ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.65 by ayin, Sat Dec 1 15:38:54 2007 UTC vs.
Revision 1.431 by root, Fri Nov 22 16:42:10 2013 UTC

		1	=encoding utf-8
		2
1	=head1 NAME	3	=head1 NAME
2		4
3	libev - a high performance full-featured event loop written in C	5	libev - a high performance full-featured event loop written in C
4		6
5	=head1 SYNOPSIS	7	=head1 SYNOPSIS
6		8
7	#include <ev.h>	9	#include <ev.h>
8		10
9	=head1 EXAMPLE PROGRAM	11	=head2 EXAMPLE PROGRAM
10		12
		13	// a single header file is required
11	#include <ev.h>	14	#include <ev.h>
12		15
		16	#include <stdio.h> // for puts
		17
		18	// every watcher type has its own typedef'd struct
		19	// with the name ev_TYPE
13	ev_io stdin_watcher;	20	ev_io stdin_watcher;
14	ev_timer timeout_watcher;	21	ev_timer timeout_watcher;
15		22
16	/* called when data readable on stdin */	23	// all watcher callbacks have a similar signature
		24	// this callback is called when data is readable on stdin
17	static void	25	static void
18	stdin_cb (EV_P_ struct ev_io *w, int revents)	26	stdin_cb (EV_P_ ev_io *w, int revents)
19	{	27	{
20	/* puts ("stdin ready"); */	28	puts ("stdin ready");
21	ev_io_stop (EV_A_ w); /* just a syntax example */	29	// for one-shot events, one must manually stop the watcher
22	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */	30	// with its corresponding stop function.
		31	ev_io_stop (EV_A_ w);
		32
		33	// this causes all nested ev_run's to stop iterating
		34	ev_break (EV_A_ EVBREAK_ALL);
23	}	35	}
24		36
		37	// another callback, this time for a time-out
25	static void	38	static void
26	timeout_cb (EV_P_ struct ev_timer *w, int revents)	39	timeout_cb (EV_P_ ev_timer *w, int revents)
27	{	40	{
28	/* puts ("timeout"); */	41	puts ("timeout");
29	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */	42	// this causes the innermost ev_run to stop iterating
		43	ev_break (EV_A_ EVBREAK_ONE);
30	}	44	}
31		45
32	int	46	int
33	main (void)	47	main (void)
34	{	48	{
35	struct ev_loop *loop = ev_default_loop (0);	49	// use the default event loop unless you have special needs
		50	struct ev_loop *loop = EV_DEFAULT;
36		51
37	/* initialise an io watcher, then start it */	52	// initialise an io watcher, then start it
		53	// this one will watch for stdin to become readable
38	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);	54	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
39	ev_io_start (loop, &stdin_watcher);	55	ev_io_start (loop, &stdin_watcher);
40		56
		57	// initialise a timer watcher, then start it
41	/* simple non-repeating 5.5 second timeout */	58	// simple non-repeating 5.5 second timeout
42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	59	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
43	ev_timer_start (loop, &timeout_watcher);	60	ev_timer_start (loop, &timeout_watcher);
44		61
45	/* loop till timeout or data ready */	62	// now wait for events to arrive
46	ev_loop (loop, 0);	63	ev_run (loop, 0);
47		64
		65	// break was called, so exit
48	return 0;	66	return 0;
49	}	67	}
50		68
51	=head1 DESCRIPTION	69	=head1 ABOUT THIS DOCUMENT
		70
		71	This document documents the libev software package.
		72
		73	The newest version of this document is also available as an html-formatted
		74	web page you might find easier to navigate when reading it for the first
		75	time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
		76
		77	While this document tries to be as complete as possible in documenting
		78	libev, its usage and the rationale behind its design, it is not a tutorial
		79	on event-based programming, nor will it introduce event-based programming
		80	with libev.
		81
		82	Familiarity with event based programming techniques in general is assumed
		83	throughout this document.
		84
		85	=head1 WHAT TO READ WHEN IN A HURRY
		86
		87	This manual tries to be very detailed, but unfortunately, this also makes
		88	it very long. If you just want to know the basics of libev, I suggest
		89	reading L</ANATOMY OF A WATCHER>, then the L</EXAMPLE PROGRAM> above and
		90	look up the missing functions in L</GLOBAL FUNCTIONS> and the C<ev_io> and
		91	C<ev_timer> sections in L</WATCHER TYPES>.
		92
		93	=head1 ABOUT LIBEV
52		94
53	Libev is an event loop: you register interest in certain events (such as a	95	Libev is an event loop: you register interest in certain events (such as a
54	file descriptor being readable or a timeout occuring), and it will manage	96	file descriptor being readable or a timeout occurring), and it will manage
55	these event sources and provide your program with events.	97	these event sources and provide your program with events.
56		98
57	To do this, it must take more or less complete control over your process	99	To do this, it must take more or less complete control over your process
58	(or thread) by executing the I<event loop> handler, and will then	100	(or thread) by executing the I<event loop> handler, and will then
59	communicate events via a callback mechanism.	101	communicate events via a callback mechanism.
…		…
61	You register interest in certain events by registering so-called I<event	103	You register interest in certain events by registering so-called I<event
62	watchers>, which are relatively small C structures you initialise with the	104	watchers>, which are relatively small C structures you initialise with the
63	details of the event, and then hand it over to libev by I<starting> the	105	details of the event, and then hand it over to libev by I<starting> the
64	watcher.	106	watcher.
65		107
66	=head1 FEATURES	108	=head2 FEATURES
67		109
68	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the	110	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
69	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms	111	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
70	for file descriptor events (C<ev_io>), the Linux C<inotify> interface	112	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
71	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers	113	(for C<ev_stat>), Linux eventfd/signalfd (for faster and cleaner
72	with customised rescheduling (C<ev_periodic>), synchronous signals	114	inter-thread wakeup (C<ev_async>)/signal handling (C<ev_signal>)) relative
73	(C<ev_signal>), process status change events (C<ev_child>), and event	115	timers (C<ev_timer>), absolute timers with customised rescheduling
74	watchers dealing with the event loop mechanism itself (C<ev_idle>,	116	(C<ev_periodic>), synchronous signals (C<ev_signal>), process status
75	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as	117	change events (C<ev_child>), and event watchers dealing with the event
76	file watchers (C<ev_stat>) and even limited support for fork events	118	loop mechanism itself (C<ev_idle>, C<ev_embed>, C<ev_prepare> and
77	(C<ev_fork>).	119	C<ev_check> watchers) as well as file watchers (C<ev_stat>) and even
		120	limited support for fork events (C<ev_fork>).
78		121
79	It also is quite fast (see this	122	It also is quite fast (see this
80	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent	123	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
81	for example).	124	for example).
82		125
83	=head1 CONVENTIONS	126	=head2 CONVENTIONS
84		127
85	Libev is very configurable. In this manual the default configuration will	128	Libev is very configurable. In this manual the default (and most common)
86	be described, which supports multiple event loops. For more info about	129	configuration will be described, which supports multiple event loops. For
87	various configuration options please have a look at B<EMBED> section in	130	more info about various configuration options please have a look at
88	this manual. If libev was configured without support for multiple event	131	B<EMBED> section in this manual. If libev was configured without support
89	loops, then all functions taking an initial argument of name C<loop>	132	for multiple event loops, then all functions taking an initial argument of
90	(which is always of type C<struct ev_loop *>) will not have this argument.	133	name C<loop> (which is always of type C<struct ev_loop *>) will not have
		134	this argument.
91		135
92	=head1 TIME REPRESENTATION	136	=head2 TIME REPRESENTATION
93		137
94	Libev represents time as a single floating point number, representing the	138	Libev represents time as a single floating point number, representing
95	(fractional) number of seconds since the (POSIX) epoch (somewhere near	139	the (fractional) number of seconds since the (POSIX) epoch (in practice
96	the beginning of 1970, details are complicated, don't ask). This type is	140	somewhere near the beginning of 1970, details are complicated, don't
97	called C<ev_tstamp>, which is what you should use too. It usually aliases	141	ask). This type is called C<ev_tstamp>, which is what you should use
98	to the C<double> type in C, and when you need to do any calculations on	142	too. It usually aliases to the C<double> type in C. When you need to do
99	it, you should treat it as such.	143	any calculations on it, you should treat it as some floating point value.
		144
		145	Unlike the name component C<stamp> might indicate, it is also used for
		146	time differences (e.g. delays) throughout libev.
		147
		148	=head1 ERROR HANDLING
		149
		150	Libev knows three classes of errors: operating system errors, usage errors
		151	and internal errors (bugs).
		152
		153	When libev catches an operating system error it cannot handle (for example
		154	a system call indicating a condition libev cannot fix), it calls the callback
		155	set via C<ev_set_syserr_cb>, which is supposed to fix the problem or
		156	abort. The default is to print a diagnostic message and to call C<abort
		157	()>.
		158
		159	When libev detects a usage error such as a negative timer interval, then
		160	it will print a diagnostic message and abort (via the C<assert> mechanism,
		161	so C<NDEBUG> will disable this checking): these are programming errors in
		162	the libev caller and need to be fixed there.
		163
		164	Libev also has a few internal error-checking C<assert>ions, and also has
		165	extensive consistency checking code. These do not trigger under normal
		166	circumstances, as they indicate either a bug in libev or worse.
		167
100		168
101	=head1 GLOBAL FUNCTIONS	169	=head1 GLOBAL FUNCTIONS
102		170
103	These functions can be called anytime, even before initialising the	171	These functions can be called anytime, even before initialising the
104	library in any way.	172	library in any way.
…		…
107		175
108	=item ev_tstamp ev_time ()	176	=item ev_tstamp ev_time ()
109		177
110	Returns the current time as libev would use it. Please note that the	178	Returns the current time as libev would use it. Please note that the
111	C<ev_now> function is usually faster and also often returns the timestamp	179	C<ev_now> function is usually faster and also often returns the timestamp
112	you actually want to know.	180	you actually want to know. Also interesting is the combination of
		181	C<ev_now_update> and C<ev_now>.
		182
		183	=item ev_sleep (ev_tstamp interval)
		184
		185	Sleep for the given interval: The current thread will be blocked
		186	until either it is interrupted or the given time interval has
		187	passed (approximately - it might return a bit earlier even if not
		188	interrupted). Returns immediately if C<< interval <= 0 >>.
		189
		190	Basically this is a sub-second-resolution C<sleep ()>.
		191
		192	The range of the C<interval> is limited - libev only guarantees to work
		193	with sleep times of up to one day (C<< interval <= 86400 >>).
113		194
114	=item int ev_version_major ()	195	=item int ev_version_major ()
115		196
116	=item int ev_version_minor ()	197	=item int ev_version_minor ()
117		198
118	You can find out the major and minor version numbers of the library	199	You can find out the major and minor ABI version numbers of the library
119	you linked against by calling the functions C<ev_version_major> and	200	you linked against by calling the functions C<ev_version_major> and
120	C<ev_version_minor>. If you want, you can compare against the global	201	C<ev_version_minor>. If you want, you can compare against the global
121	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	202	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
122	version of the library your program was compiled against.	203	version of the library your program was compiled against.
123		204
		205	These version numbers refer to the ABI version of the library, not the
		206	release version.
		207
124	Usually, it's a good idea to terminate if the major versions mismatch,	208	Usually, it's a good idea to terminate if the major versions mismatch,
125	as this indicates an incompatible change. Minor versions are usually	209	as this indicates an incompatible change. Minor versions are usually
126	compatible to older versions, so a larger minor version alone is usually	210	compatible to older versions, so a larger minor version alone is usually
127	not a problem.	211	not a problem.
128		212
129	Example: Make sure we haven't accidentally been linked against the wrong	213	Example: Make sure we haven't accidentally been linked against the wrong
130	version.	214	version (note, however, that this will not detect other ABI mismatches,
		215	such as LFS or reentrancy).
131		216
132	assert (("libev version mismatch",	217	assert (("libev version mismatch",
133	ev_version_major () == EV_VERSION_MAJOR	218	ev_version_major () == EV_VERSION_MAJOR
134	&& ev_version_minor () >= EV_VERSION_MINOR));	219	&& ev_version_minor () >= EV_VERSION_MINOR));
135		220
136	=item unsigned int ev_supported_backends ()	221	=item unsigned int ev_supported_backends ()
137		222
138	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>	223	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
139	value) compiled into this binary of libev (independent of their	224	value) compiled into this binary of libev (independent of their
…		…
141	a description of the set values.	226	a description of the set values.
142		227
143	Example: make sure we have the epoll method, because yeah this is cool and	228	Example: make sure we have the epoll method, because yeah this is cool and
144	a must have and can we have a torrent of it please!!!11	229	a must have and can we have a torrent of it please!!!11
145		230
146	assert (("sorry, no epoll, no sex",	231	assert (("sorry, no epoll, no sex",
147	ev_supported_backends () & EVBACKEND_EPOLL));	232	ev_supported_backends () & EVBACKEND_EPOLL));
148		233
149	=item unsigned int ev_recommended_backends ()	234	=item unsigned int ev_recommended_backends ()
150		235
151	Return the set of all backends compiled into this binary of libev and also	236	Return the set of all backends compiled into this binary of libev and
152	recommended for this platform. This set is often smaller than the one	237	also recommended for this platform, meaning it will work for most file
		238	descriptor types. This set is often smaller than the one returned by
153	returned by C<ev_supported_backends>, as for example kqueue is broken on	239	C<ev_supported_backends>, as for example kqueue is broken on most BSDs
154	most BSDs and will not be autodetected unless you explicitly request it	240	and will not be auto-detected unless you explicitly request it (assuming
155	(assuming you know what you are doing). This is the set of backends that	241	you know what you are doing). This is the set of backends that libev will
156	libev will probe for if you specify no backends explicitly.	242	probe for if you specify no backends explicitly.
157		243
158	=item unsigned int ev_embeddable_backends ()	244	=item unsigned int ev_embeddable_backends ()
159		245
160	Returns the set of backends that are embeddable in other event loops. This	246	Returns the set of backends that are embeddable in other event loops. This
161	is the theoretical, all-platform, value. To find which backends	247	value is platform-specific but can include backends not available on the
162	might be supported on the current system, you would need to look at	248	current system. To find which embeddable backends might be supported on
163	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	249	the current system, you would need to look at C<ev_embeddable_backends ()
164	recommended ones.	250	& ev_supported_backends ()>, likewise for recommended ones.
165		251
166	See the description of C<ev_embed> watchers for more info.	252	See the description of C<ev_embed> watchers for more info.
167		253
168	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	254	=item ev_set_allocator (void *(*cb)(void *ptr, long size) throw ())
169		255
170	Sets the allocation function to use (the prototype is similar - the	256	Sets the allocation function to use (the prototype is similar - the
171	semantics is identical - to the realloc C function). It is used to	257	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
172	allocate and free memory (no surprises here). If it returns zero when	258	used to allocate and free memory (no surprises here). If it returns zero
173	memory needs to be allocated, the library might abort or take some	259	when memory needs to be allocated (C<size != 0>), the library might abort
174	potentially destructive action. The default is your system realloc	260	or take some potentially destructive action.
175	function.	261
		262	Since some systems (at least OpenBSD and Darwin) fail to implement
		263	correct C<realloc> semantics, libev will use a wrapper around the system
		264	C<realloc> and C<free> functions by default.
176		265
177	You could override this function in high-availability programs to, say,	266	You could override this function in high-availability programs to, say,
178	free some memory if it cannot allocate memory, to use a special allocator,	267	free some memory if it cannot allocate memory, to use a special allocator,
179	or even to sleep a while and retry until some memory is available.	268	or even to sleep a while and retry until some memory is available.
180		269
181	Example: Replace the libev allocator with one that waits a bit and then	270	Example: Replace the libev allocator with one that waits a bit and then
182	retries).	271	retries (example requires a standards-compliant C<realloc>).
183		272
184	static void *	273	static void *
185	persistent_realloc (void *ptr, size_t size)	274	persistent_realloc (void *ptr, size_t size)
186	{	275	{
187	for (;;)	276	for (;;)
…		…
196	}	285	}
197		286
198	...	287	...
199	ev_set_allocator (persistent_realloc);	288	ev_set_allocator (persistent_realloc);
200		289
201	=item ev_set_syserr_cb (void (*cb)(const char *msg));	290	=item ev_set_syserr_cb (void (*cb)(const char *msg) throw ())
202		291
203	Set the callback function to call on a retryable syscall error (such	292	Set the callback function to call on a retryable system call error (such
204	as failed select, poll, epoll_wait). The message is a printable string	293	as failed select, poll, epoll_wait). The message is a printable string
205	indicating the system call or subsystem causing the problem. If this	294	indicating the system call or subsystem causing the problem. If this
206	callback is set, then libev will expect it to remedy the sitution, no	295	callback is set, then libev will expect it to remedy the situation, no
207	matter what, when it returns. That is, libev will generally retry the	296	matter what, when it returns. That is, libev will generally retry the
208	requested operation, or, if the condition doesn't go away, do bad stuff	297	requested operation, or, if the condition doesn't go away, do bad stuff
209	(such as abort).	298	(such as abort).
210		299
211	Example: This is basically the same thing that libev does internally, too.	300	Example: This is basically the same thing that libev does internally, too.
…		…
218	}	307	}
219		308
220	...	309	...
221	ev_set_syserr_cb (fatal_error);	310	ev_set_syserr_cb (fatal_error);
222		311
		312	=item ev_feed_signal (int signum)
		313
		314	This function can be used to "simulate" a signal receive. It is completely
		315	safe to call this function at any time, from any context, including signal
		316	handlers or random threads.
		317
		318	Its main use is to customise signal handling in your process, especially
		319	in the presence of threads. For example, you could block signals
		320	by default in all threads (and specifying C<EVFLAG_NOSIGMASK> when
		321	creating any loops), and in one thread, use C<sigwait> or any other
		322	mechanism to wait for signals, then "deliver" them to libev by calling
		323	C<ev_feed_signal>.
		324
223	=back	325	=back
224		326
225	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	327	=head1 FUNCTIONS CONTROLLING EVENT LOOPS
226		328
227	An event loop is described by a C<struct ev_loop *>. The library knows two	329	An event loop is described by a C<struct ev_loop *> (the C<struct> is
228	types of such loops, the I<default> loop, which supports signals and child	330	I<not> optional in this case unless libev 3 compatibility is disabled, as
229	events, and dynamically created loops which do not.	331	libev 3 had an C<ev_loop> function colliding with the struct name).
230		332
231	If you use threads, a common model is to run the default event loop	333	The library knows two types of such loops, the I<default> loop, which
232	in your main thread (or in a separate thread) and for each thread you	334	supports child process events, and dynamically created event loops which
233	create, you also create another event loop. Libev itself does no locking	335	do not.
234	whatsoever, so if you mix calls to the same event loop in different
235	threads, make sure you lock (this is usually a bad idea, though, even if
236	done correctly, because it's hideous and inefficient).
237		336
238	=over 4	337	=over 4
239		338
240	=item struct ev_loop *ev_default_loop (unsigned int flags)	339	=item struct ev_loop *ev_default_loop (unsigned int flags)
241		340
242	This will initialise the default event loop if it hasn't been initialised	341	This returns the "default" event loop object, which is what you should
243	yet and return it. If the default loop could not be initialised, returns	342	normally use when you just need "the event loop". Event loop objects and
244	false. If it already was initialised it simply returns it (and ignores the	343	the C<flags> parameter are described in more detail in the entry for
245	flags. If that is troubling you, check C<ev_backend ()> afterwards).	344	C<ev_loop_new>.
		345
		346	If the default loop is already initialised then this function simply
		347	returns it (and ignores the flags. If that is troubling you, check
		348	C<ev_backend ()> afterwards). Otherwise it will create it with the given
		349	flags, which should almost always be C<0>, unless the caller is also the
		350	one calling C<ev_run> or otherwise qualifies as "the main program".
246		351
247	If you don't know what event loop to use, use the one returned from this	352	If you don't know what event loop to use, use the one returned from this
248	function.	353	function (or via the C<EV_DEFAULT> macro).
		354
		355	Note that this function is I<not> thread-safe, so if you want to use it
		356	from multiple threads, you have to employ some kind of mutex (note also
		357	that this case is unlikely, as loops cannot be shared easily between
		358	threads anyway).
		359
		360	The default loop is the only loop that can handle C<ev_child> watchers,
		361	and to do this, it always registers a handler for C<SIGCHLD>. If this is
		362	a problem for your application you can either create a dynamic loop with
		363	C<ev_loop_new> which doesn't do that, or you can simply overwrite the
		364	C<SIGCHLD> signal handler I<after> calling C<ev_default_init>.
		365
		366	Example: This is the most typical usage.
		367
		368	if (!ev_default_loop (0))
		369	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
		370
		371	Example: Restrict libev to the select and poll backends, and do not allow
		372	environment settings to be taken into account:
		373
		374	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
		375
		376	=item struct ev_loop *ev_loop_new (unsigned int flags)
		377
		378	This will create and initialise a new event loop object. If the loop
		379	could not be initialised, returns false.
		380
		381	This function is thread-safe, and one common way to use libev with
		382	threads is indeed to create one loop per thread, and using the default
		383	loop in the "main" or "initial" thread.
249		384
250	The flags argument can be used to specify special behaviour or specific	385	The flags argument can be used to specify special behaviour or specific
251	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).	386	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
252		387
253	The following flags are supported:	388	The following flags are supported:
…		…
259	The default flags value. Use this if you have no clue (it's the right	394	The default flags value. Use this if you have no clue (it's the right
260	thing, believe me).	395	thing, believe me).
261		396
262	=item C<EVFLAG_NOENV>	397	=item C<EVFLAG_NOENV>
263		398
264	If this flag bit is ored into the flag value (or the program runs setuid	399	If this flag bit is or'ed into the flag value (or the program runs setuid
265	or setgid) then libev will I<not> look at the environment variable	400	or setgid) then libev will I<not> look at the environment variable
266	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	401	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
267	override the flags completely if it is found in the environment. This is	402	override the flags completely if it is found in the environment. This is
268	useful to try out specific backends to test their performance, or to work	403	useful to try out specific backends to test their performance, to work
269	around bugs.	404	around bugs, or to make libev threadsafe (accessing environment variables
		405	cannot be done in a threadsafe way, but usually it works if no other
		406	thread modifies them).
270		407
271	=item C<EVFLAG_FORKCHECK>	408	=item C<EVFLAG_FORKCHECK>
272		409
273	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after	410	Instead of calling C<ev_loop_fork> manually after a fork, you can also
274	a fork, you can also make libev check for a fork in each iteration by	411	make libev check for a fork in each iteration by enabling this flag.
275	enabling this flag.
276		412
277	This works by calling C<getpid ()> on every iteration of the loop,	413	This works by calling C<getpid ()> on every iteration of the loop,
278	and thus this might slow down your event loop if you do a lot of loop	414	and thus this might slow down your event loop if you do a lot of loop
279	iterations and little real work, but is usually not noticeable (on my	415	iterations and little real work, but is usually not noticeable (on my
280	Linux system for example, C<getpid> is actually a simple 5-insn sequence	416	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
281	without a syscall and thus I<very> fast, but my Linux system also has	417	without a system call and thus I<very> fast, but my GNU/Linux system also has
282	C<pthread_atfork> which is even faster).	418	C<pthread_atfork> which is even faster).
283		419
284	The big advantage of this flag is that you can forget about fork (and	420	The big advantage of this flag is that you can forget about fork (and
285	forget about forgetting to tell libev about forking) when you use this	421	forget about forgetting to tell libev about forking) when you use this
286	flag.	422	flag.
287		423
288	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>	424	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
289	environment variable.	425	environment variable.
		426
		427	=item C<EVFLAG_NOINOTIFY>
		428
		429	When this flag is specified, then libev will not attempt to use the
		430	I<inotify> API for its C<ev_stat> watchers. Apart from debugging and
		431	testing, this flag can be useful to conserve inotify file descriptors, as
		432	otherwise each loop using C<ev_stat> watchers consumes one inotify handle.
		433
		434	=item C<EVFLAG_SIGNALFD>
		435
		436	When this flag is specified, then libev will attempt to use the
		437	I<signalfd> API for its C<ev_signal> (and C<ev_child>) watchers. This API
		438	delivers signals synchronously, which makes it both faster and might make
		439	it possible to get the queued signal data. It can also simplify signal
		440	handling with threads, as long as you properly block signals in your
		441	threads that are not interested in handling them.
		442
		443	Signalfd will not be used by default as this changes your signal mask, and
		444	there are a lot of shoddy libraries and programs (glib's threadpool for
		445	example) that can't properly initialise their signal masks.
		446
		447	=item C<EVFLAG_NOSIGMASK>
		448
		449	When this flag is specified, then libev will avoid to modify the signal
		450	mask. Specifically, this means you have to make sure signals are unblocked
		451	when you want to receive them.
		452
		453	This behaviour is useful when you want to do your own signal handling, or
		454	want to handle signals only in specific threads and want to avoid libev
		455	unblocking the signals.
		456
		457	It's also required by POSIX in a threaded program, as libev calls
		458	C<sigprocmask>, whose behaviour is officially unspecified.
		459
		460	This flag's behaviour will become the default in future versions of libev.
290		461
291	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	462	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
292		463
293	This is your standard select(2) backend. Not I<completely> standard, as	464	This is your standard select(2) backend. Not I<completely> standard, as
294	libev tries to roll its own fd_set with no limits on the number of fds,	465	libev tries to roll its own fd_set with no limits on the number of fds,
295	but if that fails, expect a fairly low limit on the number of fds when	466	but if that fails, expect a fairly low limit on the number of fds when
296	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	467	using this backend. It doesn't scale too well (O(highest_fd)), but its
297	the fastest backend for a low number of fds.	468	usually the fastest backend for a low number of (low-numbered :) fds.
		469
		470	To get good performance out of this backend you need a high amount of
		471	parallelism (most of the file descriptors should be busy). If you are
		472	writing a server, you should C<accept ()> in a loop to accept as many
		473	connections as possible during one iteration. You might also want to have
		474	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		475	readiness notifications you get per iteration.
		476
		477	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		478	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		479	C<exceptfds> set on that platform).
298		480
299	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	481	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
300		482
301	And this is your standard poll(2) backend. It's more complicated than	483	And this is your standard poll(2) backend. It's more complicated
302	select, but handles sparse fds better and has no artificial limit on the	484	than select, but handles sparse fds better and has no artificial
303	number of fds you can use (except it will slow down considerably with a	485	limit on the number of fds you can use (except it will slow down
304	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	486	considerably with a lot of inactive fds). It scales similarly to select,
		487	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		488	performance tips.
		489
		490	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		491	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
305		492
306	=item C<EVBACKEND_EPOLL> (value 4, Linux)	493	=item C<EVBACKEND_EPOLL> (value 4, Linux)
307		494
		495	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
		496	kernels).
