call it on I<any> loop that you want to re-use after
		690	a fork, I<even if you do not plan to use the loop in the parent>. This is
		691	because some kernel interfaces *cough* I<kqueue> *cough* do funny things
		692	during fork.
509		693
510	On the other hand, you only need to call this function in the child	694	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	695	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.	696	you just fork+exec or create a new loop in the child, you don't have to
		697	call it at all (in fact, C<epoll> is so badly broken that it makes a
		698	difference, but libev will usually detect this case on its own and do a
		699	costly reset of the backend).
513		700
514	The function itself is quite fast and it's usually not a problem to call	701	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	702	it just in case after a fork.
516	quite nicely into a call to C<pthread_atfork>:
517		703
		704	Example: Automate calling C<ev_loop_fork> on the default loop when
		705	using pthreads.
		706
		707	static void
		708	post_fork_child (void)
		709	{
		710	ev_loop_fork (EV_DEFAULT);
		711	}
		712
		713	...
518	pthread_atfork (0, 0, ev_default_fork);	714	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		715
526	=item int ev_is_default_loop (loop)	716	=item int ev_is_default_loop (loop)
527		717
528	Returns true when the given loop actually is the default loop, false otherwise.	718	Returns true when the given loop is, in fact, the default loop, and false
		719	otherwise.
529		720
530	=item unsigned int ev_loop_count (loop)	721	=item unsigned int ev_iteration (loop)
531		722
532	Returns the count of loop iterations for the loop, which is identical to	723	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	724	to the number of times libev did poll for new events. It starts at C<0>
534	happily wraps around with enough iterations.	725	and happily wraps around with enough iterations.
535		726
536	This value can sometimes be useful as a generation counter of sorts (it	727	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	728	"ticks" the number of loop iterations), as it roughly corresponds with
538	C<ev_prepare> and C<ev_check> calls.	729	C<ev_prepare> and C<ev_check> calls - and is incremented between the
		730	prepare and check phases.
		731
		732	=item unsigned int ev_depth (loop)
		733
		734	Returns the number of times C<ev_run> was entered minus the number of
		735	times C<ev_run> was exited normally, in other words, the recursion depth.
		736
		737	Outside C<ev_run>, this number is zero. In a callback, this number is
		738	C<1>, unless C<ev_run> was invoked recursively (or from another thread),
		739	in which case it is higher.
		740
		741	Leaving C<ev_run> abnormally (setjmp/longjmp, cancelling the thread,
		742	throwing an exception etc.), doesn't count as "exit" - consider this
		743	as a hint to avoid such ungentleman-like behaviour unless it's really
		744	convenient, in which case it is fully supported.
539		745
540	=item unsigned int ev_backend (loop)	746	=item unsigned int ev_backend (loop)
541		747
542	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	748	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
543	use.	749	use.
…		…
548	received events and started processing them. This timestamp does not	754	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	755	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	756	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).	757	event occurring (or more correctly, libev finding out about it).
552		758
		759	=item ev_now_update (loop)
		760
		761	Establishes the current time by querying the kernel, updating the time
		762	returned by C<ev_now ()> in the progress. This is a costly operation and
		763	is usually done automatically within C<ev_run ()>.
		764
		765	This function is rarely useful, but when some event callback runs for a
		766	very long time without entering the event loop, updating libev's idea of
		767	the current time is a good idea.
		768
		769	See also L<The special problem of time updates> in the C<ev_timer> section.
		770
		771	=item ev_suspend (loop)
		772
		773	=item ev_resume (loop)
		774
		775	These two functions suspend and resume an event loop, for use when the
		776	loop is not used for a while and timeouts should not be processed.
		777
		778	A typical use case would be an interactive program such as a game: When
		779	the user presses C<^Z> to suspend the game and resumes it an hour later it
		780	would be best to handle timeouts as if no time had actually passed while
		781	the program was suspended. This can be achieved by calling C<ev_suspend>
		782	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		783	C<ev_resume> directly afterwards to resume timer processing.
		784
		785	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		786	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		787	will be rescheduled (that is, they will lose any events that would have
		788	occurred while suspended).
		789
		790	After calling C<ev_suspend> you B<must not> call I<any> function on the
		791	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		792	without a previous call to C<ev_suspend>.
		793
		794	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		795	event loop time (see C<ev_now_update>).
		796
553	=item ev_loop (loop, int flags)	797	=item ev_run (loop, int flags)
554		798
555	Finally, this is it, the event handler. This function usually is called	799	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	800	after you have initialised all your watchers and you want to start
557	events.	801	handling events. It will ask the operating system for any new events, call
		802	the watcher callbacks, an then repeat the whole process indefinitely: This
		803	is why event loops are called I<loops>.
558		804
559	If the flags argument is specified as C<0>, it will not return until	805	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.	806	until either no event watchers are active anymore or C<ev_break> was
		807	called.
561		808
562	Please note that an explicit C<ev_unloop> is usually better than	809	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	810	relying on all watchers to be stopped when deciding when a program has
564	finished (especially in interactive programs), but having a program that	811	finished (especially in interactive programs), but having a program
565	automatically loops as long as it has to and no longer by virtue of	812	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.	813	of relying on its watchers stopping correctly, that is truly a thing of
		814	beauty.
567		815
		816	This function is also I<mostly> exception-safe - you can break out of
		817	a C<ev_run> call by calling C<longjmp> in a callback, throwing a C++
		818	exception and so on. This does not decrement the C<ev_depth> value, nor
		819	will it clear any outstanding C<EVBREAK_ONE> breaks.
		820
568	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	821	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	822	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.	823	block your process in case there are no events and will return after one
		824	iteration of the loop. This is sometimes useful to poll and handle new
		825	events while doing lengthy calculations, to keep the program responsive.
571		826
572	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	827	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	828	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	829	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	830	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	831	user-registered callback will be called), and will return after one
		832	iteration of the loop.
		833
		834	This is useful if you are waiting for some external event in conjunction
		835	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	836	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.	837	usually a better approach for this kind of thing.
579		838
580	Here are the gory details of what C<ev_loop> does:	839	Here are the gory details of what C<ev_run> does (this is for your
		840	understanding, not a guarantee that things will work exactly like this in
		841	future versions):
581		842
		843	- Increment loop depth.
		844	- Reset the ev_break status.
582	- Before the first iteration, call any pending watchers.	845	- Before the first iteration, call any pending watchers.
		846	LOOP:
583	* If EVFLAG_FORKCHECK was used, check for a fork.	847	- If EVFLAG_FORKCHECK was used, check for a fork.
584	- If a fork was detected, queue and call all fork watchers.	848	- If a fork was detected (by any means), queue and call all fork watchers.
585	- Queue and call all prepare watchers.	849	- Queue and call all prepare watchers.
		850	- If ev_break was called, goto FINISH.
586	- If we have been forked, recreate the kernel state.	851	- If we have been forked, detach and recreate the kernel state
		852	as to not disturb the other process.
587	- Update the kernel state with all outstanding changes.	853	- Update the kernel state with all outstanding changes.
588	- Update the "event loop time".	854	- Update the "event loop time" (ev_now ()).
589	- Calculate for how long to sleep or block, if at all	855	- Calculate for how long to sleep or block, if at all
590	(active idle watchers, EVLOOP_NONBLOCK or not having	856	(active idle watchers, EVRUN_NOWAIT or not having
591	any active watchers at all will result in not sleeping).	857	any active watchers at all will result in not sleeping).
592	- Sleep if the I/O and timer collect interval say so.	858	- Sleep if the I/O and timer collect interval say so.
		859	- Increment loop iteration counter.
593	- Block the process, waiting for any events.	860	- Block the process, waiting for any events.
594	- Queue all outstanding I/O (fd) events.	861	- Queue all outstanding I/O (fd) events.
595	- Update the "event loop time" and do time jump handling.	862	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
596	- Queue all outstanding timers.	863	- Queue all expired timers.
597	- Queue all outstanding periodics.	864	- Queue all expired periodics.
598	- If no events are pending now, queue all idle watchers.	865	- Queue all idle watchers with priority higher than that of pending events.
599	- Queue all check watchers.	866	- Queue all check watchers.
600	- Call all queued watchers in reverse order (i.e. check watchers first).	867	- 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	868	Signals and child watchers are implemented as I/O watchers, and will
602	be handled here by queueing them when their watcher gets executed.	869	be handled here by queueing them when their watcher gets executed.
603	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	870	- If ev_break has been called, or EVRUN_ONCE or EVRUN_NOWAIT
604	were used, or there are no active watchers, return, otherwise	871	were used, or there are no active watchers, goto FINISH, otherwise
605	continue with step *.	872	continue with step LOOP.
		873	FINISH:
		874	- Reset the ev_break status iff it was EVBREAK_ONE.
		875	- Decrement the loop depth.
		876	- Return.
606		877
607	Example: Queue some jobs and then loop until no events are outstanding	878	Example: Queue some jobs and then loop until no events are outstanding
608	anymore.	879	anymore.
609		880
610	... queue jobs here, make sure they register event watchers as long	881	... 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..)	882	... as they still have work to do (even an idle watcher will do..)
612	ev_loop (my_loop, 0);	883	ev_run (my_loop, 0);
613	... jobs done. yeah!	884	... jobs done or somebody called break. yeah!
614		885
615	=item ev_unloop (loop, how)	886	=item ev_break (loop, how)
616		887
617	Can be used to make a call to C<ev_loop> return early (but only after it	888	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	889	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	890	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.	891	C<EVBREAK_ALL>, which will make all nested C<ev_run> calls return.
621		892
622	This "unloop state" will be cleared when entering C<ev_loop> again.	893	This "break state" will be cleared on the next call to C<ev_run>.
		894
		895	It is safe to call C<ev_break> from outside any C<ev_run> calls, too, in
		896	which case it will have no effect.
623		897
624	=item ev_ref (loop)	898	=item ev_ref (loop)
625		899
626	=item ev_unref (loop)	900	=item ev_unref (loop)
627		901
628	Ref/unref can be used to add or remove a reference count on the event	902	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	903	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	904	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	905
632	returning, ev_unref() after starting, and ev_ref() before stopping it. For	906	This is useful when you have a watcher that you never intend to
		907	unregister, but that nevertheless should not keep C<ev_run> from
		908	returning. In such a case, call C<ev_unref> after starting, and C<ev_ref>
		909	before stopping it.
		910
633	example, libev itself uses this for its internal signal pipe: It is not	911	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	912	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	913	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	914	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>	915	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,	916	before stop> (but only if the watcher wasn't active before, or was active
639	respectively).	917	before, respectively. Note also that libev might stop watchers itself
		918	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		919	in the callback).
640		920
641	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	921	Example: Create a signal watcher, but keep it from keeping C<ev_run>
642	running when nothing else is active.	922	running when nothing else is active.
643		923
644	struct ev_signal exitsig;	924	ev_signal exitsig;
645	ev_signal_init (&exitsig, sig_cb, SIGINT);	925	ev_signal_init (&exitsig, sig_cb, SIGINT);
646	ev_signal_start (loop, &exitsig);	926	ev_signal_start (loop, &exitsig);
647	evf_unref (loop);	927	ev_unref (loop);
648		928
649	Example: For some weird reason, unregister the above signal handler again.	929	Example: For some weird reason, unregister the above signal handler again.
650		930
651	ev_ref (loop);	931	ev_ref (loop);
652	ev_signal_stop (loop, &exitsig);	932	ev_signal_stop (loop, &exitsig);
653		933
654	=item ev_set_io_collect_interval (loop, ev_tstamp interval)	934	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
655		935
656	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)	936	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
657		937
658	These advanced functions influence the time that libev will spend waiting	938	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	939	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.	940	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		941	latency.
661		942
662	Setting these to a higher value (the C<interval> I<must> be >= C<0>)	943	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	944	allows libev to delay invocation of I/O and timer/periodic callbacks
664	increase efficiency of loop iterations.	945	to increase efficiency of loop iterations (or to increase power-saving
		946	opportunities).
665		947
666	The background is that sometimes your program runs just fast enough to	948	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	949	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	950	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	951	events, especially with backends like C<select ()> which have a high
670	overhead for the actual polling but can deliver many events at once.	952	overhead for the actual polling but can deliver many events at once.
671		953
672	By setting a higher I<io collect interval> you allow libev to spend more	954	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,	955	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	956	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	957	C<ev_timer>) will not be affected. Setting this to a non-null value will
676	introduce an additional C<ev_sleep ()> call into most loop iterations.	958	introduce an additional C<ev_sleep ()> call into most loop iterations. The
		959	sleep time ensures that libev will not poll for I/O events more often then
		960	once per this interval, on average (as long as the host time resolution is
		961	good enough).
677		962
678	Likewise, by setting a higher I<timeout collect interval> you allow libev	963	Likewise, by setting a higher I<timeout collect interval> you allow libev
679	to spend more time collecting timeouts, at the expense of increased	964	to spend more time collecting timeouts, at the expense of increased
680	latency (the watcher callback will be called later). C<ev_io> watchers	965	latency/jitter/inexactness (the watcher callback will be called
681	will not be affected. Setting this to a non-null value will not introduce	966	later). C<ev_io> watchers will not be affected. Setting this to a non-null
682	any overhead in libev.	967	value will not introduce any overhead in libev.
683		968
684	Many (busy) programs can usually benefit by setting the io collect	969	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	970	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	971	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>,	972	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.	973	as this approaches the timing granularity of most systems. Note that if
		974	you do transactions with the outside world and you can't increase the
		975	parallelity, then this setting will limit your transaction rate (if you
		976	need to poll once per transaction and the I/O collect interval is 0.01,
		977	then you can't do more than 100 transactions per second).
		978
		979	Setting the I<timeout collect interval> can improve the opportunity for
		980	saving power, as the program will "bundle" timer callback invocations that
		981	are "near" in time together, by delaying some, thus reducing the number of
		982	times the process sleeps and wakes up again. Another useful technique to
		983	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		984	they fire on, say, one-second boundaries only.
		985
		986	Example: we only need 0.1s timeout granularity, and we wish not to poll
		987	more often than 100 times per second:
		988
		989	ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
		990	ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
		991
		992	=item ev_invoke_pending (loop)
		993
		994	This call will simply invoke all pending watchers while resetting their
		995	pending state. Normally, C<ev_run> does this automatically when required,
		996	but when overriding the invoke callback this call comes handy. This
		997	function can be invoked from a watcher - this can be useful for example
		998	when you want to do some lengthy calculation and want to pass further
		999	event handling to another thread (you still have to make sure only one
		1000	thread executes within C<ev_invoke_pending> or C<ev_run> of course).
		1001
		1002	=item int ev_pending_count (loop)
		1003
		1004	Returns the number of pending watchers - zero indicates that no watchers
		1005	are pending.
		1006
		1007	=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
		1008
		1009	This overrides the invoke pending functionality of the loop: Instead of
		1010	invoking all pending watchers when there are any, C<ev_run> will call
		1011	this callback instead. This is useful, for example, when you want to
		1012	invoke the actual watchers inside another context (another thread etc.).
		1013
		1014	If you want to reset the callback, use C<ev_invoke_pending> as new
		1015	callback.
		1016
		1017	=item ev_set_loop_release_cb (loop, void (*release)(EV_P), void (*acquire)(EV_P))
		1018
		1019	Sometimes you want to share the same loop between multiple threads. This
		1020	can be done relatively simply by putting mutex_lock/unlock calls around
		1021	each call to a libev function.
		1022
		1023	However, C<ev_run> can run an indefinite time, so it is not feasible
		1024	to wait for it to return. One way around this is to wake up the event
		1025	loop via C<ev_break> and C<av_async_send>, another way is to set these
		1026	I<release> and I<acquire> callbacks on the loop.
		1027
		1028	When set, then C<release> will be called just before the thread is
		1029	suspended waiting for new events, and C<acquire> is called just
		1030	afterwards.
		1031
		1032	Ideally, C<release> will just call your mutex_unlock function, and
		1033	C<acquire> will just call the mutex_lock function again.
		1034
		1035	While event loop modifications are allowed between invocations of
		1036	C<release> and C<acquire> (that's their only purpose after all), no
		1037	modifications done will affect the event loop, i.e. adding watchers will
		1038	have no effect on the set of file descriptors being watched, or the time
		1039	waited. Use an C<ev_async> watcher to wake up C<ev_run> when you want it
		1040	to take note of any changes you made.
		1041
		1042	In theory, threads executing C<ev_run> will be async-cancel safe between
		1043	invocations of C<release> and C<acquire>.
		1044
		1045	See also the locking example in the C<THREADS> section later in this
		1046	document.
		1047
		1048	=item ev_set_userdata (loop, void *data)
		1049
		1050	=item void *ev_userdata (loop)
		1051
		1052	Set and retrieve a single C<void *> associated with a loop. When
		1053	C<ev_set_userdata> has never been called, then C<ev_userdata> returns
		1054	C<0>.
		1055
		1056	These two functions can be used to associate arbitrary data with a loop,
		1057	and are intended solely for the C<invoke_pending_cb>, C<release> and
		1058	C<acquire> callbacks described above, but of course can be (ab-)used for
		1059	any other purpose as well.
		1060
		1061	=item ev_verify (loop)
		1062
		1063	This function only does something when C<EV_VERIFY> support has been
		1064	compiled in, which is the default for non-minimal builds. It tries to go
		1065	through all internal structures and checks them for validity. If anything
		1066	is found to be inconsistent, it will print an error message to standard
		1067	error and call C<abort ()>.
		1068
		1069	This can be used to catch bugs inside libev itself: under normal
		1070	circumstances, this function will never abort as of course libev keeps its
		1071	data structures consistent.
689		1072
690	=back	1073	=back
691		1074
692		1075
693	=head1 ANATOMY OF A WATCHER	1076	=head1 ANATOMY OF A WATCHER
694		1077
		1078	In the following description, uppercase C<TYPE> in names stands for the
		1079	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		1080	watchers and C<ev_io_start> for I/O watchers.
		1081
695	A watcher is a structure that you create and register to record your	1082	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	1083	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:	1084	to wait for STDIN to become readable, you would create an C<ev_io> watcher
		1085	for that:
698		1086
699	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1087	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
700	{	1088	{
701	ev_io_stop (w);	1089	ev_io_stop (w);
702	ev_unloop (loop, EVUNLOOP_ALL);	1090	ev_break (loop, EVBREAK_ALL);
703	}	1091	}
704		1092
705	struct ev_loop *loop = ev_default_loop (0);	1093	struct ev_loop *loop = ev_default_loop (0);
		1094
706	struct ev_io stdin_watcher;	1095	ev_io stdin_watcher;
		1096
707	ev_init (&stdin_watcher, my_cb);	1097	ev_init (&stdin_watcher, my_cb);
708	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	1098	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
709	ev_io_start (loop, &stdin_watcher);	1099	ev_io_start (loop, &stdin_watcher);
		1100
710	ev_loop (loop, 0);	1101	ev_run (loop, 0);
711		1102
712	As you can see, you are responsible for allocating the memory for your	1103	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,	1104	watcher structures (and it is I<usually> a bad idea to do this on the
714	although this can sometimes be quite valid).	1105	stack).
715		1106
		1107	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		1108	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
		1109
716	Each watcher structure must be initialised by a call to C<ev_init	1110	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	1111	*, 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	1112	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	1113	time the event loop detects that the file descriptor given is readable
720	is readable and/or writable).	1114	and/or writable).
721		1115
722	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	1116	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	1117	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	1118	is also a macro to combine initialisation and setting in one call: C<<
725	(watcher *, callback, ...) >>.	1119	ev_TYPE_init (watcher *, callback, ...) >>.
726		1120
727	To make the watcher actually watch out for events, you have to start it	1121	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	1122	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	1123	*) >>), and you can stop watching for events at any time by calling the
730	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	1124	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
731		1125
732	As long as your watcher is active (has been started but not stopped) you	1126	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	1127	must not touch the values stored in it. Most specifically you must never
734	reinitialise it or call its C<set> macro.	1128	reinitialise it or call its C<ev_TYPE_set> macro.
735		1129
736	Each and every callback receives the event loop pointer as first, the	1130	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	1131	registered watcher structure as second, and a bitset of received events as
738	third argument.	1132	third argument.
739		1133
…		…
748	=item C<EV_WRITE>	1142	=item C<EV_WRITE>
749		1143
750	The file descriptor in the C<ev_io> watcher has become readable and/or	1144	The file descriptor in the C<ev_io> watcher has become readable and/or
751	writable.	1145	writable.
752		1146
753	=item C<EV_TIMEOUT>	1147	=item C<EV_TIMER>
754		1148
755	The C<ev_timer> watcher has timed out.	1149	The C<ev_timer> watcher has timed out.
756		1150
757	=item C<EV_PERIODIC>	1151	=item C<EV_PERIODIC>
758		1152
…		…
776		1170
777	=item C<EV_PREPARE>	1171	=item C<EV_PREPARE>
778		1172
779	=item C<EV_CHECK>	1173	=item C<EV_CHECK>
780		1174
781	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts	1175	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	1176	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	1177	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	1178	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	1179	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	1180	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
787	C<ev_loop> from blocking).	1181	C<ev_run> from blocking).
788		1182
789	=item C<EV_EMBED>	1183	=item C<EV_EMBED>
790		1184
791	The embedded event loop specified in the C<ev_embed> watcher needs attention.	1185	The embedded event loop specified in the C<ev_embed> watcher needs attention.
792		1186
793	=item C<EV_FORK>	1187	=item C<EV_FORK>
794		1188
795	The event loop has been resumed in the child process after fork (see	1189	The event loop has been resumed in the child process after fork (see
796	C<ev_fork>).	1190	C<ev_fork>).
797		1191
		1192	=item C<EV_CLEANUP>
		1193
		1194	The event loop is about to be destroyed (see C<ev_cleanup>).
		1195
798	=item C<EV_ASYNC>	1196	=item C<EV_ASYNC>
799		1197
800	The given async watcher has been asynchronously notified (see C<ev_async>).	1198	The given async watcher has been asynchronously notified (see C<ev_async>).
801		1199
		1200	=item C<EV_CUSTOM>
		1201
		1202	Not ever sent (or otherwise used) by libev itself, but can be freely used
		1203	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		1204
802	=item C<EV_ERROR>	1205	=item C<EV_ERROR>
803		1206
804	An unspecified error has occured, the watcher has been stopped. This might	1207	An unspecified error has occurred, the watcher has been stopped. This might
805	happen because the watcher could not be properly started because libev	1208	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	1209	ran out of memory, a file descriptor was found to be closed or any other
		1210	problem. Libev considers these application bugs.
		1211
807	problem. You best act on it by reporting the problem and somehow coping	1212	You best act on it by reporting the problem and somehow coping with the
808	with the watcher being stopped.	1213	watcher being stopped. Note that well-written programs should not receive
		1214	an error ever, so when your watcher receives it, this usually indicates a
		1215	bug in your program.
809		1216
810	Libev will usually signal a few "dummy" events together with an error,	1217	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	1218	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	1219	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	1220	the error from read() or write(). This will not work in multi-threaded
814	programs, though, so beware.	1221	programs, though, as the fd could already be closed and reused for another
		1222	thing, so beware.
815		1223
816	=back	1224	=back
817		1225
818	=head2 GENERIC WATCHER FUNCTIONS	1226	=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		1227
823	=over 4	1228	=over 4
824		1229
825	=item C<ev_init> (ev_TYPE *watcher, callback)	1230	=item C<ev_init> (ev_TYPE *watcher, callback)
826		1231
…		…
832	which rolls both calls into one.	1237	which rolls both calls into one.
833		1238
834	You can reinitialise a watcher at any time as long as it has been stopped	1239	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.	1240	(or never started) and there are no pending events outstanding.
836		1241
837	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	1242	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
838	int revents)>.	1243	int revents)>.
839		1244
		1245	Example: Initialise an C<ev_io> watcher in two steps.
		1246
		1247	ev_io w;
		1248	ev_init (&w, my_cb);
		1249	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1250
840	=item C<ev_TYPE_set> (ev_TYPE *, [args])	1251	=item C<ev_TYPE_set> (ev_TYPE *watcher, [args])
841		1252
842	This macro initialises the type-specific parts of a watcher. You need to	1253	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	1254	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	1255	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	1256	macro on a watcher that is active (it can be pending, however, which is a
846	difference to the C<ev_init> macro).	1257	difference to the C<ev_init> macro).
847		1258
848	Although some watcher types do not have type-specific arguments	1259	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.	1260	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
850		1261
		1262	See C<ev_init>, above, for an example.
		1263
851	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	1264	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
852		1265
853	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	1266	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	1267	calls into a single call. This is the most convenient method to initialise
855	a watcher. The same limitations apply, of course.	1268	a watcher. The same limitations apply, of course.
856		1269
		1270	Example: Initialise and set an C<ev_io> watcher in one step.
		1271
		1272	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1273
857	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	1274	=item C<ev_TYPE_start> (loop, ev_TYPE *watcher)
858		1275
859	Starts (activates) the given watcher. Only active watchers will receive	1276	Starts (activates) the given watcher. Only active watchers will receive
860	events. If the watcher is already active nothing will happen.	1277	events. If the watcher is already active nothing will happen.
861		1278
		1279	Example: Start the C<ev_io> watcher that is being abused as example in this
		1280	whole section.
		1281
		1282	ev_io_start (EV_DEFAULT_UC, &w);
		1283
862	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	1284	=item C<ev_TYPE_stop> (loop, ev_TYPE *watcher)
863		1285
864	Stops the given watcher again (if active) and clears the pending	1286	Stops the given watcher if active, and clears the pending status (whether
		1287	the watcher was active or not).
		1288
865	status. It is possible that stopped watchers are pending (for example,	1289	It is possible that stopped watchers are pending - for example,
866	non-repeating timers are being stopped when they become pending), but	1290	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	1291	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	1292	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.	1293	therefore a good idea to always call its C<ev_TYPE_stop> function.
870		1294
871	=item bool ev_is_active (ev_TYPE *watcher)	1295	=item bool ev_is_active (ev_TYPE *watcher)
872		1296
873	Returns a true value iff the watcher is active (i.e. it has been started	1297	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	1298	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)	1314	=item ev_cb_set (ev_TYPE *watcher, callback)
891		1315
892	Change the callback. You can change the callback at virtually any time	1316	Change the callback. You can change the callback at virtually any time
893	(modulo threads).	1317	(modulo threads).
894		1318
895	=item ev_set_priority (ev_TYPE *watcher, priority)	1319	=item ev_set_priority (ev_TYPE *watcher, int priority)
896		1320
897	=item int ev_priority (ev_TYPE *watcher)	1321	=item int ev_priority (ev_TYPE *watcher)
898		1322
899	Set and query the priority of the watcher. The priority is a small	1323	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>	1324	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
901	(default: C<-2>). Pending watchers with higher priority will be invoked	1325	(default: C<-2>). Pending watchers with higher priority will be invoked
902	before watchers with lower priority, but priority will not keep watchers	1326	before watchers with lower priority, but priority will not keep watchers
903	from being executed (except for C<ev_idle> watchers).	1327	from being executed (except for C<ev_idle> watchers).
904		1328
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	1329	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.	1330	you need to look at C<ev_idle> watchers, which provide this functionality.
912		1331
913	You I<must not> change the priority of a watcher as long as it is active or	1332	You I<must not> change the priority of a watcher as long as it is active or
914	pending.	1333	pending.
915		1334
		1335	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1336	fine, as long as you do not mind that the priority value you query might
		1337	or might not have been clamped to the valid range.
		1338
916	The default priority used by watchers when no priority has been set is	1339	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 :).	1340	always C<0>, which is supposed to not be too high and not be too low :).
918		1341
919	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is	1342	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	1343	priorities.
921	or might not have been adjusted to be within valid range.
922		1344
923	=item ev_invoke (loop, ev_TYPE *watcher, int revents)	1345	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
924		1346
925	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither	1347	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	1348	C<loop> nor C<revents> need to be valid as long as the watcher callback
927	can deal with that fact.	1349	can deal with that fact, as both are simply passed through to the
		1350	callback.
928		1351
929	=item int ev_clear_pending (loop, ev_TYPE *watcher)	1352	=item int ev_clear_pending (loop, ev_TYPE *watcher)
930		1353
931	If the watcher is pending, this function returns clears its pending status	1354	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	1355	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>.	1356	watcher isn't pending it does nothing and returns C<0>.
934		1357
		1358	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1359	callback to be invoked, which can be accomplished with this function.
		1360
		1361	=item ev_feed_event (loop, ev_TYPE *watcher, int revents)
		1362
		1363	Feeds the given event set into the event loop, as if the specified event
		1364	had happened for the specified watcher (which must be a pointer to an
		1365	initialised but not necessarily started event watcher). Obviously you must
		1366	not free the watcher as long as it has pending events.
		1367
		1368	Stopping the watcher, letting libev invoke it, or calling
		1369	C<ev_clear_pending> will clear the pending event, even if the watcher was
		1370	not started in the first place.
		1371
		1372	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
		1373	functions that do not need a watcher.
		1374
935	=back	1375	=back
936		1376
		1377	See also the L<ASSOCIATING CUSTOM DATA WITH A WATCHER> and L<BUILDING YOUR
		1378	OWN COMPOSITE WATCHERS> idioms.
937		1379
938	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1380	=head2 WATCHER STATES
939		1381
940	Each watcher has, by default, a member C<void *data> that you can change	1382	There are various watcher states mentioned throughout this manual -
941	and read at any time, libev will completely ignore it. This can be used	1383	active, pending and so on. In this section these states and the rules to
942	to associate arbitrary data with your watcher. If you need more data and	1384	transition between them will be described in more detail - and while these
943	don't want to allocate memory and store a pointer to it in that data	1385	rules might look complicated, they usually do "the right thing".
944	member, you can also "subclass" the watcher type and provide your own
945	data:
946		1386
947	struct my_io	1387	=over 4
948	{
949	struct ev_io io;
950	int otherfd;
951	void *somedata;
952	struct whatever *mostinteresting;
953	}
954		1388
955	And since your callback will be called with a pointer to the watcher, you	1389	=item initialiased
956	can cast it back to your own type:
957		1390
958	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1391	Before a watcher can be registered with the event loop it has to be
959	{	1392	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to
960	struct my_io *w = (struct my_io *)w_;	1393	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.
961	...
962	}
963		1394
964	More interesting and less C-conformant ways of casting your callback type	1395	In this state it is simply some block of memory that is suitable for
965	instead have been omitted.	1396	use in an event loop. It can be moved around, freed, reused etc. at
		1397	will - as long as you either keep the memory contents intact, or call
		1398	C<ev_TYPE_init> again.
966		1399
967	Another common scenario is having some data structure with multiple	1400	=item started/running/active
968	watchers:
969		1401
970	struct my_biggy	1402	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes
971	{	1403	property of the event loop, and is actively waiting for events. While in
972	int some_data;	1404	this state it cannot be accessed (except in a few documented ways), moved,
973	ev_timer t1;	1405	freed or anything else - the only legal thing is to keep a pointer to it,
974	ev_timer t2;	1406	and call libev functions on it that are documented to work on active watchers.
975	}
976		1407
977	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1408	=item pending
978	you need to use C<offsetof>:
979		1409
980	#include <stddef.h>	1410	If a watcher is active and libev determines that an event it is interested
		1411	in has occurred (such as a timer expiring), it will become pending. It will
		1412	stay in this pending state until either it is stopped or its callback is
		1413	about to be invoked, so it is not normally pending inside the watcher
		1414	callback.
981		1415
		1416	The watcher might or might not be active while it is pending (for example,
		1417	an expired non-repeating timer can be pending but no longer active). If it
		1418	is stopped, it can be freely accessed (e.g. by calling C<ev_TYPE_set>),
		1419	but it is still property of the event loop at this time, so cannot be
		1420	moved, freed or reused. And if it is active the rules described in the
		1421	previous item still apply.
		1422
		1423	It is also possible to feed an event on a watcher that is not active (e.g.
		1424	via C<ev_feed_event>), in which case it becomes pending without being
		1425	active.
		1426
		1427	=item stopped
		1428
		1429	A watcher can be stopped implicitly by libev (in which case it might still
		1430	be pending), or explicitly by calling its C<ev_TYPE_stop> function. The
		1431	latter will clear any pending state the watcher might be in, regardless
		1432	of whether it was active or not, so stopping a watcher explicitly before
		1433	freeing it is often a good idea.
		1434
		1435	While stopped (and not pending) the watcher is essentially in the
		1436	initialised state, that is, it can be reused, moved, modified in any way
		1437	you wish (but when you trash the memory block, you need to C<ev_TYPE_init>
		1438	it again).
		1439
		1440	=back
		1441
		1442	=head2 WATCHER PRIORITY MODELS
		1443
		1444	Many event loops support I<watcher priorities>, which are usually small
		1445	integers that influence the ordering of event callback invocation
		1446	between watchers in some way, all else being equal.
		1447
		1448	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1449	description for the more technical details such as the actual priority
		1450	range.
		1451
		1452	There are two common ways how these these priorities are being interpreted
		1453	by event loops:
		1454
		1455	In the more common lock-out model, higher priorities "lock out" invocation
		1456	of lower priority watchers, which means as long as higher priority
		1457	watchers receive events, lower priority watchers are not being invoked.
		1458
		1459	The less common only-for-ordering model uses priorities solely to order
		1460	callback invocation within a single event loop iteration: Higher priority
		1461	watchers are invoked before lower priority ones, but they all get invoked
		1462	before polling for new events.
		1463
		1464	Libev uses the second (only-for-ordering) model for all its watchers
		1465	except for idle watchers (which use the lock-out model).
		1466
		1467	The rationale behind this is that implementing the lock-out model for
		1468	watchers is not well supported by most kernel interfaces, and most event
		1469	libraries will just poll for the same events again and again as long as
		1470	their callbacks have not been executed, which is very inefficient in the
		1471	common case of one high-priority watcher locking out a mass of lower
		1472	priority ones.
		1473
		1474	Static (ordering) priorities are most useful when you have two or more
		1475	watchers handling the same resource: a typical usage example is having an
		1476	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1477	timeouts. Under load, data might be received while the program handles
		1478	other jobs, but since timers normally get invoked first, the timeout
		1479	handler will be executed before checking for data. In that case, giving
		1480	the timer a lower priority than the I/O watcher ensures that I/O will be
		1481	handled first even under adverse conditions (which is usually, but not
		1482	always, what you want).
		1483
		1484	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1485	will only be executed when no same or higher priority watchers have
		1486	received events, they can be used to implement the "lock-out" model when
		1487	required.
		1488
		1489	For example, to emulate how many other event libraries handle priorities,
		1490	you can associate an C<ev_idle> watcher to each such watcher, and in
		1491	the normal watcher callback, you just start the idle watcher. The real
		1492	processing is done in the idle watcher callback. This causes libev to
		1493	continuously poll and process kernel event data for the watcher, but when
		1494	the lock-out case is known to be rare (which in turn is rare :), this is
		1495	workable.
		1496
		1497	Usually, however, the lock-out model implemented that way will perform
		1498	miserably under the type of load it was designed to handle. In that case,
		1499	it might be preferable to stop the real watcher before starting the
		1500	idle watcher, so the kernel will not have to process the event in case
		1501	the actual processing will be delayed for considerable time.
		1502
		1503	Here is an example of an I/O watcher that should run at a strictly lower
		1504	priority than the default, and which should only process data when no
		1505	other events are pending:
		1506
		1507	ev_idle idle; // actual processing watcher
		1508	ev_io io; // actual event watcher
		1509
982	static void	1510	static void
983	t1_cb (EV_P_ struct ev_timer *w, int revents)	1511	io_cb (EV_P_ ev_io *w, int revents)
984	{	1512	{
985	struct my_biggy big = (struct my_biggy *	1513	// stop the I/O watcher, we received the event, but
986	(((char *)w) - offsetof (struct my_biggy, t1));	1514	// are not yet ready to handle it.
		1515	ev_io_stop (EV_A_ w);
		1516
		1517	// start the idle watcher to handle the actual event.
		1518	// it will not be executed as long as other watchers
		1519	// with the default priority are receiving events.
		1520	ev_idle_start (EV_A_ &idle);
987	}	1521	}
988		1522
989	static void	1523	static void
990	t2_cb (EV_P_ struct ev_timer *w, int revents)	1524	idle_cb (EV_P_ ev_idle *w, int revents)
991	{	1525	{
992	struct my_biggy big = (struct my_biggy *	1526	// actual processing
993	(((char *)w) - offsetof (struct my_biggy, t2));	1527	read (STDIN_FILENO, ...);
		1528
		1529	// have to start the I/O watcher again, as
		1530	// we have handled the event
		1531	ev_io_start (EV_P_ &io);
994	}	1532	}
		1533
		1534	// initialisation
		1535	ev_idle_init (&idle, idle_cb);
		1536	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1537	ev_io_start (EV_DEFAULT_ &io);
		1538
		1539	In the "real" world, it might also be beneficial to start a timer, so that
		1540	low-priority connections can not be locked out forever under load. This
		1541	enables your program to keep a lower latency for important connections
		1542	during short periods of high load, while not completely locking out less
		1543	important ones.
995		1544
996		1545
997	=head1 WATCHER TYPES	1546	=head1 WATCHER TYPES
998		1547
999	This section describes each watcher in detail, but will not repeat	1548	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	1572	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	1573	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	1574	descriptors to non-blocking mode is also usually a good idea (but not
1026	required if you know what you are doing).	1575	required if you know what you are doing).
1027		1576
1028	If you must do this, then force the use of a known-to-be-good backend
1029	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
1030	C<EVBACKEND_POLL>).
1031
1032	Another thing you have to watch out for is that it is quite easy to	1577	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	1578	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	1579	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	1580	because there is no data. It is very easy to get into this situation even
1036	lot of those (for example solaris ports), it is very easy to get into	1581	with a relatively standard program structure. Thus it is best to always
1037	this situation even with a relatively standard program structure. Thus	1582	use non-blocking I/O: An extra C<read>(2) returning C<EAGAIN> is far
1038	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.	1583	preferable to a program hanging until some data arrives.
1040		1584
1041	If you cannot run the fd in non-blocking mode (for example you should not	1585	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	1586	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	1587	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	1588	interface such as poll (fortunately in the case of Xlib, it already does
1045	its own, so its quite safe to use).	1589	this on its own, so its quite safe to use). Some people additionally
		1590	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1591	indefinitely.
		1592
		1593	But really, best use non-blocking mode.
1046		1594
1047	=head3 The special problem of disappearing file descriptors	1595	=head3 The special problem of disappearing file descriptors
1048		1596
1049	Some backends (e.g. kqueue, epoll) need to be told about closing a file	1597	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,	1598	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	1599	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	1600	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	1601	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	1602	registered with libev, there is no efficient way to see that this is, in
1055	fact, a different file descriptor.	1603	fact, a different file descriptor.
1056		1604
…		…
1074		1622
1075	There is no workaround possible except not registering events	1623	There is no workaround possible except not registering events
1076	for potentially C<dup ()>'ed file descriptors, or to resort to	1624	for potentially C<dup ()>'ed file descriptors, or to resort to
1077	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.	1625	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
1078		1626
		1627	=head3 The special problem of files
		1628
		1629	Many people try to use C<select> (or libev) on file descriptors
		1630	representing files, and expect it to become ready when their program
		1631	doesn't block on disk accesses (which can take a long time on their own).
		1632
		1633	However, this cannot ever work in the "expected" way - you get a readiness
		1634	notification as soon as the kernel knows whether and how much data is
		1635	there, and in the case of open files, that's always the case, so you
		1636	always get a readiness notification instantly, and your read (or possibly
		1637	write) will still block on the disk I/O.
		1638
		1639	Another way to view it is that in the case of sockets, pipes, character
		1640	devices and so on, there is another party (the sender) that delivers data
		1641	on its own, but in the case of files, there is no such thing: the disk
		1642	will not send data on its own, simply because it doesn't know what you
		1643	wish to read - you would first have to request some data.
		1644
		1645	Since files are typically not-so-well supported by advanced notification
		1646	mechanism, libev tries hard to emulate POSIX behaviour with respect
		1647	to files, even though you should not use it. The reason for this is
		1648	convenience: sometimes you want to watch STDIN or STDOUT, which is
		1649	usually a tty, often a pipe, but also sometimes files or special devices
		1650	(for example, C<epoll> on Linux works with F</dev/random> but not with
		1651	F</dev/urandom>), and even though the file might better be served with
		1652	asynchronous I/O instead of with non-blocking I/O, it is still useful when
		1653	it "just works" instead of freezing.
		1654
		1655	So avoid file descriptors pointing to files when you know it (e.g. use
		1656	libeio), but use them when it is convenient, e.g. for STDIN/STDOUT, or
		1657	when you rarely read from a file instead of from a socket, and want to
		1658	reuse the same code path.
		1659
1079	=head3 The special problem of fork	1660	=head3 The special problem of fork
1080		1661
1081	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit	1662	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
1082	useless behaviour. Libev fully supports fork, but needs to be told about	1663	useless behaviour. Libev fully supports fork, but needs to be told about
1083	it in the child.	1664	it in the child if you want to continue to use it in the child.
1084		1665
1085	To support fork in your programs, you either have to call	1666	To support fork in your child processes, you have to call C<ev_loop_fork
1086	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,	1667	()> after a fork in the child, enable C<EVFLAG_FORKCHECK>, or resort to
1087	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1668	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
1088	C<EVBACKEND_POLL>.
1089		1669
		1670	=head3 The special problem of SIGPIPE
		1671
		1672	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1673	when writing to a pipe whose other end has been closed, your program gets
		1674	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1675	this is sensible behaviour, for daemons, this is usually undesirable.
		1676
		1677	So when you encounter spurious, unexplained daemon exits, make sure you
		1678	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1679	somewhere, as that would have given you a big clue).
		1680
		1681	=head3 The special problem of accept()ing when you can't
		1682
		1683	Many implementations of the POSIX C<accept> function (for example,
		1684	found in post-2004 Linux) have the peculiar behaviour of not removing a
		1685	connection from the pending queue in all error cases.
		1686
		1687	For example, larger servers often run out of file descriptors (because
		1688	of resource limits), causing C<accept> to fail with C<ENFILE> but not
		1689	rejecting the connection, leading to libev signalling readiness on
		1690	the next iteration again (the connection still exists after all), and
		1691	typically causing the program to loop at 100% CPU usage.
		1692
		1693	Unfortunately, the set of errors that cause this issue differs between
		1694	operating systems, there is usually little the app can do to remedy the
		1695	situation, and no known thread-safe method of removing the connection to
		1696	cope with overload is known (to me).
		1697
		1698	One of the easiest ways to handle this situation is to just ignore it
		1699	- when the program encounters an overload, it will just loop until the
		1700	situation is over. While this is a form of busy waiting, no OS offers an
		1701	event-based way to handle this situation, so it's the best one can do.
		1702
		1703	A better way to handle the situation is to log any errors other than
		1704	C<EAGAIN> and C<EWOULDBLOCK>, making sure not to flood the log with such
		1705	messages, and continue as usual, which at least gives the user an idea of
		1706	what could be wrong ("raise the ulimit!"). For extra points one could stop
		1707	the C<ev_io> watcher on the listening fd "for a while", which reduces CPU
		1708	usage.
		1709
		1710	If your program is single-threaded, then you could also keep a dummy file
		1711	descriptor for overload situations (e.g. by opening F</dev/null>), and
		1712	when you run into C<ENFILE> or C<EMFILE>, close it, run C<accept>,
		1713	close that fd, and create a new dummy fd. This will gracefully refuse
		1714	clients under typical overload conditions.
		1715
		1716	The last way to handle it is to simply log the error and C<exit>, as
		1717	is often done with C<malloc> failures, but this results in an easy
		1718	opportunity for a DoS attack.
1090		1719
1091	=head3 Watcher-Specific Functions	1720	=head3 Watcher-Specific Functions
1092		1721
1093	=over 4	1722	=over 4
1094		1723
1095	=item ev_io_init (ev_io *, callback, int fd, int events)	1724	=item ev_io_init (ev_io *, callback, int fd, int events)
1096		1725
1097	=item ev_io_set (ev_io *, int fd, int events)	1726	=item ev_io_set (ev_io *, int fd, int events)
1098		1727
1099	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1728	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	1729	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.	1730	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
1102		1731
1103	=item int fd [read-only]	1732	=item int fd [read-only]
1104		1733
1105	The file descriptor being watched.	1734	The file descriptor being watched.
1106		1735
…		…
1114		1743
1115	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1744	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	1745	readable, but only once. Since it is likely line-buffered, you could
1117	attempt to read a whole line in the callback.	1746	attempt to read a whole line in the callback.
1118		1747
1119	static void	1748	static void
1120	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1749	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
1121	{	1750	{
1122	ev_io_stop (loop, w);	1751	ev_io_stop (loop, w);
1123	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1752	.. read from stdin here (or from w->fd) and handle any I/O errors
1124	}	1753	}
1125		1754
1126	...	1755	...
1127	struct ev_loop *loop = ev_default_init (0);	1756	struct ev_loop *loop = ev_default_init (0);
1128	struct ev_io stdin_readable;	1757	ev_io stdin_readable;
1129	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1758	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1130	ev_io_start (loop, &stdin_readable);	1759	ev_io_start (loop, &stdin_readable);
1131	ev_loop (loop, 0);	1760	ev_run (loop, 0);
1132		1761
1133		1762
1134	=head2 C<ev_timer> - relative and optionally repeating timeouts	1763	=head2 C<ev_timer> - relative and optionally repeating timeouts
1135		1764
1136	Timer watchers are simple relative timers that generate an event after a	1765	Timer watchers are simple relative timers that generate an event after a
1137	given time, and optionally repeating in regular intervals after that.	1766	given time, and optionally repeating in regular intervals after that.
1138		1767
1139	The timers are based on real time, that is, if you register an event that	1768	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	1769	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	1770	year, it will still time out after (roughly) one hour. "Roughly" because
1142	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1771	detecting time jumps is hard, and some inaccuracies are unavoidable (the
1143	monotonic clock option helps a lot here).	1772	monotonic clock option helps a lot here).
		1773
		1774	The callback is guaranteed to be invoked only I<after> its timeout has
		1775	passed (not I<at>, so on systems with very low-resolution clocks this
		1776	might introduce a small delay). If multiple timers become ready during the
		1777	same loop iteration then the ones with earlier time-out values are invoked
		1778	before ones of the same priority with later time-out values (but this is
		1779	no longer true when a callback calls C<ev_run> recursively).
		1780
		1781	=head3 Be smart about timeouts
		1782
		1783	Many real-world problems involve some kind of timeout, usually for error
		1784	recovery. A typical example is an HTTP request - if the other side hangs,
		1785	you want to raise some error after a while.
		1786
		1787	What follows are some ways to handle this problem, from obvious and
		1788	inefficient to smart and efficient.
		1789
		1790	In the following, a 60 second activity timeout is assumed - a timeout that
		1791	gets reset to 60 seconds each time there is activity (e.g. each time some
		1792	data or other life sign was received).
		1793
		1794	=over 4
		1795
		1796	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1797
		1798	This is the most obvious, but not the most simple way: In the beginning,
		1799	start the watcher:
		1800
		1801	ev_timer_init (timer, callback, 60., 0.);
		1802	ev_timer_start (loop, timer);
		1803
		1804	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1805	and start it again:
		1806
		1807	ev_timer_stop (loop, timer);
		1808	ev_timer_set (timer, 60., 0.);
		1809	ev_timer_start (loop, timer);
		1810
		1811	This is relatively simple to implement, but means that each time there is
		1812	some activity, libev will first have to remove the timer from its internal
		1813	data structure and then add it again. Libev tries to be fast, but it's
		1814	still not a constant-time operation.
		1815
		1816	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1817
		1818	This is the easiest way, and involves using C<ev_timer_again> instead of
		1819	C<ev_timer_start>.
		1820
		1821	To implement this, configure an C<ev_timer> with a C<repeat> value
		1822	of C<60> and then call C<ev_timer_again> at start and each time you
		1823	successfully read or write some data. If you go into an idle state where
		1824	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1825	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1826
		1827	That means you can ignore both the C<ev_timer_start> function and the
		1828	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1829	member and C<ev_timer_again>.