		497
308	For few fds, this backend is a bit little slower than poll and select,	498	For few fds, this backend is a bit little slower than poll and select, but
309	but it scales phenomenally better. While poll and select usually scale like	499	it scales phenomenally better. While poll and select usually scale like
310	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	500	O(total_fds) where total_fds is the total number of fds (or the highest
311	either O(1) or O(active_fds).	501	fd), epoll scales either O(1) or O(active_fds).
312		502
		503	The epoll mechanism deserves honorable mention as the most misdesigned
		504	of the more advanced event mechanisms: mere annoyances include silently
		505	dropping file descriptors, requiring a system call per change per file
		506	descriptor (and unnecessary guessing of parameters), problems with dup,
		507	returning before the timeout value, resulting in additional iterations
		508	(and only giving 5ms accuracy while select on the same platform gives
		509	0.1ms) and so on. The biggest issue is fork races, however - if a program
		510	forks then I<both> parent and child process have to recreate the epoll
		511	set, which can take considerable time (one syscall per file descriptor)
		512	and is of course hard to detect.
		513
		514	Epoll is also notoriously buggy - embedding epoll fds I<should> work,
		515	but of course I<doesn't>, and epoll just loves to report events for
		516	totally I<different> file descriptors (even already closed ones, so
		517	one cannot even remove them from the set) than registered in the set
		518	(especially on SMP systems). Libev tries to counter these spurious
		519	notifications by employing an additional generation counter and comparing
		520	that against the events to filter out spurious ones, recreating the set
		521	when required. Epoll also erroneously rounds down timeouts, but gives you
		522	no way to know when and by how much, so sometimes you have to busy-wait
		523	because epoll returns immediately despite a nonzero timeout. And last
		524	not least, it also refuses to work with some file descriptors which work
		525	perfectly fine with C<select> (files, many character devices...).
		526
		527	Epoll is truly the train wreck among event poll mechanisms, a frankenpoll,
		528	cobbled together in a hurry, no thought to design or interaction with
		529	others. Oh, the pain, will it ever stop...
		530
313	While stopping and starting an I/O watcher in the same iteration will	531	While stopping, setting and starting an I/O watcher in the same iteration
314	result in some caching, there is still a syscall per such incident	532	will result in some caching, there is still a system call per such
315	(because the fd could point to a different file description now), so its	533	incident (because the same I<file descriptor> could point to a different
316	best to avoid that. Also, dup()ed file descriptors might not work very	534	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
317	well if you register events for both fds.	535	file descriptors might not work very well if you register events for both
		536	file descriptors.
318		537
319	Please note that epoll sometimes generates spurious notifications, so you	538	Best performance from this backend is achieved by not unregistering all
320	need to use non-blocking I/O or other means to avoid blocking when no data	539	watchers for a file descriptor until it has been closed, if possible,
321	(or space) is available.	540	i.e. keep at least one watcher active per fd at all times. Stopping and
		541	starting a watcher (without re-setting it) also usually doesn't cause
		542	extra overhead. A fork can both result in spurious notifications as well
		543	as in libev having to destroy and recreate the epoll object, which can
		544	take considerable time and thus should be avoided.
		545
		546	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		547	faster than epoll for maybe up to a hundred file descriptors, depending on
		548	the usage. So sad.
		549
		550	While nominally embeddable in other event loops, this feature is broken in
		551	all kernel versions tested so far.
		552
		553	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		554	C<EVBACKEND_POLL>.
322		555
323	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	556	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
324		557
325	Kqueue deserves special mention, as at the time of this writing, it	558	Kqueue deserves special mention, as at the time of this writing, it
326	was broken on all BSDs except NetBSD (usually it doesn't work with	559	was broken on all BSDs except NetBSD (usually it doesn't work reliably
327	anything but sockets and pipes, except on Darwin, where of course its	560	with anything but sockets and pipes, except on Darwin, where of course
328	completely useless). For this reason its not being "autodetected"	561	it's completely useless). Unlike epoll, however, whose brokenness
		562	is by design, these kqueue bugs can (and eventually will) be fixed
		563	without API changes to existing programs. For this reason it's not being
329	unless you explicitly specify it explicitly in the flags (i.e. using	564	"auto-detected" unless you explicitly specify it in the flags (i.e. using
330	C<EVBACKEND_KQUEUE>).	565	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		566	system like NetBSD.
		567
		568	You still can embed kqueue into a normal poll or select backend and use it
		569	only for sockets (after having made sure that sockets work with kqueue on
		570	the target platform). See C<ev_embed> watchers for more info.
331		571
332	It scales in the same way as the epoll backend, but the interface to the	572	It scales in the same way as the epoll backend, but the interface to the
333	kernel is more efficient (which says nothing about its actual speed, of	573	kernel is more efficient (which says nothing about its actual speed, of
334	course). While starting and stopping an I/O watcher does not cause an	574	course). While stopping, setting and starting an I/O watcher does never
335	extra syscall as with epoll, it still adds up to four event changes per	575	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
336	incident, so its best to avoid that.	576	two event changes per incident. Support for C<fork ()> is very bad (you
		577	might have to leak fd's on fork, but it's more sane than epoll) and it
		578	drops fds silently in similarly hard-to-detect cases.
		579
		580	This backend usually performs well under most conditions.
		581
		582	While nominally embeddable in other event loops, this doesn't work
		583	everywhere, so you might need to test for this. And since it is broken
		584	almost everywhere, you should only use it when you have a lot of sockets
		585	(for which it usually works), by embedding it into another event loop
		586	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
		587	also broken on OS X)) and, did I mention it, using it only for sockets.
		588
		589	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		590	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		591	C<NOTE_EOF>.
337		592
338	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	593	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
339		594
340	This is not implemented yet (and might never be).	595	This is not implemented yet (and might never be, unless you send me an
		596	implementation). According to reports, C</dev/poll> only supports sockets
		597	and is not embeddable, which would limit the usefulness of this backend
		598	immensely.
341		599
342	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	600	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
343		601
344	This uses the Solaris 10 port mechanism. As with everything on Solaris,	602	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
345	it's really slow, but it still scales very well (O(active_fds)).	603	it's really slow, but it still scales very well (O(active_fds)).
346		604
347	Please note that solaris ports can result in a lot of spurious	605	While this backend scales well, it requires one system call per active
348	notifications, so you need to use non-blocking I/O or other means to avoid	606	file descriptor per loop iteration. For small and medium numbers of file
349	blocking when no data (or space) is available.	607	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		608	might perform better.
		609
		610	On the positive side, this backend actually performed fully to
		611	specification in all tests and is fully embeddable, which is a rare feat
		612	among the OS-specific backends (I vastly prefer correctness over speed
		613	hacks).
		614
		615	On the negative side, the interface is I<bizarre> - so bizarre that
		616	even sun itself gets it wrong in their code examples: The event polling
		617	function sometimes returns events to the caller even though an error
		618	occurred, but with no indication whether it has done so or not (yes, it's
		619	even documented that way) - deadly for edge-triggered interfaces where you
		620	absolutely have to know whether an event occurred or not because you have
		621	to re-arm the watcher.
		622
		623	Fortunately libev seems to be able to work around these idiocies.
		624
		625	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		626	C<EVBACKEND_POLL>.
350		627
351	=item C<EVBACKEND_ALL>	628	=item C<EVBACKEND_ALL>
352		629
353	Try all backends (even potentially broken ones that wouldn't be tried	630	Try all backends (even potentially broken ones that wouldn't be tried
354	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	631	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
355	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	632	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
356		633
		634	It is definitely not recommended to use this flag, use whatever
		635	C<ev_recommended_backends ()> returns, or simply do not specify a backend
		636	at all.
		637
		638	=item C<EVBACKEND_MASK>
		639
		640	Not a backend at all, but a mask to select all backend bits from a
		641	C<flags> value, in case you want to mask out any backends from a flags
		642	value (e.g. when modifying the C<LIBEV_FLAGS> environment variable).
		643
357	=back	644	=back
358		645
359	If one or more of these are ored into the flags value, then only these	646	If one or more of the backend flags are or'ed into the flags value,
360	backends will be tried (in the reverse order as given here). If none are	647	then only these backends will be tried (in the reverse order as listed
361	specified, most compiled-in backend will be tried, usually in reverse	648	here). If none are specified, all backends in C<ev_recommended_backends
362	order of their flag values :)	649	()> will be tried.
363
364	The most typical usage is like this:
365
366	if (!ev_default_loop (0))
367	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
368
369	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	Use whatever libev has to offer, but make sure that kqueue is used if
375	available (warning, breaks stuff, best use only with your own private
376	event loop and only if you know the OS supports your types of fds):
377
378	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
379
380	=item struct ev_loop *ev_loop_new (unsigned int flags)
381
382	Similar to C<ev_default_loop>, but always creates a new event loop that is
383	always distinct from the default loop. Unlike the default loop, it cannot
384	handle signal and child watchers, and attempts to do so will be greeted by
385	undefined behaviour (or a failed assertion if assertions are enabled).
386		650
387	Example: Try to create a event loop that uses epoll and nothing else.	651	Example: Try to create a event loop that uses epoll and nothing else.
388		652
389	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	653	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
390	if (!epoller)	654	if (!epoller)
391	fatal ("no epoll found here, maybe it hides under your chair");	655	fatal ("no epoll found here, maybe it hides under your chair");
392		656
		657	Example: Use whatever libev has to offer, but make sure that kqueue is
		658	used if available.
		659
		660	struct ev_loop *loop = ev_loop_new (ev_recommended_backends () | EVBACKEND_KQUEUE);
		661
393	=item ev_default_destroy ()	662	=item ev_loop_destroy (loop)
394		663
395	Destroys the default loop again (frees all memory and kernel state	664	Destroys an event loop object (frees all memory and kernel state
396	etc.). None of the active event watchers will be stopped in the normal	665	etc.). None of the active event watchers will be stopped in the normal
397	sense, so e.g. C<ev_is_active> might still return true. It is your	666	sense, so e.g. C<ev_is_active> might still return true. It is your
398	responsibility to either stop all watchers cleanly yoursef I<before>	667	responsibility to either stop all watchers cleanly yourself I<before>
399	calling this function, or cope with the fact afterwards (which is usually	668	calling this function, or cope with the fact afterwards (which is usually
400	the easiest thing, youc na just ignore the watchers and/or C<free ()> them	669	the easiest thing, you can just ignore the watchers and/or C<free ()> them
401	for example).	670	for example).
402		671
		672	Note that certain global state, such as signal state (and installed signal
		673	handlers), will not be freed by this function, and related watchers (such
		674	as signal and child watchers) would need to be stopped manually.
		675
		676	This function is normally used on loop objects allocated by
		677	C<ev_loop_new>, but it can also be used on the default loop returned by
		678	C<ev_default_loop>, in which case it is not thread-safe.
		679
		680	Note that it is not advisable to call this function on the default loop
		681	except in the rare occasion where you really need to free its resources.
		682	If you need dynamically allocated loops it is better to use C<ev_loop_new>
		683	and C<ev_loop_destroy>.
		684
403	=item ev_loop_destroy (loop)	685	=item ev_loop_fork (loop)
404		686
405	Like C<ev_default_destroy>, but destroys an event loop created by an	687	This function sets a flag that causes subsequent C<ev_run> iterations to
406	earlier call to C<ev_loop_new>.
407
408	=item ev_default_fork ()
409
410	This function reinitialises the kernel state for backends that have	688	reinitialise the kernel state for backends that have one. Despite the
411	one. Despite the name, you can call it anytime, but it makes most sense	689	name, you can call it anytime, but it makes most sense after forking, in
412	after forking, in either the parent or child process (or both, but that	690	the child process. You I<must> call it (or use C<EVFLAG_FORKCHECK>) in the
413	again makes little sense).	691	child before resuming or calling C<ev_run>.
414		692
415	You I<must> call this function in the child process after forking if and	693	Again, you I<have> to call it on I<any> loop that you want to re-use after
416	only if you want to use the event library in both processes. If you just	694	a fork, I<even if you do not plan to use the loop in the parent>. This is
417	fork+exec, you don't have to call it.	695	because some kernel interfaces *cough* I<kqueue> *cough* do funny things
		696	during fork.
		697
		698	On the other hand, you only need to call this function in the child
		699	process if and only if you want to use the event loop in the child. If
		700	you just fork+exec or create a new loop in the child, you don't have to
		701	call it at all (in fact, C<epoll> is so badly broken that it makes a
		702	difference, but libev will usually detect this case on its own and do a
		703	costly reset of the backend).
418		704
419	The function itself is quite fast and it's usually not a problem to call	705	The function itself is quite fast and it's usually not a problem to call
420	it just in case after a fork. To make this easy, the function will fit in	706	it just in case after a fork.
421	quite nicely into a call to C<pthread_atfork>:
422		707
		708	Example: Automate calling C<ev_loop_fork> on the default loop when
		709	using pthreads.
		710
		711	static void
		712	post_fork_child (void)
		713	{
		714	ev_loop_fork (EV_DEFAULT);
		715	}
		716
		717	...
423	pthread_atfork (0, 0, ev_default_fork);	718	pthread_atfork (0, 0, post_fork_child);
424		719
425	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use	720	=item int ev_is_default_loop (loop)
426	without calling this function, so if you force one of those backends you
427	do not need to care.
428		721
429	=item ev_loop_fork (loop)	722	Returns true when the given loop is, in fact, the default loop, and false
		723	otherwise.
430		724
431	Like C<ev_default_fork>, but acts on an event loop created by	725	=item unsigned int ev_iteration (loop)
432	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	726
433	after fork, and how you do this is entirely your own problem.	727	Returns the current iteration count for the event loop, which is identical
		728	to the number of times libev did poll for new events. It starts at C<0>
		729	and happily wraps around with enough iterations.
		730
		731	This value can sometimes be useful as a generation counter of sorts (it
		732	"ticks" the number of loop iterations), as it roughly corresponds with
		733	C<ev_prepare> and C<ev_check> calls - and is incremented between the
		734	prepare and check phases.
		735
		736	=item unsigned int ev_depth (loop)
		737
		738	Returns the number of times C<ev_run> was entered minus the number of
		739	times C<ev_run> was exited normally, in other words, the recursion depth.
		740
		741	Outside C<ev_run>, this number is zero. In a callback, this number is
		742	C<1>, unless C<ev_run> was invoked recursively (or from another thread),
		743	in which case it is higher.
		744
		745	Leaving C<ev_run> abnormally (setjmp/longjmp, cancelling the thread,
		746	throwing an exception etc.), doesn't count as "exit" - consider this
		747	as a hint to avoid such ungentleman-like behaviour unless it's really
		748	convenient, in which case it is fully supported.
434		749
435	=item unsigned int ev_backend (loop)	750	=item unsigned int ev_backend (loop)
436		751
437	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	752	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
438	use.	753	use.
…		…
441		756
442	Returns the current "event loop time", which is the time the event loop	757	Returns the current "event loop time", which is the time the event loop
443	received events and started processing them. This timestamp does not	758	received events and started processing them. This timestamp does not
444	change as long as callbacks are being processed, and this is also the base	759	change as long as callbacks are being processed, and this is also the base
445	time used for relative timers. You can treat it as the timestamp of the	760	time used for relative timers. You can treat it as the timestamp of the
446	event occuring (or more correctly, libev finding out about it).	761	event occurring (or more correctly, libev finding out about it).
447		762
		763	=item ev_now_update (loop)
		764
		765	Establishes the current time by querying the kernel, updating the time
		766	returned by C<ev_now ()> in the progress. This is a costly operation and
		767	is usually done automatically within C<ev_run ()>.
		768
		769	This function is rarely useful, but when some event callback runs for a
		770	very long time without entering the event loop, updating libev's idea of
		771	the current time is a good idea.
		772
		773	See also L</The special problem of time updates> in the C<ev_timer> section.
		774
		775	=item ev_suspend (loop)
		776
		777	=item ev_resume (loop)
		778
		779	These two functions suspend and resume an event loop, for use when the
		780	loop is not used for a while and timeouts should not be processed.
		781
		782	A typical use case would be an interactive program such as a game: When
		783	the user presses C<^Z> to suspend the game and resumes it an hour later it
		784	would be best to handle timeouts as if no time had actually passed while
		785	the program was suspended. This can be achieved by calling C<ev_suspend>
		786	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		787	C<ev_resume> directly afterwards to resume timer processing.
		788
		789	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		790	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		791	will be rescheduled (that is, they will lose any events that would have
		792	occurred while suspended).
		793
		794	After calling C<ev_suspend> you B<must not> call I<any> function on the
		795	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		796	without a previous call to C<ev_suspend>.
		797
		798	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		799	event loop time (see C<ev_now_update>).
		800
448	=item ev_loop (loop, int flags)	801	=item bool ev_run (loop, int flags)
449		802
450	Finally, this is it, the event handler. This function usually is called	803	Finally, this is it, the event handler. This function usually is called
451	after you initialised all your watchers and you want to start handling	804	after you have initialised all your watchers and you want to start
452	events.	805	handling events. It will ask the operating system for any new events, call
		806	the watcher callbacks, and then repeat the whole process indefinitely: This
		807	is why event loops are called I<loops>.
453		808
454	If the flags argument is specified as C<0>, it will not return until	809	If the flags argument is specified as C<0>, it will keep handling events
455	either no event watchers are active anymore or C<ev_unloop> was called.	810	until either no event watchers are active anymore or C<ev_break> was
		811	called.
456		812
		813	The return value is false if there are no more active watchers (which
		814	usually means "all jobs done" or "deadlock"), and true in all other cases
		815	(which usually means " you should call C<ev_run> again").
		816
457	Please note that an explicit C<ev_unloop> is usually better than	817	Please note that an explicit C<ev_break> is usually better than
458	relying on all watchers to be stopped when deciding when a program has	818	relying on all watchers to be stopped when deciding when a program has
459	finished (especially in interactive programs), but having a program that	819	finished (especially in interactive programs), but having a program
460	automatically loops as long as it has to and no longer by virtue of	820	that automatically loops as long as it has to and no longer by virtue
461	relying on its watchers stopping correctly is a thing of beauty.	821	of relying on its watchers stopping correctly, that is truly a thing of
		822	beauty.
462		823
		824	This function is I<mostly> exception-safe - you can break out of a
		825	C<ev_run> call by calling C<longjmp> in a callback, throwing a C++
		826	exception and so on. This does not decrement the C<ev_depth> value, nor
		827	will it clear any outstanding C<EVBREAK_ONE> breaks.
		828
463	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	829	A flags value of C<EVRUN_NOWAIT> will look for new events, will handle
464	those events and any outstanding ones, but will not block your process in	830	those events and any already outstanding ones, but will not wait and
465	case there are no events and will return after one iteration of the loop.	831	block your process in case there are no events and will return after one
		832	iteration of the loop. This is sometimes useful to poll and handle new
		833	events while doing lengthy calculations, to keep the program responsive.
466		834
467	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	835	A flags value of C<EVRUN_ONCE> will look for new events (waiting if
468	neccessary) and will handle those and any outstanding ones. It will block	836	necessary) and will handle those and any already outstanding ones. It
469	your process until at least one new event arrives, and will return after	837	will block your process until at least one new event arrives (which could
470	one iteration of the loop. This is useful if you are waiting for some	838	be an event internal to libev itself, so there is no guarantee that a
471	external event in conjunction with something not expressible using other	839	user-registered callback will be called), and will return after one
		840	iteration of the loop.
		841
		842	This is useful if you are waiting for some external event in conjunction
		843	with something not expressible using other libev watchers (i.e. "roll your
472	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	844	own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
473	usually a better approach for this kind of thing.	845	usually a better approach for this kind of thing.
474		846
475	Here are the gory details of what C<ev_loop> does:	847	Here are the gory details of what C<ev_run> does (this is for your
		848	understanding, not a guarantee that things will work exactly like this in
		849	future versions):
476		850
477	* If there are no active watchers (reference count is zero), return.	851	- Increment loop depth.
478	- Queue prepare watchers and then call all outstanding watchers.	852	- Reset the ev_break status.
		853	- Before the first iteration, call any pending watchers.
		854	LOOP:
		855	- If EVFLAG_FORKCHECK was used, check for a fork.
		856	- If a fork was detected (by any means), queue and call all fork watchers.
		857	- Queue and call all prepare watchers.
		858	- If ev_break was called, goto FINISH.
479	- If we have been forked, recreate the kernel state.	859	- If we have been forked, detach and recreate the kernel state
		860	as to not disturb the other process.
480	- Update the kernel state with all outstanding changes.	861	- Update the kernel state with all outstanding changes.
481	- Update the "event loop time".	862	- Update the "event loop time" (ev_now ()).
482	- Calculate for how long to block.	863	- Calculate for how long to sleep or block, if at all
		864	(active idle watchers, EVRUN_NOWAIT or not having
		865	any active watchers at all will result in not sleeping).
		866	- Sleep if the I/O and timer collect interval say so.
		867	- Increment loop iteration counter.
483	- Block the process, waiting for any events.	868	- Block the process, waiting for any events.
484	- Queue all outstanding I/O (fd) events.	869	- Queue all outstanding I/O (fd) events.
485	- Update the "event loop time" and do time jump handling.	870	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
486	- Queue all outstanding timers.	871	- Queue all expired timers.
487	- Queue all outstanding periodics.	872	- Queue all expired periodics.
488	- If no events are pending now, queue all idle watchers.	873	- Queue all idle watchers with priority higher than that of pending events.
489	- Queue all check watchers.	874	- Queue all check watchers.
490	- Call all queued watchers in reverse order (i.e. check watchers first).	875	- Call all queued watchers in reverse order (i.e. check watchers first).
491	Signals and child watchers are implemented as I/O watchers, and will	876	Signals and child watchers are implemented as I/O watchers, and will
492	be handled here by queueing them when their watcher gets executed.	877	be handled here by queueing them when their watcher gets executed.
493	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	878	- If ev_break has been called, or EVRUN_ONCE or EVRUN_NOWAIT
494	were used, return, otherwise continue with step *.	879	were used, or there are no active watchers, goto FINISH, otherwise
		880	continue with step LOOP.
		881	FINISH:
		882	- Reset the ev_break status iff it was EVBREAK_ONE.
		883	- Decrement the loop depth.
		884	- Return.
495		885
496	Example: Queue some jobs and then loop until no events are outsanding	886	Example: Queue some jobs and then loop until no events are outstanding
497	anymore.	887	anymore.
498		888
499	... queue jobs here, make sure they register event watchers as long	889	... queue jobs here, make sure they register event watchers as long
500	... as they still have work to do (even an idle watcher will do..)	890	... as they still have work to do (even an idle watcher will do..)
501	ev_loop (my_loop, 0);	891	ev_run (my_loop, 0);
502	... jobs done. yeah!	892	... jobs done or somebody called break. yeah!
503		893
504	=item ev_unloop (loop, how)	894	=item ev_break (loop, how)
505		895
506	Can be used to make a call to C<ev_loop> return early (but only after it	896	Can be used to make a call to C<ev_run> return early (but only after it
507	has processed all outstanding events). The C<how> argument must be either	897	has processed all outstanding events). The C<how> argument must be either
508	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	898	C<EVBREAK_ONE>, which will make the innermost C<ev_run> call return, or
509	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	899	C<EVBREAK_ALL>, which will make all nested C<ev_run> calls return.
		900
		901	This "break state" will be cleared on the next call to C<ev_run>.
		902
		903	It is safe to call C<ev_break> from outside any C<ev_run> calls, too, in
		904	which case it will have no effect.
510		905
511	=item ev_ref (loop)	906	=item ev_ref (loop)
512		907
513	=item ev_unref (loop)	908	=item ev_unref (loop)
514		909
515	Ref/unref can be used to add or remove a reference count on the event	910	Ref/unref can be used to add or remove a reference count on the event
516	loop: Every watcher keeps one reference, and as long as the reference	911	loop: Every watcher keeps one reference, and as long as the reference
517	count is nonzero, C<ev_loop> will not return on its own. If you have	912	count is nonzero, C<ev_run> will not return on its own.
518	a watcher you never unregister that should not keep C<ev_loop> from	913
519	returning, ev_unref() after starting, and ev_ref() before stopping it. For	914	This is useful when you have a watcher that you never intend to
		915	unregister, but that nevertheless should not keep C<ev_run> from
		916	returning. In such a case, call C<ev_unref> after starting, and C<ev_ref>
		917	before stopping it.
		918
520	example, libev itself uses this for its internal signal pipe: It is not	919	As an example, libev itself uses this for its internal signal pipe: It
521	visible to the libev user and should not keep C<ev_loop> from exiting if	920	is not visible to the libev user and should not keep C<ev_run> from
522	no event watchers registered by it are active. It is also an excellent	921	exiting if no event watchers registered by it are active. It is also an
523	way to do this for generic recurring timers or from within third-party	922	excellent way to do this for generic recurring timers or from within
524	libraries. Just remember to I<unref after start> and I<ref before stop>.	923	third-party libraries. Just remember to I<unref after start> and I<ref
		924	before stop> (but only if the watcher wasn't active before, or was active
		925	before, respectively. Note also that libev might stop watchers itself
		926	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		927	in the callback).
525		928
526	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	929	Example: Create a signal watcher, but keep it from keeping C<ev_run>
527	running when nothing else is active.	930	running when nothing else is active.
528		931
529	struct ev_signal exitsig;	932	ev_signal exitsig;
530	ev_signal_init (&exitsig, sig_cb, SIGINT);	933	ev_signal_init (&exitsig, sig_cb, SIGINT);
531	ev_signal_start (loop, &exitsig);	934	ev_signal_start (loop, &exitsig);
532	evf_unref (loop);	935	ev_unref (loop);
533		936
534	Example: For some weird reason, unregister the above signal handler again.	937	Example: For some weird reason, unregister the above signal handler again.
535		938
536	ev_ref (loop);	939	ev_ref (loop);
537	ev_signal_stop (loop, &exitsig);	940	ev_signal_stop (loop, &exitsig);
		941
		942	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		943
		944	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		945
		946	These advanced functions influence the time that libev will spend waiting
		947	for events. Both time intervals are by default C<0>, meaning that libev
		948	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		949	latency.
		950
		951	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		952	allows libev to delay invocation of I/O and timer/periodic callbacks
		953	to increase efficiency of loop iterations (or to increase power-saving
		954	opportunities).
		955
		956	The idea is that sometimes your program runs just fast enough to handle
		957	one (or very few) event(s) per loop iteration. While this makes the
		958	program responsive, it also wastes a lot of CPU time to poll for new
		959	events, especially with backends like C<select ()> which have a high
		960	overhead for the actual polling but can deliver many events at once.