		1830
		1831	At start:
		1832
		1833	ev_init (timer, callback);
		1834	timer->repeat = 60.;
		1835	ev_timer_again (loop, timer);
		1836
		1837	Each time there is some activity:
		1838
		1839	ev_timer_again (loop, timer);
		1840
		1841	It is even possible to change the time-out on the fly, regardless of
		1842	whether the watcher is active or not:
		1843
		1844	timer->repeat = 30.;
		1845	ev_timer_again (loop, timer);
		1846
		1847	This is slightly more efficient then stopping/starting the timer each time
		1848	you want to modify its timeout value, as libev does not have to completely
		1849	remove and re-insert the timer from/into its internal data structure.
		1850
		1851	It is, however, even simpler than the "obvious" way to do it.
		1852
		1853	=item 3. Let the timer time out, but then re-arm it as required.
		1854
		1855	This method is more tricky, but usually most efficient: Most timeouts are
		1856	relatively long compared to the intervals between other activity - in
		1857	our example, within 60 seconds, there are usually many I/O events with
		1858	associated activity resets.
		1859
		1860	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1861	but remember the time of last activity, and check for a real timeout only
		1862	within the callback:
		1863
		1864	ev_tstamp last_activity; // time of last activity
		1865
		1866	static void
		1867	callback (EV_P_ ev_timer *w, int revents)
		1868	{
		1869	ev_tstamp now = ev_now (EV_A);
		1870	ev_tstamp timeout = last_activity + 60.;
		1871
		1872	// if last_activity + 60. is older than now, we did time out
		1873	if (timeout < now)
		1874	{
		1875	// timeout occurred, take action
		1876	}
		1877	else
		1878	{
		1879	// callback was invoked, but there was some activity, re-arm
		1880	// the watcher to fire in last_activity + 60, which is
		1881	// guaranteed to be in the future, so "again" is positive:
		1882	w->repeat = timeout - now;
		1883	ev_timer_again (EV_A_ w);
		1884	}
		1885	}
		1886
		1887	To summarise the callback: first calculate the real timeout (defined
		1888	as "60 seconds after the last activity"), then check if that time has
		1889	been reached, which means something I<did>, in fact, time out. Otherwise
		1890	the callback was invoked too early (C<timeout> is in the future), so
		1891	re-schedule the timer to fire at that future time, to see if maybe we have
		1892	a timeout then.
		1893
		1894	Note how C<ev_timer_again> is used, taking advantage of the
		1895	C<ev_timer_again> optimisation when the timer is already running.
		1896
		1897	This scheme causes more callback invocations (about one every 60 seconds
		1898	minus half the average time between activity), but virtually no calls to
		1899	libev to change the timeout.
		1900
		1901	To start the timer, simply initialise the watcher and set C<last_activity>
		1902	to the current time (meaning we just have some activity :), then call the
		1903	callback, which will "do the right thing" and start the timer:
		1904
		1905	ev_init (timer, callback);
		1906	last_activity = ev_now (loop);
		1907	callback (loop, timer, EV_TIMER);
		1908
		1909	And when there is some activity, simply store the current time in
		1910	C<last_activity>, no libev calls at all:
		1911
		1912	last_activity = ev_now (loop);
		1913
		1914	This technique is slightly more complex, but in most cases where the
		1915	time-out is unlikely to be triggered, much more efficient.
		1916
		1917	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1918	callback :) - just change the timeout and invoke the callback, which will
		1919	fix things for you.
		1920
		1921	=item 4. Wee, just use a double-linked list for your timeouts.
		1922
		1923	If there is not one request, but many thousands (millions...), all
		1924	employing some kind of timeout with the same timeout value, then one can
		1925	do even better:
		1926
		1927	When starting the timeout, calculate the timeout value and put the timeout
		1928	at the I<end> of the list.
		1929
		1930	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1931	the list is expected to fire (for example, using the technique #3).
		1932
		1933	When there is some activity, remove the timer from the list, recalculate
		1934	the timeout, append it to the end of the list again, and make sure to
		1935	update the C<ev_timer> if it was taken from the beginning of the list.
		1936
		1937	This way, one can manage an unlimited number of timeouts in O(1) time for
		1938	starting, stopping and updating the timers, at the expense of a major
		1939	complication, and having to use a constant timeout. The constant timeout
		1940	ensures that the list stays sorted.
		1941
		1942	=back
		1943
		1944	So which method the best?
		1945
		1946	Method #2 is a simple no-brain-required solution that is adequate in most
		1947	situations. Method #3 requires a bit more thinking, but handles many cases
		1948	better, and isn't very complicated either. In most case, choosing either
		1949	one is fine, with #3 being better in typical situations.
		1950
		1951	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1952	rather complicated, but extremely efficient, something that really pays
		1953	off after the first million or so of active timers, i.e. it's usually
		1954	overkill :)
		1955
		1956	=head3 The special problem of time updates
		1957
		1958	Establishing the current time is a costly operation (it usually takes at
		1959	least two system calls): EV therefore updates its idea of the current
		1960	time only before and after C<ev_run> collects new events, which causes a
		1961	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1962	lots of events in one iteration.
1144		1963
1145	The relative timeouts are calculated relative to the C<ev_now ()>	1964	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	1965	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	1966	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	1967	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:	1968	timeout on the current time, use something like this to adjust for this:
1150		1969
1151	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1970	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
1152		1971
1153	The callback is guarenteed to be invoked only when its timeout has passed,	1972	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	1973	update of the time returned by C<ev_now ()> by calling C<ev_now_update
1155	order of execution is undefined.	1974	()>.
		1975
		1976	=head3 The special problems of suspended animation
		1977
		1978	When you leave the server world it is quite customary to hit machines that
		1979	can suspend/hibernate - what happens to the clocks during such a suspend?
		1980
		1981	Some quick tests made with a Linux 2.6.28 indicate that a suspend freezes
		1982	all processes, while the clocks (C<times>, C<CLOCK_MONOTONIC>) continue
		1983	to run until the system is suspended, but they will not advance while the
		1984	system is suspended. That means, on resume, it will be as if the program
		1985	was frozen for a few seconds, but the suspend time will not be counted
		1986	towards C<ev_timer> when a monotonic clock source is used. The real time
		1987	clock advanced as expected, but if it is used as sole clocksource, then a
		1988	long suspend would be detected as a time jump by libev, and timers would
		1989	be adjusted accordingly.
		1990
		1991	I would not be surprised to see different behaviour in different between
		1992	operating systems, OS versions or even different hardware.
		1993
		1994	The other form of suspend (job control, or sending a SIGSTOP) will see a
		1995	time jump in the monotonic clocks and the realtime clock. If the program
		1996	is suspended for a very long time, and monotonic clock sources are in use,
		1997	then you can expect C<ev_timer>s to expire as the full suspension time
		1998	will be counted towards the timers. When no monotonic clock source is in
		1999	use, then libev will again assume a timejump and adjust accordingly.
		2000
		2001	It might be beneficial for this latter case to call C<ev_suspend>
		2002	and C<ev_resume> in code that handles C<SIGTSTP>, to at least get
		2003	deterministic behaviour in this case (you can do nothing against
		2004	C<SIGSTOP>).
1156		2005
1157	=head3 Watcher-Specific Functions and Data Members	2006	=head3 Watcher-Specific Functions and Data Members
1158		2007
1159	=over 4	2008	=over 4
1160		2009
1161	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	2010	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
1162		2011
1163	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	2012	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
1164		2013
1165	Configure the timer to trigger after C<after> seconds. If C<repeat> is	2014	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	2015	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	2016	reached. If it is positive, then the timer will automatically be
1168	later, again, and again, until stopped manually.	2017	configured to trigger again C<repeat> seconds later, again, and again,
		2018	until stopped manually.
1169		2019
1170	The timer itself will do a best-effort at avoiding drift, that is, if you	2020	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	2021	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	2022	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	2023	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.	2024	do stuff) the timer will not fire more than once per event loop iteration.
1175		2025
1176	=item ev_timer_again (loop, ev_timer *)	2026	=item ev_timer_again (loop, ev_timer *)
1177		2027
1178	This will act as if the timer timed out and restart it again if it is	2028	This will act as if the timer timed out and restarts it again if it is
1179	repeating. The exact semantics are:	2029	repeating. The exact semantics are:
1180		2030
1181	If the timer is pending, its pending status is cleared.	2031	If the timer is pending, its pending status is cleared.
1182		2032
1183	If the timer is started but nonrepeating, stop it (as if it timed out).	2033	If the timer is started but non-repeating, stop it (as if it timed out).
1184		2034
1185	If the timer is repeating, either start it if necessary (with the	2035	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.	2036	C<repeat> value), or reset the running timer to the C<repeat> value.
1187		2037
1188	This sounds a bit complicated, but here is a useful and typical	2038	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	2039	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		2040
1198	That means you can ignore the C<after> value and C<ev_timer_start>	2041	=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		2042
1201	ev_timer_init (timer, callback, 0., 5.);	2043	Returns the remaining time until a timer fires. If the timer is active,
1202	ev_timer_again (loop, timer);	2044	then this time is relative to the current event loop time, otherwise it's
1203	...	2045	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		2046
1210	This is more slightly efficient then stopping/starting the timer each time	2047	That is, after an C<ev_timer_set (w, 5, 7)>, C<ev_timer_remaining> returns
1211	you want to modify its timeout value.	2048	C<5>. When the timer is started and one second passes, C<ev_timer_remaining>
		2049	will return C<4>. When the timer expires and is restarted, it will return
		2050	roughly C<7> (likely slightly less as callback invocation takes some time,
		2051	too), and so on.
1212		2052
1213	=item ev_tstamp repeat [read-write]	2053	=item ev_tstamp repeat [read-write]
1214		2054
1215	The current C<repeat> value. Will be used each time the watcher times out	2055	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),	2056	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.	2057	which is also when any modifications are taken into account.
1218		2058
1219	=back	2059	=back
1220		2060
1221	=head3 Examples	2061	=head3 Examples
1222		2062
1223	Example: Create a timer that fires after 60 seconds.	2063	Example: Create a timer that fires after 60 seconds.
1224		2064
1225	static void	2065	static void
1226	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2066	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
1227	{	2067	{
1228	.. one minute over, w is actually stopped right here	2068	.. one minute over, w is actually stopped right here
1229	}	2069	}
1230		2070
1231	struct ev_timer mytimer;	2071	ev_timer mytimer;
1232	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	2072	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1233	ev_timer_start (loop, &mytimer);	2073	ev_timer_start (loop, &mytimer);
1234		2074
1235	Example: Create a timeout timer that times out after 10 seconds of	2075	Example: Create a timeout timer that times out after 10 seconds of
1236	inactivity.	2076	inactivity.
1237		2077
1238	static void	2078	static void
1239	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2079	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
1240	{	2080	{
1241	.. ten seconds without any activity	2081	.. ten seconds without any activity
1242	}	2082	}
1243		2083
1244	struct ev_timer mytimer;	2084	ev_timer mytimer;
1245	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	2085	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1246	ev_timer_again (&mytimer); /* start timer */	2086	ev_timer_again (&mytimer); /* start timer */
1247	ev_loop (loop, 0);	2087	ev_run (loop, 0);
1248		2088
1249	// and in some piece of code that gets executed on any "activity":	2089	// and in some piece of code that gets executed on any "activity":
1250	// reset the timeout to start ticking again at 10 seconds	2090	// reset the timeout to start ticking again at 10 seconds
1251	ev_timer_again (&mytimer);	2091	ev_timer_again (&mytimer);
1252		2092
1253		2093
1254	=head2 C<ev_periodic> - to cron or not to cron?	2094	=head2 C<ev_periodic> - to cron or not to cron?
1255		2095
1256	Periodic watchers are also timers of a kind, but they are very versatile	2096	Periodic watchers are also timers of a kind, but they are very versatile
1257	(and unfortunately a bit complex).	2097	(and unfortunately a bit complex).
1258		2098
1259	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	2099	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	2100	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	2101	(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 ()	2102	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	2103	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	2104	wrist-watch).
1265	roughly 10 seconds later).
1266		2105
1267	They can also be used to implement vastly more complex timers, such as	2106	You can tell a periodic watcher to trigger after some specific point
1268	triggering an event on each midnight, local time or other, complicated,	2107	in time: for example, if you tell a periodic watcher to trigger "in 10
1269	rules.	2108	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		2109	not a delay) and then reset your system clock to January of the previous
		2110	year, then it will take a year or more to trigger the event (unlike an
		2111	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		2112	it, as it uses a relative timeout).
1270		2113
		2114	C<ev_periodic> watchers can also be used to implement vastly more complex
		2115	timers, such as triggering an event on each "midnight, local time", or
		2116	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		2117	those cannot react to time jumps.
		2118
1271	As with timers, the callback is guarenteed to be invoked only when the	2119	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	2120	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.	2121	timers become ready during the same loop iteration then the ones with
		2122	earlier time-out values are invoked before ones with later time-out values
		2123	(but this is no longer true when a callback calls C<ev_run> recursively).
1274		2124
1275	=head3 Watcher-Specific Functions and Data Members	2125	=head3 Watcher-Specific Functions and Data Members
1276		2126
1277	=over 4	2127	=over 4
1278		2128
1279	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	2129	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1280		2130
1281	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	2131	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1282		2132
1283	Lots of arguments, lets sort it out... There are basically three modes of	2133	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:	2134	operation, and we will explain them from simplest to most complex:
1285		2135
1286	=over 4	2136	=over 4
1287		2137
1288	=item * absolute timer (at = time, interval = reschedule_cb = 0)	2138	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
1289		2139
1290	In this configuration the watcher triggers an event at the wallclock time	2140	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,	2141	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	2142	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.	2143	will be stopped and invoked when the system clock reaches or surpasses
		2144	this point in time.
1294		2145
1295	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	2146	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
1296		2147
1297	In this mode the watcher will always be scheduled to time out at the next	2148	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)	2149	C<offset + N * interval> time (for some integer N, which can also be
1299	and then repeat, regardless of any time jumps.	2150	negative) and then repeat, regardless of any time jumps. The C<offset>
		2151	argument is merely an offset into the C<interval> periods.
1300		2152
1301	This can be used to create timers that do not drift with respect to system	2153	This can be used to create timers that do not drift with respect to the
1302	time:	2154	system clock, for example, here is an C<ev_periodic> that triggers each
		2155	hour, on the hour (with respect to UTC):
1303		2156
1304	ev_periodic_set (&periodic, 0., 3600., 0);	2157	ev_periodic_set (&periodic, 0., 3600., 0);
1305		2158
1306	This doesn't mean there will always be 3600 seconds in between triggers,	2159	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	2160	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	2161	full hour (UTC), or more correctly, when the system time is evenly divisible
1309	by 3600.	2162	by 3600.
1310		2163
1311	Another way to think about it (for the mathematically inclined) is that	2164	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	2165	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.	2166	time where C<time = offset (mod interval)>, regardless of any time jumps.
1314		2167
1315	For numerical stability it is preferable that the C<at> value is near	2168	The C<interval> I<MUST> be positive, and for numerical stability, the
1316	C<ev_now ()> (the current time), but there is no range requirement for	2169	interval value should be higher than C<1/8192> (which is around 100
1317	this value.	2170	microseconds) and C<offset> should be higher than C<0> and should have
		2171	at most a similar magnitude as the current time (say, within a factor of
		2172	ten). Typical values for offset are, in fact, C<0> or something between
		2173	C<0> and C<interval>, which is also the recommended range.
1318		2174
		2175	Note also that there is an upper limit to how often a timer can fire (CPU
		2176	speed for example), so if C<interval> is very small then timing stability
		2177	will of course deteriorate. Libev itself tries to be exact to be about one
		2178	millisecond (if the OS supports it and the machine is fast enough).
		2179
1319	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)	2180	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
1320		2181
1321	In this mode the values for C<interval> and C<at> are both being	2182	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	2183	ignored. Instead, each time the periodic watcher gets scheduled, the
1323	reschedule callback will be called with the watcher as first, and the	2184	reschedule callback will be called with the watcher as first, and the
1324	current time as second argument.	2185	current time as second argument.
1325		2186
1326	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	2187	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,	2188	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	2189	allowed by documentation here>.
1329	starting an C<ev_prepare> watcher, which is legal).
1330		2190
		2191	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		2192	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		2193	only event loop modification you are allowed to do).
		2194
1331	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	2195	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
1332	ev_tstamp now)>, e.g.:	2196	*w, ev_tstamp now)>, e.g.:
1333		2197
		2198	static ev_tstamp
1334	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	2199	my_rescheduler (ev_periodic *w, ev_tstamp now)
1335	{	2200	{
1336	return now + 60.;	2201	return now + 60.;
1337	}	2202	}
1338		2203
1339	It must return the next time to trigger, based on the passed time value	2204	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	2205	(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	2206	will usually be called just before the callback will be triggered, but
1342	might be called at other times, too.	2207	might be called at other times, too.
1343		2208
1344	NOTE: I<< This callback must always return a time that is later than the	2209	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.	2210	equal to the passed C<now> value >>.
1346		2211
1347	This can be used to create very complex timers, such as a timer that	2212	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	2213	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	2214	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	2215	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).	2216	reason I omitted it as an example).
1352		2217
1353	=back	2218	=back
…		…
1357	Simply stops and restarts the periodic watcher again. This is only useful	2222	Simply stops and restarts the periodic watcher again. This is only useful
1358	when you changed some parameters or the reschedule callback would return	2223	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	2224	a different time than the last time it was called (e.g. in a crond like
1360	program when the crontabs have changed).	2225	program when the crontabs have changed).
1361		2226
		2227	=item ev_tstamp ev_periodic_at (ev_periodic *)
		2228
		2229	When active, returns the absolute time that the watcher is supposed
		2230	to trigger next. This is not the same as the C<offset> argument to
		2231	C<ev_periodic_set>, but indeed works even in interval and manual
		2232	rescheduling modes.
		2233
1362	=item ev_tstamp offset [read-write]	2234	=item ev_tstamp offset [read-write]
1363		2235
1364	When repeating, this contains the offset value, otherwise this is the	2236	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>).	2237	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		2238	although libev might modify this value for better numerical stability).
1366		2239
1367	Can be modified any time, but changes only take effect when the periodic	2240	Can be modified any time, but changes only take effect when the periodic
1368	timer fires or C<ev_periodic_again> is being called.	2241	timer fires or C<ev_periodic_again> is being called.
1369		2242
1370	=item ev_tstamp interval [read-write]	2243	=item ev_tstamp interval [read-write]
1371		2244
1372	The current interval value. Can be modified any time, but changes only	2245	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	2246	take effect when the periodic timer fires or C<ev_periodic_again> is being
1374	called.	2247	called.
1375		2248
1376	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]	2249	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
1377		2250
1378	The current reschedule callback, or C<0>, if this functionality is	2251	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	2252	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.	2253	the periodic timer fires or C<ev_periodic_again> is being called.
1381		2254
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	2255	=back
1388		2256
1389	=head3 Examples	2257	=head3 Examples
1390		2258
1391	Example: Call a callback every hour, or, more precisely, whenever the	2259	Example: Call a callback every hour, or, more precisely, whenever the
1392	system clock is divisible by 3600. The callback invocation times have	2260	system time is divisible by 3600. The callback invocation times have
1393	potentially a lot of jittering, but good long-term stability.	2261	potentially a lot of jitter, but good long-term stability.
1394		2262
1395	static void	2263	static void
1396	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	2264	clock_cb (struct ev_loop *loop, ev_periodic *w, int revents)
1397	{	2265	{
1398	... its now a full hour (UTC, or TAI or whatever your clock follows)	2266	... its now a full hour (UTC, or TAI or whatever your clock follows)
1399	}	2267	}
1400		2268
1401	struct ev_periodic hourly_tick;	2269	ev_periodic hourly_tick;
1402	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	2270	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1403	ev_periodic_start (loop, &hourly_tick);	2271	ev_periodic_start (loop, &hourly_tick);
1404		2272
1405	Example: The same as above, but use a reschedule callback to do it:	2273	Example: The same as above, but use a reschedule callback to do it:
1406		2274
1407	#include <math.h>	2275	#include <math.h>
1408		2276
1409	static ev_tstamp	2277	static ev_tstamp
1410	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	2278	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
1411	{	2279	{
1412	return fmod (now, 3600.) + 3600.;	2280	return now + (3600. - fmod (now, 3600.));
1413	}	2281	}
1414		2282
1415	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	2283	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1416		2284
1417	Example: Call a callback every hour, starting now:	2285	Example: Call a callback every hour, starting now:
1418		2286
1419	struct ev_periodic hourly_tick;	2287	ev_periodic hourly_tick;
1420	ev_periodic_init (&hourly_tick, clock_cb,	2288	ev_periodic_init (&hourly_tick, clock_cb,
1421	fmod (ev_now (loop), 3600.), 3600., 0);	2289	fmod (ev_now (loop), 3600.), 3600., 0);
1422	ev_periodic_start (loop, &hourly_tick);	2290	ev_periodic_start (loop, &hourly_tick);
1423		2291
1424		2292
1425	=head2 C<ev_signal> - signal me when a signal gets signalled!	2293	=head2 C<ev_signal> - signal me when a signal gets signalled!
1426		2294
1427	Signal watchers will trigger an event when the process receives a specific	2295	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	2296	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	2297	will try its best to deliver signals synchronously, i.e. as part of the
1430	normal event processing, like any other event.	2298	normal event processing, like any other event.
1431		2299
		2300	If you want signals to be delivered truly asynchronously, just use
		2301	C<sigaction> as you would do without libev and forget about sharing
		2302	the signal. You can even use C<ev_async> from a signal handler to
		2303	synchronously wake up an event loop.
		2304
1432	You can configure as many watchers as you like per signal. Only when the	2305	You can configure as many watchers as you like for the same signal, but
		2306	only within the same loop, i.e. you can watch for C<SIGINT> in your
		2307	default loop and for C<SIGIO> in another loop, but you cannot watch for
		2308	C<SIGINT> in both the default loop and another loop at the same time. At
		2309	the moment, C<SIGCHLD> is permanently tied to the default loop.
		2310
1433	first watcher gets started will libev actually register a signal watcher	2311	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	2312	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	2313	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		2314
1439	If possible and supported, libev will install its handlers with	2315	If possible and supported, libev will install its handlers with
1440	C<SA_RESTART> behaviour enabled, so syscalls should not be unduly	2316	C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should
1441	interrupted. If you have a problem with syscalls getting interrupted by	2317	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	2318	interrupted by signals you can block all signals in an C<ev_check> watcher
1443	them in an C<ev_prepare> watcher.	2319	and unblock them in an C<ev_prepare> watcher.
		2320
		2321	=head3 The special problem of inheritance over fork/execve/pthread_create
		2322
		2323	Both the signal mask (C<sigprocmask>) and the signal disposition
		2324	(C<sigaction>) are unspecified after starting a signal watcher (and after
		2325	stopping it again), that is, libev might or might not block the signal,
		2326	and might or might not set or restore the installed signal handler (but
		2327	see C<EVFLAG_NOSIGMASK>).
		2328
		2329	While this does not matter for the signal disposition (libev never
		2330	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
		2331	C<execve>), this matters for the signal mask: many programs do not expect
		2332	certain signals to be blocked.
		2333
		2334	This means that before calling C<exec> (from the child) you should reset
		2335	the signal mask to whatever "default" you expect (all clear is a good
		2336	choice usually).
		2337
		2338	The simplest way to ensure that the signal mask is reset in the child is
		2339	to install a fork handler with C<pthread_atfork> that resets it. That will
		2340	catch fork calls done by libraries (such as the libc) as well.
		2341
		2342	In current versions of libev, the signal will not be blocked indefinitely
		2343	unless you use the C<signalfd> API (C<EV_SIGNALFD>). While this reduces
		2344	the window of opportunity for problems, it will not go away, as libev
		2345	I<has> to modify the signal mask, at least temporarily.
		2346
		2347	So I can't stress this enough: I<If you do not reset your signal mask when
		2348	you expect it to be empty, you have a race condition in your code>. This
		2349	is not a libev-specific thing, this is true for most event libraries.
		2350
		2351	=head3 The special problem of threads signal handling
		2352
		2353	POSIX threads has problematic signal handling semantics, specifically,
		2354	a lot of functionality (sigfd, sigwait etc.) only really works if all
		2355	threads in a process block signals, which is hard to achieve.
		2356
		2357	When you want to use sigwait (or mix libev signal handling with your own
		2358	for the same signals), you can tackle this problem by globally blocking
		2359	all signals before creating any threads (or creating them with a fully set
		2360	sigprocmask) and also specifying the C<EVFLAG_NOSIGMASK> when creating
		2361	loops. Then designate one thread as "signal receiver thread" which handles
		2362	these signals. You can pass on any signals that libev might be interested
		2363	in by calling C<ev_feed_signal>.