		961
		962	By setting a higher I<io collect interval> you allow libev to spend more
		963	time collecting I/O events, so you can handle more events per iteration,
		964	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		965	C<ev_timer>) will not be affected. Setting this to a non-null value will
		966	introduce an additional C<ev_sleep ()> call into most loop iterations. The
		967	sleep time ensures that libev will not poll for I/O events more often then
		968	once per this interval, on average (as long as the host time resolution is
		969	good enough).
		970
		971	Likewise, by setting a higher I<timeout collect interval> you allow libev
		972	to spend more time collecting timeouts, at the expense of increased
		973	latency/jitter/inexactness (the watcher callback will be called
		974	later). C<ev_io> watchers will not be affected. Setting this to a non-null
		975	value will not introduce any overhead in libev.
		976
		977	Many (busy) programs can usually benefit by setting the I/O collect
		978	interval to a value near C<0.1> or so, which is often enough for
		979	interactive servers (of course not for games), likewise for timeouts. It
		980	usually doesn't make much sense to set it to a lower value than C<0.01>,
		981	as this approaches the timing granularity of most systems. Note that if
		982	you do transactions with the outside world and you can't increase the
		983	parallelity, then this setting will limit your transaction rate (if you
		984	need to poll once per transaction and the I/O collect interval is 0.01,
		985	then you can't do more than 100 transactions per second).
		986
		987	Setting the I<timeout collect interval> can improve the opportunity for
		988	saving power, as the program will "bundle" timer callback invocations that
		989	are "near" in time together, by delaying some, thus reducing the number of
		990	times the process sleeps and wakes up again. Another useful technique to
		991	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		992	they fire on, say, one-second boundaries only.
		993
		994	Example: we only need 0.1s timeout granularity, and we wish not to poll
		995	more often than 100 times per second:
		996
		997	ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
		998	ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
		999
		1000	=item ev_invoke_pending (loop)
		1001
		1002	This call will simply invoke all pending watchers while resetting their
		1003	pending state. Normally, C<ev_run> does this automatically when required,
		1004	but when overriding the invoke callback this call comes handy. This
		1005	function can be invoked from a watcher - this can be useful for example
		1006	when you want to do some lengthy calculation and want to pass further
		1007	event handling to another thread (you still have to make sure only one
		1008	thread executes within C<ev_invoke_pending> or C<ev_run> of course).
		1009
		1010	=item int ev_pending_count (loop)
		1011
		1012	Returns the number of pending watchers - zero indicates that no watchers
		1013	are pending.
		1014
		1015	=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
		1016
		1017	This overrides the invoke pending functionality of the loop: Instead of
		1018	invoking all pending watchers when there are any, C<ev_run> will call
		1019	this callback instead. This is useful, for example, when you want to
		1020	invoke the actual watchers inside another context (another thread etc.).
		1021
		1022	If you want to reset the callback, use C<ev_invoke_pending> as new
		1023	callback.
		1024
		1025	=item ev_set_loop_release_cb (loop, void (*release)(EV_P) throw (), void (*acquire)(EV_P) throw ())
		1026
		1027	Sometimes you want to share the same loop between multiple threads. This
		1028	can be done relatively simply by putting mutex_lock/unlock calls around
		1029	each call to a libev function.
		1030
		1031	However, C<ev_run> can run an indefinite time, so it is not feasible
		1032	to wait for it to return. One way around this is to wake up the event
		1033	loop via C<ev_break> and C<ev_async_send>, another way is to set these
		1034	I<release> and I<acquire> callbacks on the loop.
		1035
		1036	When set, then C<release> will be called just before the thread is
		1037	suspended waiting for new events, and C<acquire> is called just
		1038	afterwards.
		1039
		1040	Ideally, C<release> will just call your mutex_unlock function, and
		1041	C<acquire> will just call the mutex_lock function again.
		1042
		1043	While event loop modifications are allowed between invocations of
		1044	C<release> and C<acquire> (that's their only purpose after all), no
		1045	modifications done will affect the event loop, i.e. adding watchers will
		1046	have no effect on the set of file descriptors being watched, or the time
		1047	waited. Use an C<ev_async> watcher to wake up C<ev_run> when you want it
		1048	to take note of any changes you made.
		1049
		1050	In theory, threads executing C<ev_run> will be async-cancel safe between
		1051	invocations of C<release> and C<acquire>.
		1052
		1053	See also the locking example in the C<THREADS> section later in this
		1054	document.
		1055
		1056	=item ev_set_userdata (loop, void *data)
		1057
		1058	=item void *ev_userdata (loop)
		1059
		1060	Set and retrieve a single C<void *> associated with a loop. When
		1061	C<ev_set_userdata> has never been called, then C<ev_userdata> returns
		1062	C<0>.
		1063
		1064	These two functions can be used to associate arbitrary data with a loop,
		1065	and are intended solely for the C<invoke_pending_cb>, C<release> and
		1066	C<acquire> callbacks described above, but of course can be (ab-)used for
		1067	any other purpose as well.
		1068
		1069	=item ev_verify (loop)
		1070
		1071	This function only does something when C<EV_VERIFY> support has been
		1072	compiled in, which is the default for non-minimal builds. It tries to go
		1073	through all internal structures and checks them for validity. If anything
		1074	is found to be inconsistent, it will print an error message to standard
		1075	error and call C<abort ()>.
		1076
		1077	This can be used to catch bugs inside libev itself: under normal
		1078	circumstances, this function will never abort as of course libev keeps its
		1079	data structures consistent.
538		1080
539	=back	1081	=back
540		1082
541		1083
542	=head1 ANATOMY OF A WATCHER	1084	=head1 ANATOMY OF A WATCHER
543		1085
		1086	In the following description, uppercase C<TYPE> in names stands for the
		1087	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		1088	watchers and C<ev_io_start> for I/O watchers.
		1089
544	A watcher is a structure that you create and register to record your	1090	A watcher is an opaque structure that you allocate and register to record
545	interest in some event. For instance, if you want to wait for STDIN to	1091	your interest in some event. To make a concrete example, imagine you want
546	become readable, you would create an C<ev_io> watcher for that:	1092	to wait for STDIN to become readable, you would create an C<ev_io> watcher
		1093	for that:
547		1094
548	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1095	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
549	{	1096	{
550	ev_io_stop (w);	1097	ev_io_stop (w);
551	ev_unloop (loop, EVUNLOOP_ALL);	1098	ev_break (loop, EVBREAK_ALL);
552	}	1099	}
553		1100
554	struct ev_loop *loop = ev_default_loop (0);	1101	struct ev_loop *loop = ev_default_loop (0);
		1102
555	struct ev_io stdin_watcher;	1103	ev_io stdin_watcher;
		1104
556	ev_init (&stdin_watcher, my_cb);	1105	ev_init (&stdin_watcher, my_cb);
557	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	1106	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
558	ev_io_start (loop, &stdin_watcher);	1107	ev_io_start (loop, &stdin_watcher);
		1108
559	ev_loop (loop, 0);	1109	ev_run (loop, 0);
560		1110
561	As you can see, you are responsible for allocating the memory for your	1111	As you can see, you are responsible for allocating the memory for your
562	watcher structures (and it is usually a bad idea to do this on the stack,	1112	watcher structures (and it is I<usually> a bad idea to do this on the
563	although this can sometimes be quite valid).	1113	stack).
564		1114
		1115	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		1116	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
		1117
565	Each watcher structure must be initialised by a call to C<ev_init	1118	Each watcher structure must be initialised by a call to C<ev_init (watcher
566	(watcher *, callback)>, which expects a callback to be provided. This	1119	*, callback)>, which expects a callback to be provided. This callback is
567	callback gets invoked each time the event occurs (or, in the case of io	1120	invoked each time the event occurs (or, in the case of I/O watchers, each
568	watchers, each time the event loop detects that the file descriptor given	1121	time the event loop detects that the file descriptor given is readable
569	is readable and/or writable).	1122	and/or writable).
570		1123
571	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	1124	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
572	with arguments specific to this watcher type. There is also a macro	1125	macro to configure it, with arguments specific to the watcher type. There
573	to combine initialisation and setting in one call: C<< ev_<type>_init	1126	is also a macro to combine initialisation and setting in one call: C<<
574	(watcher *, callback, ...) >>.	1127	ev_TYPE_init (watcher *, callback, ...) >>.
575		1128
576	To make the watcher actually watch out for events, you have to start it	1129	To make the watcher actually watch out for events, you have to start it
577	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	1130	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
578	*) >>), and you can stop watching for events at any time by calling the	1131	*) >>), and you can stop watching for events at any time by calling the
579	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	1132	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
580		1133
581	As long as your watcher is active (has been started but not stopped) you	1134	As long as your watcher is active (has been started but not stopped) you
582	must not touch the values stored in it. Most specifically you must never	1135	must not touch the values stored in it. Most specifically you must never
583	reinitialise it or call its C<set> macro.	1136	reinitialise it or call its C<ev_TYPE_set> macro.
584		1137
585	Each and every callback receives the event loop pointer as first, the	1138	Each and every callback receives the event loop pointer as first, the
586	registered watcher structure as second, and a bitset of received events as	1139	registered watcher structure as second, and a bitset of received events as
587	third argument.	1140	third argument.
588		1141
…		…
597	=item C<EV_WRITE>	1150	=item C<EV_WRITE>
598		1151
599	The file descriptor in the C<ev_io> watcher has become readable and/or	1152	The file descriptor in the C<ev_io> watcher has become readable and/or
600	writable.	1153	writable.
601		1154
602	=item C<EV_TIMEOUT>	1155	=item C<EV_TIMER>
603		1156
604	The C<ev_timer> watcher has timed out.	1157	The C<ev_timer> watcher has timed out.
605		1158
606	=item C<EV_PERIODIC>	1159	=item C<EV_PERIODIC>
607		1160
…		…
625		1178
626	=item C<EV_PREPARE>	1179	=item C<EV_PREPARE>
627		1180
628	=item C<EV_CHECK>	1181	=item C<EV_CHECK>
629		1182
630	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts	1183	All C<ev_prepare> watchers are invoked just I<before> C<ev_run> starts to
631	to gather new events, and all C<ev_check> watchers are invoked just after	1184	gather new events, and all C<ev_check> watchers are queued (not invoked)
632	C<ev_loop> has gathered them, but before it invokes any callbacks for any	1185	just after C<ev_run> has gathered them, but before it queues any callbacks
		1186	for any received events. That means C<ev_prepare> watchers are the last
		1187	watchers invoked before the event loop sleeps or polls for new events, and
		1188	C<ev_check> watchers will be invoked before any other watchers of the same
		1189	or lower priority within an event loop iteration.
		1190
633	received events. Callbacks of both watcher types can start and stop as	1191	Callbacks of both watcher types can start and stop as many watchers as
634	many watchers as they want, and all of them will be taken into account	1192	they want, and all of them will be taken into account (for example, a
635	(for example, a C<ev_prepare> watcher might start an idle watcher to keep	1193	C<ev_prepare> watcher might start an idle watcher to keep C<ev_run> from
636	C<ev_loop> from blocking).	1194	blocking).
637		1195
638	=item C<EV_EMBED>	1196	=item C<EV_EMBED>
639		1197
640	The embedded event loop specified in the C<ev_embed> watcher needs attention.	1198	The embedded event loop specified in the C<ev_embed> watcher needs attention.
641		1199
642	=item C<EV_FORK>	1200	=item C<EV_FORK>
643		1201
644	The event loop has been resumed in the child process after fork (see	1202	The event loop has been resumed in the child process after fork (see
645	C<ev_fork>).	1203	C<ev_fork>).
646		1204
		1205	=item C<EV_CLEANUP>
		1206
		1207	The event loop is about to be destroyed (see C<ev_cleanup>).
		1208
		1209	=item C<EV_ASYNC>
		1210
		1211	The given async watcher has been asynchronously notified (see C<ev_async>).
		1212
		1213	=item C<EV_CUSTOM>
		1214
		1215	Not ever sent (or otherwise used) by libev itself, but can be freely used
		1216	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		1217
647	=item C<EV_ERROR>	1218	=item C<EV_ERROR>
648		1219
649	An unspecified error has occured, the watcher has been stopped. This might	1220	An unspecified error has occurred, the watcher has been stopped. This might
650	happen because the watcher could not be properly started because libev	1221	happen because the watcher could not be properly started because libev
651	ran out of memory, a file descriptor was found to be closed or any other	1222	ran out of memory, a file descriptor was found to be closed or any other
		1223	problem. Libev considers these application bugs.
		1224
652	problem. You best act on it by reporting the problem and somehow coping	1225	You best act on it by reporting the problem and somehow coping with the
653	with the watcher being stopped.	1226	watcher being stopped. Note that well-written programs should not receive
		1227	an error ever, so when your watcher receives it, this usually indicates a
		1228	bug in your program.
654		1229
655	Libev will usually signal a few "dummy" events together with an error,	1230	Libev will usually signal a few "dummy" events together with an error, for
656	for example it might indicate that a fd is readable or writable, and if	1231	example it might indicate that a fd is readable or writable, and if your
657	your callbacks is well-written it can just attempt the operation and cope	1232	callbacks is well-written it can just attempt the operation and cope with
658	with the error from read() or write(). This will not work in multithreaded	1233	the error from read() or write(). This will not work in multi-threaded
659	programs, though, so beware.	1234	programs, though, as the fd could already be closed and reused for another
		1235	thing, so beware.
660		1236
661	=back	1237	=back
662		1238
663	=head2 GENERIC WATCHER FUNCTIONS	1239	=head2 GENERIC WATCHER FUNCTIONS
664
665	In the following description, C<TYPE> stands for the watcher type,
666	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
667		1240
668	=over 4	1241	=over 4
669		1242
670	=item C<ev_init> (ev_TYPE *watcher, callback)	1243	=item C<ev_init> (ev_TYPE *watcher, callback)
671		1244
…		…
677	which rolls both calls into one.	1250	which rolls both calls into one.
678		1251
679	You can reinitialise a watcher at any time as long as it has been stopped	1252	You can reinitialise a watcher at any time as long as it has been stopped
680	(or never started) and there are no pending events outstanding.	1253	(or never started) and there are no pending events outstanding.
681		1254
682	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	1255	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
683	int revents)>.	1256	int revents)>.
684		1257
		1258	Example: Initialise an C<ev_io> watcher in two steps.
		1259
		1260	ev_io w;
		1261	ev_init (&w, my_cb);
		1262	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1263
685	=item C<ev_TYPE_set> (ev_TYPE *, [args])	1264	=item C<ev_TYPE_set> (ev_TYPE *watcher, [args])
686		1265
687	This macro initialises the type-specific parts of a watcher. You need to	1266	This macro initialises the type-specific parts of a watcher. You need to
688	call C<ev_init> at least once before you call this macro, but you can	1267	call C<ev_init> at least once before you call this macro, but you can
689	call C<ev_TYPE_set> any number of times. You must not, however, call this	1268	call C<ev_TYPE_set> any number of times. You must not, however, call this
690	macro on a watcher that is active (it can be pending, however, which is a	1269	macro on a watcher that is active (it can be pending, however, which is a
691	difference to the C<ev_init> macro).	1270	difference to the C<ev_init> macro).
692		1271
693	Although some watcher types do not have type-specific arguments	1272	Although some watcher types do not have type-specific arguments
694	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	1273	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
695		1274
		1275	See C<ev_init>, above, for an example.
		1276
696	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	1277	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
697		1278
698	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	1279	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
699	calls into a single call. This is the most convinient method to initialise	1280	calls into a single call. This is the most convenient method to initialise
700	a watcher. The same limitations apply, of course.	1281	a watcher. The same limitations apply, of course.
701		1282
		1283	Example: Initialise and set an C<ev_io> watcher in one step.
		1284
		1285	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1286
702	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	1287	=item C<ev_TYPE_start> (loop, ev_TYPE *watcher)
703		1288
704	Starts (activates) the given watcher. Only active watchers will receive	1289	Starts (activates) the given watcher. Only active watchers will receive
705	events. If the watcher is already active nothing will happen.	1290	events. If the watcher is already active nothing will happen.
706		1291
		1292	Example: Start the C<ev_io> watcher that is being abused as example in this
		1293	whole section.
		1294
		1295	ev_io_start (EV_DEFAULT_UC, &w);
		1296
707	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	1297	=item C<ev_TYPE_stop> (loop, ev_TYPE *watcher)
708		1298
709	Stops the given watcher again (if active) and clears the pending	1299	Stops the given watcher if active, and clears the pending status (whether
		1300	the watcher was active or not).
		1301
710	status. It is possible that stopped watchers are pending (for example,	1302	It is possible that stopped watchers are pending - for example,
711	non-repeating timers are being stopped when they become pending), but	1303	non-repeating timers are being stopped when they become pending - but
712	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	1304	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
713	you want to free or reuse the memory used by the watcher it is therefore a	1305	pending. If you want to free or reuse the memory used by the watcher it is
714	good idea to always call its C<ev_TYPE_stop> function.	1306	therefore a good idea to always call its C<ev_TYPE_stop> function.
715		1307
716	=item bool ev_is_active (ev_TYPE *watcher)	1308	=item bool ev_is_active (ev_TYPE *watcher)
717		1309
718	Returns a true value iff the watcher is active (i.e. it has been started	1310	Returns a true value iff the watcher is active (i.e. it has been started
719	and not yet been stopped). As long as a watcher is active you must not modify	1311	and not yet been stopped). As long as a watcher is active you must not modify
…		…
722	=item bool ev_is_pending (ev_TYPE *watcher)	1314	=item bool ev_is_pending (ev_TYPE *watcher)
723		1315
724	Returns a true value iff the watcher is pending, (i.e. it has outstanding	1316	Returns a true value iff the watcher is pending, (i.e. it has outstanding
725	events but its callback has not yet been invoked). As long as a watcher	1317	events but its callback has not yet been invoked). As long as a watcher
726	is pending (but not active) you must not call an init function on it (but	1318	is pending (but not active) you must not call an init function on it (but
727	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to	1319	C<ev_TYPE_set> is safe), you must not change its priority, and you must
728	libev (e.g. you cnanot C<free ()> it).	1320	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		1321	it).
729		1322
730	=item callback ev_cb (ev_TYPE *watcher)	1323	=item callback ev_cb (ev_TYPE *watcher)
731		1324
732	Returns the callback currently set on the watcher.	1325	Returns the callback currently set on the watcher.
733		1326
734	=item ev_cb_set (ev_TYPE *watcher, callback)	1327	=item ev_set_cb (ev_TYPE *watcher, callback)
735		1328
736	Change the callback. You can change the callback at virtually any time	1329	Change the callback. You can change the callback at virtually any time
737	(modulo threads).	1330	(modulo threads).
738		1331
		1332	=item ev_set_priority (ev_TYPE *watcher, int priority)
		1333
		1334	=item int ev_priority (ev_TYPE *watcher)
		1335
		1336	Set and query the priority of the watcher. The priority is a small
		1337	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		1338	(default: C<-2>). Pending watchers with higher priority will be invoked
		1339	before watchers with lower priority, but priority will not keep watchers
		1340	from being executed (except for C<ev_idle> watchers).
		1341
		1342	If you need to suppress invocation when higher priority events are pending
		1343	you need to look at C<ev_idle> watchers, which provide this functionality.
		1344
		1345	You I<must not> change the priority of a watcher as long as it is active or
		1346	pending.
		1347
		1348	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1349	fine, as long as you do not mind that the priority value you query might
		1350	or might not have been clamped to the valid range.
		1351
		1352	The default priority used by watchers when no priority has been set is
		1353	always C<0>, which is supposed to not be too high and not be too low :).
		1354
		1355	See L</WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
		1356	priorities.
		1357
		1358	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		1359
		1360	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		1361	C<loop> nor C<revents> need to be valid as long as the watcher callback
		1362	can deal with that fact, as both are simply passed through to the
		1363	callback.
		1364
		1365	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		1366
		1367	If the watcher is pending, this function clears its pending status and
		1368	returns its C<revents> bitset (as if its callback was invoked). If the
		1369	watcher isn't pending it does nothing and returns C<0>.
		1370
		1371	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1372	callback to be invoked, which can be accomplished with this function.
		1373
		1374	=item ev_feed_event (loop, ev_TYPE *watcher, int revents)
		1375
		1376	Feeds the given event set into the event loop, as if the specified event
		1377	had happened for the specified watcher (which must be a pointer to an
		1378	initialised but not necessarily started event watcher). Obviously you must
		1379	not free the watcher as long as it has pending events.
		1380
		1381	Stopping the watcher, letting libev invoke it, or calling
		1382	C<ev_clear_pending> will clear the pending event, even if the watcher was
		1383	not started in the first place.
		1384
		1385	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
		1386	functions that do not need a watcher.
		1387
739	=back	1388	=back
740		1389
		1390	See also the L</ASSOCIATING CUSTOM DATA WITH A WATCHER> and L</BUILDING YOUR
		1391	OWN COMPOSITE WATCHERS> idioms.
741		1392
742	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1393	=head2 WATCHER STATES
743		1394
744	Each watcher has, by default, a member C<void *data> that you can change	1395	There are various watcher states mentioned throughout this manual -
745	and read at any time, libev will completely ignore it. This can be used	1396	active, pending and so on. In this section these states and the rules to
746	to associate arbitrary data with your watcher. If you need more data and	1397	transition between them will be described in more detail - and while these
747	don't want to allocate memory and store a pointer to it in that data	1398	rules might look complicated, they usually do "the right thing".
748	member, you can also "subclass" the watcher type and provide your own
749	data:
750		1399
751	struct my_io	1400	=over 4
752	{
753	struct ev_io io;
754	int otherfd;
755	void *somedata;
756	struct whatever *mostinteresting;
757	}
758		1401
759	And since your callback will be called with a pointer to the watcher, you	1402	=item initialised
760	can cast it back to your own type:
761		1403
762	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1404	Before a watcher can be registered with the event loop it has to be
763	{	1405	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to
764	struct my_io *w = (struct my_io *)w_;	1406	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.
765	...
766	}
767		1407
768	More interesting and less C-conformant ways of casting your callback type	1408	In this state it is simply some block of memory that is suitable for
769	instead have been omitted.	1409	use in an event loop. It can be moved around, freed, reused etc. at
		1410	will - as long as you either keep the memory contents intact, or call
		1411	C<ev_TYPE_init> again.
770		1412
771	Another common scenario is having some data structure with multiple	1413	=item started/running/active
772	watchers:
773		1414
774	struct my_biggy	1415	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes
775	{	1416	property of the event loop, and is actively waiting for events. While in
776	int some_data;	1417	this state it cannot be accessed (except in a few documented ways), moved,
777	ev_timer t1;	1418	freed or anything else - the only legal thing is to keep a pointer to it,
778	ev_timer t2;	1419	and call libev functions on it that are documented to work on active watchers.
779	}
780		1420
781	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1421	=item pending
782	you need to use C<offsetof>:
783		1422
784	#include <stddef.h>	1423	If a watcher is active and libev determines that an event it is interested
		1424	in has occurred (such as a timer expiring), it will become pending. It will
		1425	stay in this pending state until either it is stopped or its callback is
		1426	about to be invoked, so it is not normally pending inside the watcher
		1427	callback.
785		1428
		1429	The watcher might or might not be active while it is pending (for example,
		1430	an expired non-repeating timer can be pending but no longer active). If it
		1431	is stopped, it can be freely accessed (e.g. by calling C<ev_TYPE_set>),
		1432	but it is still property of the event loop at this time, so cannot be
		1433	moved, freed or reused. And if it is active the rules described in the
		1434	previous item still apply.
		1435
		1436	It is also possible to feed an event on a watcher that is not active (e.g.
		1437	via C<ev_feed_event>), in which case it becomes pending without being
		1438	active.
		1439
		1440	=item stopped
		1441
		1442	A watcher can be stopped implicitly by libev (in which case it might still
		1443	be pending), or explicitly by calling its C<ev_TYPE_stop> function. The
		1444	latter will clear any pending state the watcher might be in, regardless
		1445	of whether it was active or not, so stopping a watcher explicitly before
		1446	freeing it is often a good idea.
		1447
		1448	While stopped (and not pending) the watcher is essentially in the
		1449	initialised state, that is, it can be reused, moved, modified in any way
		1450	you wish (but when you trash the memory block, you need to C<ev_TYPE_init>
		1451	it again).
		1452
		1453	=back
		1454
		1455	=head2 WATCHER PRIORITY MODELS
		1456
		1457	Many event loops support I<watcher priorities>, which are usually small
		1458	integers that influence the ordering of event callback invocation
		1459	between watchers in some way, all else being equal.
		1460
		1461	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1462	description for the more technical details such as the actual priority
		1463	range.
		1464
		1465	There are two common ways how these these priorities are being interpreted
		1466	by event loops:
		1467
		1468	In the more common lock-out model, higher priorities "lock out" invocation
		1469	of lower priority watchers, which means as long as higher priority
		1470	watchers receive events, lower priority watchers are not being invoked.
		1471
		1472	The less common only-for-ordering model uses priorities solely to order
		1473	callback invocation within a single event loop iteration: Higher priority
		1474	watchers are invoked before lower priority ones, but they all get invoked
		1475	before polling for new events.
		1476
		1477	Libev uses the second (only-for-ordering) model for all its watchers
		1478	except for idle watchers (which use the lock-out model).
		1479
		1480	The rationale behind this is that implementing the lock-out model for
		1481	watchers is not well supported by most kernel interfaces, and most event
		1482	libraries will just poll for the same events again and again as long as
		1483	their callbacks have not been executed, which is very inefficient in the
		1484	common case of one high-priority watcher locking out a mass of lower
		1485	priority ones.
		1486
		1487	Static (ordering) priorities are most useful when you have two or more
		1488	watchers handling the same resource: a typical usage example is having an
		1489	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1490	timeouts. Under load, data might be received while the program handles
		1491	other jobs, but since timers normally get invoked first, the timeout
		1492	handler will be executed before checking for data. In that case, giving
		1493	the timer a lower priority than the I/O watcher ensures that I/O will be
		1494	handled first even under adverse conditions (which is usually, but not
		1495	always, what you want).
		1496
		1497	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1498	will only be executed when no same or higher priority watchers have
		1499	received events, they can be used to implement the "lock-out" model when
		1500	required.
		1501
		1502	For example, to emulate how many other event libraries handle priorities,
		1503	you can associate an C<ev_idle> watcher to each such watcher, and in
		1504	the normal watcher callback, you just start the idle watcher. The real
		1505	processing is done in the idle watcher callback. This causes libev to
		1506	continuously poll and process kernel event data for the watcher, but when
		1507	the lock-out case is known to be rare (which in turn is rare :), this is
		1508	workable.
		1509
		1510	Usually, however, the lock-out model implemented that way will perform
		1511	miserably under the type of load it was designed to handle. In that case,
		1512	it might be preferable to stop the real watcher before starting the
		1513	idle watcher, so the kernel will not have to process the event in case
		1514	the actual processing will be delayed for considerable time.