1444		2364
1445	=head3 Watcher-Specific Functions and Data Members	2365	=head3 Watcher-Specific Functions and Data Members
1446		2366
1447	=over 4	2367	=over 4
1448		2368
…		…
1459		2379
1460	=back	2380	=back
1461		2381
1462	=head3 Examples	2382	=head3 Examples
1463		2383
1464	Example: Try to exit cleanly on SIGINT and SIGTERM.	2384	Example: Try to exit cleanly on SIGINT.
1465		2385
1466	static void	2386	static void
1467	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	2387	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
1468	{	2388	{
1469	ev_unloop (loop, EVUNLOOP_ALL);	2389	ev_break (loop, EVBREAK_ALL);
1470	}	2390	}
1471		2391
1472	struct ev_signal signal_watcher;	2392	ev_signal signal_watcher;
1473	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	2393	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1474	ev_signal_start (loop, &sigint_cb);	2394	ev_signal_start (loop, &signal_watcher);
1475		2395
1476		2396
1477	=head2 C<ev_child> - watch out for process status changes	2397	=head2 C<ev_child> - watch out for process status changes
1478		2398
1479	Child watchers trigger when your process receives a SIGCHLD in response to	2399	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	2400	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	2401	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	2402	has been forked (which implies it might have already exited), as long
1483	loop isn't entered (or is continued from a watcher).	2403	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2404	forking and then immediately registering a watcher for the child is fine,
		2405	but forking and registering a watcher a few event loop iterations later or
		2406	in the next callback invocation is not.
1484		2407
1485	Only the default event loop is capable of handling signals, and therefore	2408	Only the default event loop is capable of handling signals, and therefore
1486	you can only rgeister child watchers in the default event loop.	2409	you can only register child watchers in the default event loop.
		2410
		2411	Due to some design glitches inside libev, child watchers will always be
		2412	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2413	libev)
1487		2414
1488	=head3 Process Interaction	2415	=head3 Process Interaction
1489		2416
1490	Libev grabs C<SIGCHLD> as soon as the default event loop is	2417	Libev grabs C<SIGCHLD> as soon as the default event loop is
1491	initialised. This is necessary to guarantee proper behaviour even if	2418	initialised. This is necessary to guarantee proper behaviour even if the
1492	the first child watcher is started after the child exits. The occurance	2419	first child watcher is started after the child exits. The occurrence
1493	of C<SIGCHLD> is recorded asynchronously, but child reaping is done	2420	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
1494	synchronously as part of the event loop processing. Libev always reaps all	2421	synchronously as part of the event loop processing. Libev always reaps all
1495	children, even ones not watched.	2422	children, even ones not watched.
1496		2423
1497	=head3 Overriding the Built-In Processing	2424	=head3 Overriding the Built-In Processing
…		…
1501	handler, you can override it easily by installing your own handler for	2428	handler, you can override it easily by installing your own handler for
1502	C<SIGCHLD> after initialising the default loop, and making sure the	2429	C<SIGCHLD> after initialising the default loop, and making sure the
1503	default loop never gets destroyed. You are encouraged, however, to use an	2430	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	2431	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.	2432	that, so other libev users can use C<ev_child> watchers freely.
		2433
		2434	=head3 Stopping the Child Watcher
		2435
		2436	Currently, the child watcher never gets stopped, even when the
		2437	child terminates, so normally one needs to stop the watcher in the
		2438	callback. Future versions of libev might stop the watcher automatically
		2439	when a child exit is detected (calling C<ev_child_stop> twice is not a
		2440	problem).
1506		2441
1507	=head3 Watcher-Specific Functions and Data Members	2442	=head3 Watcher-Specific Functions and Data Members
1508		2443
1509	=over 4	2444	=over 4
1510		2445
…		…
1539	=head3 Examples	2474	=head3 Examples
1540		2475
1541	Example: C<fork()> a new process and install a child handler to wait for	2476	Example: C<fork()> a new process and install a child handler to wait for
1542	its completion.	2477	its completion.
1543		2478
1544	ev_child cw;	2479	ev_child cw;
1545		2480
1546	static void	2481	static void
1547	child_cb (EV_P_ struct ev_child *w, int revents)	2482	child_cb (EV_P_ ev_child *w, int revents)
1548	{	2483	{
1549	ev_child_stop (EV_A_ w);	2484	ev_child_stop (EV_A_ w);
1550	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);	2485	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1551	}	2486	}
1552		2487
1553	pid_t pid = fork ();	2488	pid_t pid = fork ();
1554		2489
1555	if (pid < 0)	2490	if (pid < 0)
1556	// error	2491	// error
1557	else if (pid == 0)	2492	else if (pid == 0)
1558	{	2493	{
1559	// the forked child executes here	2494	// the forked child executes here
1560	exit (1);	2495	exit (1);
1561	}	2496	}
1562	else	2497	else
1563	{	2498	{
1564	ev_child_init (&cw, child_cb, pid, 0);	2499	ev_child_init (&cw, child_cb, pid, 0);
1565	ev_child_start (EV_DEFAULT_ &cw);	2500	ev_child_start (EV_DEFAULT_ &cw);
1566	}	2501	}
1567		2502
1568		2503
1569	=head2 C<ev_stat> - did the file attributes just change?	2504	=head2 C<ev_stat> - did the file attributes just change?
1570		2505
1571	This watches a filesystem path for attribute changes. That is, it calls	2506	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	2507	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.	2508	and sees if it changed compared to the last time, invoking the callback if
		2509	it did.
1574		2510
1575	The path does not need to exist: changing from "path exists" to "path does	2511	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	2512	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	2513	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	2514	C<st_nlink> field being zero (which is otherwise always forced to be at
1579	the stat buffer having unspecified contents.	2515	least one) and all the other fields of the stat buffer having unspecified
		2516	contents.
1580		2517
1581	The path I<should> be absolute and I<must not> end in a slash. If it is	2518	The path I<must not> end in a slash or contain special components such as
		2519	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.	2520	your working directory changes, then the behaviour is undefined.
1583		2521
1584	Since there is no standard to do this, the portable implementation simply	2522	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	2523	portable implementation simply calls C<stat(2)> regularly on the path
1586	can specify a recommended polling interval for this case. If you specify	2524	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,	2525	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	2526	recommended!) then a I<suitable, unspecified default> value will be used
1589	five seconds, although this might change dynamically). Libev will also	2527	(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	2528	change dynamically). Libev will also impose a minimum interval which is
1591	usually overkill.	2529	currently around C<0.1>, but that's usually overkill.
1592		2530
1593	This watcher type is not meant for massive numbers of stat watchers,	2531	This watcher type is not meant for massive numbers of stat watchers,
1594	as even with OS-supported change notifications, this can be	2532	as even with OS-supported change notifications, this can be
1595	resource-intensive.	2533	resource-intensive.
1596		2534
1597	At the time of this writing, only the Linux inotify interface is	2535	At the time of this writing, the only OS-specific interface implemented
1598	implemented (implementing kqueue support is left as an exercise for the	2536	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	2537	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	2538	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		2539
1605	=head3 Inotify	2540	=head3 ABI Issues (Largefile Support)
1606		2541
		2542	Libev by default (unless the user overrides this) uses the default
		2543	compilation environment, which means that on systems with large file
		2544	support disabled by default, you get the 32 bit version of the stat
		2545	structure. When using the library from programs that change the ABI to
		2546	use 64 bit file offsets the programs will fail. In that case you have to
		2547	compile libev with the same flags to get binary compatibility. This is
		2548	obviously the case with any flags that change the ABI, but the problem is
		2549	most noticeably displayed with ev_stat and large file support.
		2550
		2551	The solution for this is to lobby your distribution maker to make large
		2552	file interfaces available by default (as e.g. FreeBSD does) and not
		2553	optional. Libev cannot simply switch on large file support because it has
		2554	to exchange stat structures with application programs compiled using the
		2555	default compilation environment.
		2556
		2557	=head3 Inotify and Kqueue
		2558
1607	When C<inotify (7)> support has been compiled into libev (generally only	2559	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	2560	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	2561	inotify descriptor will be created lazily when the first C<ev_stat>
1610	when the first C<ev_stat> watcher is being started.	2562	watcher is being started.
1611		2563
1612	Inotify presense does not change the semantics of C<ev_stat> watchers	2564	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	2565	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	2566	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.	2567	there are many cases where libev has to resort to regular C<stat> polling,
		2568	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2569	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2570	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2571	xfs are fully working) libev usually gets away without polling.
1616		2572
1617	(There is no support for kqueue, as apparently it cannot be used to	2573	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	2574	implement this functionality, due to the requirement of having a file
1619	descriptor open on the object at all times).	2575	descriptor open on the object at all times, and detecting renames, unlinks
		2576	etc. is difficult.
		2577
		2578	=head3 C<stat ()> is a synchronous operation
		2579
		2580	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2581	the process. The exception are C<ev_stat> watchers - those call C<stat
		2582	()>, which is a synchronous operation.
		2583
		2584	For local paths, this usually doesn't matter: unless the system is very
		2585	busy or the intervals between stat's are large, a stat call will be fast,
		2586	as the path data is usually in memory already (except when starting the
		2587	watcher).
		2588
		2589	For networked file systems, calling C<stat ()> can block an indefinite
		2590	time due to network issues, and even under good conditions, a stat call
		2591	often takes multiple milliseconds.
		2592
		2593	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2594	paths, although this is fully supported by libev.
1620		2595
1621	=head3 The special problem of stat time resolution	2596	=head3 The special problem of stat time resolution
1622		2597
1623	The C<stat ()> syscall only supports full-second resolution portably, and	2598	The C<stat ()> system call only supports full-second resolution portably,
1624	even on systems where the resolution is higher, many filesystems still	2599	and even on systems where the resolution is higher, most file systems
1625	only support whole seconds.	2600	still only support whole seconds.
1626		2601
1627	That means that, if the time is the only thing that changes, you might	2602	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	2603	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	2604	calls your callback, which does something. When there is another update
1630	the same second, C<ev_stat> will be unable to detect it.	2605	within the same second, C<ev_stat> will be unable to detect unless the
		2606	stat data does change in other ways (e.g. file size).
1631		2607
1632	The solution to this is to delay acting on a change for a second (or till	2608	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>	2609	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>	2610	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	2611	ev_timer_again (loop, w)>).
1636	systems.	2612
		2613	The C<.02> offset is added to work around small timing inconsistencies
		2614	of some operating systems (where the second counter of the current time
		2615	might be be delayed. One such system is the Linux kernel, where a call to
		2616	C<gettimeofday> might return a timestamp with a full second later than
		2617	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		2618	update file times then there will be a small window where the kernel uses
		2619	the previous second to update file times but libev might already execute
		2620	the timer callback).
1637		2621
1638	=head3 Watcher-Specific Functions and Data Members	2622	=head3 Watcher-Specific Functions and Data Members
1639		2623
1640	=over 4	2624	=over 4
1641		2625
…		…
1647	C<path>. The C<interval> is a hint on how quickly a change is expected to	2631	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	2632	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	2633	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.	2634	path for as long as the watcher is active.
1651		2635
1652	The callback will be receive C<EV_STAT> when a change was detected,	2636	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	2637	relative to the attributes at the time the watcher was started (or the
1654	last change was detected).	2638	last change was detected).
1655		2639
1656	=item ev_stat_stat (loop, ev_stat *)	2640	=item ev_stat_stat (loop, ev_stat *)
1657		2641
1658	Updates the stat buffer immediately with new values. If you change the	2642	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	2643	watched path in your callback, you could call this function to avoid
1660	detecting this change (while introducing a race condition). Can also be	2644	detecting this change (while introducing a race condition if you are not
1661	useful simply to find out the new values.	2645	the only one changing the path). Can also be useful simply to find out the
		2646	new values.
1662		2647
1663	=item ev_statdata attr [read-only]	2648	=item ev_statdata attr [read-only]
1664		2649
1665	The most-recently detected attributes of the file. Although the type is of	2650	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	2651	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	2652	suitable for your system, but you can only rely on the POSIX-standardised
		2653	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.	2654	some error while C<stat>ing the file.
1669		2655
1670	=item ev_statdata prev [read-only]	2656	=item ev_statdata prev [read-only]
1671		2657
1672	The previous attributes of the file. The callback gets invoked whenever	2658	The previous attributes of the file. The callback gets invoked whenever
1673	C<prev> != C<attr>.	2659	C<prev> != C<attr>, or, more precisely, one or more of these members
		2660	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		2661	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1674		2662
1675	=item ev_tstamp interval [read-only]	2663	=item ev_tstamp interval [read-only]
1676		2664
1677	The specified interval.	2665	The specified interval.
1678		2666
1679	=item const char *path [read-only]	2667	=item const char *path [read-only]
1680		2668
1681	The filesystem path that is being watched.	2669	The file system path that is being watched.
1682		2670
1683	=back	2671	=back
1684		2672
1685	=head3 Examples	2673	=head3 Examples
1686		2674
1687	Example: Watch C</etc/passwd> for attribute changes.	2675	Example: Watch C</etc/passwd> for attribute changes.
1688		2676
1689	static void	2677	static void
1690	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	2678	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1691	{	2679	{
1692	/* /etc/passwd changed in some way */	2680	/* /etc/passwd changed in some way */
1693	if (w->attr.st_nlink)	2681	if (w->attr.st_nlink)
1694	{	2682	{
1695	printf ("passwd current size %ld\n", (long)w->attr.st_size);	2683	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1696	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	2684	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1697	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	2685	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1698	}	2686	}
1699	else	2687	else
1700	/* you shalt not abuse printf for puts */	2688	/* you shalt not abuse printf for puts */
1701	puts ("wow, /etc/passwd is not there, expect problems. "	2689	puts ("wow, /etc/passwd is not there, expect problems. "
1702	"if this is windows, they already arrived\n");	2690	"if this is windows, they already arrived\n");
1703	}	2691	}
1704		2692
1705	...	2693	...
1706	ev_stat passwd;	2694	ev_stat passwd;
1707		2695
1708	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);	2696	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1709	ev_stat_start (loop, &passwd);	2697	ev_stat_start (loop, &passwd);
1710		2698
1711	Example: Like above, but additionally use a one-second delay so we do not	2699	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	2700	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	2701	one might do the work both on C<ev_stat> callback invocation I<and> on
1714	C<ev_timer> callback invocation).	2702	C<ev_timer> callback invocation).
1715		2703
1716	static ev_stat passwd;	2704	static ev_stat passwd;
1717	static ev_timer timer;	2705	static ev_timer timer;
1718		2706
1719	static void	2707	static void
1720	timer_cb (EV_P_ ev_timer *w, int revents)	2708	timer_cb (EV_P_ ev_timer *w, int revents)
1721	{	2709	{
1722	ev_timer_stop (EV_A_ w);	2710	ev_timer_stop (EV_A_ w);
1723		2711
1724	/* now it's one second after the most recent passwd change */	2712	/* now it's one second after the most recent passwd change */
1725	}	2713	}
1726		2714
1727	static void	2715	static void
1728	stat_cb (EV_P_ ev_stat *w, int revents)	2716	stat_cb (EV_P_ ev_stat *w, int revents)
1729	{	2717	{
1730	/* reset the one-second timer */	2718	/* reset the one-second timer */
1731	ev_timer_again (EV_A_ &timer);	2719	ev_timer_again (EV_A_ &timer);
1732	}	2720	}
1733		2721
1734	...	2722	...
1735	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);	2723	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
1736	ev_stat_start (loop, &passwd);	2724	ev_stat_start (loop, &passwd);
1737	ev_timer_init (&timer, timer_cb, 0., 1.01);	2725	ev_timer_init (&timer, timer_cb, 0., 1.02);
1738		2726
1739		2727
1740	=head2 C<ev_idle> - when you've got nothing better to do...	2728	=head2 C<ev_idle> - when you've got nothing better to do...
1741		2729
1742	Idle watchers trigger events when no other events of the same or higher	2730	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	2731	priority are pending (prepare, check and other idle watchers do not count
1744	count).	2732	as receiving "events").
1745		2733
1746	That is, as long as your process is busy handling sockets or timeouts	2734	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	2735	(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	2736	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	2737	are pending), the idle watchers are being called once per event loop
…		…
1760		2748
1761	=head3 Watcher-Specific Functions and Data Members	2749	=head3 Watcher-Specific Functions and Data Members
1762		2750
1763	=over 4	2751	=over 4
1764		2752
1765	=item ev_idle_init (ev_signal *, callback)	2753	=item ev_idle_init (ev_idle *, callback)
1766		2754
1767	Initialises and configures the idle watcher - it has no parameters of any	2755	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,	2756	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1769	believe me.	2757	believe me.
1770		2758
…		…
1773	=head3 Examples	2761	=head3 Examples
1774		2762
1775	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	2763	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1776	callback, free it. Also, use no error checking, as usual.	2764	callback, free it. Also, use no error checking, as usual.
1777		2765
1778	static void	2766	static void
1779	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	2767	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
1780	{	2768	{
1781	free (w);	2769	free (w);
1782	// now do something you wanted to do when the program has	2770	// now do something you wanted to do when the program has
1783	// no longer anything immediate to do.	2771	// no longer anything immediate to do.
1784	}	2772	}
1785		2773
1786	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	2774	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
1787	ev_idle_init (idle_watcher, idle_cb);	2775	ev_idle_init (idle_watcher, idle_cb);
1788	ev_idle_start (loop, idle_cb);	2776	ev_idle_start (loop, idle_watcher);
1789		2777
1790		2778
1791	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2779	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1792		2780
1793	Prepare and check watchers are usually (but not always) used in tandem:	2781	Prepare and check watchers are usually (but not always) used in pairs:
1794	prepare watchers get invoked before the process blocks and check watchers	2782	prepare watchers get invoked before the process blocks and check watchers
1795	afterwards.	2783	afterwards.
1796		2784
1797	You I<must not> call C<ev_loop> or similar functions that enter	2785	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>	2786	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	2787	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	2788	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,	2789	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	2790	C<ev_check> so if you have one watcher of each kind they will always be
1803	called in pairs bracketing the blocking call.	2791	called in pairs bracketing the blocking call.
1804		2792
1805	Their main purpose is to integrate other event mechanisms into libev and	2793	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	2794	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	2795	variable changes, implement your own watchers, integrate net-snmp or a
1808	coroutine library and lots more. They are also occasionally useful if	2796	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,	2797	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>	2798	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1811	watcher).	2799	watcher).
1812		2800
1813	This is done by examining in each prepare call which file descriptors need	2801	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	2802	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	2803	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	2804	libraries provide exactly this functionality). Then, in the check watcher,
1817	any events that occured (by checking the pending status of all watchers	2805	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	2806	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,	2807	I/O and timer callbacks will never actually be called (but must be valid
1820	because you never know, you know?).	2808	nevertheless, because you never know, you know?).
1821		2809
1822	As another example, the Perl Coro module uses these hooks to integrate	2810	As another example, the Perl Coro module uses these hooks to integrate
1823	coroutines into libev programs, by yielding to other active coroutines	2811	coroutines into libev programs, by yielding to other active coroutines
1824	during each prepare and only letting the process block if no coroutines	2812	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	2813	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	2816	loop from blocking if lower-priority coroutines are active, thus mapping
1829	low-priority coroutines to idle/background tasks).	2817	low-priority coroutines to idle/background tasks).
1830		2818
1831	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	2819	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	2820	priority, to ensure that they are being run before any other watchers
		2821	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2822
1833	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	2823	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	2824	activate ("feed") events into libev. While libev fully supports this, they
1835	supports this, they will be called before other C<ev_check> watchers	2825	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	2826	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	2827	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	2828	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
1839	coexist peacefully with others).	2829	others).
1840		2830
1841	=head3 Watcher-Specific Functions and Data Members	2831	=head3 Watcher-Specific Functions and Data Members
1842		2832
1843	=over 4	2833	=over 4
1844		2834
…		…
1846		2836
1847	=item ev_check_init (ev_check *, callback)	2837	=item ev_check_init (ev_check *, callback)
1848		2838
1849	Initialises and configures the prepare or check watcher - they have no	2839	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>	2840	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.	2841	macros, but using them is utterly, utterly, utterly and completely
		2842	pointless.
1852		2843
1853	=back	2844	=back
1854		2845
1855	=head3 Examples	2846	=head3 Examples
1856		2847
1857	There are a number of principal ways to embed other event loops or modules	2848	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	2849	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	2850	(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>	2851	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	2852	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
1862	into the Glib event loop).	2853	Glib event loop).
1863		2854
1864	Method 1: Add IO watchers and a timeout watcher in a prepare handler,	2855	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	2856	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	2857	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	2858	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.	2859	the callbacks for the IO/timeout watchers might not have been called yet.
1869		2860
1870	static ev_io iow [nfd];	2861	static ev_io iow [nfd];
1871	static ev_timer tw;	2862	static ev_timer tw;
1872		2863
1873	static void	2864	static void
1874	io_cb (ev_loop *loop, ev_io *w, int revents)	2865	io_cb (struct ev_loop *loop, ev_io *w, int revents)
1875	{	2866	{
1876	}	2867	}
1877		2868
1878	// create io watchers for each fd and a timer before blocking	2869	// create io watchers for each fd and a timer before blocking
1879	static void	2870	static void
1880	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2871	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
1881	{	2872	{
1882	int timeout = 3600000;	2873	int timeout = 3600000;
1883	struct pollfd fds [nfd];	2874	struct pollfd fds [nfd];
1884	// actual code will need to loop here and realloc etc.	2875	// actual code will need to loop here and realloc etc.
1885	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2876	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1886		2877
1887	/* the callback is illegal, but won't be called as we stop during check */	2878	/* the callback is illegal, but won't be called as we stop during check */
1888	ev_timer_init (&tw, 0, timeout * 1e-3);	2879	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
1889	ev_timer_start (loop, &tw);	2880	ev_timer_start (loop, &tw);
1890		2881
1891	// create one ev_io per pollfd	2882	// create one ev_io per pollfd
1892	for (int i = 0; i < nfd; ++i)	2883	for (int i = 0; i < nfd; ++i)
1893	{	2884	{
1894	ev_io_init (iow + i, io_cb, fds [i].fd,	2885	ev_io_init (iow + i, io_cb, fds [i].fd,
1895	((fds [i].events & POLLIN ? EV_READ : 0)	2886	((fds [i].events & POLLIN ? EV_READ : 0)
1896	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	2887	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1897		2888
1898	fds [i].revents = 0;	2889	fds [i].revents = 0;
1899	ev_io_start (loop, iow + i);	2890	ev_io_start (loop, iow + i);
1900	}	2891	}
1901	}	2892	}
1902		2893
1903	// stop all watchers after blocking	2894	// stop all watchers after blocking
1904	static void	2895	static void
1905	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2896	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
1906	{	2897	{
1907	ev_timer_stop (loop, &tw);	2898	ev_timer_stop (loop, &tw);
1908		2899
1909	for (int i = 0; i < nfd; ++i)	2900	for (int i = 0; i < nfd; ++i)
1910	{	2901	{
1911	// set the relevant poll flags	2902	// set the relevant poll flags
1912	// could also call adns_processreadable etc. here	2903	// could also call adns_processreadable etc. here
1913	struct pollfd *fd = fds + i;	2904	struct pollfd *fd = fds + i;
1914	int revents = ev_clear_pending (iow + i);	2905	int revents = ev_clear_pending (iow + i);
1915	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;	2906	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1916	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;	2907	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1917		2908
1918	// now stop the watcher	2909	// now stop the watcher
1919	ev_io_stop (loop, iow + i);	2910	ev_io_stop (loop, iow + i);
1920	}	2911	}
1921		2912
1922	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	2913	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1923	}	2914	}
1924		2915
1925	Method 2: This would be just like method 1, but you run C<adns_afterpoll>	2916	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.	2917	in the prepare watcher and would dispose of the check watcher.
1927		2918
1928	Method 3: If the module to be embedded supports explicit event	2919	Method 3: If the module to be embedded supports explicit event
1929	notification (adns does), you can also make use of the actual watcher	2920	notification (libadns does), you can also make use of the actual watcher
1930	callbacks, and only destroy/create the watchers in the prepare watcher.	2921	callbacks, and only destroy/create the watchers in the prepare watcher.