		1515
		1516	Here is an example of an I/O watcher that should run at a strictly lower
		1517	priority than the default, and which should only process data when no
		1518	other events are pending:
		1519
		1520	ev_idle idle; // actual processing watcher
		1521	ev_io io; // actual event watcher
		1522
786	static void	1523	static void
787	t1_cb (EV_P_ struct ev_timer *w, int revents)	1524	io_cb (EV_P_ ev_io *w, int revents)
788	{	1525	{
789	struct my_biggy big = (struct my_biggy *	1526	// stop the I/O watcher, we received the event, but
790	(((char *)w) - offsetof (struct my_biggy, t1));	1527	// are not yet ready to handle it.
		1528	ev_io_stop (EV_A_ w);
		1529
		1530	// start the idle watcher to handle the actual event.
		1531	// it will not be executed as long as other watchers
		1532	// with the default priority are receiving events.
		1533	ev_idle_start (EV_A_ &idle);
791	}	1534	}
792		1535
793	static void	1536	static void
794	t2_cb (EV_P_ struct ev_timer *w, int revents)	1537	idle_cb (EV_P_ ev_idle *w, int revents)
795	{	1538	{
796	struct my_biggy big = (struct my_biggy *	1539	// actual processing
797	(((char *)w) - offsetof (struct my_biggy, t2));	1540	read (STDIN_FILENO, ...);
		1541
		1542	// have to start the I/O watcher again, as
		1543	// we have handled the event
		1544	ev_io_start (EV_P_ &io);
798	}	1545	}
		1546
		1547	// initialisation
		1548	ev_idle_init (&idle, idle_cb);
		1549	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1550	ev_io_start (EV_DEFAULT_ &io);
		1551
		1552	In the "real" world, it might also be beneficial to start a timer, so that
		1553	low-priority connections can not be locked out forever under load. This
		1554	enables your program to keep a lower latency for important connections
		1555	during short periods of high load, while not completely locking out less
		1556	important ones.
799		1557
800		1558
801	=head1 WATCHER TYPES	1559	=head1 WATCHER TYPES
802		1560
803	This section describes each watcher in detail, but will not repeat	1561	This section describes each watcher in detail, but will not repeat
…		…
827	In general you can register as many read and/or write event watchers per	1585	In general you can register as many read and/or write event watchers per
828	fd as you want (as long as you don't confuse yourself). Setting all file	1586	fd as you want (as long as you don't confuse yourself). Setting all file
829	descriptors to non-blocking mode is also usually a good idea (but not	1587	descriptors to non-blocking mode is also usually a good idea (but not
830	required if you know what you are doing).	1588	required if you know what you are doing).
831		1589
832	You have to be careful with dup'ed file descriptors, though. Some backends
833	(the linux epoll backend is a notable example) cannot handle dup'ed file
834	descriptors correctly if you register interest in two or more fds pointing
835	to the same underlying file/socket/etc. description (that is, they share
836	the same underlying "file open").
837
838	If you must do this, then force the use of a known-to-be-good backend
839	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
840	C<EVBACKEND_POLL>).
841
842	Another thing you have to watch out for is that it is quite easy to	1590	Another thing you have to watch out for is that it is quite easy to
843	receive "spurious" readyness notifications, that is your callback might	1591	receive "spurious" readiness notifications, that is, your callback might
844	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1592	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
845	because there is no data. Not only are some backends known to create a	1593	because there is no data. It is very easy to get into this situation even
846	lot of those (for example solaris ports), it is very easy to get into	1594	with a relatively standard program structure. Thus it is best to always
847	this situation even with a relatively standard program structure. Thus	1595	use non-blocking I/O: An extra C<read>(2) returning C<EAGAIN> is far
848	it is best to always use non-blocking I/O: An extra C<read>(2) returning
849	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1596	preferable to a program hanging until some data arrives.
850		1597
851	If you cannot run the fd in non-blocking mode (for example you should not	1598	If you cannot run the fd in non-blocking mode (for example you should
852	play around with an Xlib connection), then you have to seperately re-test	1599	not play around with an Xlib connection), then you have to separately
853	wether a file descriptor is really ready with a known-to-be good interface	1600	re-test whether a file descriptor is really ready with a known-to-be good
854	such as poll (fortunately in our Xlib example, Xlib already does this on	1601	interface such as poll (fortunately in the case of Xlib, it already does
855	its own, so its quite safe to use).	1602	this on its own, so its quite safe to use). Some people additionally
		1603	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1604	indefinitely.
		1605
		1606	But really, best use non-blocking mode.
		1607
		1608	=head3 The special problem of disappearing file descriptors
		1609
		1610	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1611	descriptor (either due to calling C<close> explicitly or any other means,
		1612	such as C<dup2>). The reason is that you register interest in some file
		1613	descriptor, but when it goes away, the operating system will silently drop
		1614	this interest. If another file descriptor with the same number then is
		1615	registered with libev, there is no efficient way to see that this is, in
		1616	fact, a different file descriptor.
		1617
		1618	To avoid having to explicitly tell libev about such cases, libev follows
		1619	the following policy: Each time C<ev_io_set> is being called, libev
		1620	will assume that this is potentially a new file descriptor, otherwise
		1621	it is assumed that the file descriptor stays the same. That means that
		1622	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1623	descriptor even if the file descriptor number itself did not change.
		1624
		1625	This is how one would do it normally anyway, the important point is that
		1626	the libev application should not optimise around libev but should leave
		1627	optimisations to libev.
		1628
		1629	=head3 The special problem of dup'ed file descriptors
		1630
		1631	Some backends (e.g. epoll), cannot register events for file descriptors,
		1632	but only events for the underlying file descriptions. That means when you
		1633	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1634	events for them, only one file descriptor might actually receive events.
		1635
		1636	There is no workaround possible except not registering events
		1637	for potentially C<dup ()>'ed file descriptors, or to resort to
		1638	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1639
		1640	=head3 The special problem of files
		1641
		1642	Many people try to use C<select> (or libev) on file descriptors
		1643	representing files, and expect it to become ready when their program
		1644	doesn't block on disk accesses (which can take a long time on their own).
		1645
		1646	However, this cannot ever work in the "expected" way - you get a readiness
		1647	notification as soon as the kernel knows whether and how much data is
		1648	there, and in the case of open files, that's always the case, so you
		1649	always get a readiness notification instantly, and your read (or possibly
		1650	write) will still block on the disk I/O.
		1651
		1652	Another way to view it is that in the case of sockets, pipes, character
		1653	devices and so on, there is another party (the sender) that delivers data
		1654	on its own, but in the case of files, there is no such thing: the disk
		1655	will not send data on its own, simply because it doesn't know what you
		1656	wish to read - you would first have to request some data.
		1657
		1658	Since files are typically not-so-well supported by advanced notification
		1659	mechanism, libev tries hard to emulate POSIX behaviour with respect
		1660	to files, even though you should not use it. The reason for this is
		1661	convenience: sometimes you want to watch STDIN or STDOUT, which is
		1662	usually a tty, often a pipe, but also sometimes files or special devices
		1663	(for example, C<epoll> on Linux works with F</dev/random> but not with
		1664	F</dev/urandom>), and even though the file might better be served with
		1665	asynchronous I/O instead of with non-blocking I/O, it is still useful when
		1666	it "just works" instead of freezing.
		1667
		1668	So avoid file descriptors pointing to files when you know it (e.g. use
		1669	libeio), but use them when it is convenient, e.g. for STDIN/STDOUT, or
		1670	when you rarely read from a file instead of from a socket, and want to
		1671	reuse the same code path.
		1672
		1673	=head3 The special problem of fork
		1674
		1675	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1676	useless behaviour. Libev fully supports fork, but needs to be told about
		1677	it in the child if you want to continue to use it in the child.
		1678
		1679	To support fork in your child processes, you have to call C<ev_loop_fork
		1680	()> after a fork in the child, enable C<EVFLAG_FORKCHECK>, or resort to
		1681	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1682
		1683	=head3 The special problem of SIGPIPE
		1684
		1685	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1686	when writing to a pipe whose other end has been closed, your program gets
		1687	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1688	this is sensible behaviour, for daemons, this is usually undesirable.
		1689
		1690	So when you encounter spurious, unexplained daemon exits, make sure you
		1691	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1692	somewhere, as that would have given you a big clue).
		1693
		1694	=head3 The special problem of accept()ing when you can't
		1695
		1696	Many implementations of the POSIX C<accept> function (for example,
		1697	found in post-2004 Linux) have the peculiar behaviour of not removing a
		1698	connection from the pending queue in all error cases.
		1699
		1700	For example, larger servers often run out of file descriptors (because
		1701	of resource limits), causing C<accept> to fail with C<ENFILE> but not
		1702	rejecting the connection, leading to libev signalling readiness on
		1703	the next iteration again (the connection still exists after all), and
		1704	typically causing the program to loop at 100% CPU usage.
		1705
		1706	Unfortunately, the set of errors that cause this issue differs between
		1707	operating systems, there is usually little the app can do to remedy the
		1708	situation, and no known thread-safe method of removing the connection to
		1709	cope with overload is known (to me).
		1710
		1711	One of the easiest ways to handle this situation is to just ignore it
		1712	- when the program encounters an overload, it will just loop until the
		1713	situation is over. While this is a form of busy waiting, no OS offers an
		1714	event-based way to handle this situation, so it's the best one can do.
		1715
		1716	A better way to handle the situation is to log any errors other than
		1717	C<EAGAIN> and C<EWOULDBLOCK>, making sure not to flood the log with such
		1718	messages, and continue as usual, which at least gives the user an idea of
		1719	what could be wrong ("raise the ulimit!"). For extra points one could stop
		1720	the C<ev_io> watcher on the listening fd "for a while", which reduces CPU
		1721	usage.
		1722
		1723	If your program is single-threaded, then you could also keep a dummy file
		1724	descriptor for overload situations (e.g. by opening F</dev/null>), and
		1725	when you run into C<ENFILE> or C<EMFILE>, close it, run C<accept>,
		1726	close that fd, and create a new dummy fd. This will gracefully refuse
		1727	clients under typical overload conditions.
		1728
		1729	The last way to handle it is to simply log the error and C<exit>, as
		1730	is often done with C<malloc> failures, but this results in an easy
		1731	opportunity for a DoS attack.
		1732
		1733	=head3 Watcher-Specific Functions
856		1734
857	=over 4	1735	=over 4
858		1736
859	=item ev_io_init (ev_io *, callback, int fd, int events)	1737	=item ev_io_init (ev_io *, callback, int fd, int events)
860		1738
861	=item ev_io_set (ev_io *, int fd, int events)	1739	=item ev_io_set (ev_io *, int fd, int events)
862		1740
863	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1741	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
864	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or	1742	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
865	C<EV_READ | EV_WRITE> to receive the given events.	1743	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
866		1744
867	=item int fd [read-only]	1745	=item int fd [read-only]
868		1746
869	The file descriptor being watched.	1747	The file descriptor being watched.
870		1748
871	=item int events [read-only]	1749	=item int events [read-only]
872		1750
873	The events being watched.	1751	The events being watched.
874		1752
875	=back	1753	=back
		1754
		1755	=head3 Examples
876		1756
877	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1757	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
878	readable, but only once. Since it is likely line-buffered, you could	1758	readable, but only once. Since it is likely line-buffered, you could
879	attempt to read a whole line in the callback.	1759	attempt to read a whole line in the callback.
880		1760
881	static void	1761	static void
882	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1762	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
883	{	1763	{
884	ev_io_stop (loop, w);	1764	ev_io_stop (loop, w);
885	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1765	.. read from stdin here (or from w->fd) and handle any I/O errors
886	}	1766	}
887		1767
888	...	1768	...
889	struct ev_loop *loop = ev_default_init (0);	1769	struct ev_loop *loop = ev_default_init (0);
890	struct ev_io stdin_readable;	1770	ev_io stdin_readable;
891	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1771	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
892	ev_io_start (loop, &stdin_readable);	1772	ev_io_start (loop, &stdin_readable);
893	ev_loop (loop, 0);	1773	ev_run (loop, 0);
894		1774
895		1775
896	=head2 C<ev_timer> - relative and optionally repeating timeouts	1776	=head2 C<ev_timer> - relative and optionally repeating timeouts
897		1777
898	Timer watchers are simple relative timers that generate an event after a	1778	Timer watchers are simple relative timers that generate an event after a
899	given time, and optionally repeating in regular intervals after that.	1779	given time, and optionally repeating in regular intervals after that.
900		1780
901	The timers are based on real time, that is, if you register an event that	1781	The timers are based on real time, that is, if you register an event that
902	times out after an hour and you reset your system clock to last years	1782	times out after an hour and you reset your system clock to January last
903	time, it will still time out after (roughly) and hour. "Roughly" because	1783	year, it will still time out after (roughly) one hour. "Roughly" because
904	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1784	detecting time jumps is hard, and some inaccuracies are unavoidable (the
905	monotonic clock option helps a lot here).	1785	monotonic clock option helps a lot here).
		1786
		1787	The callback is guaranteed to be invoked only I<after> its timeout has
		1788	passed (not I<at>, so on systems with very low-resolution clocks this
		1789	might introduce a small delay, see "the special problem of being too
		1790	early", below). If multiple timers become ready during the same loop
		1791	iteration then the ones with earlier time-out values are invoked before
		1792	ones of the same priority with later time-out values (but this is no
		1793	longer true when a callback calls C<ev_run> recursively).
		1794
		1795	=head3 Be smart about timeouts
		1796
		1797	Many real-world problems involve some kind of timeout, usually for error
		1798	recovery. A typical example is an HTTP request - if the other side hangs,
		1799	you want to raise some error after a while.
		1800
		1801	What follows are some ways to handle this problem, from obvious and
		1802	inefficient to smart and efficient.
		1803
		1804	In the following, a 60 second activity timeout is assumed - a timeout that
		1805	gets reset to 60 seconds each time there is activity (e.g. each time some
		1806	data or other life sign was received).
		1807
		1808	=over 4
		1809
		1810	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1811
		1812	This is the most obvious, but not the most simple way: In the beginning,
		1813	start the watcher:
		1814
		1815	ev_timer_init (timer, callback, 60., 0.);
		1816	ev_timer_start (loop, timer);
		1817
		1818	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1819	and start it again:
		1820
		1821	ev_timer_stop (loop, timer);
		1822	ev_timer_set (timer, 60., 0.);
		1823	ev_timer_start (loop, timer);
		1824
		1825	This is relatively simple to implement, but means that each time there is
		1826	some activity, libev will first have to remove the timer from its internal
		1827	data structure and then add it again. Libev tries to be fast, but it's
		1828	still not a constant-time operation.
		1829
		1830	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1831
		1832	This is the easiest way, and involves using C<ev_timer_again> instead of
		1833	C<ev_timer_start>.
		1834
		1835	To implement this, configure an C<ev_timer> with a C<repeat> value
		1836	of C<60> and then call C<ev_timer_again> at start and each time you
		1837	successfully read or write some data. If you go into an idle state where
		1838	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1839	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1840
		1841	That means you can ignore both the C<ev_timer_start> function and the
		1842	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1843	member and C<ev_timer_again>.
		1844
		1845	At start:
		1846
		1847	ev_init (timer, callback);
		1848	timer->repeat = 60.;
		1849	ev_timer_again (loop, timer);
		1850
		1851	Each time there is some activity:
		1852
		1853	ev_timer_again (loop, timer);
		1854
		1855	It is even possible to change the time-out on the fly, regardless of
		1856	whether the watcher is active or not:
		1857
		1858	timer->repeat = 30.;
		1859	ev_timer_again (loop, timer);
		1860
		1861	This is slightly more efficient then stopping/starting the timer each time
		1862	you want to modify its timeout value, as libev does not have to completely
		1863	remove and re-insert the timer from/into its internal data structure.
		1864
		1865	It is, however, even simpler than the "obvious" way to do it.
		1866
		1867	=item 3. Let the timer time out, but then re-arm it as required.
		1868
		1869	This method is more tricky, but usually most efficient: Most timeouts are
		1870	relatively long compared to the intervals between other activity - in
		1871	our example, within 60 seconds, there are usually many I/O events with
		1872	associated activity resets.
		1873
		1874	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1875	but remember the time of last activity, and check for a real timeout only
		1876	within the callback:
		1877
		1878	ev_tstamp timeout = 60.;
		1879	ev_tstamp last_activity; // time of last activity
		1880	ev_timer timer;
		1881
		1882	static void
		1883	callback (EV_P_ ev_timer *w, int revents)
		1884	{
		1885	// calculate when the timeout would happen
		1886	ev_tstamp after = last_activity - ev_now (EV_A) + timeout;
		1887
		1888	// if negative, it means we the timeout already occurred
		1889	if (after < 0.)
		1890	{
		1891	// timeout occurred, take action
		1892	}
		1893	else
		1894	{
		1895	// callback was invoked, but there was some recent
		1896	// activity. simply restart the timer to time out
		1897	// after "after" seconds, which is the earliest time
		1898	// the timeout can occur.
		1899	ev_timer_set (w, after, 0.);
		1900	ev_timer_start (EV_A_ w);
		1901	}
		1902	}
		1903
		1904	To summarise the callback: first calculate in how many seconds the
		1905	timeout will occur (by calculating the absolute time when it would occur,
		1906	C<last_activity + timeout>, and subtracting the current time, C<ev_now
		1907	(EV_A)> from that).
		1908
		1909	If this value is negative, then we are already past the timeout, i.e. we
		1910	timed out, and need to do whatever is needed in this case.
		1911
		1912	Otherwise, we now the earliest time at which the timeout would trigger,
		1913	and simply start the timer with this timeout value.
		1914
		1915	In other words, each time the callback is invoked it will check whether
		1916	the timeout occurred. If not, it will simply reschedule itself to check
		1917	again at the earliest time it could time out. Rinse. Repeat.
		1918
		1919	This scheme causes more callback invocations (about one every 60 seconds
		1920	minus half the average time between activity), but virtually no calls to
		1921	libev to change the timeout.
		1922
		1923	To start the machinery, simply initialise the watcher and set
		1924	C<last_activity> to the current time (meaning there was some activity just
		1925	now), then call the callback, which will "do the right thing" and start
		1926	the timer:
		1927
		1928	last_activity = ev_now (EV_A);
		1929	ev_init (&timer, callback);
		1930	callback (EV_A_ &timer, 0);
		1931
		1932	When there is some activity, simply store the current time in
		1933	C<last_activity>, no libev calls at all:
		1934
		1935	if (activity detected)
		1936	last_activity = ev_now (EV_A);
		1937
		1938	When your timeout value changes, then the timeout can be changed by simply
		1939	providing a new value, stopping the timer and calling the callback, which
		1940	will again do the right thing (for example, time out immediately :).
		1941
		1942	timeout = new_value;
		1943	ev_timer_stop (EV_A_ &timer);
		1944	callback (EV_A_ &timer, 0);
		1945
		1946	This technique is slightly more complex, but in most cases where the
		1947	time-out is unlikely to be triggered, much more efficient.
		1948
		1949	=item 4. Wee, just use a double-linked list for your timeouts.
		1950
		1951	If there is not one request, but many thousands (millions...), all
		1952	employing some kind of timeout with the same timeout value, then one can
		1953	do even better:
		1954
		1955	When starting the timeout, calculate the timeout value and put the timeout
		1956	at the I<end> of the list.
		1957
		1958	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1959	the list is expected to fire (for example, using the technique #3).
		1960
		1961	When there is some activity, remove the timer from the list, recalculate
		1962	the timeout, append it to the end of the list again, and make sure to
		1963	update the C<ev_timer> if it was taken from the beginning of the list.
		1964
		1965	This way, one can manage an unlimited number of timeouts in O(1) time for
		1966	starting, stopping and updating the timers, at the expense of a major
		1967	complication, and having to use a constant timeout. The constant timeout
		1968	ensures that the list stays sorted.
		1969
		1970	=back
		1971
		1972	So which method the best?
		1973
		1974	Method #2 is a simple no-brain-required solution that is adequate in most
		1975	situations. Method #3 requires a bit more thinking, but handles many cases
		1976	better, and isn't very complicated either. In most case, choosing either
		1977	one is fine, with #3 being better in typical situations.
		1978
		1979	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1980	rather complicated, but extremely efficient, something that really pays
		1981	off after the first million or so of active timers, i.e. it's usually
		1982	overkill :)
		1983
		1984	=head3 The special problem of being too early
		1985
		1986	If you ask a timer to call your callback after three seconds, then
		1987	you expect it to be invoked after three seconds - but of course, this
		1988	cannot be guaranteed to infinite precision. Less obviously, it cannot be
		1989	guaranteed to any precision by libev - imagine somebody suspending the
		1990	process with a STOP signal for a few hours for example.
		1991
		1992	So, libev tries to invoke your callback as soon as possible I<after> the
		1993	delay has occurred, but cannot guarantee this.
		1994
		1995	A less obvious failure mode is calling your callback too early: many event
		1996	loops compare timestamps with a "elapsed delay >= requested delay", but
		1997	this can cause your callback to be invoked much earlier than you would
		1998	expect.
		1999
		2000	To see why, imagine a system with a clock that only offers full second
		2001	resolution (think windows if you can't come up with a broken enough OS
		2002	yourself). If you schedule a one-second timer at the time 500.9, then the
		2003	event loop will schedule your timeout to elapse at a system time of 500
		2004	(500.9 truncated to the resolution) + 1, or 501.
		2005
		2006	If an event library looks at the timeout 0.1s later, it will see "501 >=
		2007	501" and invoke the callback 0.1s after it was started, even though a
		2008	one-second delay was requested - this is being "too early", despite best
		2009	intentions.
		2010
		2011	This is the reason why libev will never invoke the callback if the elapsed
		2012	delay equals the requested delay, but only when the elapsed delay is
		2013	larger than the requested delay. In the example above, libev would only invoke
		2014	the callback at system time 502, or 1.1s after the timer was started.
		2015
		2016	So, while libev cannot guarantee that your callback will be invoked
		2017	exactly when requested, it I<can> and I<does> guarantee that the requested
		2018	delay has actually elapsed, or in other words, it always errs on the "too
		2019	late" side of things.
		2020
		2021	=head3 The special problem of time updates
		2022
		2023	Establishing the current time is a costly operation (it usually takes
		2024	at least one system call): EV therefore updates its idea of the current
		2025	time only before and after C<ev_run> collects new events, which causes a
		2026	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		2027	lots of events in one iteration.
906		2028
907	The relative timeouts are calculated relative to the C<ev_now ()>	2029	The relative timeouts are calculated relative to the C<ev_now ()>
908	time. This is usually the right thing as this timestamp refers to the time	2030	time. This is usually the right thing as this timestamp refers to the time
909	of the event triggering whatever timeout you are modifying/starting. If	2031	of the event triggering whatever timeout you are modifying/starting. If
910	you suspect event processing to be delayed and you I<need> to base the timeout	2032	you suspect event processing to be delayed and you I<need> to base the
911	on the current time, use something like this to adjust for this:	2033	timeout on the current time, use something like this to adjust for this:
912		2034
913	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	2035	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
914		2036
915	The callback is guarenteed to be invoked only when its timeout has passed,	2037	If the event loop is suspended for a long time, you can also force an
916	but if multiple timers become ready during the same loop iteration then	2038	update of the time returned by C<ev_now ()> by calling C<ev_now_update
917	order of execution is undefined.	2039	()>.
		2040
		2041	=head3 The special problem of unsynchronised clocks
		2042
		2043	Modern systems have a variety of clocks - libev itself uses the normal
		2044	"wall clock" clock and, if available, the monotonic clock (to avoid time
		2045	jumps).
		2046
		2047	Neither of these clocks is synchronised with each other or any other clock
		2048	on the system, so C<ev_time ()> might return a considerably different time
		2049	than C<gettimeofday ()> or C<time ()>. On a GNU/Linux system, for example,
		2050	a call to C<gettimeofday> might return a second count that is one higher
		2051	than a directly following call to C<time>.
		2052
		2053	The moral of this is to only compare libev-related timestamps with
		2054	C<ev_time ()> and C<ev_now ()>, at least if you want better precision than
		2055	a second or so.
		2056
		2057	One more problem arises due to this lack of synchronisation: if libev uses
		2058	the system monotonic clock and you compare timestamps from C<ev_time>
		2059	or C<ev_now> from when you started your timer and when your callback is
		2060	invoked, you will find that sometimes the callback is a bit "early".
		2061
		2062	This is because C<ev_timer>s work in real time, not wall clock time, so
		2063	libev makes sure your callback is not invoked before the delay happened,
		2064	I<measured according to the real time>, not the system clock.
		2065
		2066	If your timeouts are based on a physical timescale (e.g. "time out this
		2067	connection after 100 seconds") then this shouldn't bother you as it is
		2068	exactly the right behaviour.
		2069
		2070	If you want to compare wall clock/system timestamps to your timers, then
		2071	you need to use C<ev_periodic>s, as these are based on the wall clock
		2072	time, where your comparisons will always generate correct results.
		2073
		2074	=head3 The special problems of suspended animation
		2075
		2076	When you leave the server world it is quite customary to hit machines that
		2077	can suspend/hibernate - what happens to the clocks during such a suspend?
		2078
		2079	Some quick tests made with a Linux 2.6.28 indicate that a suspend freezes
		2080	all processes, while the clocks (C<times>, C<CLOCK_MONOTONIC>) continue
		2081	to run until the system is suspended, but they will not advance while the
		2082	system is suspended. That means, on resume, it will be as if the program
		2083	was frozen for a few seconds, but the suspend time will not be counted
		2084	towards C<ev_timer> when a monotonic clock source is used. The real time
		2085	clock advanced as expected, but if it is used as sole clocksource, then a
		2086	long suspend would be detected as a time jump by libev, and timers would
		2087	be adjusted accordingly.
		2088
		2089	I would not be surprised to see different behaviour in different between
		2090	operating systems, OS versions or even different hardware.
		2091
		2092	The other form of suspend (job control, or sending a SIGSTOP) will see a
		2093	time jump in the monotonic clocks and the realtime clock. If the program
		2094	is suspended for a very long time, and monotonic clock sources are in use,
		2095	then you can expect C<ev_timer>s to expire as the full suspension time
		2096	will be counted towards the timers. When no monotonic clock source is in
		2097	use, then libev will again assume a timejump and adjust accordingly.
		2098
		2099	It might be beneficial for this latter case to call C<ev_suspend>
		2100	and C<ev_resume> in code that handles C<SIGTSTP>, to at least get
		2101	deterministic behaviour in this case (you can do nothing against
		2102	C<SIGSTOP>).
		2103
		2104	=head3 Watcher-Specific Functions and Data Members
918		2105
919	=over 4	2106	=over 4
920		2107
921	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	2108	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
922		2109
923	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	2110	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
924		2111
925	Configure the timer to trigger after C<after> seconds. If C<repeat> is	2112	Configure the timer to trigger after C<after> seconds. If C<repeat>
926	C<0.>, then it will automatically be stopped. If it is positive, then the	2113	is C<0.>, then it will automatically be stopped once the timeout is
927	timer will automatically be configured to trigger again C<repeat> seconds	2114	reached. If it is positive, then the timer will automatically be
928	later, again, and again, until stopped manually.	2115	configured to trigger again C<repeat> seconds later, again, and again,
		2116	until stopped manually.
929		2117
930	The timer itself will do a best-effort at avoiding drift, that is, if you	2118	The timer itself will do a best-effort at avoiding drift, that is, if
931	configure a timer to trigger every 10 seconds, then it will trigger at	2119	you configure a timer to trigger every 10 seconds, then it will normally
932	exactly 10 second intervals. If, however, your program cannot keep up with	2120	trigger at exactly 10 second intervals. If, however, your program cannot
933	the timer (because it takes longer than those 10 seconds to do stuff) the	2121	keep up with the timer (because it takes longer than those 10 seconds to
934	timer will not fire more than once per event loop iteration.	2122	do stuff) the timer will not fire more than once per event loop iteration.
935		2123
936	=item ev_timer_again (loop)	2124	=item ev_timer_again (loop, ev_timer *)
937		2125
938	This will act as if the timer timed out and restart it again if it is	2126	This will act as if the timer timed out, and restarts it again if it is
939	repeating. The exact semantics are:	2127	repeating. It basically works like calling C<ev_timer_stop>, updating the
		2128	timeout to the C<repeat> value and calling C<ev_timer_start>.
940		2129
		2130	The exact semantics are as in the following rules, all of which will be
		2131	applied to the watcher:
		2132
		2133	=over 4
		2134
941	If the timer is pending, its pending status is cleared.	2135	=item If the timer is pending, the pending status is always cleared.
942		2136
943	If the timer is started but nonrepeating, stop it (as if it timed out).	2137	=item If the timer is started but non-repeating, stop it (as if it timed
		2138	out, without invoking it).
944		2139
945	If the timer is repeating, either start it if necessary (with the	2140	=item If the timer is repeating, make the C<repeat> value the new timeout
946	C<repeat> value), or reset the running timer to the C<repeat> value.	2141	and start the timer, if necessary.
947		2142
948	This sounds a bit complicated, but here is a useful and typical	2143	=back
949	example: Imagine you have a tcp connection and you want a so-called idle
950	timeout, that is, you want to be called when there have been, say, 60
951	seconds of inactivity on the socket. The easiest way to do this is to
952	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
953	C<ev_timer_again> each time you successfully read or write some data. If
954	you go into an idle state where you do not expect data to travel on the
955	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
956	automatically restart it if need be.
957		2144
958	That means you can ignore the C<after> value and C<ev_timer_start>	2145	This sounds a bit complicated, see L</Be smart about timeouts>, above, for a
959	altogether and only ever use the C<repeat> value and C<ev_timer_again>:	2146	usage example.
960		2147
961	ev_timer_init (timer, callback, 0., 5.);	2148	=item ev_tstamp ev_timer_remaining (loop, ev_timer *)
962	ev_timer_again (loop, timer);
963	...
964	timer->again = 17.;
965	ev_timer_again (loop, timer);
966	...
967	timer->again = 10.;
968	ev_timer_again (loop, timer);
969		2149
970	This is more slightly efficient then stopping/starting the timer each time	2150	Returns the remaining time until a timer fires. If the timer is active,
971	you want to modify its timeout value.	2151	then this time is relative to the current event loop time, otherwise it's
		2152	the timeout value currently configured.
		2153
		2154	That is, after an C<ev_timer_set (w, 5, 7)>, C<ev_timer_remaining> returns
		2155	C<5>. When the timer is started and one second passes, C<ev_timer_remaining>
		2156	will return C<4>. When the timer expires and is restarted, it will return
		2157	roughly C<7> (likely slightly less as callback invocation takes some time,
		2158	too), and so on.
972		2159
973	=item ev_tstamp repeat [read-write]	2160	=item ev_tstamp repeat [read-write]
974		2161
975	The current C<repeat> value. Will be used each time the watcher times out	2162	The current C<repeat> value. Will be used each time the watcher times out
976	or C<ev_timer_again> is called and determines the next timeout (if any),	2163	or C<ev_timer_again> is called, and determines the next timeout (if any),
977	which is also when any modifications are taken into account.	2164	which is also when any modifications are taken into account.
978		2165
979	=back	2166	=back
980		2167
		2168	=head3 Examples
		2169
981	Example: Create a timer that fires after 60 seconds.	2170	Example: Create a timer that fires after 60 seconds.
982		2171
983	static void	2172	static void
984	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2173	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
985	{	2174	{
986	.. one minute over, w is actually stopped right here	2175	.. one minute over, w is actually stopped right here
987	}	2176	}
988		2177
989	struct ev_timer mytimer;	2178	ev_timer mytimer;
990	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	2179	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
991	ev_timer_start (loop, &mytimer);	2180	ev_timer_start (loop, &mytimer);
992		2181
993	Example: Create a timeout timer that times out after 10 seconds of	2182	Example: Create a timeout timer that times out after 10 seconds of
994	inactivity.	2183	inactivity.
995		2184
996	static void	2185	static void
997	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	2186	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
998	{	2187	{
999	.. ten seconds without any activity	2188	.. ten seconds without any activity
1000	}	2189	}
1001		2190
1002	struct ev_timer mytimer;	2191	ev_timer mytimer;
1003	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	2192	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1004	ev_timer_again (&mytimer); /* start timer */	2193	ev_timer_again (&mytimer); /* start timer */
1005	ev_loop (loop, 0);	2194	ev_run (loop, 0);
1006		2195
1007	// and in some piece of code that gets executed on any "activity":	2196	// and in some piece of code that gets executed on any "activity":
1008	// reset the timeout to start ticking again at 10 seconds	2197	// reset the timeout to start ticking again at 10 seconds
1009	ev_timer_again (&mytimer);	2198	ev_timer_again (&mytimer);
1010		2199
1011		2200
1012	=head2 C<ev_periodic> - to cron or not to cron?	2201	=head2 C<ev_periodic> - to cron or not to cron?
1013		2202
1014	Periodic watchers are also timers of a kind, but they are very versatile	2203	Periodic watchers are also timers of a kind, but they are very versatile
1015	(and unfortunately a bit complex).	2204	(and unfortunately a bit complex).
1016		2205
1017	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	2206	Unlike C<ev_timer>, periodic watchers are not based on real time (or
1018	but on wallclock time (absolute time). You can tell a periodic watcher	2207	relative time, the physical time that passes) but on wall clock time
1019	to trigger "at" some specific point in time. For example, if you tell a	2208	(absolute time, the thing you can read on your calender or clock). The
1020	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	2209	difference is that wall clock time can run faster or slower than real
1021	+ 10.>) and then reset your system clock to the last year, then it will	2210	time, and time jumps are not uncommon (e.g. when you adjust your
1022	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	2211	wrist-watch).
1023	roughly 10 seconds later and of course not if you reset your system time
1024	again).
1025		2212
1026	They can also be used to implement vastly more complex timers, such as	2213	You can tell a periodic watcher to trigger after some specific point
		2214	in time: for example, if you tell a periodic watcher to trigger "in 10
		2215	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		2216	not a delay) and then reset your system clock to January of the previous
		2217	year, then it will take a year or more to trigger the event (unlike an
		2218	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		2219	it, as it uses a relative timeout).
		2220
		2221	C<ev_periodic> watchers can also be used to implement vastly more complex
1027	triggering an event on eahc midnight, local time.	2222	timers, such as triggering an event on each "midnight, local time", or
		2223	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		2224	those cannot react to time jumps.
1028		2225
1029	As with timers, the callback is guarenteed to be invoked only when the	2226	As with timers, the callback is guaranteed to be invoked only when the
1030	time (C<at>) has been passed, but if multiple periodic timers become ready	2227	point in time where it is supposed to trigger has passed. If multiple
1031	during the same loop iteration then order of execution is undefined.	2228	timers become ready during the same loop iteration then the ones with
		2229	earlier time-out values are invoked before ones with later time-out values
		2230	(but this is no longer true when a callback calls C<ev_run> recursively).
		2231
		2232	=head3 Watcher-Specific Functions and Data Members
1032		2233
1033	=over 4	2234	=over 4
1034		2235
1035	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	2236	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1036		2237
1037	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	2238	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
1038		2239
1039	Lots of arguments, lets sort it out... There are basically three modes of	2240	Lots of arguments, let's sort it out... There are basically three modes of
1040	operation, and we will explain them from simplest to complex:	2241	operation, and we will explain them from simplest to most complex:
1041		2242
1042	=over 4	2243	=over 4
1043		2244
1044	=item * absolute timer (interval = reschedule_cb = 0)	2245	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
1045		2246
1046	In this configuration the watcher triggers an event at the wallclock time	2247	In this configuration the watcher triggers an event after the wall clock
1047	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	2248	time C<offset> has passed. It will not repeat and will not adjust when a
1048	that is, if it is to be run at January 1st 2011 then it will run when the	2249	time jump occurs, that is, if it is to be run at January 1st 2011 then it
1049	system time reaches or surpasses this time.	2250	will be stopped and invoked when the system clock reaches or surpasses
		2251	this point in time.
1050		2252
1051	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	2253	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
1052		2254
1053	In this mode the watcher will always be scheduled to time out at the next	2255	In this mode the watcher will always be scheduled to time out at the next
1054	C<at + N * interval> time (for some integer N) and then repeat, regardless	2256	C<offset + N * interval> time (for some integer N, which can also be
1055	of any time jumps.	2257	negative) and then repeat, regardless of any time jumps. The C<offset>
		2258	argument is merely an offset into the C<interval> periods.
1056		2259
1057	This can be used to create timers that do not drift with respect to system	2260	This can be used to create timers that do not drift with respect to the
1058	time:	2261	system clock, for example, here is an C<ev_periodic> that triggers each
		2262	hour, on the hour (with respect to UTC):
1059		2263
1060	ev_periodic_set (&periodic, 0., 3600., 0);	2264	ev_periodic_set (&periodic, 0., 3600., 0);
1061		2265
1062	This doesn't mean there will always be 3600 seconds in between triggers,	2266	This doesn't mean there will always be 3600 seconds in between triggers,
1063	but only that the the callback will be called when the system time shows a	2267	but only that the callback will be called when the system time shows a
1064	full hour (UTC), or more correctly, when the system time is evenly divisible	2268	full hour (UTC), or more correctly, when the system time is evenly divisible
1065	by 3600.	2269	by 3600.
1066		2270
1067	Another way to think about it (for the mathematically inclined) is that	2271	Another way to think about it (for the mathematically inclined) is that
1068	C<ev_periodic> will try to run the callback in this mode at the next possible	2272	C<ev_periodic> will try to run the callback in this mode at the next possible
1069	time where C<time = at (mod interval)>, regardless of any time jumps.	2273	time where C<time = offset (mod interval)>, regardless of any time jumps.
1070		2274
		2275	The C<interval> I<MUST> be positive, and for numerical stability, the
		2276	interval value should be higher than C<1/8192> (which is around 100
		2277	microseconds) and C<offset> should be higher than C<0> and should have
		2278	at most a similar magnitude as the current time (say, within a factor of
		2279	ten). Typical values for offset are, in fact, C<0> or something between
		2280	C<0> and C<interval>, which is also the recommended range.
		2281
		2282	Note also that there is an upper limit to how often a timer can fire (CPU
		2283	speed for example), so if C<interval> is very small then timing stability
		2284	will of course deteriorate. Libev itself tries to be exact to be about one
		2285	millisecond (if the OS supports it and the machine is fast enough).
		2286
1071	=item * manual reschedule mode (reschedule_cb = callback)	2287	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
1072		2288
1073	In this mode the values for C<interval> and C<at> are both being	2289	In this mode the values for C<interval> and C<offset> are both being
1074	ignored. Instead, each time the periodic watcher gets scheduled, the	2290	ignored. Instead, each time the periodic watcher gets scheduled, the
1075	reschedule callback will be called with the watcher as first, and the	2291	reschedule callback will be called with the watcher as first, and the
1076	current time as second argument.	2292	current time as second argument.
1077		2293
1078	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	2294	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
1079	ever, or make any event loop modifications>. If you need to stop it,	2295	or make ANY other event loop modifications whatsoever, unless explicitly
1080	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by	2296	allowed by documentation here>.
1081	starting a prepare watcher).
1082		2297
		2298	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		2299	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		2300	only event loop modification you are allowed to do).
		2301
1083	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	2302	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
1084	ev_tstamp now)>, e.g.:	2303	*w, ev_tstamp now)>, e.g.:
1085		2304
		2305	static ev_tstamp
1086	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	2306	my_rescheduler (ev_periodic *w, ev_tstamp now)
1087	{	2307	{
1088	return now + 60.;	2308	return now + 60.;
1089	}	2309	}
1090		2310
1091	It must return the next time to trigger, based on the passed time value	2311	It must return the next time to trigger, based on the passed time value
1092	(that is, the lowest time value larger than to the second argument). It	2312	(that is, the lowest time value larger than to the second argument). It
1093	will usually be called just before the callback will be triggered, but	2313	will usually be called just before the callback will be triggered, but
1094	might be called at other times, too.	2314	might be called at other times, too.
1095		2315
1096	NOTE: I<< This callback must always return a time that is later than the	2316	NOTE: I<< This callback must always return a time that is higher than or
1097	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	2317	equal to the passed C<now> value >>.
1098		2318
1099	This can be used to create very complex timers, such as a timer that	2319	This can be used to create very complex timers, such as a timer that
1100	triggers on each midnight, local time. To do this, you would calculate the	2320	triggers on "next midnight, local time". To do this, you would calculate the
1101	next midnight after C<now> and return the timestamp value for this. How	2321	next midnight after C<now> and return the timestamp value for this. How
1102	you do this is, again, up to you (but it is not trivial, which is the main	2322	you do this is, again, up to you (but it is not trivial, which is the main
1103	reason I omitted it as an example).	2323	reason I omitted it as an example).
1104		2324
1105	=back	2325	=back
…		…
1109	Simply stops and restarts the periodic watcher again. This is only useful	2329	Simply stops and restarts the periodic watcher again. This is only useful
1110	when you changed some parameters or the reschedule callback would return	2330	when you changed some parameters or the reschedule callback would return
1111	a different time than the last time it was called (e.g. in a crond like	2331	a different time than the last time it was called (e.g. in a crond like
1112	program when the crontabs have changed).	2332	program when the crontabs have changed).
1113		2333
		2334	=item ev_tstamp ev_periodic_at (ev_periodic *)
		2335
		2336	When active, returns the absolute time that the watcher is supposed
		2337	to trigger next. This is not the same as the C<offset> argument to
		2338	C<ev_periodic_set>, but indeed works even in interval and manual
		2339	rescheduling modes.
		2340
		2341	=item ev_tstamp offset [read-write]
		2342
		2343	When repeating, this contains the offset value, otherwise this is the
		2344	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		2345	although libev might modify this value for better numerical stability).
		2346
		2347	Can be modified any time, but changes only take effect when the periodic
		2348	timer fires or C<ev_periodic_again> is being called.
		2349
1114	=item ev_tstamp interval [read-write]	2350	=item ev_tstamp interval [read-write]
1115		2351
1116	The current interval value. Can be modified any time, but changes only	2352	The current interval value. Can be modified any time, but changes only
1117	take effect when the periodic timer fires or C<ev_periodic_again> is being	2353	take effect when the periodic timer fires or C<ev_periodic_again> is being
1118	called.	2354	called.
1119		2355
1120	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]	2356	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
1121		2357
1122	The current reschedule callback, or C<0>, if this functionality is	2358	The current reschedule callback, or C<0>, if this functionality is
1123	switched off. Can be changed any time, but changes only take effect when	2359	switched off. Can be changed any time, but changes only take effect when
1124	the periodic timer fires or C<ev_periodic_again> is being called.	2360	the periodic timer fires or C<ev_periodic_again> is being called.
1125		2361
1126	=back	2362	=back
1127		2363
		2364	=head3 Examples
		2365
1128	Example: Call a callback every hour, or, more precisely, whenever the	2366	Example: Call a callback every hour, or, more precisely, whenever the
1129	system clock is divisible by 3600. The callback invocation times have	2367	system time is divisible by 3600. The callback invocation times have
1130	potentially a lot of jittering, but good long-term stability.	2368	potentially a lot of jitter, but good long-term stability.
1131		2369
1132	static void	2370	static void
1133	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	2371	clock_cb (struct ev_loop *loop, ev_periodic *w, int revents)
1134	{	2372	{
1135	... its now a full hour (UTC, or TAI or whatever your clock follows)	2373	... its now a full hour (UTC, or TAI or whatever your clock follows)
1136	}	2374	}
1137		2375
1138	struct ev_periodic hourly_tick;	2376	ev_periodic hourly_tick;
1139	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	2377	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1140	ev_periodic_start (loop, &hourly_tick);	2378	ev_periodic_start (loop, &hourly_tick);
1141		2379
1142	Example: The same as above, but use a reschedule callback to do it:	2380	Example: The same as above, but use a reschedule callback to do it:
1143		2381
1144	#include <math.h>	2382	#include <math.h>
1145		2383
1146	static ev_tstamp	2384	static ev_tstamp
1147	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	2385	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
1148	{	2386	{
1149	return fmod (now, 3600.) + 3600.;	2387	return now + (3600. - fmod (now, 3600.));
1150	}	2388	}
1151		2389
1152	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	2390	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1153		2391
1154	Example: Call a callback every hour, starting now:	2392	Example: Call a callback every hour, starting now:
1155		2393
1156	struct ev_periodic hourly_tick;	2394	ev_periodic hourly_tick;
1157	ev_periodic_init (&hourly_tick, clock_cb,	2395	ev_periodic_init (&hourly_tick, clock_cb,
1158	fmod (ev_now (loop), 3600.), 3600., 0);	2396	fmod (ev_now (loop), 3600.), 3600., 0);
1159	ev_periodic_start (loop, &hourly_tick);	2397	ev_periodic_start (loop, &hourly_tick);
1160		2398
1161		2399
1162	=head2 C<ev_signal> - signal me when a signal gets signalled!	2400	=head2 C<ev_signal> - signal me when a signal gets signalled!
1163		2401
1164	Signal watchers will trigger an event when the process receives a specific	2402	Signal watchers will trigger an event when the process receives a specific
1165	signal one or more times. Even though signals are very asynchronous, libev	2403	signal one or more times. Even though signals are very asynchronous, libev
1166	will try it's best to deliver signals synchronously, i.e. as part of the	2404	will try its best to deliver signals synchronously, i.e. as part of the
1167	normal event processing, like any other event.	2405	normal event processing, like any other event.
1168		2406
		2407	If you want signals to be delivered truly asynchronously, just use
		2408	C<sigaction> as you would do without libev and forget about sharing
		2409	the signal. You can even use C<ev_async> from a signal handler to
		2410	synchronously wake up an event loop.
		2411
1169	You can configure as many watchers as you like per signal. Only when the	2412	You can configure as many watchers as you like for the same signal, but
1170	first watcher gets started will libev actually register a signal watcher	2413	only within the same loop, i.e. you can watch for C<SIGINT> in your
1171	with the kernel (thus it coexists with your own signal handlers as long	2414	default loop and for C<SIGIO> in another loop, but you cannot watch for
1172	as you don't register any with libev). Similarly, when the last signal	2415	C<SIGINT> in both the default loop and another loop at the same time. At
1173	watcher for a signal is stopped libev will reset the signal handler to	2416	the moment, C<SIGCHLD> is permanently tied to the default loop.
1174	SIG_DFL (regardless of what it was set to before).	2417
		2418	Only after the first watcher for a signal is started will libev actually
		2419	register something with the kernel. It thus coexists with your own signal
		2420	handlers as long as you don't register any with libev for the same signal.
		2421
		2422	If possible and supported, libev will install its handlers with
		2423	C<SA_RESTART> (or equivalent) behaviour enabled, so system calls should
		2424	not be unduly interrupted. If you have a problem with system calls getting
		2425	interrupted by signals you can block all signals in an C<ev_check> watcher
		2426	and unblock them in an C<ev_prepare> watcher.
		2427
		2428	=head3 The special problem of inheritance over fork/execve/pthread_create
		2429
		2430	Both the signal mask (C<sigprocmask>) and the signal disposition
		2431	(C<sigaction>) are unspecified after starting a signal watcher (and after
		2432	stopping it again), that is, libev might or might not block the signal,
		2433	and might or might not set or restore the installed signal handler (but
		2434	see C<EVFLAG_NOSIGMASK>).
		2435
		2436	While this does not matter for the signal disposition (libev never
		2437	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
		2438	C<execve>), this matters for the signal mask: many programs do not expect
		2439	certain signals to be blocked.
		2440
		2441	This means that before calling C<exec> (from the child) you should reset
		2442	the signal mask to whatever "default" you expect (all clear is a good
		2443	choice usually).
		2444
		2445	The simplest way to ensure that the signal mask is reset in the child is
		2446	to install a fork handler with C<pthread_atfork> that resets it. That will
		2447	catch fork calls done by libraries (such as the libc) as well.
		2448
		2449	In current versions of libev, the signal will not be blocked indefinitely
		2450	unless you use the C<signalfd> API (C<EV_SIGNALFD>). While this reduces
		2451	the window of opportunity for problems, it will not go away, as libev
		2452	I<has> to modify the signal mask, at least temporarily.
		2453
		2454	So I can't stress this enough: I<If you do not reset your signal mask when
		2455	you expect it to be empty, you have a race condition in your code>. This
		2456	is not a libev-specific thing, this is true for most event libraries.
		2457
		2458	=head3 The special problem of threads signal handling
		2459
		2460	POSIX threads has problematic signal handling semantics, specifically,
		2461	a lot of functionality (sigfd, sigwait etc.) only really works if all
		2462	threads in a process block signals, which is hard to achieve.
		2463
		2464	When you want to use sigwait (or mix libev signal handling with your own
		2465	for the same signals), you can tackle this problem by globally blocking
		2466	all signals before creating any threads (or creating them with a fully set
		2467	sigprocmask) and also specifying the C<EVFLAG_NOSIGMASK> when creating
		2468	loops. Then designate one thread as "signal receiver thread" which handles
		2469	these signals. You can pass on any signals that libev might be interested
		2470	in by calling C<ev_feed_signal>.
		2471
		2472	=head3 Watcher-Specific Functions and Data Members
1175		2473
1176	=over 4	2474	=over 4
1177		2475
1178	=item ev_signal_init (ev_signal *, callback, int signum)	2476	=item ev_signal_init (ev_signal *, callback, int signum)
1179		2477
…		…
1186		2484
1187	The signal the watcher watches out for.	2485	The signal the watcher watches out for.
1188		2486
1189	=back	2487	=back
1190		2488
		2489	=head3 Examples
		2490
		2491	Example: Try to exit cleanly on SIGINT.
		2492
		2493	static void
		2494	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
		2495	{
		2496	ev_break (loop, EVBREAK_ALL);
		2497	}
		2498
		2499	ev_signal signal_watcher;
		2500	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		2501	ev_signal_start (loop, &signal_watcher);
		2502
1191		2503
1192	=head2 C<ev_child> - watch out for process status changes	2504	=head2 C<ev_child> - watch out for process status changes
1193		2505
1194	Child watchers trigger when your process receives a SIGCHLD in response to	2506	Child watchers trigger when your process receives a SIGCHLD in response to
1195	some child status changes (most typically when a child of yours dies).	2507	some child status changes (most typically when a child of yours dies or
		2508	exits). It is permissible to install a child watcher I<after> the child
		2509	has been forked (which implies it might have already exited), as long
		2510	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2511	forking and then immediately registering a watcher for the child is fine,
		2512	but forking and registering a watcher a few event loop iterations later or
		2513	in the next callback invocation is not.
		2514
		2515	Only the default event loop is capable of handling signals, and therefore
		2516	you can only register child watchers in the default event loop.
		2517
		2518	Due to some design glitches inside libev, child watchers will always be
		2519	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2520	libev)
		2521
		2522	=head3 Process Interaction
		2523
		2524	Libev grabs C<SIGCHLD> as soon as the default event loop is
		2525	initialised. This is necessary to guarantee proper behaviour even if the
		2526	first child watcher is started after the child exits. The occurrence
		2527	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		2528	synchronously as part of the event loop processing. Libev always reaps all
		2529	children, even ones not watched.
		2530
		2531	=head3 Overriding the Built-In Processing
		2532
		2533	Libev offers no special support for overriding the built-in child
		2534	processing, but if your application collides with libev's default child
		2535	handler, you can override it easily by installing your own handler for
		2536	C<SIGCHLD> after initialising the default loop, and making sure the
		2537	default loop never gets destroyed. You are encouraged, however, to use an
		2538	event-based approach to child reaping and thus use libev's support for
		2539	that, so other libev users can use C<ev_child> watchers freely.
		2540
		2541	=head3 Stopping the Child Watcher
		2542
		2543	Currently, the child watcher never gets stopped, even when the
		2544	child terminates, so normally one needs to stop the watcher in the
		2545	callback. Future versions of libev might stop the watcher automatically
		2546	when a child exit is detected (calling C<ev_child_stop> twice is not a
		2547	problem).
		2548
		2549	=head3 Watcher-Specific Functions and Data Members
1196		2550
1197	=over 4	2551	=over 4
1198		2552
1199	=item ev_child_init (ev_child *, callback, int pid)	2553	=item ev_child_init (ev_child *, callback, int pid, int trace)
1200		2554
1201	=item ev_child_set (ev_child *, int pid)	2555	=item ev_child_set (ev_child *, int pid, int trace)
1202		2556
1203	Configures the watcher to wait for status changes of process C<pid> (or	2557	Configures the watcher to wait for status changes of process C<pid> (or
1204	I<any> process if C<pid> is specified as C<0>). The callback can look	2558	I<any> process if C<pid> is specified as C<0>). The callback can look
1205	at the C<rstatus> member of the C<ev_child> watcher structure to see	2559	at the C<rstatus> member of the C<ev_child> watcher structure to see
1206	the status word (use the macros from C<sys/wait.h> and see your systems	2560	the status word (use the macros from C<sys/wait.h> and see your systems
1207	C<waitpid> documentation). The C<rpid> member contains the pid of the	2561	C<waitpid> documentation). The C<rpid> member contains the pid of the
1208	process causing the status change.	2562	process causing the status change. C<trace> must be either C<0> (only
		2563	activate the watcher when the process terminates) or C<1> (additionally
		2564	activate the watcher when the process is stopped or continued).
1209		2565
1210	=item int pid [read-only]	2566	=item int pid [read-only]
1211		2567
1212	The process id this watcher watches out for, or C<0>, meaning any process id.	2568	The process id this watcher watches out for, or C<0>, meaning any process id.