1931		2922
1932	static void	2923	static void
1933	timer_cb (EV_P_ ev_timer *w, int revents)	2924	timer_cb (EV_P_ ev_timer *w, int revents)
1934	{	2925	{
1935	adns_state ads = (adns_state)w->data;	2926	adns_state ads = (adns_state)w->data;
1936	update_now (EV_A);	2927	update_now (EV_A);
1937		2928
1938	adns_processtimeouts (ads, &tv_now);	2929	adns_processtimeouts (ads, &tv_now);
1939	}	2930	}
1940		2931
1941	static void	2932	static void
1942	io_cb (EV_P_ ev_io *w, int revents)	2933	io_cb (EV_P_ ev_io *w, int revents)
1943	{	2934	{
1944	adns_state ads = (adns_state)w->data;	2935	adns_state ads = (adns_state)w->data;
1945	update_now (EV_A);	2936	update_now (EV_A);
1946		2937
1947	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);	2938	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
1948	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);	2939	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
1949	}	2940	}
1950		2941
1951	// do not ever call adns_afterpoll	2942	// do not ever call adns_afterpoll
1952		2943
1953	Method 4: Do not use a prepare or check watcher because the module you	2944	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	2945	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	2946	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	2947	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
1957	this.	2948	this approach, effectively embedding EV as a client into the horrible
		2949	libglib event loop.
1958		2950
1959	static gint	2951	static gint
1960	event_poll_func (GPollFD *fds, guint nfds, gint timeout)	2952	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
1961	{	2953	{
1962	int got_events = 0;	2954	int got_events = 0;
1963		2955
1964	for (n = 0; n < nfds; ++n)	2956	for (n = 0; n < nfds; ++n)
1965	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events	2957	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
1966		2958
1967	if (timeout >= 0)	2959	if (timeout >= 0)
1968	// create/start timer	2960	// create/start timer
1969		2961
1970	// poll	2962	// poll
1971	ev_loop (EV_A_ 0);	2963	ev_run (EV_A_ 0);
1972		2964
1973	// stop timer again	2965	// stop timer again
1974	if (timeout >= 0)	2966	if (timeout >= 0)
1975	ev_timer_stop (EV_A_ &to);	2967	ev_timer_stop (EV_A_ &to);
1976		2968
1977	// stop io watchers again - their callbacks should have set	2969	// stop io watchers again - their callbacks should have set
1978	for (n = 0; n < nfds; ++n)	2970	for (n = 0; n < nfds; ++n)
1979	ev_io_stop (EV_A_ iow [n]);	2971	ev_io_stop (EV_A_ iow [n]);
1980		2972
1981	return got_events;	2973	return got_events;
1982	}	2974	}
1983		2975
1984		2976
1985	=head2 C<ev_embed> - when one backend isn't enough...	2977	=head2 C<ev_embed> - when one backend isn't enough...
1986		2978
1987	This is a rather advanced watcher type that lets you embed one event loop	2979	This is a rather advanced watcher type that lets you embed one event loop
…		…
1993	prioritise I/O.	2985	prioritise I/O.
1994		2986
1995	As an example for a bug workaround, the kqueue backend might only support	2987	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	2988	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	2989	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	2990	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	2991	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	2992	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.	2993	C<kevent>, but at least you can use both mechanisms for what they are
		2994	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
2002		2995
2003	As for prioritising I/O: rarely you have the case where some fds have	2996	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	2997	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	2998	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	2999	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.	3000	the rest in a second one, and embed the second one in the first.
2008		3001
2009	As long as the watcher is active, the callback will be invoked every time	3002	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	3003	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	3004	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	3005	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	3006	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	3007	to give the embedded loop strictly lower priority for example).
2015	embedded loop sweep.
2016		3008
2017	As long as the watcher is started it will automatically handle events. The	3009	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	3010	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		3011
2022	Also, there have not currently been made special provisions for forking:	3012	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,	3013	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	3014	embedding loop forks. In other cases, the user is responsible for calling
2025	yourself.	3015	C<ev_loop_fork> on the embedded loop.
2026		3016
2027	Unfortunately, not all backends are embeddable, only the ones returned by	3017	Unfortunately, not all backends are embeddable: only the ones returned by
2028	C<ev_embeddable_backends> are, which, unfortunately, does not include any	3018	C<ev_embeddable_backends> are, which, unfortunately, does not include any
2029	portable one.	3019	portable one.
2030		3020
2031	So when you want to use this feature you will always have to be prepared	3021	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	3022	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	3023	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.	3024	create it, and if that fails, use the normal loop for everything.
2035		3025
		3026	=head3 C<ev_embed> and fork
		3027
		3028	While the C<ev_embed> watcher is running, forks in the embedding loop will
		3029	automatically be applied to the embedded loop as well, so no special
		3030	fork handling is required in that case. When the watcher is not running,
		3031	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		3032	as applicable.
		3033
2036	=head3 Watcher-Specific Functions and Data Members	3034	=head3 Watcher-Specific Functions and Data Members
2037		3035
2038	=over 4	3036	=over 4
2039		3037
2040	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	3038	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
…		…
2043		3041
2044	Configures the watcher to embed the given loop, which must be	3042	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	3043	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	3044	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,	3045	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).	3046	if you do not want that, you need to temporarily stop the embed watcher).
2049		3047
2050	=item ev_embed_sweep (loop, ev_embed *)	3048	=item ev_embed_sweep (loop, ev_embed *)
2051		3049
2052	Make a single, non-blocking sweep over the embedded loop. This works	3050	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	3051	similarly to C<ev_run (embedded_loop, EVRUN_NOWAIT)>, but in the most
2054	apropriate way for embedded loops.	3052	appropriate way for embedded loops.
2055		3053
2056	=item struct ev_loop *other [read-only]	3054	=item struct ev_loop *other [read-only]
2057		3055
2058	The embedded event loop.	3056	The embedded event loop.
2059		3057
…		…
2061		3059
2062	=head3 Examples	3060	=head3 Examples
2063		3061
2064	Example: Try to get an embeddable event loop and embed it into the default	3062	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	3063	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	3064	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	3065	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
2068	used).	3066	used).
2069		3067
2070	struct ev_loop *loop_hi = ev_default_init (0);	3068	struct ev_loop *loop_hi = ev_default_init (0);
2071	struct ev_loop *loop_lo = 0;	3069	struct ev_loop *loop_lo = 0;
2072	struct ev_embed embed;	3070	ev_embed embed;
2073		3071
2074	// see if there is a chance of getting one that works	3072	// see if there is a chance of getting one that works
2075	// (remember that a flags value of 0 means autodetection)	3073	// (remember that a flags value of 0 means autodetection)
2076	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()	3074	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
2077	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())	3075	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
2078	: 0;	3076	: 0;
2079		3077
2080	// if we got one, then embed it, otherwise default to loop_hi	3078	// if we got one, then embed it, otherwise default to loop_hi
2081	if (loop_lo)	3079	if (loop_lo)
2082	{	3080	{
2083	ev_embed_init (&embed, 0, loop_lo);	3081	ev_embed_init (&embed, 0, loop_lo);
2084	ev_embed_start (loop_hi, &embed);	3082	ev_embed_start (loop_hi, &embed);
2085	}	3083	}
2086	else	3084	else
2087	loop_lo = loop_hi;	3085	loop_lo = loop_hi;
2088		3086
2089	Example: Check if kqueue is available but not recommended and create	3087	Example: Check if kqueue is available but not recommended and create
2090	a kqueue backend for use with sockets (which usually work with any	3088	a kqueue backend for use with sockets (which usually work with any
2091	kqueue implementation). Store the kqueue/socket-only event loop in	3089	kqueue implementation). Store the kqueue/socket-only event loop in
2092	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).	3090	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
2093		3091
2094	struct ev_loop *loop = ev_default_init (0);	3092	struct ev_loop *loop = ev_default_init (0);
2095	struct ev_loop *loop_socket = 0;	3093	struct ev_loop *loop_socket = 0;
2096	struct ev_embed embed;	3094	ev_embed embed;
2097		3095
2098	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)	3096	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
2099	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))	3097	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
2100	{	3098	{
2101	ev_embed_init (&embed, 0, loop_socket);	3099	ev_embed_init (&embed, 0, loop_socket);
2102	ev_embed_start (loop, &embed);	3100	ev_embed_start (loop, &embed);
2103	}	3101	}
2104		3102
2105	if (!loop_socket)	3103	if (!loop_socket)
2106	loop_socket = loop;	3104	loop_socket = loop;
2107		3105
2108	// now use loop_socket for all sockets, and loop for everything else	3106	// now use loop_socket for all sockets, and loop for everything else
2109		3107
2110		3108
2111	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	3109	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
2112		3110
2113	Fork watchers are called when a C<fork ()> was detected (usually because	3111	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,	3114	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	3115	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	3116	C<ev_default_fork> cheats and calls it in the wrong process, the fork
2119	handlers will be invoked, too, of course.	3117	handlers will be invoked, too, of course.
2120		3118
		3119	=head3 The special problem of life after fork - how is it possible?
		3120
		3121	Most uses of C<fork()> consist of forking, then some simple calls to set
		3122	up/change the process environment, followed by a call to C<exec()>. This
		3123	sequence should be handled by libev without any problems.
		3124
		3125	This changes when the application actually wants to do event handling
		3126	in the child, or both parent in child, in effect "continuing" after the
		3127	fork.
		3128
		3129	The default mode of operation (for libev, with application help to detect
		3130	forks) is to duplicate all the state in the child, as would be expected
		3131	when I<either> the parent I<or> the child process continues.
		3132
		3133	When both processes want to continue using libev, then this is usually the
		3134	wrong result. In that case, usually one process (typically the parent) is
		3135	supposed to continue with all watchers in place as before, while the other
		3136	process typically wants to start fresh, i.e. without any active watchers.
		3137
		3138	The cleanest and most efficient way to achieve that with libev is to
		3139	simply create a new event loop, which of course will be "empty", and
		3140	use that for new watchers. This has the advantage of not touching more
		3141	memory than necessary, and thus avoiding the copy-on-write, and the
		3142	disadvantage of having to use multiple event loops (which do not support
		3143	signal watchers).
		3144
		3145	When this is not possible, or you want to use the default loop for
		3146	other reasons, then in the process that wants to start "fresh", call
		3147	C<ev_loop_destroy (EV_DEFAULT)> followed by C<ev_default_loop (...)>.
		3148	Destroying the default loop will "orphan" (not stop) all registered
		3149	watchers, so you have to be careful not to execute code that modifies
		3150	those watchers. Note also that in that case, you have to re-register any
		3151	signal watchers.
		3152
2121	=head3 Watcher-Specific Functions and Data Members	3153	=head3 Watcher-Specific Functions and Data Members
2122		3154
2123	=over 4	3155	=over 4
2124		3156
2125	=item ev_fork_init (ev_signal *, callback)	3157	=item ev_fork_init (ev_fork *, callback)
2126		3158
2127	Initialises and configures the fork watcher - it has no parameters of any	3159	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,	3160	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
2129	believe me.	3161	really.
2130		3162
2131	=back	3163	=back
2132		3164
2133		3165
		3166	=head2 C<ev_cleanup> - even the best things end
		3167
		3168	Cleanup watchers are called just before the event loop is being destroyed
		3169	by a call to C<ev_loop_destroy>.
		3170
		3171	While there is no guarantee that the event loop gets destroyed, cleanup
		3172	watchers provide a convenient method to install cleanup hooks for your
		3173	program, worker threads and so on - you just to make sure to destroy the
		3174	loop when you want them to be invoked.
		3175
		3176	Cleanup watchers are invoked in the same way as any other watcher. Unlike
		3177	all other watchers, they do not keep a reference to the event loop (which
		3178	makes a lot of sense if you think about it). Like all other watchers, you
		3179	can call libev functions in the callback, except C<ev_cleanup_start>.
		3180
		3181	=head3 Watcher-Specific Functions and Data Members
		3182
		3183	=over 4
		3184
		3185	=item ev_cleanup_init (ev_cleanup *, callback)
		3186
		3187	Initialises and configures the cleanup watcher - it has no parameters of
		3188	any kind. There is a C<ev_cleanup_set> macro, but using it is utterly
		3189	pointless, I assure you.
		3190
		3191	=back
		3192
		3193	Example: Register an atexit handler to destroy the default loop, so any
		3194	cleanup functions are called.
		3195
		3196	static void
		3197	program_exits (void)
		3198	{
		3199	ev_loop_destroy (EV_DEFAULT_UC);
		3200	}
		3201
		3202	...
		3203	atexit (program_exits);
		3204
		3205
2134	=head2 C<ev_async> - how to wake up another event loop	3206	=head2 C<ev_async> - how to wake up an event loop
2135		3207
2136	In general, you cannot use an C<ev_loop> from multiple threads or other	3208	In general, you cannot use an C<ev_loop> from multiple threads or other
2137	asynchronous sources such as signal handlers (as opposed to multiple event	3209	asynchronous sources such as signal handlers (as opposed to multiple event
2138	loops - those are of course safe to use in different threads).	3210	loops - those are of course safe to use in different threads).
2139		3211
2140	Sometimes, however, you need to wake up another event loop you do not	3212	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	3213	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	3214	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	3215	it by calling C<ev_async_send>, which is thread- and signal safe.
2144	safe.
2145		3216
2146	This functionality is very similar to C<ev_signal> watchers, as signals,	3217	This functionality is very similar to C<ev_signal> watchers, as signals,
2147	too, are asynchronous in nature, and signals, too, will be compressed	3218	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	3219	(i.e. the number of callback invocations may be less than the number of
2149	C<ev_async_sent> calls).	3220	C<ev_async_sent> calls). In fact, you could use signal watchers as a kind
2150		3221	of "global async watchers" by using a watcher on an otherwise unused
2151	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not	3222	signal, and C<ev_feed_signal> to signal this watcher from another thread,
2152	just the default loop.	3223	even without knowing which loop owns the signal.
2153		3224
2154	=head3 Queueing	3225	=head3 Queueing
2155		3226
2156	C<ev_async> does not support queueing of data in any way. The reason	3227	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	3228	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	3229	multiple-writer-single-reader queue that works in all cases and doesn't
2159	need elaborate support such as pthreads.	3230	need elaborate support such as pthreads or unportable memory access
		3231	semantics.
2160		3232
2161	That means that if you want to queue data, you have to provide your own	3233	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	3234	queue. But at least I can tell you how to implement locking around your
2163	queue:	3235	queue:
2164		3236
2165	=over 4	3237	=over 4
2166		3238
2167	=item queueing from a signal handler context	3239	=item queueing from a signal handler context
2168		3240
2169	To implement race-free queueing, you simply add to the queue in the signal	3241	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	3242	handler but you block the signal handler in the watcher callback. Here is
2171	some fictitiuous SIGUSR1 handler:	3243	an example that does that for some fictitious SIGUSR1 handler:
2172		3244
2173	static ev_async mysig;	3245	static ev_async mysig;
2174		3246
2175	static void	3247	static void
2176	sigusr1_handler (void)	3248	sigusr1_handler (void)
…		…
2242	=over 4	3314	=over 4
2243		3315
2244	=item ev_async_init (ev_async *, callback)	3316	=item ev_async_init (ev_async *, callback)
2245		3317
2246	Initialises and configures the async watcher - it has no parameters of any	3318	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,	3319	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
2248	believe me.	3320	trust me.
2249		3321
2250	=item ev_async_send (loop, ev_async *)	3322	=item ev_async_send (loop, ev_async *)
2251		3323
2252	Sends/signals/activates the given C<ev_async> watcher, that is, feeds	3324	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	3325	an C<EV_ASYNC> event on the watcher into the event loop, and instantly
		3326	returns.
		3327
2254	C<ev_feed_event>, this call is safe to do in other threads, signal or	3328	Unlike C<ev_feed_event>, this call is safe to do from other threads,
2255	similar contexts (see the dicusssion of C<EV_ATOMIC_T> in the embedding	3329	signal or similar contexts (see the discussion of C<EV_ATOMIC_T> in the
2256	section below on what exactly this means).	3330	embedding section below on what exactly this means).
2257		3331
2258	This call incurs the overhead of a syscall only once per loop iteration,	3332	Note that, as with other watchers in libev, multiple events might get
2259	so while the overhead might be noticable, it doesn't apply to repeated	3333	compressed into a single callback invocation (another way to look at
2260	calls to C<ev_async_send>.	3334	this is that C<ev_async> watchers are level-triggered: they are set on
		3335	C<ev_async_send>, reset when the event loop detects that).
		3336
		3337	This call incurs the overhead of at most one extra system call per event
		3338	loop iteration, if the event loop is blocked, and no syscall at all if
		3339	the event loop (or your program) is processing events. That means that
		3340	repeated calls are basically free (there is no need to avoid calls for
		3341	performance reasons) and that the overhead becomes smaller (typically
		3342	zero) under load.
		3343
		3344	=item bool = ev_async_pending (ev_async *)
		3345
		3346	Returns a non-zero value when C<ev_async_send> has been called on the
		3347	watcher but the event has not yet been processed (or even noted) by the
		3348	event loop.
		3349
		3350	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		3351	the loop iterates next and checks for the watcher to have become active,
		3352	it will reset the flag again. C<ev_async_pending> can be used to very
		3353	quickly check whether invoking the loop might be a good idea.
		3354
		3355	Not that this does I<not> check whether the watcher itself is pending,
		3356	only whether it has been requested to make this watcher pending: there
		3357	is a time window between the event loop checking and resetting the async
		3358	notification, and the callback being invoked.
2261		3359
2262	=back	3360	=back
2263		3361
2264		3362
2265	=head1 OTHER FUNCTIONS	3363	=head1 OTHER FUNCTIONS
…		…
2269	=over 4	3367	=over 4
2270		3368
2271	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	3369	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2272		3370
2273	This function combines a simple timer and an I/O watcher, calls your	3371	This function combines a simple timer and an I/O watcher, calls your
2274	callback on whichever event happens first and automatically stop both	3372	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	3373	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	3374	or timeout without having to allocate/configure/start/stop/free one or
2277	more watchers yourself.	3375	more watchers yourself.
2278		3376
2279	If C<fd> is less than 0, then no I/O watcher will be started and events	3377	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	3378	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2281	C<events> set will be craeted and started.	3379	the given C<fd> and C<events> set will be created and started.
2282		3380
2283	If C<timeout> is less than 0, then no timeout watcher will be	3381	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	3382	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	3383	repeat = 0) will be started. C<0> is a valid timeout.
2286	dubious value.
2287		3384
2288	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	3385	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	3386	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>	3387	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMER>) and the C<arg>
2291	value passed to C<ev_once>:	3388	value passed to C<ev_once>. Note that it is possible to receive I<both>
		3389	a timeout and an io event at the same time - you probably should give io
		3390	events precedence.
2292		3391
		3392	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		3393
2293	static void stdin_ready (int revents, void *arg)	3394	static void stdin_ready (int revents, void *arg)
2294	{	3395	{
2295	if (revents & EV_TIMEOUT)
2296	/* doh, nothing entered */;
2297	else if (revents & EV_READ)	3396	if (revents & EV_READ)
2298	/* stdin might have data for us, joy! */;	3397	/* stdin might have data for us, joy! */;
		3398	else if (revents & EV_TIMER)
		3399	/* doh, nothing entered */;
2299	}	3400	}
2300		3401
2301	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	3402	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2302		3403
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)	3404	=item ev_feed_fd_event (loop, int fd, int revents)
2310		3405
2311	Feed an event on the given fd, as if a file descriptor backend detected	3406	Feed an event on the given fd, as if a file descriptor backend detected
2312	the given events it.	3407	the given events it.
2313		3408
2314	=item ev_feed_signal_event (ev_loop *loop, int signum)	3409	=item ev_feed_signal_event (loop, int signum)
2315		3410
2316	Feed an event as if the given signal occured (C<loop> must be the default	3411	Feed an event as if the given signal occurred. See also C<ev_feed_signal>,
2317	loop!).	3412	which is async-safe.
2318		3413
2319	=back	3414	=back
		3415
		3416
		3417	=head1 COMMON OR USEFUL IDIOMS (OR BOTH)
		3418
		3419	This section explains some common idioms that are not immediately
		3420	obvious. Note that examples are sprinkled over the whole manual, and this
		3421	section only contains stuff that wouldn't fit anywhere else.
		3422
		3423	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
		3424
		3425	Each watcher has, by default, a C<void *data> member that you can read
		3426	or modify at any time: libev will completely ignore it. This can be used
		3427	to associate arbitrary data with your watcher. If you need more data and
		3428	don't want to allocate memory separately and store a pointer to it in that
		3429	data member, you can also "subclass" the watcher type and provide your own
		3430	data:
		3431
		3432	struct my_io
		3433	{
		3434	ev_io io;
		3435	int otherfd;
		3436	void *somedata;
		3437	struct whatever *mostinteresting;
		3438	};
		3439
		3440	...
		3441	struct my_io w;
		3442	ev_io_init (&w.io, my_cb, fd, EV_READ);
		3443
		3444	And since your callback will be called with a pointer to the watcher, you
		3445	can cast it back to your own type:
		3446
		3447	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
		3448	{
		3449	struct my_io *w = (struct my_io *)w_;
		3450	...
		3451	}
		3452
		3453	More interesting and less C-conformant ways of casting your callback
		3454	function type instead have been omitted.
		3455
		3456	=head2 BUILDING YOUR OWN COMPOSITE WATCHERS
		3457
		3458	Another common scenario is to use some data structure with multiple
		3459	embedded watchers, in effect creating your own watcher that combines
		3460	multiple libev event sources into one "super-watcher":
		3461
		3462	struct my_biggy
		3463	{
		3464	int some_data;
		3465	ev_timer t1;
		3466	ev_timer t2;
		3467	}
		3468
		3469	In this case getting the pointer to C<my_biggy> is a bit more
		3470	complicated: Either you store the address of your C<my_biggy> struct in
		3471	the C<data> member of the watcher (for woozies or C++ coders), or you need
		3472	to use some pointer arithmetic using C<offsetof> inside your watchers (for
		3473	real programmers):
		3474
		3475	#include <stddef.h>
		3476
		3477	static void
		3478	t1_cb (EV_P_ ev_timer *w, int revents)
		3479	{
		3480	struct my_biggy big = (struct my_biggy *)
		3481	(((char *)w) - offsetof (struct my_biggy, t1));
		3482	}
		3483
		3484	static void
		3485	t2_cb (EV_P_ ev_timer *w, int revents)
		3486	{
		3487	struct my_biggy big = (struct my_biggy *)
		3488	(((char *)w) - offsetof (struct my_biggy, t2));
		3489	}
		3490
		3491	=head2 MODEL/NESTED EVENT LOOP INVOCATIONS AND EXIT CONDITIONS
		3492
		3493	Often (especially in GUI toolkits) there are places where you have
		3494	I<modal> interaction, which is most easily implemented by recursively
		3495	invoking C<ev_run>.
		3496
		3497	This brings the problem of exiting - a callback might want to finish the
		3498	main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but
		3499	a modal "Are you sure?" dialog is still waiting), or just the nested one
		3500	and not the main one (e.g. user clocked "Ok" in a modal dialog), or some
		3501	other combination: In these cases, C<ev_break> will not work alone.
		3502
		3503	The solution is to maintain "break this loop" variable for each C<ev_run>
		3504	invocation, and use a loop around C<ev_run> until the condition is
		3505	triggered, using C<EVRUN_ONCE>:
		3506
		3507	// main loop
		3508	int exit_main_loop = 0;
		3509
		3510	while (!exit_main_loop)
		3511	ev_run (EV_DEFAULT_ EVRUN_ONCE);
		3512
		3513	// in a model watcher
		3514	int exit_nested_loop = 0;
		3515
		3516	while (!exit_nested_loop)
		3517	ev_run (EV_A_ EVRUN_ONCE);
		3518
		3519	To exit from any of these loops, just set the corresponding exit variable:
		3520
		3521	// exit modal loop
		3522	exit_nested_loop = 1;
		3523
		3524	// exit main program, after modal loop is finished
		3525	exit_main_loop = 1;
		3526
		3527	// exit both
		3528	exit_main_loop = exit_nested_loop = 1;
		3529
		3530	=head2 THREAD LOCKING EXAMPLE
		3531
		3532	Here is a fictitious example of how to run an event loop in a different
		3533	thread from where callbacks are being invoked and watchers are
		3534	created/added/removed.