1213		2569
…		…
1220	The process exit/trace status caused by C<rpid> (see your systems	2576	The process exit/trace status caused by C<rpid> (see your systems
1221	C<waitpid> and C<sys/wait.h> documentation for details).	2577	C<waitpid> and C<sys/wait.h> documentation for details).
1222		2578
1223	=back	2579	=back
1224		2580
1225	Example: Try to exit cleanly on SIGINT and SIGTERM.	2581	=head3 Examples
1226		2582
		2583	Example: C<fork()> a new process and install a child handler to wait for
		2584	its completion.
		2585
		2586	ev_child cw;
		2587
1227	static void	2588	static void
1228	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	2589	child_cb (EV_P_ ev_child *w, int revents)
1229	{	2590	{
1230	ev_unloop (loop, EVUNLOOP_ALL);	2591	ev_child_stop (EV_A_ w);
		2592	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1231	}	2593	}
1232		2594
1233	struct ev_signal signal_watcher;	2595	pid_t pid = fork ();
1234	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	2596
1235	ev_signal_start (loop, &sigint_cb);	2597	if (pid < 0)
		2598	// error
		2599	else if (pid == 0)
		2600	{
		2601	// the forked child executes here
		2602	exit (1);
		2603	}
		2604	else
		2605	{
		2606	ev_child_init (&cw, child_cb, pid, 0);
		2607	ev_child_start (EV_DEFAULT_ &cw);
		2608	}
1236		2609
1237		2610
1238	=head2 C<ev_stat> - did the file attributes just change?	2611	=head2 C<ev_stat> - did the file attributes just change?
1239		2612
1240	This watches a filesystem path for attribute changes. That is, it calls	2613	This watches a file system path for attribute changes. That is, it calls
1241	C<stat> regularly (or when the OS says it changed) and sees if it changed	2614	C<stat> on that path in regular intervals (or when the OS says it changed)
1242	compared to the last time, invoking the callback if it did.	2615	and sees if it changed compared to the last time, invoking the callback
		2616	if it did. Starting the watcher C<stat>'s the file, so only changes that
		2617	happen after the watcher has been started will be reported.
1243		2618
1244	The path does not need to exist: changing from "path exists" to "path does	2619	The path does not need to exist: changing from "path exists" to "path does
1245	not exist" is a status change like any other. The condition "path does	2620	not exist" is a status change like any other. The condition "path does not
1246	not exist" is signified by the C<st_nlink> field being zero (which is	2621	exist" (or more correctly "path cannot be stat'ed") is signified by the
1247	otherwise always forced to be at least one) and all the other fields of	2622	C<st_nlink> field being zero (which is otherwise always forced to be at
1248	the stat buffer having unspecified contents.	2623	least one) and all the other fields of the stat buffer having unspecified
		2624	contents.
1249		2625
1250	The path I<should> be absolute and I<must not> end in a slash. If it is	2626	The path I<must not> end in a slash or contain special components such as
		2627	C<.> or C<..>. The path I<should> be absolute: If it is relative and
1251	relative and your working directory changes, the behaviour is undefined.	2628	your working directory changes, then the behaviour is undefined.
1252		2629
1253	Since there is no standard to do this, the portable implementation simply	2630	Since there is no portable change notification interface available, the
1254	calls C<stat (2)> regularly on the path to see if it changed somehow. You	2631	portable implementation simply calls C<stat(2)> regularly on the path
1255	can specify a recommended polling interval for this case. If you specify	2632	to see if it changed somehow. You can specify a recommended polling
1256	a polling interval of C<0> (highly recommended!) then a I<suitable,	2633	interval for this case. If you specify a polling interval of C<0> (highly
1257	unspecified default> value will be used (which you can expect to be around	2634	recommended!) then a I<suitable, unspecified default> value will be used
1258	five seconds, although this might change dynamically). Libev will also	2635	(which you can expect to be around five seconds, although this might
1259	impose a minimum interval which is currently around C<0.1>, but thats	2636	change dynamically). Libev will also impose a minimum interval which is
1260	usually overkill.	2637	currently around C<0.1>, but that's usually overkill.
1261		2638
1262	This watcher type is not meant for massive numbers of stat watchers,	2639	This watcher type is not meant for massive numbers of stat watchers,
1263	as even with OS-supported change notifications, this can be	2640	as even with OS-supported change notifications, this can be
1264	resource-intensive.	2641	resource-intensive.
1265		2642
1266	At the time of this writing, only the Linux inotify interface is	2643	At the time of this writing, the only OS-specific interface implemented
1267	implemented (implementing kqueue support is left as an exercise for the	2644	is the Linux inotify interface (implementing kqueue support is left as an
1268	reader). Inotify will be used to give hints only and should not change the	2645	exercise for the reader. Note, however, that the author sees no way of
1269	semantics of C<ev_stat> watchers, which means that libev sometimes needs	2646	implementing C<ev_stat> semantics with kqueue, except as a hint).
1270	to fall back to regular polling again even with inotify, but changes are	2647
1271	usually detected immediately, and if the file exists there will be no	2648	=head3 ABI Issues (Largefile Support)
1272	polling.	2649
		2650	Libev by default (unless the user overrides this) uses the default
		2651	compilation environment, which means that on systems with large file
		2652	support disabled by default, you get the 32 bit version of the stat
		2653	structure. When using the library from programs that change the ABI to
		2654	use 64 bit file offsets the programs will fail. In that case you have to
		2655	compile libev with the same flags to get binary compatibility. This is
		2656	obviously the case with any flags that change the ABI, but the problem is
		2657	most noticeably displayed with ev_stat and large file support.
		2658
		2659	The solution for this is to lobby your distribution maker to make large
		2660	file interfaces available by default (as e.g. FreeBSD does) and not
		2661	optional. Libev cannot simply switch on large file support because it has
		2662	to exchange stat structures with application programs compiled using the
		2663	default compilation environment.
		2664
		2665	=head3 Inotify and Kqueue
		2666
		2667	When C<inotify (7)> support has been compiled into libev and present at
		2668	runtime, it will be used to speed up change detection where possible. The
		2669	inotify descriptor will be created lazily when the first C<ev_stat>
		2670	watcher is being started.
		2671
		2672	Inotify presence does not change the semantics of C<ev_stat> watchers
		2673	except that changes might be detected earlier, and in some cases, to avoid
		2674	making regular C<stat> calls. Even in the presence of inotify support
		2675	there are many cases where libev has to resort to regular C<stat> polling,
		2676	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2677	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2678	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2679	xfs are fully working) libev usually gets away without polling.
		2680
		2681	There is no support for kqueue, as apparently it cannot be used to
		2682	implement this functionality, due to the requirement of having a file
		2683	descriptor open on the object at all times, and detecting renames, unlinks
		2684	etc. is difficult.
		2685
		2686	=head3 C<stat ()> is a synchronous operation
		2687
		2688	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2689	the process. The exception are C<ev_stat> watchers - those call C<stat
		2690	()>, which is a synchronous operation.
		2691
		2692	For local paths, this usually doesn't matter: unless the system is very
		2693	busy or the intervals between stat's are large, a stat call will be fast,
		2694	as the path data is usually in memory already (except when starting the
		2695	watcher).
		2696
		2697	For networked file systems, calling C<stat ()> can block an indefinite
		2698	time due to network issues, and even under good conditions, a stat call
		2699	often takes multiple milliseconds.
		2700
		2701	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2702	paths, although this is fully supported by libev.
		2703
		2704	=head3 The special problem of stat time resolution
		2705
		2706	The C<stat ()> system call only supports full-second resolution portably,
		2707	and even on systems where the resolution is higher, most file systems
		2708	still only support whole seconds.
		2709
		2710	That means that, if the time is the only thing that changes, you can
		2711	easily miss updates: on the first update, C<ev_stat> detects a change and
		2712	calls your callback, which does something. When there is another update
		2713	within the same second, C<ev_stat> will be unable to detect unless the
		2714	stat data does change in other ways (e.g. file size).
		2715
		2716	The solution to this is to delay acting on a change for slightly more
		2717	than a second (or till slightly after the next full second boundary), using
		2718	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		2719	ev_timer_again (loop, w)>).
		2720
		2721	The C<.02> offset is added to work around small timing inconsistencies
		2722	of some operating systems (where the second counter of the current time
		2723	might be be delayed. One such system is the Linux kernel, where a call to
		2724	C<gettimeofday> might return a timestamp with a full second later than
		2725	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		2726	update file times then there will be a small window where the kernel uses
		2727	the previous second to update file times but libev might already execute
		2728	the timer callback).
		2729
		2730	=head3 Watcher-Specific Functions and Data Members
1273		2731
1274	=over 4	2732	=over 4
1275		2733
1276	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)	2734	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
1277		2735
…		…
1281	C<path>. The C<interval> is a hint on how quickly a change is expected to	2739	C<path>. The C<interval> is a hint on how quickly a change is expected to
1282	be detected and should normally be specified as C<0> to let libev choose	2740	be detected and should normally be specified as C<0> to let libev choose
1283	a suitable value. The memory pointed to by C<path> must point to the same	2741	a suitable value. The memory pointed to by C<path> must point to the same
1284	path for as long as the watcher is active.	2742	path for as long as the watcher is active.
1285		2743
1286	The callback will be receive C<EV_STAT> when a change was detected,	2744	The callback will receive an C<EV_STAT> event when a change was detected,
1287	relative to the attributes at the time the watcher was started (or the	2745	relative to the attributes at the time the watcher was started (or the
1288	last change was detected).	2746	last change was detected).
1289		2747
1290	=item ev_stat_stat (ev_stat *)	2748	=item ev_stat_stat (loop, ev_stat *)
1291		2749
1292	Updates the stat buffer immediately with new values. If you change the	2750	Updates the stat buffer immediately with new values. If you change the
1293	watched path in your callback, you could call this fucntion to avoid	2751	watched path in your callback, you could call this function to avoid
1294	detecting this change (while introducing a race condition). Can also be	2752	detecting this change (while introducing a race condition if you are not
1295	useful simply to find out the new values.	2753	the only one changing the path). Can also be useful simply to find out the
		2754	new values.
1296		2755
1297	=item ev_statdata attr [read-only]	2756	=item ev_statdata attr [read-only]
1298		2757
1299	The most-recently detected attributes of the file. Although the type is of	2758	The most-recently detected attributes of the file. Although the type is
1300	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	2759	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1301	suitable for your system. If the C<st_nlink> member is C<0>, then there	2760	suitable for your system, but you can only rely on the POSIX-standardised
		2761	members to be present. If the C<st_nlink> member is C<0>, then there was
1302	was some error while C<stat>ing the file.	2762	some error while C<stat>ing the file.
1303		2763
1304	=item ev_statdata prev [read-only]	2764	=item ev_statdata prev [read-only]
1305		2765
1306	The previous attributes of the file. The callback gets invoked whenever	2766	The previous attributes of the file. The callback gets invoked whenever
1307	C<prev> != C<attr>.	2767	C<prev> != C<attr>, or, more precisely, one or more of these members
		2768	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		2769	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1308		2770
1309	=item ev_tstamp interval [read-only]	2771	=item ev_tstamp interval [read-only]
1310		2772
1311	The specified interval.	2773	The specified interval.
1312		2774
1313	=item const char *path [read-only]	2775	=item const char *path [read-only]
1314		2776
1315	The filesystem path that is being watched.	2777	The file system path that is being watched.
1316		2778
1317	=back	2779	=back
1318		2780
		2781	=head3 Examples
		2782
1319	Example: Watch C</etc/passwd> for attribute changes.	2783	Example: Watch C</etc/passwd> for attribute changes.
1320		2784
1321	static void	2785	static void
1322	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	2786	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1323	{	2787	{
1324	/* /etc/passwd changed in some way */	2788	/* /etc/passwd changed in some way */
1325	if (w->attr.st_nlink)	2789	if (w->attr.st_nlink)
1326	{	2790	{
1327	printf ("passwd current size %ld\n", (long)w->attr.st_size);	2791	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1328	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	2792	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1329	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	2793	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1330	}	2794	}
1331	else	2795	else
1332	/* you shalt not abuse printf for puts */	2796	/* you shalt not abuse printf for puts */
1333	puts ("wow, /etc/passwd is not there, expect problems. "	2797	puts ("wow, /etc/passwd is not there, expect problems. "
1334	"if this is windows, they already arrived\n");	2798	"if this is windows, they already arrived\n");
1335	}	2799	}
1336		2800
1337	...	2801	...
1338	ev_stat passwd;	2802	ev_stat passwd;
1339		2803
1340	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");	2804	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1341	ev_stat_start (loop, &passwd);	2805	ev_stat_start (loop, &passwd);
		2806
		2807	Example: Like above, but additionally use a one-second delay so we do not
		2808	miss updates (however, frequent updates will delay processing, too, so
		2809	one might do the work both on C<ev_stat> callback invocation I<and> on
		2810	C<ev_timer> callback invocation).
		2811
		2812	static ev_stat passwd;
		2813	static ev_timer timer;
		2814
		2815	static void
		2816	timer_cb (EV_P_ ev_timer *w, int revents)
		2817	{
		2818	ev_timer_stop (EV_A_ w);
		2819
		2820	/* now it's one second after the most recent passwd change */
		2821	}
		2822
		2823	static void
		2824	stat_cb (EV_P_ ev_stat *w, int revents)
		2825	{
		2826	/* reset the one-second timer */
		2827	ev_timer_again (EV_A_ &timer);
		2828	}
		2829
		2830	...
		2831	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		2832	ev_stat_start (loop, &passwd);
		2833	ev_timer_init (&timer, timer_cb, 0., 1.02);
1342		2834
1343		2835
1344	=head2 C<ev_idle> - when you've got nothing better to do...	2836	=head2 C<ev_idle> - when you've got nothing better to do...
1345		2837
1346	Idle watchers trigger events when there are no other events are pending	2838	Idle watchers trigger events when no other events of the same or higher
1347	(prepare, check and other idle watchers do not count). That is, as long	2839	priority are pending (prepare, check and other idle watchers do not count
1348	as your process is busy handling sockets or timeouts (or even signals,	2840	as receiving "events").
1349	imagine) it will not be triggered. But when your process is idle all idle	2841
1350	watchers are being called again and again, once per event loop iteration -	2842	That is, as long as your process is busy handling sockets or timeouts
		2843	(or even signals, imagine) of the same or higher priority it will not be
		2844	triggered. But when your process is idle (or only lower-priority watchers
		2845	are pending), the idle watchers are being called once per event loop
1351	until stopped, that is, or your process receives more events and becomes	2846	iteration - until stopped, that is, or your process receives more events
1352	busy.	2847	and becomes busy again with higher priority stuff.
1353		2848
1354	The most noteworthy effect is that as long as any idle watchers are	2849	The most noteworthy effect is that as long as any idle watchers are
1355	active, the process will not block when waiting for new events.	2850	active, the process will not block when waiting for new events.
1356		2851
1357	Apart from keeping your process non-blocking (which is a useful	2852	Apart from keeping your process non-blocking (which is a useful
1358	effect on its own sometimes), idle watchers are a good place to do	2853	effect on its own sometimes), idle watchers are a good place to do
1359	"pseudo-background processing", or delay processing stuff to after the	2854	"pseudo-background processing", or delay processing stuff to after the
1360	event loop has handled all outstanding events.	2855	event loop has handled all outstanding events.
1361		2856
		2857	=head3 Abusing an C<ev_idle> watcher for its side-effect
		2858
		2859	As long as there is at least one active idle watcher, libev will never
		2860	sleep unnecessarily. Or in other words, it will loop as fast as possible.
		2861	For this to work, the idle watcher doesn't need to be invoked at all - the
		2862	lowest priority will do.
		2863
		2864	This mode of operation can be useful together with an C<ev_check> watcher,
		2865	to do something on each event loop iteration - for example to balance load
		2866	between different connections.
		2867
		2868	See L</Abusing an ev_check watcher for its side-effect> for a longer
		2869	example.
		2870
		2871	=head3 Watcher-Specific Functions and Data Members
		2872
1362	=over 4	2873	=over 4
1363		2874
1364	=item ev_idle_init (ev_signal *, callback)	2875	=item ev_idle_init (ev_idle *, callback)
1365		2876
1366	Initialises and configures the idle watcher - it has no parameters of any	2877	Initialises and configures the idle watcher - it has no parameters of any
1367	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2878	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1368	believe me.	2879	believe me.
1369		2880
1370	=back	2881	=back
1371		2882
		2883	=head3 Examples
		2884
1372	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	2885	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1373	callback, free it. Also, use no error checking, as usual.	2886	callback, free it. Also, use no error checking, as usual.
1374		2887
1375	static void	2888	static void
1376	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	2889	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
1377	{	2890	{
		2891	// stop the watcher
		2892	ev_idle_stop (loop, w);
		2893
		2894	// now we can free it
1378	free (w);	2895	free (w);
		2896
1379	// now do something you wanted to do when the program has	2897	// now do something you wanted to do when the program has
1380	// no longer asnything immediate to do.	2898	// no longer anything immediate to do.
1381	}	2899	}
1382		2900
1383	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	2901	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
1384	ev_idle_init (idle_watcher, idle_cb);	2902	ev_idle_init (idle_watcher, idle_cb);
1385	ev_idle_start (loop, idle_cb);	2903	ev_idle_start (loop, idle_watcher);
1386		2904
1387		2905
1388	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2906	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1389		2907
1390	Prepare and check watchers are usually (but not always) used in tandem:	2908	Prepare and check watchers are often (but not always) used in pairs:
1391	prepare watchers get invoked before the process blocks and check watchers	2909	prepare watchers get invoked before the process blocks and check watchers
1392	afterwards.	2910	afterwards.
1393		2911
1394	You I<must not> call C<ev_loop> or similar functions that enter	2912	You I<must not> call C<ev_run> or similar functions that enter
1395	the current event loop from either C<ev_prepare> or C<ev_check>	2913	the current event loop from either C<ev_prepare> or C<ev_check>
1396	watchers. Other loops than the current one are fine, however. The	2914	watchers. Other loops than the current one are fine, however. The
1397	rationale behind this is that you do not need to check for recursion in	2915	rationale behind this is that you do not need to check for recursion in
1398	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,	2916	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
1399	C<ev_check> so if you have one watcher of each kind they will always be	2917	C<ev_check> so if you have one watcher of each kind they will always be
1400	called in pairs bracketing the blocking call.	2918	called in pairs bracketing the blocking call.
1401		2919
1402	Their main purpose is to integrate other event mechanisms into libev and	2920	Their main purpose is to integrate other event mechanisms into libev and
1403	their use is somewhat advanced. This could be used, for example, to track	2921	their use is somewhat advanced. They could be used, for example, to track
1404	variable changes, implement your own watchers, integrate net-snmp or a	2922	variable changes, implement your own watchers, integrate net-snmp or a
1405	coroutine library and lots more. They are also occasionally useful if	2923	coroutine library and lots more. They are also occasionally useful if
1406	you cache some data and want to flush it before blocking (for example,	2924	you cache some data and want to flush it before blocking (for example,
1407	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>	2925	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1408	watcher).	2926	watcher).
1409		2927
1410	This is done by examining in each prepare call which file descriptors need	2928	This is done by examining in each prepare call which file descriptors
1411	to be watched by the other library, registering C<ev_io> watchers for	2929	need to be watched by the other library, registering C<ev_io> watchers
1412	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2930	for them and starting an C<ev_timer> watcher for any timeouts (many
1413	provide just this functionality). Then, in the check watcher you check for	2931	libraries provide exactly this functionality). Then, in the check watcher,
1414	any events that occured (by checking the pending status of all watchers	2932	you check for any events that occurred (by checking the pending status
1415	and stopping them) and call back into the library. The I/O and timer	2933	of all watchers and stopping them) and call back into the library. The
1416	callbacks will never actually be called (but must be valid nevertheless,	2934	I/O and timer callbacks will never actually be called (but must be valid
1417	because you never know, you know?).	2935	nevertheless, because you never know, you know?).
1418		2936
1419	As another example, the Perl Coro module uses these hooks to integrate	2937	As another example, the Perl Coro module uses these hooks to integrate
1420	coroutines into libev programs, by yielding to other active coroutines	2938	coroutines into libev programs, by yielding to other active coroutines
1421	during each prepare and only letting the process block if no coroutines	2939	during each prepare and only letting the process block if no coroutines
1422	are ready to run (it's actually more complicated: it only runs coroutines	2940	are ready to run (it's actually more complicated: it only runs coroutines
1423	with priority higher than or equal to the event loop and one coroutine	2941	with priority higher than or equal to the event loop and one coroutine
1424	of lower priority, but only once, using idle watchers to keep the event	2942	of lower priority, but only once, using idle watchers to keep the event
1425	loop from blocking if lower-priority coroutines are active, thus mapping	2943	loop from blocking if lower-priority coroutines are active, thus mapping
1426	low-priority coroutines to idle/background tasks).	2944	low-priority coroutines to idle/background tasks).
1427		2945
		2946	When used for this purpose, it is recommended to give C<ev_check> watchers
		2947	highest (C<EV_MAXPRI>) priority, to ensure that they are being run before
		2948	any other watchers after the poll (this doesn't matter for C<ev_prepare>
		2949	watchers).
		2950
		2951	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
		2952	activate ("feed") events into libev. While libev fully supports this, they
		2953	might get executed before other C<ev_check> watchers did their job. As
		2954	C<ev_check> watchers are often used to embed other (non-libev) event
		2955	loops those other event loops might be in an unusable state until their
		2956	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
		2957	others).
		2958
		2959	=head3 Abusing an C<ev_check> watcher for its side-effect
		2960
		2961	C<ev_check> (and less often also C<ev_prepare>) watchers can also be
		2962	useful because they are called once per event loop iteration. For
		2963	example, if you want to handle a large number of connections fairly, you
		2964	normally only do a bit of work for each active connection, and if there
		2965	is more work to do, you wait for the next event loop iteration, so other
		2966	connections have a chance of making progress.
		2967
		2968	Using an C<ev_check> watcher is almost enough: it will be called on the
		2969	next event loop iteration. However, that isn't as soon as possible -
		2970	without external events, your C<ev_check> watcher will not be invoked.
		2971
		2972	This is where C<ev_idle> watchers come in handy - all you need is a
		2973	single global idle watcher that is active as long as you have one active
		2974	C<ev_check> watcher. The C<ev_idle> watcher makes sure the event loop
		2975	will not sleep, and the C<ev_check> watcher makes sure a callback gets
		2976	invoked. Neither watcher alone can do that.
		2977
		2978	=head3 Watcher-Specific Functions and Data Members
		2979
1428	=over 4	2980	=over 4
1429		2981
1430	=item ev_prepare_init (ev_prepare *, callback)	2982	=item ev_prepare_init (ev_prepare *, callback)
1431		2983
1432	=item ev_check_init (ev_check *, callback)	2984	=item ev_check_init (ev_check *, callback)
1433		2985
1434	Initialises and configures the prepare or check watcher - they have no	2986	Initialises and configures the prepare or check watcher - they have no
1435	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2987	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1436	macros, but using them is utterly, utterly and completely pointless.	2988	macros, but using them is utterly, utterly, utterly and completely
		2989	pointless.
1437		2990
1438	=back	2991	=back
1439		2992
1440	Example: To include a library such as adns, you would add IO watchers	2993	=head3 Examples
1441	and a timeout watcher in a prepare handler, as required by libadns, and	2994
		2995	There are a number of principal ways to embed other event loops or modules
		2996	into libev. Here are some ideas on how to include libadns into libev
		2997	(there is a Perl module named C<EV::ADNS> that does this, which you could
		2998	use as a working example. Another Perl module named C<EV::Glib> embeds a
		2999	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		3000	Glib event loop).
		3001
		3002	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1442	in a check watcher, destroy them and call into libadns. What follows is	3003	and in a check watcher, destroy them and call into libadns. What follows
1443	pseudo-code only of course:	3004	is pseudo-code only of course. This requires you to either use a low
		3005	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		3006	the callbacks for the IO/timeout watchers might not have been called yet.
1444		3007
1445	static ev_io iow [nfd];	3008	static ev_io iow [nfd];
1446	static ev_timer tw;	3009	static ev_timer tw;
1447		3010
1448	static void	3011	static void
1449	io_cb (ev_loop *loop, ev_io *w, int revents)	3012	io_cb (struct ev_loop *loop, ev_io *w, int revents)
1450	{	3013	{
1451	// set the relevant poll flags
1452	// could also call adns_processreadable etc. here
1453	struct pollfd *fd = (struct pollfd *)w->data;
1454	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1455	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1456	}	3014	}
1457		3015
1458	// create io watchers for each fd and a timer before blocking	3016	// create io watchers for each fd and a timer before blocking
1459	static void	3017	static void
1460	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	3018	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
1461	{	3019	{
1462	int timeout = 3600000;	3020	int timeout = 3600000;
1463	struct pollfd fds [nfd];	3021	struct pollfd fds [nfd];
1464	// actual code will need to loop here and realloc etc.	3022	// actual code will need to loop here and realloc etc.
1465	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	3023	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1466		3024
1467	/* the callback is illegal, but won't be called as we stop during check */	3025	/* the callback is illegal, but won't be called as we stop during check */
1468	ev_timer_init (&tw, 0, timeout * 1e-3);	3026	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
1469	ev_timer_start (loop, &tw);	3027	ev_timer_start (loop, &tw);
1470		3028
1471	// create on ev_io per pollfd	3029	// create one ev_io per pollfd
1472	for (int i = 0; i < nfd; ++i)	3030	for (int i = 0; i < nfd; ++i)
1473	{	3031	{
1474	ev_io_init (iow + i, io_cb, fds [i].fd,	3032	ev_io_init (iow + i, io_cb, fds [i].fd,
1475	((fds [i].events & POLLIN ? EV_READ : 0)	3033	((fds [i].events & POLLIN ? EV_READ : 0)
1476	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	3034	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1477		3035
1478	fds [i].revents = 0;	3036	fds [i].revents = 0;
1479	iow [i].data = fds + i;
1480	ev_io_start (loop, iow + i);	3037	ev_io_start (loop, iow + i);
1481	}	3038	}
1482	}	3039	}
1483		3040
1484	// stop all watchers after blocking	3041	// stop all watchers after blocking
1485	static void	3042	static void
1486	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	3043	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
1487	{	3044	{
1488	ev_timer_stop (loop, &tw);	3045	ev_timer_stop (loop, &tw);
1489		3046
1490	for (int i = 0; i < nfd; ++i)	3047	for (int i = 0; i < nfd; ++i)
		3048	{
		3049	// set the relevant poll flags
		3050	// could also call adns_processreadable etc. here
		3051	struct pollfd *fd = fds + i;
		3052	int revents = ev_clear_pending (iow + i);
		3053	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		3054	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		3055
		3056	// now stop the watcher
1491	ev_io_stop (loop, iow + i);	3057	ev_io_stop (loop, iow + i);
		3058	}
1492		3059
1493	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	3060	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1494	}	3061	}
		3062
		3063	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		3064	in the prepare watcher and would dispose of the check watcher.