		3535
		3536	For a real-world example, see the C<EV::Loop::Async> perl module,
		3537	which uses exactly this technique (which is suited for many high-level
		3538	languages).
		3539
		3540	The example uses a pthread mutex to protect the loop data, a condition
		3541	variable to wait for callback invocations, an async watcher to notify the
		3542	event loop thread and an unspecified mechanism to wake up the main thread.
		3543
		3544	First, you need to associate some data with the event loop:
		3545
		3546	typedef struct {
		3547	mutex_t lock; /* global loop lock */
		3548	ev_async async_w;
		3549	thread_t tid;
		3550	cond_t invoke_cv;
		3551	} userdata;
		3552
		3553	void prepare_loop (EV_P)
		3554	{
		3555	// for simplicity, we use a static userdata struct.
		3556	static userdata u;
		3557
		3558	ev_async_init (&u->async_w, async_cb);
		3559	ev_async_start (EV_A_ &u->async_w);
		3560
		3561	pthread_mutex_init (&u->lock, 0);
		3562	pthread_cond_init (&u->invoke_cv, 0);
		3563
		3564	// now associate this with the loop
		3565	ev_set_userdata (EV_A_ u);
		3566	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		3567	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		3568
		3569	// then create the thread running ev_run
		3570	pthread_create (&u->tid, 0, l_run, EV_A);
		3571	}
		3572
		3573	The callback for the C<ev_async> watcher does nothing: the watcher is used
		3574	solely to wake up the event loop so it takes notice of any new watchers
		3575	that might have been added:
		3576
		3577	static void
		3578	async_cb (EV_P_ ev_async *w, int revents)
		3579	{
		3580	// just used for the side effects
		3581	}
		3582
		3583	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		3584	protecting the loop data, respectively.
		3585
		3586	static void
		3587	l_release (EV_P)
		3588	{
		3589	userdata *u = ev_userdata (EV_A);
		3590	pthread_mutex_unlock (&u->lock);
		3591	}
		3592
		3593	static void
		3594	l_acquire (EV_P)
		3595	{
		3596	userdata *u = ev_userdata (EV_A);
		3597	pthread_mutex_lock (&u->lock);
		3598	}
		3599
		3600	The event loop thread first acquires the mutex, and then jumps straight
		3601	into C<ev_run>:
		3602
		3603	void *
		3604	l_run (void *thr_arg)
		3605	{
		3606	struct ev_loop *loop = (struct ev_loop *)thr_arg;
		3607
		3608	l_acquire (EV_A);
		3609	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		3610	ev_run (EV_A_ 0);
		3611	l_release (EV_A);
		3612
		3613	return 0;
		3614	}
		3615
		3616	Instead of invoking all pending watchers, the C<l_invoke> callback will
		3617	signal the main thread via some unspecified mechanism (signals? pipe
		3618	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		3619	have been called (in a while loop because a) spurious wakeups are possible
		3620	and b) skipping inter-thread-communication when there are no pending
		3621	watchers is very beneficial):
		3622
		3623	static void
		3624	l_invoke (EV_P)
		3625	{
		3626	userdata *u = ev_userdata (EV_A);
		3627
		3628	while (ev_pending_count (EV_A))
		3629	{
		3630	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		3631	pthread_cond_wait (&u->invoke_cv, &u->lock);
		3632	}
		3633	}
		3634
		3635	Now, whenever the main thread gets told to invoke pending watchers, it
		3636	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		3637	thread to continue:
		3638
		3639	static void
		3640	real_invoke_pending (EV_P)
		3641	{
		3642	userdata *u = ev_userdata (EV_A);
		3643
		3644	pthread_mutex_lock (&u->lock);
		3645	ev_invoke_pending (EV_A);
		3646	pthread_cond_signal (&u->invoke_cv);
		3647	pthread_mutex_unlock (&u->lock);
		3648	}
		3649
		3650	Whenever you want to start/stop a watcher or do other modifications to an
		3651	event loop, you will now have to lock:
		3652
		3653	ev_timer timeout_watcher;
		3654	userdata *u = ev_userdata (EV_A);
		3655
		3656	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		3657
		3658	pthread_mutex_lock (&u->lock);
		3659	ev_timer_start (EV_A_ &timeout_watcher);
		3660	ev_async_send (EV_A_ &u->async_w);
		3661	pthread_mutex_unlock (&u->lock);
		3662
		3663	Note that sending the C<ev_async> watcher is required because otherwise
		3664	an event loop currently blocking in the kernel will have no knowledge
		3665	about the newly added timer. By waking up the loop it will pick up any new
		3666	watchers in the next event loop iteration.
		3667
		3668	=head2 THREADS, COROUTINES, CONTINUATIONS, QUEUES... INSTEAD OF CALLBACKS
		3669
		3670	While the overhead of a callback that e.g. schedules a thread is small, it
		3671	is still an overhead. If you embed libev, and your main usage is with some
		3672	kind of threads or coroutines, you might want to customise libev so that
		3673	doesn't need callbacks anymore.
		3674
		3675	Imagine you have coroutines that you can switch to using a function
		3676	C<switch_to (coro)>, that libev runs in a coroutine called C<libev_coro>
		3677	and that due to some magic, the currently active coroutine is stored in a
		3678	global called C<current_coro>. Then you can build your own "wait for libev
		3679	event" primitive by changing C<EV_CB_DECLARE> and C<EV_CB_INVOKE> (note
		3680	the differing C<;> conventions):
		3681
		3682	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3683	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb)
		3684
		3685	That means instead of having a C callback function, you store the
		3686	coroutine to switch to in each watcher, and instead of having libev call
		3687	your callback, you instead have it switch to that coroutine.
		3688
		3689	A coroutine might now wait for an event with a function called
		3690	C<wait_for_event>. (the watcher needs to be started, as always, but it doesn't
		3691	matter when, or whether the watcher is active or not when this function is
		3692	called):
		3693
		3694	void
		3695	wait_for_event (ev_watcher *w)
		3696	{
		3697	ev_cb_set (w) = current_coro;
		3698	switch_to (libev_coro);
		3699	}
		3700
		3701	That basically suspends the coroutine inside C<wait_for_event> and
		3702	continues the libev coroutine, which, when appropriate, switches back to
		3703	this or any other coroutine. I am sure if you sue this your own :)
		3704
		3705	You can do similar tricks if you have, say, threads with an event queue -
		3706	instead of storing a coroutine, you store the queue object and instead of
		3707	switching to a coroutine, you push the watcher onto the queue and notify
		3708	any waiters.
		3709
		3710	To embed libev, see L<EMBEDDING>, but in short, it's easiest to create two
		3711	files, F<my_ev.h> and F<my_ev.c> that include the respective libev files:
		3712
		3713	// my_ev.h
		3714	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3715	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb);
		3716	#include "../libev/ev.h"
		3717
		3718	// my_ev.c
		3719	#define EV_H "my_ev.h"
		3720	#include "../libev/ev.c"
		3721
		3722	And then use F<my_ev.h> when you would normally use F<ev.h>, and compile
		3723	F<my_ev.c> into your project. When properly specifying include paths, you
		3724	can even use F<ev.h> as header file name directly.
2320		3725
2321		3726
2322	=head1 LIBEVENT EMULATION	3727	=head1 LIBEVENT EMULATION
2323		3728
2324	Libev offers a compatibility emulation layer for libevent. It cannot	3729	Libev offers a compatibility emulation layer for libevent. It cannot
2325	emulate the internals of libevent, so here are some usage hints:	3730	emulate the internals of libevent, so here are some usage hints:
2326		3731
2327	=over 4	3732	=over 4
		3733
		3734	=item * Only the libevent-1.4.1-beta API is being emulated.
		3735
		3736	This was the newest libevent version available when libev was implemented,
		3737	and is still mostly unchanged in 2010.
2328		3738
2329	=item * Use it by including <event.h>, as usual.	3739	=item * Use it by including <event.h>, as usual.
2330		3740
2331	=item * The following members are fully supported: ev_base, ev_callback,	3741	=item * The following members are fully supported: ev_base, ev_callback,
2332	ev_arg, ev_fd, ev_res, ev_events.	3742	ev_arg, ev_fd, ev_res, ev_events.
…		…
2337		3747
2338	=item * Priorities are not currently supported. Initialising priorities	3748	=item * Priorities are not currently supported. Initialising priorities
2339	will fail and all watchers will have the same priority, even though there	3749	will fail and all watchers will have the same priority, even though there
2340	is an ev_pri field.	3750	is an ev_pri field.
2341		3751
		3752	=item * In libevent, the last base created gets the signals, in libev, the
		3753	base that registered the signal gets the signals.
		3754
2342	=item * Other members are not supported.	3755	=item * Other members are not supported.
2343		3756
2344	=item * The libev emulation is I<not> ABI compatible to libevent, you need	3757	=item * The libev emulation is I<not> ABI compatible to libevent, you need
2345	to use the libev header file and library.	3758	to use the libev header file and library.
2346		3759
2347	=back	3760	=back
2348		3761
2349	=head1 C++ SUPPORT	3762	=head1 C++ SUPPORT
2350		3763
2351	Libev comes with some simplistic wrapper classes for C++ that mainly allow	3764	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	3765	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.	3766	the callback model to a model using method callbacks on objects.
2354		3767
2355	To use it,	3768	To use it,
2356		3769
2357	#include <ev++.h>	3770	#include <ev++.h>
2358		3771
2359	This automatically includes F<ev.h> and puts all of its definitions (many	3772	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	3773	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	3774	put into the C<ev> namespace. It should support all the same embedding
2362	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.	3775	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++	3777	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	3778	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	3779	that the watcher is associated with (or no additional members at all if
2367	you disable C<EV_MULTIPLICITY> when embedding libev).	3780	you disable C<EV_MULTIPLICITY> when embedding libev).
2368		3781
2369	Currently, functions, and static and non-static member functions can be	3782	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	3783	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	3784	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	3785	you need support for other types of functors please contact the author
2373	it).	3786	(preferably after implementing it).
2374		3787
2375	Here is a list of things available in the C<ev> namespace:	3788	Here is a list of things available in the C<ev> namespace:
2376		3789
2377	=over 4	3790	=over 4
2378		3791
…		…
2396		3809
2397	=over 4	3810	=over 4
2398		3811
2399	=item ev::TYPE::TYPE ()	3812	=item ev::TYPE::TYPE ()
2400		3813
2401	=item ev::TYPE::TYPE (struct ev_loop *)	3814	=item ev::TYPE::TYPE (loop)
2402		3815
2403	=item ev::TYPE::~TYPE	3816	=item ev::TYPE::~TYPE
2404		3817
2405	The constructor (optionally) takes an event loop to associate the watcher	3818	The constructor (optionally) takes an event loop to associate the watcher
2406	with. If it is omitted, it will use C<EV_DEFAULT>.	3819	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	3842	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.	3843	thunking function, making it as fast as a direct C callback.
2431		3844
2432	Example: simple class declaration and watcher initialisation	3845	Example: simple class declaration and watcher initialisation
2433		3846
2434	struct myclass	3847	struct myclass
2435	{	3848	{
2436	void io_cb (ev::io &w, int revents) { }	3849	void io_cb (ev::io &w, int revents) { }
2437	}	3850	}
2438		3851
2439	myclass obj;	3852	myclass obj;
2440	ev::io iow;	3853	ev::io iow;
2441	iow.set <myclass, &myclass::io_cb> (&obj);	3854	iow.set <myclass, &myclass::io_cb> (&obj);
		3855
		3856	=item w->set (object *)
		3857
		3858	This is a variation of a method callback - leaving out the method to call
		3859	will default the method to C<operator ()>, which makes it possible to use
		3860	functor objects without having to manually specify the C<operator ()> all
		3861	the time. Incidentally, you can then also leave out the template argument
		3862	list.
		3863
		3864	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3865	int revents)>.
		3866
		3867	See the method-C<set> above for more details.
		3868
		3869	Example: use a functor object as callback.
		3870
		3871	struct myfunctor
		3872	{
		3873	void operator() (ev::io &w, int revents)
		3874	{
		3875	...
		3876	}
		3877	}
		3878
		3879	myfunctor f;
		3880
		3881	ev::io w;
		3882	w.set (&f);
2442		3883
2443	=item w->set<function> (void *data = 0)	3884	=item w->set<function> (void *data = 0)
2444		3885
2445	Also sets a callback, but uses a static method or plain function as	3886	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	3887	callback. The optional C<data> argument will be stored in the watcher's
…		…
2448		3889
2449	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.	3890	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
2450		3891
2451	See the method-C<set> above for more details.	3892	See the method-C<set> above for more details.
2452		3893
2453	Example:	3894	Example: Use a plain function as callback.
2454		3895
2455	static void io_cb (ev::io &w, int revents) { }	3896	static void io_cb (ev::io &w, int revents) { }
2456	iow.set <io_cb> ();	3897	iow.set <io_cb> ();
2457		3898
2458	=item w->set (struct ev_loop *)	3899	=item w->set (loop)
2459		3900
2460	Associates a different C<struct ev_loop> with this watcher. You can only	3901	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).	3902	do this when the watcher is inactive (and not pending either).
2462		3903
2463	=item w->set ([args])	3904	=item w->set ([arguments])
2464		3905
2465	Basically the same as C<ev_TYPE_set>, with the same args. Must be	3906	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	3907	method or a suitable start method must be called at least once. Unlike the
2467	automatically stopped and restarted when reconfiguring it with this	3908	C counterpart, an active watcher gets automatically stopped and restarted
2468	method.	3909	when reconfiguring it with this method.
2469		3910
2470	=item w->start ()	3911	=item w->start ()
2471		3912
2472	Starts the watcher. Note that there is no C<loop> argument, as the	3913	Starts the watcher. Note that there is no C<loop> argument, as the
2473	constructor already stores the event loop.	3914	constructor already stores the event loop.
2474		3915
		3916	=item w->start ([arguments])
		3917
		3918	Instead of calling C<set> and C<start> methods separately, it is often
		3919	convenient to wrap them in one call. Uses the same type of arguments as
		3920	the configure C<set> method of the watcher.
		3921
2475	=item w->stop ()	3922	=item w->stop ()
2476		3923
2477	Stops the watcher if it is active. Again, no C<loop> argument.	3924	Stops the watcher if it is active. Again, no C<loop> argument.
2478		3925
2479	=item w->again () (C<ev::timer>, C<ev::periodic> only)	3926	=item w->again () (C<ev::timer>, C<ev::periodic> only)
…		…
2491		3938
2492	=back	3939	=back
2493		3940
2494	=back	3941	=back
2495		3942
2496	Example: Define a class with an IO and idle watcher, start one of them in	3943	Example: Define a class with two I/O and idle watchers, start the I/O
2497	the constructor.	3944	watchers in the constructor.
2498		3945
2499	class myclass	3946	class myclass
2500	{	3947	{
2501	ev::io io; void io_cb (ev::io &w, int revents);	3948	ev::io io ; void io_cb (ev::io &w, int revents);
		3949	ev::io io2 ; void io2_cb (ev::io &w, int revents);
2502	ev:idle idle void idle_cb (ev::idle &w, int revents);	3950	ev::idle idle; void idle_cb (ev::idle &w, int revents);
2503		3951
2504	myclass (int fd)	3952	myclass (int fd)
2505	{	3953	{
2506	io .set <myclass, &myclass::io_cb > (this);	3954	io .set <myclass, &myclass::io_cb > (this);
		3955	io2 .set <myclass, &myclass::io2_cb > (this);
2507	idle.set <myclass, &myclass::idle_cb> (this);	3956	idle.set <myclass, &myclass::idle_cb> (this);
2508		3957
2509	io.start (fd, ev::READ);	3958	io.set (fd, ev::WRITE); // configure the watcher
		3959	io.start (); // start it whenever convenient
		3960
		3961	io2.start (fd, ev::READ); // set + start in one call
2510	}	3962	}
2511	};	3963	};
2512		3964
2513		3965
2514	=head1 OTHER LANGUAGE BINDINGS	3966	=head1 OTHER LANGUAGE BINDINGS
2515		3967
2516	Libev does not offer other language bindings itself, but bindings for a	3968	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	3969	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	3970	any interesting language binding in addition to the ones listed here, drop
2519	me a note.	3971	me a note.
2520		3972
2521	=over 4	3973	=over 4
2522		3974
2523	=item Perl	3975	=item Perl
2524		3976
2525	The EV module implements the full libev API and is actually used to test	3977	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,	3978	libev. EV is developed together with libev. Apart from the EV core module,
2527	there are additional modules that implement libev-compatible interfaces	3979	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	3980	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>).	3981	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3982	and C<EV::Glib>).
2530		3983
2531	It can be found and installed via CPAN, its homepage is found at	3984	It can be found and installed via CPAN, its homepage is at
2532	L<http://software.schmorp.de/pkg/EV>.	3985	L<http://software.schmorp.de/pkg/EV>.
2533		3986
		3987	=item Python
		3988
		3989	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		3990	seems to be quite complete and well-documented.
		3991
2534	=item Ruby	3992	=item Ruby
2535		3993
2536	Tony Arcieri has written a ruby extension that offers access to a subset	3994	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	3995	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	3996	more on top of it. It can be found via gem servers. Its homepage is at
2539	L<http://rev.rubyforge.org/>.	3997	L<http://rev.rubyforge.org/>.
2540		3998
		3999	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		4000	makes rev work even on mingw.
		4001
		4002	=item Haskell
		4003
		4004	A haskell binding to libev is available at
		4005	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		4006
2541	=item D	4007	=item D
2542		4008
2543	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to	4009	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>.	4010	be found at L<http://www.llucax.com.ar/proj/ev.d/index.html>.
		4011
		4012	=item Ocaml
		4013
		4014	Erkki Seppala has written Ocaml bindings for libev, to be found at
		4015	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		4016
		4017	=item Lua
		4018
		4019	Brian Maher has written a partial interface to libev for lua (at the
		4020	time of this writing, only C<ev_io> and C<ev_timer>), to be found at
		4021	L<http://github.com/brimworks/lua-ev>.
2545		4022
2546	=back	4023	=back
2547		4024
2548		4025
2549	=head1 MACRO MAGIC	4026	=head1 MACRO MAGIC
2550		4027
2551	Libev can be compiled with a variety of options, the most fundamantal	4028	Libev can be compiled with a variety of options, the most fundamental
2552	of which is C<EV_MULTIPLICITY>. This option determines whether (most)	4029	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
2553	functions and callbacks have an initial C<struct ev_loop *> argument.	4030	functions and callbacks have an initial C<struct ev_loop *> argument.
2554		4031
2555	To make it easier to write programs that cope with either variant, the	4032	To make it easier to write programs that cope with either variant, the
2556	following macros are defined:	4033	following macros are defined:
…		…
2561		4038
2562	This provides the loop I<argument> for functions, if one is required ("ev	4039	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,	4040	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:	4041	C<EV_A_> is used when other arguments are following. Example:
2565		4042
2566	ev_unref (EV_A);	4043	ev_unref (EV_A);
2567	ev_timer_add (EV_A_ watcher);	4044	ev_timer_add (EV_A_ watcher);
2568	ev_loop (EV_A_ 0);	4045	ev_run (EV_A_ 0);
2569		4046
2570	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	4047	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
2571	which is often provided by the following macro.	4048	which is often provided by the following macro.
2572		4049
2573	=item C<EV_P>, C<EV_P_>	4050	=item C<EV_P>, C<EV_P_>
2574		4051
2575	This provides the loop I<parameter> for functions, if one is required ("ev	4052	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,	4053	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:	4054	C<EV_P_> is used when other parameters are following. Example:
2578		4055
2579	// this is how ev_unref is being declared	4056	// this is how ev_unref is being declared
2580	static void ev_unref (EV_P);	4057	static void ev_unref (EV_P);
2581		4058
2582	// this is how you can declare your typical callback	4059	// this is how you can declare your typical callback
2583	static void cb (EV_P_ ev_timer *w, int revents)	4060	static void cb (EV_P_ ev_timer *w, int revents)
2584		4061
2585	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	4062	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
2586	suitable for use with C<EV_A>.	4063	suitable for use with C<EV_A>.
2587		4064
2588	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	4065	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
2589		4066
2590	Similar to the other two macros, this gives you the value of the default	4067	Similar to the other two macros, this gives you the value of the default
2591	loop, if multiple loops are supported ("ev loop default").	4068	loop, if multiple loops are supported ("ev loop default").
		4069
		4070	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		4071
		4072	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		4073	default loop has been initialised (C<UC> == unchecked). Their behaviour
		4074	is undefined when the default loop has not been initialised by a previous
		4075	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		4076
		4077	It is often prudent to use C<EV_DEFAULT> when initialising the first
		4078	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
2592		4079
2593	=back	4080	=back
2594		4081
2595	Example: Declare and initialise a check watcher, utilising the above	4082	Example: Declare and initialise a check watcher, utilising the above
2596	macros so it will work regardless of whether multiple loops are supported	4083	macros so it will work regardless of whether multiple loops are supported
2597	or not.	4084	or not.
2598		4085
2599	static void	4086	static void
2600	check_cb (EV_P_ ev_timer *w, int revents)	4087	check_cb (EV_P_ ev_timer *w, int revents)
2601	{	4088	{
2602	ev_check_stop (EV_A_ w);	4089	ev_check_stop (EV_A_ w);
2603	}	4090	}
2604		4091
2605	ev_check check;	4092	ev_check check;
2606	ev_check_init (&check, check_cb);	4093	ev_check_init (&check, check_cb);
2607	ev_check_start (EV_DEFAULT_ &check);	4094	ev_check_start (EV_DEFAULT_ &check);
2608	ev_loop (EV_DEFAULT_ 0);	4095	ev_run (EV_DEFAULT_ 0);
2609		4096
2610	=head1 EMBEDDING	4097	=head1 EMBEDDING
2611		4098
2612	Libev can (and often is) directly embedded into host	4099	Libev can (and often is) directly embedded into host
2613	applications. Examples of applications that embed it include the Deliantra	4100	applications. Examples of applications that embed it include the Deliantra
…		…
2620	libev somewhere in your source tree).	4107	libev somewhere in your source tree).
2621		4108
2622	=head2 FILESETS	4109	=head2 FILESETS
2623		4110
2624	Depending on what features you need you need to include one or more sets of files	4111	Depending on what features you need you need to include one or more sets of files
2625	in your app.	4112	in your application.
2626		4113
2627	=head3 CORE EVENT LOOP	4114	=head3 CORE EVENT LOOP
2628		4115
2629	To include only the libev core (all the C<ev_*> functions), with manual	4116	To include only the libev core (all the C<ev_*> functions), with manual
2630	configuration (no autoconf):	4117	configuration (no autoconf):
2631		4118
2632	#define EV_STANDALONE 1	4119	#define EV_STANDALONE 1
2633	#include "ev.c"	4120	#include "ev.c"
2634		4121
2635	This will automatically include F<ev.h>, too, and should be done in a	4122	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	4123	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	4124	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	4125	done by writing a wrapper around F<ev.h> that you can include instead and
2639	where you can put other configuration options):	4126	where you can put other configuration options):
2640		4127
2641	#define EV_STANDALONE 1	4128	#define EV_STANDALONE 1
2642	#include "ev.h"	4129	#include "ev.h"
2643		4130
2644	Both header files and implementation files can be compiled with a C++	4131	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	4132	compiler (at least, that's a stated goal, and breakage will be treated
2646	as a bug).	4133	as a bug).
2647		4134
2648	You need the following files in your source tree, or in a directory	4135	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):	4136	in your include path (e.g. in libev/ when using -Ilibev):
2650		4137
2651	ev.h	4138	ev.h
2652	ev.c	4139	ev.c
2653	ev_vars.h	4140	ev_vars.h
2654	ev_wrap.h	4141	ev_wrap.h
2655		4142
2656	ev_win32.c required on win32 platforms only	4143	ev_win32.c required on win32 platforms only
2657		4144
2658	ev_select.c only when select backend is enabled (which is enabled by default)	4145	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)	4146	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)	4147	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)	4148	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)	4149	ev_port.c only when the solaris port backend is enabled (disabled by default)
2663		4150
2664	F<ev.c> includes the backend files directly when enabled, so you only need	4151	F<ev.c> includes the backend files directly when enabled, so you only need
2665	to compile this single file.	4152	to compile this single file.