		3065
		3066	Method 3: If the module to be embedded supports explicit event
		3067	notification (libadns does), you can also make use of the actual watcher
		3068	callbacks, and only destroy/create the watchers in the prepare watcher.
		3069
		3070	static void
		3071	timer_cb (EV_P_ ev_timer *w, int revents)
		3072	{
		3073	adns_state ads = (adns_state)w->data;
		3074	update_now (EV_A);
		3075
		3076	adns_processtimeouts (ads, &tv_now);
		3077	}
		3078
		3079	static void
		3080	io_cb (EV_P_ ev_io *w, int revents)
		3081	{
		3082	adns_state ads = (adns_state)w->data;
		3083	update_now (EV_A);
		3084
		3085	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		3086	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		3087	}
		3088
		3089	// do not ever call adns_afterpoll
		3090
		3091	Method 4: Do not use a prepare or check watcher because the module you
		3092	want to embed is not flexible enough to support it. Instead, you can
		3093	override their poll function. The drawback with this solution is that the
		3094	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
		3095	this approach, effectively embedding EV as a client into the horrible
		3096	libglib event loop.
		3097
		3098	static gint
		3099	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		3100	{
		3101	int got_events = 0;
		3102
		3103	for (n = 0; n < nfds; ++n)
		3104	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		3105
		3106	if (timeout >= 0)
		3107	// create/start timer
		3108
		3109	// poll
		3110	ev_run (EV_A_ 0);
		3111
		3112	// stop timer again
		3113	if (timeout >= 0)
		3114	ev_timer_stop (EV_A_ &to);
		3115
		3116	// stop io watchers again - their callbacks should have set
		3117	for (n = 0; n < nfds; ++n)
		3118	ev_io_stop (EV_A_ iow [n]);
		3119
		3120	return got_events;
		3121	}
1495		3122
1496		3123
1497	=head2 C<ev_embed> - when one backend isn't enough...	3124	=head2 C<ev_embed> - when one backend isn't enough...
1498		3125
1499	This is a rather advanced watcher type that lets you embed one event loop	3126	This is a rather advanced watcher type that lets you embed one event loop
…		…
1505	prioritise I/O.	3132	prioritise I/O.
1506		3133
1507	As an example for a bug workaround, the kqueue backend might only support	3134	As an example for a bug workaround, the kqueue backend might only support
1508	sockets on some platform, so it is unusable as generic backend, but you	3135	sockets on some platform, so it is unusable as generic backend, but you
1509	still want to make use of it because you have many sockets and it scales	3136	still want to make use of it because you have many sockets and it scales
1510	so nicely. In this case, you would create a kqueue-based loop and embed it	3137	so nicely. In this case, you would create a kqueue-based loop and embed
1511	into your default loop (which might use e.g. poll). Overall operation will	3138	it into your default loop (which might use e.g. poll). Overall operation
1512	be a bit slower because first libev has to poll and then call kevent, but	3139	will be a bit slower because first libev has to call C<poll> and then
1513	at least you can use both at what they are best.	3140	C<kevent>, but at least you can use both mechanisms for what they are
		3141	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
1514		3142
1515	As for prioritising I/O: rarely you have the case where some fds have	3143	As for prioritising I/O: under rare circumstances you have the case where
1516	to be watched and handled very quickly (with low latency), and even	3144	some fds have to be watched and handled very quickly (with low latency),
1517	priorities and idle watchers might have too much overhead. In this case	3145	and even priorities and idle watchers might have too much overhead. In
1518	you would put all the high priority stuff in one loop and all the rest in	3146	this case you would put all the high priority stuff in one loop and all
1519	a second one, and embed the second one in the first.	3147	the rest in a second one, and embed the second one in the first.
1520		3148
1521	As long as the watcher is active, the callback will be invoked every time	3149	As long as the watcher is active, the callback will be invoked every
1522	there might be events pending in the embedded loop. The callback must then	3150	time there might be events pending in the embedded loop. The callback
1523	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke	3151	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
1524	their callbacks (you could also start an idle watcher to give the embedded	3152	sweep and invoke their callbacks (the callback doesn't need to invoke the
1525	loop strictly lower priority for example). You can also set the callback	3153	C<ev_embed_sweep> function directly, it could also start an idle watcher
1526	to C<0>, in which case the embed watcher will automatically execute the	3154	to give the embedded loop strictly lower priority for example).
1527	embedded loop sweep.
1528		3155
1529	As long as the watcher is started it will automatically handle events. The	3156	You can also set the callback to C<0>, in which case the embed watcher
1530	callback will be invoked whenever some events have been handled. You can	3157	will automatically execute the embedded loop sweep whenever necessary.
1531	set the callback to C<0> to avoid having to specify one if you are not
1532	interested in that.
1533		3158
1534	Also, there have not currently been made special provisions for forking:	3159	Fork detection will be handled transparently while the C<ev_embed> watcher
1535	when you fork, you not only have to call C<ev_loop_fork> on both loops,	3160	is active, i.e., the embedded loop will automatically be forked when the
1536	but you will also have to stop and restart any C<ev_embed> watchers	3161	embedding loop forks. In other cases, the user is responsible for calling
1537	yourself.	3162	C<ev_loop_fork> on the embedded loop.
1538		3163
1539	Unfortunately, not all backends are embeddable, only the ones returned by	3164	Unfortunately, not all backends are embeddable: only the ones returned by
1540	C<ev_embeddable_backends> are, which, unfortunately, does not include any	3165	C<ev_embeddable_backends> are, which, unfortunately, does not include any
1541	portable one.	3166	portable one.
1542		3167
1543	So when you want to use this feature you will always have to be prepared	3168	So when you want to use this feature you will always have to be prepared
1544	that you cannot get an embeddable loop. The recommended way to get around	3169	that you cannot get an embeddable loop. The recommended way to get around
1545	this is to have a separate variables for your embeddable loop, try to	3170	this is to have a separate variables for your embeddable loop, try to
1546	create it, and if that fails, use the normal loop for everything:	3171	create it, and if that fails, use the normal loop for everything.
1547		3172
1548	struct ev_loop *loop_hi = ev_default_init (0);	3173	=head3 C<ev_embed> and fork
1549	struct ev_loop *loop_lo = 0;
1550	struct ev_embed embed;
1551
1552	// see if there is a chance of getting one that works
1553	// (remember that a flags value of 0 means autodetection)
1554	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1555	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1556	: 0;
1557		3174
1558	// if we got one, then embed it, otherwise default to loop_hi	3175	While the C<ev_embed> watcher is running, forks in the embedding loop will
1559	if (loop_lo)	3176	automatically be applied to the embedded loop as well, so no special
1560	{	3177	fork handling is required in that case. When the watcher is not running,
1561	ev_embed_init (&embed, 0, loop_lo);	3178	however, it is still the task of the libev user to call C<ev_loop_fork ()>
1562	ev_embed_start (loop_hi, &embed);	3179	as applicable.
1563	}	3180
1564	else	3181	=head3 Watcher-Specific Functions and Data Members
1565	loop_lo = loop_hi;
1566		3182
1567	=over 4	3183	=over 4
1568		3184
1569	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	3185	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
1570		3186
1571	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)	3187	=item ev_embed_set (ev_embed *, struct ev_loop *embedded_loop)
1572		3188
1573	Configures the watcher to embed the given loop, which must be	3189	Configures the watcher to embed the given loop, which must be
1574	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be	3190	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1575	invoked automatically, otherwise it is the responsibility of the callback	3191	invoked automatically, otherwise it is the responsibility of the callback
1576	to invoke it (it will continue to be called until the sweep has been done,	3192	to invoke it (it will continue to be called until the sweep has been done,
1577	if you do not want thta, you need to temporarily stop the embed watcher).	3193	if you do not want that, you need to temporarily stop the embed watcher).
1578		3194
1579	=item ev_embed_sweep (loop, ev_embed *)	3195	=item ev_embed_sweep (loop, ev_embed *)
1580		3196
1581	Make a single, non-blocking sweep over the embedded loop. This works	3197	Make a single, non-blocking sweep over the embedded loop. This works
1582	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	3198	similarly to C<ev_run (embedded_loop, EVRUN_NOWAIT)>, but in the most
1583	apropriate way for embedded loops.	3199	appropriate way for embedded loops.
1584		3200
1585	=item struct ev_loop *loop [read-only]	3201	=item struct ev_loop *other [read-only]
1586		3202
1587	The embedded event loop.	3203	The embedded event loop.
1588		3204
1589	=back	3205	=back
		3206
		3207	=head3 Examples
		3208
		3209	Example: Try to get an embeddable event loop and embed it into the default
		3210	event loop. If that is not possible, use the default loop. The default
		3211	loop is stored in C<loop_hi>, while the embeddable loop is stored in
		3212	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
		3213	used).
		3214
		3215	struct ev_loop *loop_hi = ev_default_init (0);
		3216	struct ev_loop *loop_lo = 0;
		3217	ev_embed embed;
		3218
		3219	// see if there is a chance of getting one that works
		3220	// (remember that a flags value of 0 means autodetection)
		3221	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		3222	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		3223	: 0;
		3224
		3225	// if we got one, then embed it, otherwise default to loop_hi
		3226	if (loop_lo)
		3227	{
		3228	ev_embed_init (&embed, 0, loop_lo);
		3229	ev_embed_start (loop_hi, &embed);
		3230	}
		3231	else
		3232	loop_lo = loop_hi;
		3233
		3234	Example: Check if kqueue is available but not recommended and create
		3235	a kqueue backend for use with sockets (which usually work with any
		3236	kqueue implementation). Store the kqueue/socket-only event loop in
		3237	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		3238
		3239	struct ev_loop *loop = ev_default_init (0);
		3240	struct ev_loop *loop_socket = 0;
		3241	ev_embed embed;
		3242
		3243	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		3244	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		3245	{
		3246	ev_embed_init (&embed, 0, loop_socket);
		3247	ev_embed_start (loop, &embed);
		3248	}
		3249
		3250	if (!loop_socket)
		3251	loop_socket = loop;
		3252
		3253	// now use loop_socket for all sockets, and loop for everything else
1590		3254
1591		3255
1592	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	3256	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
1593		3257
1594	Fork watchers are called when a C<fork ()> was detected (usually because	3258	Fork watchers are called when a C<fork ()> was detected (usually because
1595	whoever is a good citizen cared to tell libev about it by calling	3259	whoever is a good citizen cared to tell libev about it by calling
1596	C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the	3260	C<ev_loop_fork>). The invocation is done before the event loop blocks next
1597	event loop blocks next and before C<ev_check> watchers are being called,	3261	and before C<ev_check> watchers are being called, and only in the child
1598	and only in the child after the fork. If whoever good citizen calling	3262	after the fork. If whoever good citizen calling C<ev_default_fork> cheats
1599	C<ev_default_fork> cheats and calls it in the wrong process, the fork	3263	and calls it in the wrong process, the fork handlers will be invoked, too,
1600	handlers will be invoked, too, of course.	3264	of course.
		3265
		3266	=head3 The special problem of life after fork - how is it possible?
		3267
		3268	Most uses of C<fork()> consist of forking, then some simple calls to set
		3269	up/change the process environment, followed by a call to C<exec()>. This
		3270	sequence should be handled by libev without any problems.
		3271
		3272	This changes when the application actually wants to do event handling
		3273	in the child, or both parent in child, in effect "continuing" after the
		3274	fork.
		3275
		3276	The default mode of operation (for libev, with application help to detect
		3277	forks) is to duplicate all the state in the child, as would be expected
		3278	when I<either> the parent I<or> the child process continues.
		3279
		3280	When both processes want to continue using libev, then this is usually the
		3281	wrong result. In that case, usually one process (typically the parent) is
		3282	supposed to continue with all watchers in place as before, while the other
		3283	process typically wants to start fresh, i.e. without any active watchers.
		3284
		3285	The cleanest and most efficient way to achieve that with libev is to
		3286	simply create a new event loop, which of course will be "empty", and
		3287	use that for new watchers. This has the advantage of not touching more
		3288	memory than necessary, and thus avoiding the copy-on-write, and the
		3289	disadvantage of having to use multiple event loops (which do not support
		3290	signal watchers).
		3291
		3292	When this is not possible, or you want to use the default loop for
		3293	other reasons, then in the process that wants to start "fresh", call
		3294	C<ev_loop_destroy (EV_DEFAULT)> followed by C<ev_default_loop (...)>.
		3295	Destroying the default loop will "orphan" (not stop) all registered
		3296	watchers, so you have to be careful not to execute code that modifies
		3297	those watchers. Note also that in that case, you have to re-register any
		3298	signal watchers.
		3299
		3300	=head3 Watcher-Specific Functions and Data Members
1601		3301
1602	=over 4	3302	=over 4
1603		3303
1604	=item ev_fork_init (ev_signal *, callback)	3304	=item ev_fork_init (ev_fork *, callback)
1605		3305
1606	Initialises and configures the fork watcher - it has no parameters of any	3306	Initialises and configures the fork watcher - it has no parameters of any
1607	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,	3307	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
1608	believe me.	3308	really.
1609		3309
1610	=back	3310	=back
1611		3311
1612		3312
		3313	=head2 C<ev_cleanup> - even the best things end
		3314
		3315	Cleanup watchers are called just before the event loop is being destroyed
		3316	by a call to C<ev_loop_destroy>.
		3317
		3318	While there is no guarantee that the event loop gets destroyed, cleanup
		3319	watchers provide a convenient method to install cleanup hooks for your
		3320	program, worker threads and so on - you just to make sure to destroy the
		3321	loop when you want them to be invoked.
		3322
		3323	Cleanup watchers are invoked in the same way as any other watcher. Unlike
		3324	all other watchers, they do not keep a reference to the event loop (which
		3325	makes a lot of sense if you think about it). Like all other watchers, you
		3326	can call libev functions in the callback, except C<ev_cleanup_start>.
		3327
		3328	=head3 Watcher-Specific Functions and Data Members
		3329
		3330	=over 4
		3331
		3332	=item ev_cleanup_init (ev_cleanup *, callback)
		3333
		3334	Initialises and configures the cleanup watcher - it has no parameters of
		3335	any kind. There is a C<ev_cleanup_set> macro, but using it is utterly
		3336	pointless, I assure you.
		3337
		3338	=back
		3339
		3340	Example: Register an atexit handler to destroy the default loop, so any
		3341	cleanup functions are called.
		3342
		3343	static void
		3344	program_exits (void)
		3345	{
		3346	ev_loop_destroy (EV_DEFAULT_UC);
		3347	}
		3348
		3349	...
		3350	atexit (program_exits);
		3351
		3352
		3353	=head2 C<ev_async> - how to wake up an event loop
		3354
		3355	In general, you cannot use an C<ev_loop> from multiple threads or other
		3356	asynchronous sources such as signal handlers (as opposed to multiple event
		3357	loops - those are of course safe to use in different threads).
		3358
		3359	Sometimes, however, you need to wake up an event loop you do not control,
		3360	for example because it belongs to another thread. This is what C<ev_async>
		3361	watchers do: as long as the C<ev_async> watcher is active, you can signal
		3362	it by calling C<ev_async_send>, which is thread- and signal safe.
		3363
		3364	This functionality is very similar to C<ev_signal> watchers, as signals,
		3365	too, are asynchronous in nature, and signals, too, will be compressed
		3366	(i.e. the number of callback invocations may be less than the number of
		3367	C<ev_async_send> calls). In fact, you could use signal watchers as a kind
		3368	of "global async watchers" by using a watcher on an otherwise unused
		3369	signal, and C<ev_feed_signal> to signal this watcher from another thread,
		3370	even without knowing which loop owns the signal.
		3371
		3372	=head3 Queueing
		3373
		3374	C<ev_async> does not support queueing of data in any way. The reason
		3375	is that the author does not know of a simple (or any) algorithm for a
		3376	multiple-writer-single-reader queue that works in all cases and doesn't
		3377	need elaborate support such as pthreads or unportable memory access
		3378	semantics.
		3379
		3380	That means that if you want to queue data, you have to provide your own
		3381	queue. But at least I can tell you how to implement locking around your
		3382	queue:
		3383
		3384	=over 4
		3385
		3386	=item queueing from a signal handler context
		3387
		3388	To implement race-free queueing, you simply add to the queue in the signal
		3389	handler but you block the signal handler in the watcher callback. Here is
		3390	an example that does that for some fictitious SIGUSR1 handler:
		3391
		3392	static ev_async mysig;
		3393
		3394	static void
		3395	sigusr1_handler (void)
		3396	{
		3397	sometype data;
		3398
		3399	// no locking etc.
		3400	queue_put (data);
		3401	ev_async_send (EV_DEFAULT_ &mysig);
		3402	}
		3403
		3404	static void
		3405	mysig_cb (EV_P_ ev_async *w, int revents)
		3406	{
		3407	sometype data;
		3408	sigset_t block, prev;
		3409
		3410	sigemptyset (&block);
		3411	sigaddset (&block, SIGUSR1);
		3412	sigprocmask (SIG_BLOCK, &block, &prev);
		3413
		3414	while (queue_get (&data))
		3415	process (data);
		3416
		3417	if (sigismember (&prev, SIGUSR1)
		3418	sigprocmask (SIG_UNBLOCK, &block, 0);
		3419	}
		3420
		3421	(Note: pthreads in theory requires you to use C<pthread_setmask>
		3422	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		3423	either...).
		3424
		3425	=item queueing from a thread context
		3426
		3427	The strategy for threads is different, as you cannot (easily) block
		3428	threads but you can easily preempt them, so to queue safely you need to
		3429	employ a traditional mutex lock, such as in this pthread example:
		3430
		3431	static ev_async mysig;
		3432	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		3433
		3434	static void
		3435	otherthread (void)
		3436	{
		3437	// only need to lock the actual queueing operation
		3438	pthread_mutex_lock (&mymutex);
		3439	queue_put (data);
		3440	pthread_mutex_unlock (&mymutex);
		3441
		3442	ev_async_send (EV_DEFAULT_ &mysig);
		3443	}
		3444
		3445	static void
		3446	mysig_cb (EV_P_ ev_async *w, int revents)
		3447	{
		3448	pthread_mutex_lock (&mymutex);
		3449
		3450	while (queue_get (&data))
		3451	process (data);
		3452
		3453	pthread_mutex_unlock (&mymutex);
		3454	}
		3455
		3456	=back
		3457
		3458
		3459	=head3 Watcher-Specific Functions and Data Members
		3460
		3461	=over 4
		3462
		3463	=item ev_async_init (ev_async *, callback)
		3464
		3465	Initialises and configures the async watcher - it has no parameters of any
		3466	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
		3467	trust me.
		3468
		3469	=item ev_async_send (loop, ev_async *)
		3470
		3471	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		3472	an C<EV_ASYNC> event on the watcher into the event loop, and instantly
		3473	returns.
		3474
		3475	Unlike C<ev_feed_event>, this call is safe to do from other threads,
		3476	signal or similar contexts (see the discussion of C<EV_ATOMIC_T> in the
		3477	embedding section below on what exactly this means).
		3478
		3479	Note that, as with other watchers in libev, multiple events might get
		3480	compressed into a single callback invocation (another way to look at
		3481	this is that C<ev_async> watchers are level-triggered: they are set on
		3482	C<ev_async_send>, reset when the event loop detects that).
		3483
		3484	This call incurs the overhead of at most one extra system call per event
		3485	loop iteration, if the event loop is blocked, and no syscall at all if
		3486	the event loop (or your program) is processing events. That means that
		3487	repeated calls are basically free (there is no need to avoid calls for
		3488	performance reasons) and that the overhead becomes smaller (typically
		3489	zero) under load.
		3490
		3491	=item bool = ev_async_pending (ev_async *)
		3492
		3493	Returns a non-zero value when C<ev_async_send> has been called on the
		3494	watcher but the event has not yet been processed (or even noted) by the
		3495	event loop.
		3496
		3497	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		3498	the loop iterates next and checks for the watcher to have become active,
		3499	it will reset the flag again. C<ev_async_pending> can be used to very
		3500	quickly check whether invoking the loop might be a good idea.
		3501
		3502	Not that this does I<not> check whether the watcher itself is pending,
		3503	only whether it has been requested to make this watcher pending: there
		3504	is a time window between the event loop checking and resetting the async
		3505	notification, and the callback being invoked.
		3506
		3507	=back
		3508
		3509
1613	=head1 OTHER FUNCTIONS	3510	=head1 OTHER FUNCTIONS
1614		3511
1615	There are some other functions of possible interest. Described. Here. Now.	3512	There are some other functions of possible interest. Described. Here. Now.
1616		3513
1617	=over 4	3514	=over 4
1618		3515
1619	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	3516	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
1620		3517
1621	This function combines a simple timer and an I/O watcher, calls your	3518	This function combines a simple timer and an I/O watcher, calls your
1622	callback on whichever event happens first and automatically stop both	3519	callback on whichever event happens first and automatically stops both
1623	watchers. This is useful if you want to wait for a single event on an fd	3520	watchers. This is useful if you want to wait for a single event on an fd
1624	or timeout without having to allocate/configure/start/stop/free one or	3521	or timeout without having to allocate/configure/start/stop/free one or
1625	more watchers yourself.	3522	more watchers yourself.
1626		3523
1627	If C<fd> is less than 0, then no I/O watcher will be started and events	3524	If C<fd> is less than 0, then no I/O watcher will be started and the
1628	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	3525	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
1629	C<events> set will be craeted and started.	3526	the given C<fd> and C<events> set will be created and started.
1630		3527
1631	If C<timeout> is less than 0, then no timeout watcher will be	3528	If C<timeout> is less than 0, then no timeout watcher will be
1632	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	3529	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
1633	repeat = 0) will be started. While C<0> is a valid timeout, it is of	3530	repeat = 0) will be started. C<0> is a valid timeout.
1634	dubious value.
1635		3531
1636	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	3532	The callback has the type C<void (*cb)(int revents, void *arg)> and is
1637	passed an C<revents> set like normal event callbacks (a combination of	3533	passed an C<revents> set like normal event callbacks (a combination of
1638	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	3534	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMER>) and the C<arg>
1639	value passed to C<ev_once>:	3535	value passed to C<ev_once>. Note that it is possible to receive I<both>
		3536	a timeout and an io event at the same time - you probably should give io
		3537	events precedence.
1640		3538
		3539	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		3540
1641	static void stdin_ready (int revents, void *arg)	3541	static void stdin_ready (int revents, void *arg)
		3542	{
		3543	if (revents & EV_READ)
		3544	/* stdin might have data for us, joy! */;
		3545	else if (revents & EV_TIMER)
		3546	/* doh, nothing entered */;
		3547	}
		3548
		3549	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
		3550
		3551	=item ev_feed_fd_event (loop, int fd, int revents)
		3552
		3553	Feed an event on the given fd, as if a file descriptor backend detected
		3554	the given events.
		3555
		3556	=item ev_feed_signal_event (loop, int signum)
		3557
		3558	Feed an event as if the given signal occurred. See also C<ev_feed_signal>,
		3559	which is async-safe.
		3560
		3561	=back
		3562
		3563
		3564	=head1 COMMON OR USEFUL IDIOMS (OR BOTH)
		3565
		3566	This section explains some common idioms that are not immediately
		3567	obvious. Note that examples are sprinkled over the whole manual, and this
		3568	section only contains stuff that wouldn't fit anywhere else.
		3569
		3570	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
		3571
		3572	Each watcher has, by default, a C<void *data> member that you can read
		3573	or modify at any time: libev will completely ignore it. This can be used
		3574	to associate arbitrary data with your watcher. If you need more data and
		3575	don't want to allocate memory separately and store a pointer to it in that
		3576	data member, you can also "subclass" the watcher type and provide your own
		3577	data:
		3578
		3579	struct my_io
		3580	{
		3581	ev_io io;
		3582	int otherfd;
		3583	void *somedata;
		3584	struct whatever *mostinteresting;
		3585	};
		3586
		3587	...
		3588	struct my_io w;
		3589	ev_io_init (&w.io, my_cb, fd, EV_READ);
		3590
		3591	And since your callback will be called with a pointer to the watcher, you
		3592	can cast it back to your own type:
		3593
		3594	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
		3595	{
		3596	struct my_io *w = (struct my_io *)w_;
		3597	...
		3598	}
		3599
		3600	More interesting and less C-conformant ways of casting your callback
		3601	function type instead have been omitted.
		3602
		3603	=head2 BUILDING YOUR OWN COMPOSITE WATCHERS
		3604
		3605	Another common scenario is to use some data structure with multiple
		3606	embedded watchers, in effect creating your own watcher that combines
		3607	multiple libev event sources into one "super-watcher":
		3608
		3609	struct my_biggy
		3610	{
		3611	int some_data;
		3612	ev_timer t1;
		3613	ev_timer t2;
		3614	}
		3615
		3616	In this case getting the pointer to C<my_biggy> is a bit more
		3617	complicated: Either you store the address of your C<my_biggy> struct in
		3618	the C<data> member of the watcher (for woozies or C++ coders), or you need
		3619	to use some pointer arithmetic using C<offsetof> inside your watchers (for
		3620	real programmers):
		3621
		3622	#include <stddef.h>
		3623
		3624	static void
		3625	t1_cb (EV_P_ ev_timer *w, int revents)
		3626	{
		3627	struct my_biggy big = (struct my_biggy *)
		3628	(((char *)w) - offsetof (struct my_biggy, t1));
		3629	}
		3630
		3631	static void
		3632	t2_cb (EV_P_ ev_timer *w, int revents)
		3633	{
		3634	struct my_biggy big = (struct my_biggy *)
		3635	(((char *)w) - offsetof (struct my_biggy, t2));
		3636	}
		3637
		3638	=head2 AVOIDING FINISHING BEFORE RETURNING
		3639
		3640	Often you have structures like this in event-based programs:
		3641
		3642	callback ()
1642	{	3643	{
1643	if (revents & EV_TIMEOUT)	3644	free (request);
1644	/* doh, nothing entered */;
1645	else if (revents & EV_READ)
1646	/* stdin might have data for us, joy! */;
1647	}	3645	}
1648		3646
1649	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	3647	request = start_new_request (..., callback);
1650		3648
1651	=item ev_feed_event (ev_loop *, watcher *, int revents)	3649	The intent is to start some "lengthy" operation. The C<request> could be
		3650	used to cancel the operation, or do other things with it.
1652		3651
1653	Feeds the given event set into the event loop, as if the specified event	3652	It's not uncommon to have code paths in C<start_new_request> that
1654	had happened for the specified watcher (which must be a pointer to an	3653	immediately invoke the callback, for example, to report errors. Or you add
1655	initialised but not necessarily started event watcher).	3654	some caching layer that finds that it can skip the lengthy aspects of the
		3655	operation and simply invoke the callback with the result.