2666		4153
2667	=head3 LIBEVENT COMPATIBILITY API	4154	=head3 LIBEVENT COMPATIBILITY API
2668		4155
2669	To include the libevent compatibility API, also include:	4156	To include the libevent compatibility API, also include:
2670		4157
2671	#include "event.c"	4158	#include "event.c"
2672		4159
2673	in the file including F<ev.c>, and:	4160	in the file including F<ev.c>, and:
2674		4161
2675	#include "event.h"	4162	#include "event.h"
2676		4163
2677	in the files that want to use the libevent API. This also includes F<ev.h>.	4164	in the files that want to use the libevent API. This also includes F<ev.h>.
2678		4165
2679	You need the following additional files for this:	4166	You need the following additional files for this:
2680		4167
2681	event.h	4168	event.h
2682	event.c	4169	event.c
2683		4170
2684	=head3 AUTOCONF SUPPORT	4171	=head3 AUTOCONF SUPPORT
2685		4172
2686	Instead of using C<EV_STANDALONE=1> and providing your config in	4173	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	4174	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	4175	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
2689	include F<config.h> and configure itself accordingly.	4176	include F<config.h> and configure itself accordingly.
2690		4177
2691	For this of course you need the m4 file:	4178	For this of course you need the m4 file:
2692		4179
2693	libev.m4	4180	libev.m4
2694		4181
2695	=head2 PREPROCESSOR SYMBOLS/MACROS	4182	=head2 PREPROCESSOR SYMBOLS/MACROS
2696		4183
2697	Libev can be configured via a variety of preprocessor symbols you have to define	4184	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	4185	define before including (or compiling) any of its files. The default in
2699	and only include the select backend.	4186	the absence of autoconf is documented for every option.
		4187
		4188	Symbols marked with "(h)" do not change the ABI, and can have different
		4189	values when compiling libev vs. including F<ev.h>, so it is permissible
		4190	to redefine them before including F<ev.h> without breaking compatibility
		4191	to a compiled library. All other symbols change the ABI, which means all
		4192	users of libev and the libev code itself must be compiled with compatible
		4193	settings.
2700		4194
2701	=over 4	4195	=over 4
2702		4196
		4197	=item EV_COMPAT3 (h)
		4198
		4199	Backwards compatibility is a major concern for libev. This is why this
		4200	release of libev comes with wrappers for the functions and symbols that
		4201	have been renamed between libev version 3 and 4.
		4202
		4203	You can disable these wrappers (to test compatibility with future
		4204	versions) by defining C<EV_COMPAT3> to C<0> when compiling your
		4205	sources. This has the additional advantage that you can drop the C<struct>
		4206	from C<struct ev_loop> declarations, as libev will provide an C<ev_loop>
		4207	typedef in that case.
		4208
		4209	In some future version, the default for C<EV_COMPAT3> will become C<0>,
		4210	and in some even more future version the compatibility code will be
		4211	removed completely.
		4212
2703	=item EV_STANDALONE	4213	=item EV_STANDALONE (h)
2704		4214
2705	Must always be C<1> if you do not use autoconf configuration, which	4215	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	4216	keeps libev from including F<config.h>, and it also defines dummy
2707	implementations for some libevent functions (such as logging, which is not	4217	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	4218	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.	4219	F<event.h> that are not directly supported by the libev core alone.
2710		4220
		4221	In standalone mode, libev will still try to automatically deduce the
		4222	configuration, but has to be more conservative.
		4223
		4224	=item EV_USE_FLOOR
		4225
		4226	If defined to be C<1>, libev will use the C<floor ()> function for its
		4227	periodic reschedule calculations, otherwise libev will fall back on a
		4228	portable (slower) implementation. If you enable this, you usually have to
		4229	link against libm or something equivalent. Enabling this when the C<floor>
		4230	function is not available will fail, so the safe default is to not enable
		4231	this.
		4232
2711	=item EV_USE_MONOTONIC	4233	=item EV_USE_MONOTONIC
2712		4234
2713	If defined to be C<1>, libev will try to detect the availability of the	4235	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	4236	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	4237	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	4238	you usually have to link against librt or something similar. Enabling it
2717	the functionality isn't available is safe, though, although you have	4239	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>	4240	to make sure you link against any libraries where the C<clock_gettime>
2719	function is hiding in (often F<-lrt>).	4241	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
2720		4242
2721	=item EV_USE_REALTIME	4243	=item EV_USE_REALTIME
2722		4244
2723	If defined to be C<1>, libev will try to detect the availability of the	4245	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	4246	real-time clock option at compile time (and assume its availability
2725	runtime if successful). Otherwise no use of the realtime clock option will	4247	at runtime if successful). Otherwise no use of the real-time clock
2726	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	4248	option will be attempted. This effectively replaces C<gettimeofday>
2727	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the	4249	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
2728	note about libraries in the description of C<EV_USE_MONOTONIC>, though.	4250	correctness. See the note about libraries in the description of
		4251	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		4252	C<EV_USE_CLOCK_SYSCALL>.
		4253
		4254	=item EV_USE_CLOCK_SYSCALL
		4255
		4256	If defined to be C<1>, libev will try to use a direct syscall instead
		4257	of calling the system-provided C<clock_gettime> function. This option
		4258	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		4259	unconditionally pulls in C<libpthread>, slowing down single-threaded
		4260	programs needlessly. Using a direct syscall is slightly slower (in
		4261	theory), because no optimised vdso implementation can be used, but avoids
		4262	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		4263	higher, as it simplifies linking (no need for C<-lrt>).
2729		4264
2730	=item EV_USE_NANOSLEEP	4265	=item EV_USE_NANOSLEEP
2731		4266
2732	If defined to be C<1>, libev will assume that C<nanosleep ()> is available	4267	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 ()>.	4268	and will use it for delays. Otherwise it will use C<select ()>.
2734		4269
		4270	=item EV_USE_EVENTFD
		4271
		4272	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		4273	available and will probe for kernel support at runtime. This will improve
		4274	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		4275	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		4276	2.7 or newer, otherwise disabled.
		4277
2735	=item EV_USE_SELECT	4278	=item EV_USE_SELECT
2736		4279
2737	If undefined or defined to be C<1>, libev will compile in support for the	4280	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	4281	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	4282	other method takes over, select will be it. Otherwise the select backend
2740	will not be compiled in.	4283	will not be compiled in.
2741		4284
2742	=item EV_SELECT_USE_FD_SET	4285	=item EV_SELECT_USE_FD_SET
2743		4286
2744	If defined to C<1>, then the select backend will use the system C<fd_set>	4287	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	4288	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	4289	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	4290	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	4291	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	4292	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
2750	influence the size of the C<fd_set> used.	4293	configures the maximum size of the C<fd_set>.
2751		4294
2752	=item EV_SELECT_IS_WINSOCKET	4295	=item EV_SELECT_IS_WINSOCKET
2753		4296
2754	When defined to C<1>, the select backend will assume that	4297	When defined to C<1>, the select backend will assume that
2755	select/socket/connect etc. don't understand file descriptors but	4298	select/socket/connect etc. don't understand file descriptors but
…		…
2757	be used is the winsock select). This means that it will call	4300	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,	4301	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	4302	it is assumed that all these functions actually work on fds, even
2760	on win32. Should not be defined on non-win32 platforms.	4303	on win32. Should not be defined on non-win32 platforms.
2761		4304
2762	=item EV_FD_TO_WIN32_HANDLE	4305	=item EV_FD_TO_WIN32_HANDLE(fd)
2763		4306
2764	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map	4307	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	4308	file descriptors to socket handles. When not defining this symbol (the
2766	default), then libev will call C<_get_osfhandle>, which is usually	4309	default), then libev will call C<_get_osfhandle>, which is usually
2767	correct. In some cases, programs use their own file descriptor management,	4310	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.	4311	in which case they can provide this function to map fds to socket handles.
2769		4312
		4313	=item EV_WIN32_HANDLE_TO_FD(handle)
		4314
		4315	If C<EV_SELECT_IS_WINSOCKET> then libev maps handles to file descriptors
		4316	using the standard C<_open_osfhandle> function. For programs implementing
		4317	their own fd to handle mapping, overwriting this function makes it easier
		4318	to do so. This can be done by defining this macro to an appropriate value.
		4319
		4320	=item EV_WIN32_CLOSE_FD(fd)
		4321
		4322	If programs implement their own fd to handle mapping on win32, then this
		4323	macro can be used to override the C<close> function, useful to unregister
		4324	file descriptors again. Note that the replacement function has to close
		4325	the underlying OS handle.
		4326
2770	=item EV_USE_POLL	4327	=item EV_USE_POLL
2771		4328
2772	If defined to be C<1>, libev will compile in support for the C<poll>(2)	4329	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	4330	backend. Otherwise it will be enabled on non-win32 platforms. It
2774	takes precedence over select.	4331	takes precedence over select.
2775		4332
2776	=item EV_USE_EPOLL	4333	=item EV_USE_EPOLL
2777		4334
2778	If defined to be C<1>, libev will compile in support for the Linux	4335	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,	4336	C<epoll>(7) backend. Its availability will be detected at runtime,
2780	otherwise another method will be used as fallback. This is the	4337	otherwise another method will be used as fallback. This is the preferred
2781	preferred backend for GNU/Linux systems.	4338	backend for GNU/Linux systems. If undefined, it will be enabled if the
		4339	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2782		4340
2783	=item EV_USE_KQUEUE	4341	=item EV_USE_KQUEUE
2784		4342
2785	If defined to be C<1>, libev will compile in support for the BSD style	4343	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,	4344	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	4357	otherwise another method will be used as fallback. This is the preferred
2800	backend for Solaris 10 systems.	4358	backend for Solaris 10 systems.
2801		4359
2802	=item EV_USE_DEVPOLL	4360	=item EV_USE_DEVPOLL
2803		4361
2804	reserved for future expansion, works like the USE symbols above.	4362	Reserved for future expansion, works like the USE symbols above.
2805		4363
2806	=item EV_USE_INOTIFY	4364	=item EV_USE_INOTIFY
2807		4365
2808	If defined to be C<1>, libev will compile in support for the Linux inotify	4366	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	4367	interface to speed up C<ev_stat> watchers. Its actual availability will
2810	be detected at runtime.	4368	be detected at runtime. If undefined, it will be enabled if the headers
		4369	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2811		4370
2812	=item EV_ATOMIC_T	4371	=item EV_ATOMIC_T
2813		4372
2814	Libev requires an integer type (suitable for storing C<0> or C<1>) whose	4373	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	4374	access is atomic and serialised with respect to other threads or signal
2816	type is easily found in the C language, so you can provide your own type	4375	contexts. No such type is easily found in the C language, so you can
2817	that you know is safe for your purposes. It is used both for signal handler "locking"	4376	provide your own type that you know is safe for your purposes. It is used
2818	as well as for signal and thread safety in C<ev_async> watchers.	4377	both for signal handler "locking" as well as for signal and thread safety
		4378	in C<ev_async> watchers.
2819		4379
2820	In the absense of this define, libev will use C<sig_atomic_t volatile>	4380	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.	4381	(from F<signal.h>), which is usually good enough on most platforms.
2822		4382
2823	=item EV_H	4383	=item EV_H (h)
2824		4384
2825	The name of the F<ev.h> header file used to include it. The default if	4385	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	4386	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.	4387	used to virtually rename the F<ev.h> header file in case of conflicts.
2828		4388
2829	=item EV_CONFIG_H	4389	=item EV_CONFIG_H (h)
2830		4390
2831	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	4391	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	4392	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2833	C<EV_H>, above.	4393	C<EV_H>, above.
2834		4394
2835	=item EV_EVENT_H	4395	=item EV_EVENT_H (h)
2836		4396
2837	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	4397	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">.	4398	of how the F<event.h> header can be found, the default is C<"event.h">.
2839		4399
2840	=item EV_PROTOTYPES	4400	=item EV_PROTOTYPES (h)
2841		4401
2842	If defined to be C<0>, then F<ev.h> will not define any function	4402	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	4403	prototypes, but still define all the structs and other symbols. This is
2844	occasionally useful if you want to provide your own wrapper functions	4404	occasionally useful if you want to provide your own wrapper functions
2845	around libev functions.	4405	around libev functions.
…		…
2864	When doing priority-based operations, libev usually has to linearly search	4424	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	4425	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	4426	and time, so using the defaults of five priorities (-2 .. +2) is usually
2867	fine.	4427	fine.
2868		4428
2869	If your embedding app does not need any priorities, defining these both to	4429	If your embedding application does not need any priorities, defining these
2870	C<0> will save some memory and cpu.	4430	both to C<0> will save some memory and CPU.
2871		4431
2872	=item EV_PERIODIC_ENABLE	4432	=item EV_PERIODIC_ENABLE, EV_IDLE_ENABLE, EV_EMBED_ENABLE, EV_STAT_ENABLE,
		4433	EV_PREPARE_ENABLE, EV_CHECK_ENABLE, EV_FORK_ENABLE, EV_SIGNAL_ENABLE,
		4434	EV_ASYNC_ENABLE, EV_CHILD_ENABLE.
2873		4435
2874	If undefined or defined to be C<1>, then periodic timers are supported. If	4436	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	4437	the respective watcher type is supported. If defined to be C<0>, then it
2876	code.	4438	is not. Disabling watcher types mainly saves code size.
2877		4439
2878	=item EV_IDLE_ENABLE	4440	=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		4441
2906	If you need to shave off some kilobytes of code at the expense of some	4442	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	4443	speed (but with the full API), you can define this symbol to request
2908	some inlining decisions, saves roughly 30% codesize of amd64.	4444	certain subsets of functionality. The default is to enable all features
		4445	that can be enabled on the platform.
		4446
		4447	A typical way to use this symbol is to define it to C<0> (or to a bitset
		4448	with some broad features you want) and then selectively re-enable
		4449	additional parts you want, for example if you want everything minimal,
		4450	but multiple event loop support, async and child watchers and the poll
		4451	backend, use this:
		4452
		4453	#define EV_FEATURES 0
		4454	#define EV_MULTIPLICITY 1
		4455	#define EV_USE_POLL 1
		4456	#define EV_CHILD_ENABLE 1
		4457	#define EV_ASYNC_ENABLE 1
		4458
		4459	The actual value is a bitset, it can be a combination of the following
		4460	values:
		4461
		4462	=over 4
		4463
		4464	=item C<1> - faster/larger code
		4465
		4466	Use larger code to speed up some operations.
		4467
		4468	Currently this is used to override some inlining decisions (enlarging the
		4469	code size by roughly 30% on amd64).
		4470
		4471	When optimising for size, use of compiler flags such as C<-Os> with
		4472	gcc is recommended, as well as C<-DNDEBUG>, as libev contains a number of
		4473	assertions.
		4474
		4475	=item C<2> - faster/larger data structures
		4476
		4477	Replaces the small 2-heap for timer management by a faster 4-heap, larger
		4478	hash table sizes and so on. This will usually further increase code size
		4479	and can additionally have an effect on the size of data structures at
		4480	runtime.
		4481
		4482	=item C<4> - full API configuration
		4483
		4484	This enables priorities (sets C<EV_MAXPRI>=2 and C<EV_MINPRI>=-2), and
		4485	enables multiplicity (C<EV_MULTIPLICITY>=1).
		4486
		4487	=item C<8> - full API
		4488
		4489	This enables a lot of the "lesser used" API functions. See C<ev.h> for
		4490	details on which parts of the API are still available without this
		4491	feature, and do not complain if this subset changes over time.
		4492
		4493	=item C<16> - enable all optional watcher types
		4494
		4495	Enables all optional watcher types. If you want to selectively enable
		4496	only some watcher types other than I/O and timers (e.g. prepare,
		4497	embed, async, child...) you can enable them manually by defining
		4498	C<EV_watchertype_ENABLE> to C<1> instead.
		4499
		4500	=item C<32> - enable all backends
		4501
		4502	This enables all backends - without this feature, you need to enable at
		4503	least one backend manually (C<EV_USE_SELECT> is a good choice).
		4504
		4505	=item C<64> - enable OS-specific "helper" APIs
		4506
		4507	Enable inotify, eventfd, signalfd and similar OS-specific helper APIs by
		4508	default.
		4509
		4510	=back
		4511
		4512	Compiling with C<gcc -Os -DEV_STANDALONE -DEV_USE_EPOLL=1 -DEV_FEATURES=0>
		4513	reduces the compiled size of libev from 24.7Kb code/2.8Kb data to 6.5Kb
		4514	code/0.3Kb data on my GNU/Linux amd64 system, while still giving you I/O
		4515	watchers, timers and monotonic clock support.
		4516
		4517	With an intelligent-enough linker (gcc+binutils are intelligent enough
		4518	when you use C<-Wl,--gc-sections -ffunction-sections>) functions unused by
		4519	your program might be left out as well - a binary starting a timer and an
		4520	I/O watcher then might come out at only 5Kb.
		4521
		4522	=item EV_AVOID_STDIO
		4523
		4524	If this is set to C<1> at compiletime, then libev will avoid using stdio
		4525	functions (printf, scanf, perror etc.). This will increase the code size
		4526	somewhat, but if your program doesn't otherwise depend on stdio and your
		4527	libc allows it, this avoids linking in the stdio library which is quite
		4528	big.
		4529
		4530	Note that error messages might become less precise when this option is
		4531	enabled.
		4532
		4533	=item EV_NSIG
		4534
		4535	The highest supported signal number, +1 (or, the number of
		4536	signals): Normally, libev tries to deduce the maximum number of signals
		4537	automatically, but sometimes this fails, in which case it can be
		4538	specified. Also, using a lower number than detected (C<32> should be
		4539	good for about any system in existence) can save some memory, as libev
		4540	statically allocates some 12-24 bytes per signal number.
2909		4541
2910	=item EV_PID_HASHSIZE	4542	=item EV_PID_HASHSIZE
2911		4543
2912	C<ev_child> watchers use a small hash table to distribute workload by	4544	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	4545	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	4546	usually more than enough. If you need to manage thousands of children you
2915	increase this value (I<must> be a power of two).	4547	might want to increase this value (I<must> be a power of two).
2916		4548
2917	=item EV_INOTIFY_HASHSIZE	4549	=item EV_INOTIFY_HASHSIZE
2918		4550
2919	C<ev_stat> watchers use a small hash table to distribute workload by	4551	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>),	4552	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>	4553	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	4554	C<ev_stat> watchers you might want to increase this value (I<must> be a
2923	two).	4555	power of two).
		4556
		4557	=item EV_USE_4HEAP
		4558
		4559	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4560	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		4561	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		4562	faster performance with many (thousands) of watchers.
		4563
		4564	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4565	will be C<0>.
		4566
		4567	=item EV_HEAP_CACHE_AT
		4568
		4569	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4570	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		4571	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		4572	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		4573	but avoids random read accesses on heap changes. This improves performance
		4574	noticeably with many (hundreds) of watchers.
		4575
		4576	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4577	will be C<0>.
		4578
		4579	=item EV_VERIFY
		4580
		4581	Controls how much internal verification (see C<ev_verify ()>) will
		4582	be done: If set to C<0>, no internal verification code will be compiled
		4583	in. If set to C<1>, then verification code will be compiled in, but not
		4584	called. If set to C<2>, then the internal verification code will be
		4585	called once per loop, which can slow down libev. If set to C<3>, then the
		4586	verification code will be called very frequently, which will slow down
		4587	libev considerably.
		4588
		4589	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4590	will be C<0>.
2924		4591
2925	=item EV_COMMON	4592	=item EV_COMMON
2926		4593
2927	By default, all watchers have a C<void *data> member. By redefining	4594	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	4595	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,	4596	members. You have to define it each time you include one of the files,
2930	though, and it must be identical each time.	4597	though, and it must be identical each time.
2931		4598
2932	For example, the perl EV module uses something like this:	4599	For example, the perl EV module uses something like this:
2933		4600
2934	#define EV_COMMON \	4601	#define EV_COMMON \
2935	SV *self; /* contains this struct */ \	4602	SV *self; /* contains this struct */ \
2936	SV *cb_sv, *fh /* note no trailing ";" */	4603	SV *cb_sv, *fh /* note no trailing ";" */
2937		4604
2938	=item EV_CB_DECLARE (type)	4605	=item EV_CB_DECLARE (type)
2939		4606
2940	=item EV_CB_INVOKE (watcher, revents)	4607	=item EV_CB_INVOKE (watcher, revents)
2941		4608
…		…
2946	definition and a statement, respectively. See the F<ev.h> header file for	4613	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	4614	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	4615	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++.	4616	method calls instead of plain function calls in C++.
2950		4617
		4618	=back
		4619
2951	=head2 EXPORTED API SYMBOLS	4620	=head2 EXPORTED API SYMBOLS
2952		4621
2953	If you need to re-export the API (e.g. via a dll) and you need a list of	4622	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	4623	exported symbols, you can use the provided F<Symbol.*> files which list
2955	all public symbols, one per line:	4624	all public symbols, one per line:
2956		4625
2957	Symbols.ev for libev proper	4626	Symbols.ev for libev proper
2958	Symbols.event for the libevent emulation	4627	Symbols.event for the libevent emulation
2959		4628
2960	This can also be used to rename all public symbols to avoid clashes with	4629	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	4630	multiple versions of libev linked together (which is obviously bad in
2962	itself, but sometimes it is inconvinient to avoid this).	4631	itself, but sometimes it is inconvenient to avoid this).
2963		4632
2964	A sed command like this will create wrapper C<#define>'s that you need to	4633	A sed command like this will create wrapper C<#define>'s that you need to
2965	include before including F<ev.h>:	4634	include before including F<ev.h>:
2966		4635
2967	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h	4636	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
…		…
2984	file.	4653	file.
2985		4654
2986	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	4655	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:	4656	that everybody includes and which overrides some configure choices:
2988		4657
2989	#define EV_MINIMAL 1	4658	#define EV_FEATURES 8
2990	#define EV_USE_POLL 0	4659	#define EV_USE_SELECT 1
2991	#define EV_MULTIPLICITY 0
2992	#define EV_PERIODIC_ENABLE 0	4660	#define EV_PREPARE_ENABLE 1
		4661	#define EV_IDLE_ENABLE 1
2993	#define EV_STAT_ENABLE 0	4662	#define EV_SIGNAL_ENABLE 1
2994	#define EV_FORK_ENABLE 0	4663	#define EV_CHILD_ENABLE 1
		4664	#define EV_USE_STDEXCEPT 0
2995	#define EV_CONFIG_H <config.h>	4665	#define EV_CONFIG_H <config.h>
2996	#define EV_MINPRI 0
2997	#define EV_MAXPRI 0
2998		4666
2999	#include "ev++.h"	4667	#include "ev++.h"
3000		4668
3001	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	4669	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
3002		4670
3003	#include "ev_cpp.h"	4671	#include "ev_cpp.h"
3004	#include "ev.c"	4672	#include "ev.c"
3005		4673
		4674	=head1 INTERACTION WITH OTHER PROGRAMS, LIBRARIES OR THE ENVIRONMENT
3006		4675
3007	=head1 COMPLEXITIES	4676	=head2 THREADS AND COROUTINES
3008		4677
3009	In this section the complexities of (many of) the algorithms used inside	4678	=head3 THREADS
3010	libev will be explained. For complexity discussions about backends see the
3011	documentation for C<ev_default_init>.
3012		4679
3013	All of the following are about amortised time: If an array needs to be	4680	All libev functions are reentrant and thread-safe unless explicitly
3014	extended, libev needs to realloc and move the whole array, but this	4681	documented otherwise, but libev implements no locking itself. This means
3015	happens asymptotically never with higher number of elements, so O(1) might	4682	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	4683	are no concurrent calls into any libev function with the same loop
3017	it is much faster and asymptotically approaches constant time.	4684	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		4685	of course): libev guarantees that different event loops share no data
		4686	structures that need any locking.
		4687
		4688	Or to put it differently: calls with different loop parameters can be done
		4689	concurrently from multiple threads, calls with the same loop parameter
		4690	must be done serially (but can be done from different threads, as long as
		4691	only one thread ever is inside a call at any point in time, e.g. by using
		4692	a mutex per loop).
		4693
		4694	Specifically to support threads (and signal handlers), libev implements
		4695	so-called C<ev_async> watchers, which allow some limited form of
		4696	concurrency on the same event loop, namely waking it up "from the
		4697	outside".
		4698
		4699	If you want to know which design (one loop, locking, or multiple loops
		4700	without or something else still) is best for your problem, then I cannot
		4701	help you, but here is some generic advice:
3018		4702
3019	=over 4	4703	=over 4
3020		4704
3021	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	4705	=item * most applications have a main thread: use the default libev loop
		4706	in that thread, or create a separate thread running only the default loop.
3022		4707
3023	This means that, when you have a watcher that triggers in one hour and	4708	This helps integrating other libraries or software modules that use libev
3024	there are 100 watchers that would trigger before that then inserting will	4709	themselves and don't care/know about threading.
3025	have to skip roughly seven (C<ld 100>) of these watchers.
3026		4710
3027	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	4711	=item * one loop per thread is usually a good model.
3028		4712
3029	That means that changing a timer costs less than removing/adding them	4713	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.	4714	exists, but it is always a good start.
3031		4715
3032	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)	4716	=item * other models exist, such as the leader/follower pattern, where one
		4717	loop is handed through multiple threads in a kind of round-robin fashion.
3033		4718
3034	These just add the watcher into an array or at the head of a list.	4719	Choosing a model is hard - look around, learn, know that usually you can do
		4720	better than you currently do :-)
3035		4721
3036	=item Stopping check/prepare/idle/fork/async watchers: O(1)	4722	=item * often you need to talk to some other thread which blocks in the
		4723	event loop.
3037		4724
3038	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	4725	C<ev_async> watchers can be used to wake them up from other threads safely
		4726	(or from signal contexts...).
3039		4727
3040	These watchers are stored in lists then need to be walked to find the	4728	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	4729	work in the default loop by registering the signal watcher with the
3042	have many watchers waiting for the same fd or signal).	4730	default loop and triggering an C<ev_async> watcher from the default loop
3043		4731	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		4732
3074	=back	4733	=back
3075		4734
		4735	See also L<THREAD LOCKING EXAMPLE>.
3076		4736
3077	=head1 Win32 platform limitations and workarounds	4737	=head3 COROUTINES
		4738
		4739	Libev is very accommodating to coroutines ("cooperative threads"):
		4740	libev fully supports nesting calls to its functions from different
		4741	coroutines (e.g. you can call C<ev_run> on the same loop from two
		4742	different coroutines, and switch freely between both coroutines running
		4743	the loop, as long as you don't confuse yourself). The only exception is
		4744	that you must not do this from C<ev_periodic> reschedule callbacks.
		4745
		4746	Care has been taken to ensure that libev does not keep local state inside
		4747	C<ev_run>, and other calls do not usually allow for coroutine switches as
		4748	they do not call any callbacks.
		4749
		4750	=head2 COMPILER WARNINGS
		4751
		4752	Depending on your compiler and compiler settings, you might get no or a
		4753	lot of warnings when compiling libev code. Some people are apparently
		4754	scared by this.
		4755
		4756	However, these are unavoidable for many reasons. For one, each compiler
		4757	has different warnings, and each user has different tastes regarding
		4758	warning options. "Warn-free" code therefore cannot be a goal except when
		4759	targeting a specific compiler and compiler-version.
		4760
		4761	Another reason is that some compiler warnings require elaborate
		4762	workarounds, or other changes to the code that make it less clear and less
		4763	maintainable.
		4764
		4765	And of course, some compiler warnings are just plain stupid, or simply
		4766	wrong (because they don't actually warn about the condition their message
		4767	seems to warn about). For example, certain older gcc versions had some
		4768	warnings that resulted in an extreme number of false positives. These have
		4769	been fixed, but some people still insist on making code warn-free with
		4770	such buggy versions.
		4771
		4772	While libev is written to generate as few warnings as possible,
		4773	"warn-free" code is not a goal, and it is recommended not to build libev
		4774	with any compiler warnings enabled unless you are prepared to cope with
		4775	them (e.g. by ignoring them). Remember that warnings are just that:
		4776	warnings, not errors, or proof of bugs.
		4777
		4778
		4779	=head2 VALGRIND
		4780
		4781	Valgrind has a special section here because it is a popular tool that is
		4782	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		4783
		4784	If you think you found a bug (memory leak, uninitialised data access etc.)
		4785	in libev, then check twice: If valgrind reports something like:
		4786
		4787	==2274== definitely lost: 0 bytes in 0 blocks.
		4788	==2274== possibly lost: 0 bytes in 0 blocks.
		4789	==2274== still reachable: 256 bytes in 1 blocks.
		4790
		4791	Then there is no memory leak, just as memory accounted to global variables
		4792	is not a memleak - the memory is still being referenced, and didn't leak.
		4793
		4794	Similarly, under some circumstances, valgrind might report kernel bugs
		4795	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		4796	although an acceptable workaround has been found here), or it might be
		4797	confused.
		4798
		4799	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		4800	make it into some kind of religion.
		4801
		4802	If you are unsure about something, feel free to contact the mailing list
		4803	with the full valgrind report and an explanation on why you think this
		4804	is a bug in libev (best check the archives, too :). However, don't be
		4805	annoyed when you get a brisk "this is no bug" answer and take the chance
		4806	of learning how to interpret valgrind properly.
		4807
		4808	If you need, for some reason, empty reports from valgrind for your project
		4809	I suggest using suppression lists.
		4810
		4811
		4812	=head1 PORTABILITY NOTES
		4813
		4814	=head2 GNU/LINUX 32 BIT LIMITATIONS
		4815
		4816	GNU/Linux is the only common platform that supports 64 bit file/large file
		4817	interfaces but I<disables> them by default.
		4818
		4819	That means that libev compiled in the default environment doesn't support
		4820	files larger than 2GiB or so, which mainly affects C<ev_stat> watchers.
		4821
		4822	Unfortunately, many programs try to work around this GNU/Linux issue
		4823	by enabling the large file API, which makes them incompatible with the
		4824	standard libev compiled for their system.
		4825
		4826	Likewise, libev cannot enable the large file API itself as this would
		4827	suddenly make it incompatible to the default compile time environment,
		4828	i.e. all programs not using special compile switches.
		4829
		4830	=head2 OS/X AND DARWIN BUGS
		4831
		4832	The whole thing is a bug if you ask me - basically any system interface
		4833	you touch is broken, whether it is locales, poll, kqueue or even the
		4834	OpenGL drivers.
		4835
		4836	=head3 C<kqueue> is buggy
		4837
		4838	The kqueue syscall is broken in all known versions - most versions support
		4839	only sockets, many support pipes.
		4840
		4841	Libev tries to work around this by not using C<kqueue> by default on this
		4842	rotten platform, but of course you can still ask for it when creating a
		4843	loop - embedding a socket-only kqueue loop into a select-based one is
		4844	probably going to work well.
		4845
		4846	=head3 C<poll> is buggy
		4847
		4848	Instead of fixing C<kqueue>, Apple replaced their (working) C<poll>
		4849	implementation by something calling C<kqueue> internally around the 10.5.6
		4850	release, so now C<kqueue> I<and> C<poll> are broken.
		4851
		4852	Libev tries to work around this by not using C<poll> by default on
		4853	this rotten platform, but of course you can still ask for it when creating
		4854	a loop.
		4855
		4856	=head3 C<select> is buggy
		4857
		4858	All that's left is C<select>, and of course Apple found a way to fuck this
		4859	one up as well: On OS/X, C<select> actively limits the number of file
		4860	descriptors you can pass in to 1024 - your program suddenly crashes when
		4861	you use more.
		4862
		4863	There is an undocumented "workaround" for this - defining
		4864	C<_DARWIN_UNLIMITED_SELECT>, which libev tries to use, so select I<should>
		4865	work on OS/X.
		4866
		4867	=head2 SOLARIS PROBLEMS AND WORKAROUNDS
		4868
		4869	=head3 C<errno> reentrancy
		4870
		4871	The default compile environment on Solaris is unfortunately so
		4872	thread-unsafe that you can't even use components/libraries compiled
		4873	without C<-D_REENTRANT> in a threaded program, which, of course, isn't
		4874	defined by default. A valid, if stupid, implementation choice.
		4875
		4876	If you want to use libev in threaded environments you have to make sure
		4877	it's compiled with C<_REENTRANT> defined.
		4878
		4879	=head3 Event port backend
		4880
		4881	The scalable event interface for Solaris is called "event
		4882	ports". Unfortunately, this mechanism is very buggy in all major
		4883	releases. If you run into high CPU usage, your program freezes or you get
		4884	a large number of spurious wakeups, make sure you have all the relevant
		4885	and latest kernel patches applied. No, I don't know which ones, but there
		4886	are multiple ones to apply, and afterwards, event ports actually work
		4887	great.
		4888
		4889	If you can't get it to work, you can try running the program by setting
		4890	the environment variable C<LIBEV_FLAGS=3> to only allow C<poll> and
		4891	C<select> backends.
		4892
		4893	=head2 AIX POLL BUG
		4894
		4895	AIX unfortunately has a broken C<poll.h> header. Libev works around
		4896	this by trying to avoid the poll backend altogether (i.e. it's not even
		4897	compiled in), which normally isn't a big problem as C<select> works fine
		4898	with large bitsets on AIX, and AIX is dead anyway.
		4899
		4900	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		4901
		4902	=head3 General issues
3078		4903
3079	Win32 doesn't support any of the standards (e.g. POSIX) that libev	4904	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	4905	requires, and its I/O model is fundamentally incompatible with the POSIX
3081	model. Libev still offers limited functionality on this platform in	4906	model. Libev still offers limited functionality on this platform in
3082	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	4907	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
3083	descriptors. This only applies when using Win32 natively, not when using	4908	descriptors. This only applies when using Win32 natively, not when using
3084	e.g. cygwin.	4909	e.g. cygwin. Actually, it only applies to the microsofts own compilers,
		4910	as every compiler comes with a slightly differently broken/incompatible
		4911	environment.
		4912
		4913	Lifting these limitations would basically require the full
		4914	re-implementation of the I/O system. If you are into this kind of thing,
		4915	then note that glib does exactly that for you in a very portable way (note
		4916	also that glib is the slowest event library known to man).
3085		4917
3086	There is no supported compilation method available on windows except	4918	There is no supported compilation method available on windows except
3087	embedding it into other applications.	4919	embedding it into other applications.
3088		4920
		4921	Sensible signal handling is officially unsupported by Microsoft - libev
		4922	tries its best, but under most conditions, signals will simply not work.
		4923
		4924	Not a libev limitation but worth mentioning: windows apparently doesn't
		4925	accept large writes: instead of resulting in a partial write, windows will
		4926	either accept everything or return C<ENOBUFS> if the buffer is too large,
		4927	so make sure you only write small amounts into your sockets (less than a
		4928	megabyte seems safe, but this apparently depends on the amount of memory
		4929	available).
		4930
3089	Due to the many, low, and arbitrary limits on the win32 platform and the	4931	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	4932	the abysmal performance of winsockets, using a large number of sockets
3091	recommended (and not reasonable). If your program needs to use more than	4933	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	4934	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	4935	different implementation for windows, as libev offers the POSIX readiness
3094	be implemented efficiently on windows (microsoft monopoly games).	4936	notification model, which cannot be implemented efficiently on windows
		4937	(due to Microsoft monopoly games).
3095		4938
3096	=over 4	4939	A typical way to use libev under windows is to embed it (see the embedding
		4940	section for details) and use the following F<evwrap.h> header file instead
		4941	of F<ev.h>:
3097		4942
		4943	#define EV_STANDALONE /* keeps ev from requiring config.h */
		4944	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		4945
		4946	#include "ev.h"
		4947
		4948	And compile the following F<evwrap.c> file into your project (make sure
		4949	you do I<not> compile the F<ev.c> or any other embedded source files!):
		4950
		4951	#include "evwrap.h"
		4952	#include "ev.c"
		4953
3098	=item The winsocket select function	4954	=head3 The winsocket C<select> function
3099		4955
3100	The winsocket C<select> function doesn't follow POSIX in that it requires	4956	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	4957	requires socket I<handles> and not socket I<file descriptors> (it is
3102	very inefficient, and also requires a mapping from file descriptors	4958	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>,	4959	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	4960	C runtime provides the function C<_open_osfhandle> for this). See the
3105	symbols for more info.	4961	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		4962	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
3106		4963
3107	The configuration for a "naked" win32 using the microsoft runtime	4964	The configuration for a "naked" win32 using the Microsoft runtime
3108	libraries and raw winsocket select is:	4965	libraries and raw winsocket select is:
3109		4966
3110	#define EV_USE_SELECT 1	4967	#define EV_USE_SELECT 1
3111	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */	4968	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
3112		4969
3113	Note that winsockets handling of fd sets is O(n), so you can easily get a	4970	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.	4971	complexity in the O(n²) range when using win32.
3115		4972
3116	=item Limited number of file descriptors	4973	=head3 Limited number of file descriptors
3117		4974
3118	Windows has numerous arbitrary (and low) limits on things. Early versions	4975	Windows has numerous arbitrary (and low) limits on things.
3119	of winsocket's select only supported waiting for a max. of C<64> handles	4976
		4977	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	4978	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	4979	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).	4980	recommends spawning a chain of threads and wait for 63 handles and the
		4981	previous thread in each. Sounds great!).
3123		4982
3124	Newer versions support more handles, but you need to define C<FD_SETSIZE>	4983	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	4984	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	4985	call (which might be in libev or elsewhere, for example, perl and many
3127	select emulation on windows).	4986	other interpreters do their own select emulation on windows).
3128		4987
3129	Another limit is the number of file descriptors in the microsoft runtime	4988	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	4989	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	4990	fetish or something like this inside Microsoft). You can increase this
3132	C<_setmaxstdio>, which can increase this limit to C<2048> (another	4991	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	4992	(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	4993	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	4994	(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	4995	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.	4996	the cost of calling select (O(n²)) will likely make this unworkable.
		4997
		4998	=head2 PORTABILITY REQUIREMENTS
		4999
		5000	In addition to a working ISO-C implementation and of course the
		5001	backend-specific APIs, libev relies on a few additional extensions:
		5002
		5003	=over 4
		5004
		5005	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		5006	calling conventions regardless of C<ev_watcher_type *>.
		5007
		5008	Libev assumes not only that all watcher pointers have the same internal
		5009	structure (guaranteed by POSIX but not by ISO C for example), but it also
		5010	assumes that the same (machine) code can be used to call any watcher
		5011	callback: The watcher callbacks have different type signatures, but libev
		5012	calls them using an C<ev_watcher *> internally.
		5013
		5014	=item pointer accesses must be thread-atomic
		5015
		5016	Accessing a pointer value must be atomic, it must both be readable and
		5017	writable in one piece - this is the case on all current architectures.
		5018
		5019	=item C<sig_atomic_t volatile> must be thread-atomic as well
		5020
		5021	The type C<sig_atomic_t volatile> (or whatever is defined as
		5022	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		5023	threads. This is not part of the specification for C<sig_atomic_t>, but is
		5024	believed to be sufficiently portable.
		5025
		5026	=item C<sigprocmask> must work in a threaded environment
		5027
		5028	Libev uses C<sigprocmask> to temporarily block signals. This is not
		5029	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		5030	pthread implementations will either allow C<sigprocmask> in the "main
		5031	thread" or will block signals process-wide, both behaviours would
		5032	be compatible with libev. Interaction between C<sigprocmask> and
		5033	C<pthread_sigmask> could complicate things, however.
		5034
		5035	The most portable way to handle signals is to block signals in all threads
		5036	except the initial one, and run the default loop in the initial thread as
		5037	well.
		5038
		5039	=item C<long> must be large enough for common memory allocation sizes
		5040
		5041	To improve portability and simplify its API, libev uses C<long> internally
		5042	instead of C<size_t> when allocating its data structures. On non-POSIX
		5043	systems (Microsoft...) this might be unexpectedly low, but is still at
		5044	least 31 bits everywhere, which is enough for hundreds of millions of
		5045	watchers.
		5046
		5047	=item C<double> must hold a time value in seconds with enough accuracy
		5048
		5049	The type C<double> is used to represent timestamps. It is required to
		5050	have at least 51 bits of mantissa (and 9 bits of exponent), which is
		5051	good enough for at least into the year 4000 with millisecond accuracy
		5052	(the design goal for libev). This requirement is overfulfilled by
		5053	implementations using IEEE 754, which is basically all existing ones.
		5054
		5055	With IEEE 754 doubles, you get microsecond accuracy until at least the
		5056	year 2255 (and millisecond accuray till the year 287396 - by then, libev
		5057	is either obsolete or somebody patched it to use C<long double> or
		5058	something like that, just kidding).
3140		5059
3141	=back	5060	=back
3142		5061
		5062	If you know of other additional requirements drop me a note.
		5063
		5064
		5065	=head1 ALGORITHMIC COMPLEXITIES
		5066
		5067	In this section the complexities of (many of) the algorithms used inside
		5068	libev will be documented. For complexity discussions about backends see
		5069	the documentation for C<ev_default_init>.
		5070
		5071	All of the following are about amortised time: If an array needs to be
		5072	extended, libev needs to realloc and move the whole array, but this
		5073	happens asymptotically rarer with higher number of elements, so O(1) might
		5074	mean that libev does a lengthy realloc operation in rare cases, but on
		5075	average it is much faster and asymptotically approaches constant time.
		5076
		5077	=over 4
		5078
		5079	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		5080
		5081	This means that, when you have a watcher that triggers in one hour and
		5082	there are 100 watchers that would trigger before that, then inserting will
		5083	have to skip roughly seven (C<ld 100>) of these watchers.
		5084
		5085	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		5086
		5087	That means that changing a timer costs less than removing/adding them,
		5088	as only the relative motion in the event queue has to be paid for.
		5089
		5090	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
		5091
		5092	These just add the watcher into an array or at the head of a list.
		5093
		5094	=item Stopping check/prepare/idle/fork/async watchers: O(1)
		5095
		5096	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		5097
		5098	These watchers are stored in lists, so they need to be walked to find the
		5099	correct watcher to remove. The lists are usually short (you don't usually
		5100	have many watchers waiting for the same fd or signal: one is typical, two
		5101	is rare).
		5102
		5103	=item Finding the next timer in each loop iteration: O(1)
		5104
		5105	By virtue of using a binary or 4-heap, the next timer is always found at a
		5106	fixed position in the storage array.
		5107
		5108	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		5109
		5110	A change means an I/O watcher gets started or stopped, which requires
		5111	libev to recalculate its status (and possibly tell the kernel, depending
		5112	on backend and whether C<ev_io_set> was used).
		5113
		5114	=item Activating one watcher (putting it into the pending state): O(1)
		5115
		5116	=item Priority handling: O(number_of_priorities)
		5117
		5118	Priorities are implemented by allocating some space for each
		5119	priority. When doing priority-based operations, libev usually has to
		5120	linearly search all the priorities, but starting/stopping and activating
		5121	watchers becomes O(1) with respect to priority handling.
		5122
		5123	=item Sending an ev_async: O(1)
		5124
		5125	=item Processing ev_async_send: O(number_of_async_watchers)
		5126
		5127	=item Processing signals: O(max_signal_number)
		5128
		5129	Sending involves a system call I<iff> there were no other C<ev_async_send>
		5130	calls in the current loop iteration and the loop is currently
		5131	blocked. Checking for async and signal events involves iterating over all
		5132	running async watchers or all signal numbers.
		5133
		5134	=back
		5135
		5136
		5137	=head1 PORTING FROM LIBEV 3.X TO 4.X
		5138
		5139	The major version 4 introduced some incompatible changes to the API.
		5140
		5141	At the moment, the C<ev.h> header file provides compatibility definitions
		5142	for all changes, so most programs should still compile. The compatibility
		5143	layer might be removed in later versions of libev, so better update to the
		5144	new API early than late.
		5145
		5146	=over 4
		5147
		5148	=item C<EV_COMPAT3> backwards compatibility mechanism
		5149
		5150	The backward compatibility mechanism can be controlled by
		5151	C<EV_COMPAT3>. See L<PREPROCESSOR SYMBOLS/MACROS> in the L<EMBEDDING>
		5152	section.
		5153
		5154	=item C<ev_default_destroy> and C<ev_default_fork> have been removed
		5155
		5156	These calls can be replaced easily by their C<ev_loop_xxx> counterparts:
		5157
		5158	ev_loop_destroy (EV_DEFAULT_UC);
		5159	ev_loop_fork (EV_DEFAULT);
		5160
		5161	=item function/symbol renames
		5162
		5163	A number of functions and symbols have been renamed:
		5164
		5165	ev_loop => ev_run
		5166	EVLOOP_NONBLOCK => EVRUN_NOWAIT
		5167	EVLOOP_ONESHOT => EVRUN_ONCE
		5168
		5169	ev_unloop => ev_break
		5170	EVUNLOOP_CANCEL => EVBREAK_CANCEL
		5171	EVUNLOOP_ONE => EVBREAK_ONE
		5172	EVUNLOOP_ALL => EVBREAK_ALL
		5173
		5174	EV_TIMEOUT => EV_TIMER
		5175
		5176	ev_loop_count => ev_iteration
		5177	ev_loop_depth => ev_depth
		5178	ev_loop_verify => ev_verify
		5179
		5180	Most functions working on C<struct ev_loop> objects don't have an
		5181	C<ev_loop_> prefix, so it was removed; C<ev_loop>, C<ev_unloop> and
		5182	associated constants have been renamed to not collide with the C<struct
		5183	ev_loop> anymore and C<EV_TIMER> now follows the same naming scheme
		5184	as all other watcher types. Note that C<ev_loop_fork> is still called
		5185	C<ev_loop_fork> because it would otherwise clash with the C<ev_fork>
		5186	typedef.
		5187
		5188	=item C<EV_MINIMAL> mechanism replaced by C<EV_FEATURES>
		5189
		5190	The preprocessor symbol C<EV_MINIMAL> has been replaced by a different
		5191	mechanism, C<EV_FEATURES>. Programs using C<EV_MINIMAL> usually compile
		5192	and work, but the library code will of course be larger.
		5193
		5194	=back
		5195
		5196
		5197	=head1 GLOSSARY
		5198
		5199	=over 4
		5200
		5201	=item active
		5202
		5203	A watcher is active as long as it has been started and not yet stopped.
		5204	See L<WATCHER STATES> for details.
		5205
		5206	=item application
		5207
		5208	In this document, an application is whatever is using libev.
		5209
		5210	=item backend
		5211
		5212	The part of the code dealing with the operating system interfaces.
		5213
		5214	=item callback
		5215
		5216	The address of a function that is called when some event has been
		5217	detected. Callbacks are being passed the event loop, the watcher that
		5218	received the event, and the actual event bitset.
		5219
		5220	=item callback/watcher invocation
		5221
		5222	The act of calling the callback associated with a watcher.
		5223
		5224	=item event
		5225
		5226	A change of state of some external event, such as data now being available
		5227	for reading on a file descriptor, time having passed or simply not having
		5228	any other events happening anymore.
		5229
		5230	In libev, events are represented as single bits (such as C<EV_READ> or
		5231	C<EV_TIMER>).
		5232
		5233	=item event library
		5234
		5235	A software package implementing an event model and loop.
		5236
		5237	=item event loop
		5238
		5239	An entity that handles and processes external events and converts them
		5240	into callback invocations.
		5241
		5242	=item event model
		5243
		5244	The model used to describe how an event loop handles and processes
		5245	watchers and events.
		5246
		5247	=item pending
		5248
		5249	A watcher is pending as soon as the corresponding event has been
		5250	detected. See L<WATCHER STATES> for details.
		5251
		5252	=item real time
		5253
		5254	The physical time that is observed. It is apparently strictly monotonic :)
		5255
		5256	=item wall-clock time
		5257
		5258	The time and date as shown on clocks. Unlike real time, it can actually
		5259	be wrong and jump forwards and backwards, e.g. when you adjust your
		5260	clock.
		5261
		5262	=item watcher
		5263
		5264	A data structure that describes interest in certain events. Watchers need
		5265	to be started (attached to an event loop) before they can receive events.
		5266
		5267	=back
3143		5268
3144	=head1 AUTHOR	5269	=head1 AUTHOR
3145		5270
3146	Marc Lehmann <libev@schmorp.de>.	5271	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael
		5272	Magnusson and Emanuele Giaquinta, and minor corrections by many others.
3147		5273

Diff Legend

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