1656		3656
1657	=item ev_feed_fd_event (ev_loop *, int fd, int revents)	3657	The problem here is that this will happen I<before> C<start_new_request>
		3658	has returned, so C<request> is not set.
1658		3659
1659	Feed an event on the given fd, as if a file descriptor backend detected	3660	Even if you pass the request by some safer means to the callback, you
1660	the given events it.	3661	might want to do something to the request after starting it, such as
		3662	canceling it, which probably isn't working so well when the callback has
		3663	already been invoked.
1661		3664
1662	=item ev_feed_signal_event (ev_loop *loop, int signum)	3665	A common way around all these issues is to make sure that
		3666	C<start_new_request> I<always> returns before the callback is invoked. If
		3667	C<start_new_request> immediately knows the result, it can artificially
		3668	delay invoking the callback by using a C<prepare> or C<idle> watcher for
		3669	example, or more sneakily, by reusing an existing (stopped) watcher and
		3670	pushing it into the pending queue:
1663		3671
1664	Feed an event as if the given signal occured (C<loop> must be the default	3672	ev_set_cb (watcher, callback);
1665	loop!).	3673	ev_feed_event (EV_A_ watcher, 0);
1666		3674
1667	=back	3675	This way, C<start_new_request> can safely return before the callback is
		3676	invoked, while not delaying callback invocation too much.
		3677
		3678	=head2 MODEL/NESTED EVENT LOOP INVOCATIONS AND EXIT CONDITIONS
		3679
		3680	Often (especially in GUI toolkits) there are places where you have
		3681	I<modal> interaction, which is most easily implemented by recursively
		3682	invoking C<ev_run>.
		3683
		3684	This brings the problem of exiting - a callback might want to finish the
		3685	main C<ev_run> call, but not the nested one (e.g. user clicked "Quit", but
		3686	a modal "Are you sure?" dialog is still waiting), or just the nested one
		3687	and not the main one (e.g. user clocked "Ok" in a modal dialog), or some
		3688	other combination: In these cases, a simple C<ev_break> will not work.
		3689
		3690	The solution is to maintain "break this loop" variable for each C<ev_run>
		3691	invocation, and use a loop around C<ev_run> until the condition is
		3692	triggered, using C<EVRUN_ONCE>:
		3693
		3694	// main loop
		3695	int exit_main_loop = 0;
		3696
		3697	while (!exit_main_loop)
		3698	ev_run (EV_DEFAULT_ EVRUN_ONCE);
		3699
		3700	// in a modal watcher
		3701	int exit_nested_loop = 0;
		3702
		3703	while (!exit_nested_loop)
		3704	ev_run (EV_A_ EVRUN_ONCE);
		3705
		3706	To exit from any of these loops, just set the corresponding exit variable:
		3707
		3708	// exit modal loop
		3709	exit_nested_loop = 1;
		3710
		3711	// exit main program, after modal loop is finished
		3712	exit_main_loop = 1;
		3713
		3714	// exit both
		3715	exit_main_loop = exit_nested_loop = 1;
		3716
		3717	=head2 THREAD LOCKING EXAMPLE
		3718
		3719	Here is a fictitious example of how to run an event loop in a different
		3720	thread from where callbacks are being invoked and watchers are
		3721	created/added/removed.
		3722
		3723	For a real-world example, see the C<EV::Loop::Async> perl module,
		3724	which uses exactly this technique (which is suited for many high-level
		3725	languages).
		3726
		3727	The example uses a pthread mutex to protect the loop data, a condition
		3728	variable to wait for callback invocations, an async watcher to notify the
		3729	event loop thread and an unspecified mechanism to wake up the main thread.
		3730
		3731	First, you need to associate some data with the event loop:
		3732
		3733	typedef struct {
		3734	mutex_t lock; /* global loop lock */
		3735	ev_async async_w;
		3736	thread_t tid;
		3737	cond_t invoke_cv;
		3738	} userdata;
		3739
		3740	void prepare_loop (EV_P)
		3741	{
		3742	// for simplicity, we use a static userdata struct.
		3743	static userdata u;
		3744
		3745	ev_async_init (&u->async_w, async_cb);
		3746	ev_async_start (EV_A_ &u->async_w);
		3747
		3748	pthread_mutex_init (&u->lock, 0);
		3749	pthread_cond_init (&u->invoke_cv, 0);
		3750
		3751	// now associate this with the loop
		3752	ev_set_userdata (EV_A_ u);
		3753	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		3754	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		3755
		3756	// then create the thread running ev_run
		3757	pthread_create (&u->tid, 0, l_run, EV_A);
		3758	}
		3759
		3760	The callback for the C<ev_async> watcher does nothing: the watcher is used
		3761	solely to wake up the event loop so it takes notice of any new watchers
		3762	that might have been added:
		3763
		3764	static void
		3765	async_cb (EV_P_ ev_async *w, int revents)
		3766	{
		3767	// just used for the side effects
		3768	}
		3769
		3770	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		3771	protecting the loop data, respectively.
		3772
		3773	static void
		3774	l_release (EV_P)
		3775	{
		3776	userdata *u = ev_userdata (EV_A);
		3777	pthread_mutex_unlock (&u->lock);
		3778	}
		3779
		3780	static void
		3781	l_acquire (EV_P)
		3782	{
		3783	userdata *u = ev_userdata (EV_A);
		3784	pthread_mutex_lock (&u->lock);
		3785	}
		3786
		3787	The event loop thread first acquires the mutex, and then jumps straight
		3788	into C<ev_run>:
		3789
		3790	void *
		3791	l_run (void *thr_arg)
		3792	{
		3793	struct ev_loop *loop = (struct ev_loop *)thr_arg;
		3794
		3795	l_acquire (EV_A);
		3796	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		3797	ev_run (EV_A_ 0);
		3798	l_release (EV_A);
		3799
		3800	return 0;
		3801	}
		3802
		3803	Instead of invoking all pending watchers, the C<l_invoke> callback will
		3804	signal the main thread via some unspecified mechanism (signals? pipe
		3805	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		3806	have been called (in a while loop because a) spurious wakeups are possible
		3807	and b) skipping inter-thread-communication when there are no pending
		3808	watchers is very beneficial):
		3809
		3810	static void
		3811	l_invoke (EV_P)
		3812	{
		3813	userdata *u = ev_userdata (EV_A);
		3814
		3815	while (ev_pending_count (EV_A))
		3816	{
		3817	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		3818	pthread_cond_wait (&u->invoke_cv, &u->lock);
		3819	}
		3820	}
		3821
		3822	Now, whenever the main thread gets told to invoke pending watchers, it
		3823	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		3824	thread to continue:
		3825
		3826	static void
		3827	real_invoke_pending (EV_P)
		3828	{
		3829	userdata *u = ev_userdata (EV_A);
		3830
		3831	pthread_mutex_lock (&u->lock);
		3832	ev_invoke_pending (EV_A);
		3833	pthread_cond_signal (&u->invoke_cv);
		3834	pthread_mutex_unlock (&u->lock);
		3835	}
		3836
		3837	Whenever you want to start/stop a watcher or do other modifications to an
		3838	event loop, you will now have to lock:
		3839
		3840	ev_timer timeout_watcher;
		3841	userdata *u = ev_userdata (EV_A);
		3842
		3843	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		3844
		3845	pthread_mutex_lock (&u->lock);
		3846	ev_timer_start (EV_A_ &timeout_watcher);
		3847	ev_async_send (EV_A_ &u->async_w);
		3848	pthread_mutex_unlock (&u->lock);
		3849
		3850	Note that sending the C<ev_async> watcher is required because otherwise
		3851	an event loop currently blocking in the kernel will have no knowledge
		3852	about the newly added timer. By waking up the loop it will pick up any new
		3853	watchers in the next event loop iteration.
		3854
		3855	=head2 THREADS, COROUTINES, CONTINUATIONS, QUEUES... INSTEAD OF CALLBACKS
		3856
		3857	While the overhead of a callback that e.g. schedules a thread is small, it
		3858	is still an overhead. If you embed libev, and your main usage is with some
		3859	kind of threads or coroutines, you might want to customise libev so that
		3860	doesn't need callbacks anymore.
		3861
		3862	Imagine you have coroutines that you can switch to using a function
		3863	C<switch_to (coro)>, that libev runs in a coroutine called C<libev_coro>
		3864	and that due to some magic, the currently active coroutine is stored in a
		3865	global called C<current_coro>. Then you can build your own "wait for libev
		3866	event" primitive by changing C<EV_CB_DECLARE> and C<EV_CB_INVOKE> (note
		3867	the differing C<;> conventions):
		3868
		3869	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3870	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb)
		3871
		3872	That means instead of having a C callback function, you store the
		3873	coroutine to switch to in each watcher, and instead of having libev call
		3874	your callback, you instead have it switch to that coroutine.
		3875
		3876	A coroutine might now wait for an event with a function called
		3877	C<wait_for_event>. (the watcher needs to be started, as always, but it doesn't
		3878	matter when, or whether the watcher is active or not when this function is
		3879	called):
		3880
		3881	void
		3882	wait_for_event (ev_watcher *w)
		3883	{
		3884	ev_set_cb (w, current_coro);
		3885	switch_to (libev_coro);
		3886	}
		3887
		3888	That basically suspends the coroutine inside C<wait_for_event> and
		3889	continues the libev coroutine, which, when appropriate, switches back to
		3890	this or any other coroutine.
		3891
		3892	You can do similar tricks if you have, say, threads with an event queue -
		3893	instead of storing a coroutine, you store the queue object and instead of
		3894	switching to a coroutine, you push the watcher onto the queue and notify
		3895	any waiters.
		3896
		3897	To embed libev, see L</EMBEDDING>, but in short, it's easiest to create two
		3898	files, F<my_ev.h> and F<my_ev.c> that include the respective libev files:
		3899
		3900	// my_ev.h
		3901	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3902	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb);
		3903	#include "../libev/ev.h"
		3904
		3905	// my_ev.c
		3906	#define EV_H "my_ev.h"
		3907	#include "../libev/ev.c"
		3908
		3909	And then use F<my_ev.h> when you would normally use F<ev.h>, and compile
		3910	F<my_ev.c> into your project. When properly specifying include paths, you
		3911	can even use F<ev.h> as header file name directly.
1668		3912
1669		3913
1670	=head1 LIBEVENT EMULATION	3914	=head1 LIBEVENT EMULATION
1671		3915
1672	Libev offers a compatibility emulation layer for libevent. It cannot	3916	Libev offers a compatibility emulation layer for libevent. It cannot
1673	emulate the internals of libevent, so here are some usage hints:	3917	emulate the internals of libevent, so here are some usage hints:
1674		3918
1675	=over 4	3919	=over 4
		3920
		3921	=item * Only the libevent-1.4.1-beta API is being emulated.
		3922
		3923	This was the newest libevent version available when libev was implemented,
		3924	and is still mostly unchanged in 2010.
1676		3925
1677	=item * Use it by including <event.h>, as usual.	3926	=item * Use it by including <event.h>, as usual.
1678		3927
1679	=item * The following members are fully supported: ev_base, ev_callback,	3928	=item * The following members are fully supported: ev_base, ev_callback,
1680	ev_arg, ev_fd, ev_res, ev_events.	3929	ev_arg, ev_fd, ev_res, ev_events.
…		…
1685		3934
1686	=item * Priorities are not currently supported. Initialising priorities	3935	=item * Priorities are not currently supported. Initialising priorities
1687	will fail and all watchers will have the same priority, even though there	3936	will fail and all watchers will have the same priority, even though there
1688	is an ev_pri field.	3937	is an ev_pri field.
1689		3938
		3939	=item * In libevent, the last base created gets the signals, in libev, the
		3940	base that registered the signal gets the signals.
		3941
1690	=item * Other members are not supported.	3942	=item * Other members are not supported.
1691		3943
1692	=item * The libev emulation is I<not> ABI compatible to libevent, you need	3944	=item * The libev emulation is I<not> ABI compatible to libevent, you need
1693	to use the libev header file and library.	3945	to use the libev header file and library.
1694		3946
1695	=back	3947	=back
1696		3948
1697	=head1 C++ SUPPORT	3949	=head1 C++ SUPPORT
1698		3950
		3951	=head2 C API
		3952
		3953	The normal C API should work fine when used from C++: both ev.h and the
		3954	libev sources can be compiled as C++. Therefore, code that uses the C API
		3955	will work fine.
		3956
		3957	Proper exception specifications might have to be added to callbacks passed
		3958	to libev: exceptions may be thrown only from watcher callbacks, all
		3959	other callbacks (allocator, syserr, loop acquire/release and periodic
		3960	reschedule callbacks) must not throw exceptions, and might need a C<throw
		3961	()> specification. If you have code that needs to be compiled as both C
		3962	and C++ you can use the C<EV_THROW> macro for this:
		3963
		3964	static void
		3965	fatal_error (const char *msg) EV_THROW
		3966	{
		3967	perror (msg);
		3968	abort ();
		3969	}
		3970
		3971	...
		3972	ev_set_syserr_cb (fatal_error);
		3973
		3974	The only API functions that can currently throw exceptions are C<ev_run>,
		3975	C<ev_invoke>, C<ev_invoke_pending> and C<ev_loop_destroy> (the latter
		3976	because it runs cleanup watchers).
		3977
		3978	Throwing exceptions in watcher callbacks is only supported if libev itself
		3979	is compiled with a C++ compiler or your C and C++ environments allow
		3980	throwing exceptions through C libraries (most do).
		3981
		3982	=head2 C++ API
		3983
1699	Libev comes with some simplistic wrapper classes for C++ that mainly allow	3984	Libev comes with some simplistic wrapper classes for C++ that mainly allow
1700	you to use some convinience methods to start/stop watchers and also change	3985	you to use some convenience methods to start/stop watchers and also change
1701	the callback model to a model using method callbacks on objects.	3986	the callback model to a model using method callbacks on objects.
1702		3987
1703	To use it,	3988	To use it,
1704		3989
1705	#include <ev++.h>	3990	#include <ev++.h>
1706		3991
1707	(it is not installed by default). This automatically includes F<ev.h>	3992	This automatically includes F<ev.h> and puts all of its definitions (many
1708	and puts all of its definitions (many of them macros) into the global	3993	of them macros) into the global namespace. All C++ specific things are
1709	namespace. All C++ specific things are put into the C<ev> namespace.	3994	put into the C<ev> namespace. It should support all the same embedding
		3995	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
1710		3996
1711	It should support all the same embedding options as F<ev.h>, most notably	3997	Care has been taken to keep the overhead low. The only data member the C++
1712	C<EV_MULTIPLICITY>.	3998	classes add (compared to plain C-style watchers) is the event loop pointer
		3999	that the watcher is associated with (or no additional members at all if
		4000	you disable C<EV_MULTIPLICITY> when embedding libev).
		4001
		4002	Currently, functions, static and non-static member functions and classes
		4003	with C<operator ()> can be used as callbacks. Other types should be easy
		4004	to add as long as they only need one additional pointer for context. If
		4005	you need support for other types of functors please contact the author
		4006	(preferably after implementing it).
		4007
		4008	For all this to work, your C++ compiler either has to use the same calling
		4009	conventions as your C compiler (for static member functions), or you have
		4010	to embed libev and compile libev itself as C++.
1713		4011
1714	Here is a list of things available in the C<ev> namespace:	4012	Here is a list of things available in the C<ev> namespace:
1715		4013
1716	=over 4	4014	=over 4
1717		4015
…		…
1727	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.	4025	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.
1728		4026
1729	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of	4027	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of
1730	the same name in the C<ev> namespace, with the exception of C<ev_signal>	4028	the same name in the C<ev> namespace, with the exception of C<ev_signal>
1731	which is called C<ev::sig> to avoid clashes with the C<signal> macro	4029	which is called C<ev::sig> to avoid clashes with the C<signal> macro
1732	defines by many implementations.	4030	defined by many implementations.
1733		4031
1734	All of those classes have these methods:	4032	All of those classes have these methods:
1735		4033
1736	=over 4	4034	=over 4
1737		4035
1738	=item ev::TYPE::TYPE (object *, object::method *)	4036	=item ev::TYPE::TYPE ()
1739		4037
1740	=item ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)	4038	=item ev::TYPE::TYPE (loop)
1741		4039
1742	=item ev::TYPE::~TYPE	4040	=item ev::TYPE::~TYPE
1743		4041
1744	The constructor takes a pointer to an object and a method pointer to	4042	The constructor (optionally) takes an event loop to associate the watcher
1745	the event handler callback to call in this class. The constructor calls	4043	with. If it is omitted, it will use C<EV_DEFAULT>.
1746	C<ev_init> for you, which means you have to call the C<set> method	4044
1747	before starting it. If you do not specify a loop then the constructor	4045	The constructor calls C<ev_init> for you, which means you have to call the
1748	automatically associates the default loop with this watcher.	4046	C<set> method before starting it.
		4047
		4048	It will not set a callback, however: You have to call the templated C<set>
		4049	method to set a callback before you can start the watcher.
		4050
		4051	(The reason why you have to use a method is a limitation in C++ which does
		4052	not allow explicit template arguments for constructors).
1749		4053
1750	The destructor automatically stops the watcher if it is active.	4054	The destructor automatically stops the watcher if it is active.
1751		4055
		4056	=item w->set<class, &class::method> (object *)
		4057
		4058	This method sets the callback method to call. The method has to have a
		4059	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		4060	first argument and the C<revents> as second. The object must be given as
		4061	parameter and is stored in the C<data> member of the watcher.
		4062
		4063	This method synthesizes efficient thunking code to call your method from
		4064	the C callback that libev requires. If your compiler can inline your
		4065	callback (i.e. it is visible to it at the place of the C<set> call and
		4066	your compiler is good :), then the method will be fully inlined into the
		4067	thunking function, making it as fast as a direct C callback.
		4068
		4069	Example: simple class declaration and watcher initialisation
		4070
		4071	struct myclass
		4072	{
		4073	void io_cb (ev::io &w, int revents) { }
		4074	}
		4075
		4076	myclass obj;
		4077	ev::io iow;
		4078	iow.set <myclass, &myclass::io_cb> (&obj);
		4079
		4080	=item w->set (object *)
		4081
		4082	This is a variation of a method callback - leaving out the method to call
		4083	will default the method to C<operator ()>, which makes it possible to use
		4084	functor objects without having to manually specify the C<operator ()> all
		4085	the time. Incidentally, you can then also leave out the template argument
		4086	list.
		4087
		4088	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		4089	int revents)>.
		4090
		4091	See the method-C<set> above for more details.
		4092
		4093	Example: use a functor object as callback.
		4094
		4095	struct myfunctor
		4096	{
		4097	void operator() (ev::io &w, int revents)
		4098	{
		4099	...
		4100	}
		4101	}
		4102
		4103	myfunctor f;
		4104
		4105	ev::io w;
		4106	w.set (&f);
		4107
		4108	=item w->set<function> (void *data = 0)
		4109
		4110	Also sets a callback, but uses a static method or plain function as
		4111	callback. The optional C<data> argument will be stored in the watcher's
		4112	C<data> member and is free for you to use.
		4113
		4114	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		4115
		4116	See the method-C<set> above for more details.
		4117
		4118	Example: Use a plain function as callback.
		4119
		4120	static void io_cb (ev::io &w, int revents) { }
		4121	iow.set <io_cb> ();
		4122
1752	=item w->set (struct ev_loop *)	4123	=item w->set (loop)
1753		4124
1754	Associates a different C<struct ev_loop> with this watcher. You can only	4125	Associates a different C<struct ev_loop> with this watcher. You can only
1755	do this when the watcher is inactive (and not pending either).	4126	do this when the watcher is inactive (and not pending either).
1756		4127
1757	=item w->set ([args])	4128	=item w->set ([arguments])
1758		4129
1759	Basically the same as C<ev_TYPE_set>, with the same args. Must be	4130	Basically the same as C<ev_TYPE_set> (except for C<ev::embed> watchers>),
		4131	with the same arguments. Either this method or a suitable start method
1760	called at least once. Unlike the C counterpart, an active watcher gets	4132	must be called at least once. Unlike the C counterpart, an active watcher
1761	automatically stopped and restarted.	4133	gets automatically stopped and restarted when reconfiguring it with this
		4134	method.
		4135
		4136	For C<ev::embed> watchers this method is called C<set_embed>, to avoid
		4137	clashing with the C<set (loop)> method.
1762		4138
1763	=item w->start ()	4139	=item w->start ()
1764		4140
1765	Starts the watcher. Note that there is no C<loop> argument as the	4141	Starts the watcher. Note that there is no C<loop> argument, as the
1766	constructor already takes the loop.	4142	constructor already stores the event loop.
		4143
		4144	=item w->start ([arguments])
		4145
		4146	Instead of calling C<set> and C<start> methods separately, it is often
		4147	convenient to wrap them in one call. Uses the same type of arguments as
		4148	the configure C<set> method of the watcher.
1767		4149
1768	=item w->stop ()	4150	=item w->stop ()
1769		4151
1770	Stops the watcher if it is active. Again, no C<loop> argument.	4152	Stops the watcher if it is active. Again, no C<loop> argument.
1771		4153
1772	=item w->again () C<ev::timer>, C<ev::periodic> only	4154	=item w->again () (C<ev::timer>, C<ev::periodic> only)
1773		4155
1774	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding	4156	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
1775	C<ev_TYPE_again> function.	4157	C<ev_TYPE_again> function.
1776		4158
1777	=item w->sweep () C<ev::embed> only	4159	=item w->sweep () (C<ev::embed> only)
1778		4160
1779	Invokes C<ev_embed_sweep>.	4161	Invokes C<ev_embed_sweep>.
1780		4162
1781	=item w->update () C<ev::stat> only	4163	=item w->update () (C<ev::stat> only)
1782		4164
1783	Invokes C<ev_stat_stat>.	4165	Invokes C<ev_stat_stat>.
1784		4166
1785	=back	4167	=back
1786		4168
1787	=back	4169	=back
1788		4170
1789	Example: Define a class with an IO and idle watcher, start one of them in	4171	Example: Define a class with two I/O and idle watchers, start the I/O
1790	the constructor.	4172	watchers in the constructor.
1791		4173
1792	class myclass	4174	class myclass
1793	{	4175	{
1794	ev_io io; void io_cb (ev::io &w, int revents);	4176	ev::io io ; void io_cb (ev::io &w, int revents);
		4177	ev::io io2 ; void io2_cb (ev::io &w, int revents);
1795	ev_idle idle void idle_cb (ev::idle &w, int revents);	4178	ev::idle idle; void idle_cb (ev::idle &w, int revents);
1796		4179
1797	myclass ();	4180	myclass (int fd)
		4181	{
		4182	io .set <myclass, &myclass::io_cb > (this);
		4183	io2 .set <myclass, &myclass::io2_cb > (this);
		4184	idle.set <myclass, &myclass::idle_cb> (this);
		4185
		4186	io.set (fd, ev::WRITE); // configure the watcher
		4187	io.start (); // start it whenever convenient
		4188
		4189	io2.start (fd, ev::READ); // set + start in one call
		4190	}
1798	}	4191	};
1799		4192
1800	myclass::myclass (int fd)	4193
1801	: io (this, &myclass::io_cb),	4194	=head1 OTHER LANGUAGE BINDINGS
1802	idle (this, &myclass::idle_cb)	4195
1803	{	4196	Libev does not offer other language bindings itself, but bindings for a
1804	io.start (fd, ev::READ);	4197	number of languages exist in the form of third-party packages. If you know
1805	}	4198	any interesting language binding in addition to the ones listed here, drop
		4199	me a note.
		4200
		4201	=over 4
		4202
		4203	=item Perl
		4204
		4205	The EV module implements the full libev API and is actually used to test
		4206	libev. EV is developed together with libev. Apart from the EV core module,
		4207	there are additional modules that implement libev-compatible interfaces
		4208	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
		4209	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		4210	and C<EV::Glib>).
		4211
		4212	It can be found and installed via CPAN, its homepage is at
		4213	L<http://software.schmorp.de/pkg/EV>.
		4214
		4215	=item Python
		4216
		4217	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		4218	seems to be quite complete and well-documented.
		4219
		4220	=item Ruby
		4221
		4222	Tony Arcieri has written a ruby extension that offers access to a subset
		4223	of the libev API and adds file handle abstractions, asynchronous DNS and
		4224	more on top of it. It can be found via gem servers. Its homepage is at
		4225	L<http://rev.rubyforge.org/>.
		4226
		4227	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		4228	makes rev work even on mingw.
		4229
		4230	=item Haskell
		4231
		4232	A haskell binding to libev is available at
		4233	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		4234
		4235	=item D
		4236
		4237	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		4238	be found at L<http://www.llucax.com.ar/proj/ev.d/index.html>.
		4239
		4240	=item Ocaml
		4241
		4242	Erkki Seppala has written Ocaml bindings for libev, to be found at
		4243	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		4244
		4245	=item Lua
		4246
		4247	Brian Maher has written a partial interface to libev for lua (at the
		4248	time of this writing, only C<ev_io> and C<ev_timer>), to be found at
		4249	L<http://github.com/brimworks/lua-ev>.
		4250
		4251	=item Javascript
		4252
		4253	Node.js (L<http://nodejs.org>) uses libev as the underlying event library.
		4254
		4255	=item Others
		4256
		4257	There are others, and I stopped counting.
		4258
		4259	=back
1806		4260
1807		4261
1808	=head1 MACRO MAGIC	4262	=head1 MACRO MAGIC
1809		4263
1810	Libev can be compiled with a variety of options, the most fundemantal is	4264	Libev can be compiled with a variety of options, the most fundamental
1811	C<EV_MULTIPLICITY>. This option determines wether (most) functions and	4265	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
1812	callbacks have an initial C<struct ev_loop *> argument.	4266	functions and callbacks have an initial C<struct ev_loop *> argument.
1813		4267
1814	To make it easier to write programs that cope with either variant, the	4268	To make it easier to write programs that cope with either variant, the
1815	following macros are defined:	4269	following macros are defined:
1816		4270
1817	=over 4	4271	=over 4
…		…
1820		4274
1821	This provides the loop I<argument> for functions, if one is required ("ev	4275	This provides the loop I<argument> for functions, if one is required ("ev
1822	loop argument"). The C<EV_A> form is used when this is the sole argument,	4276	loop argument"). The C<EV_A> form is used when this is the sole argument,
1823	C<EV_A_> is used when other arguments are following. Example:	4277	C<EV_A_> is used when other arguments are following. Example:
1824		4278
1825	ev_unref (EV_A);	4279	ev_unref (EV_A);
1826	ev_timer_add (EV_A_ watcher);	4280	ev_timer_add (EV_A_ watcher);
1827	ev_loop (EV_A_ 0);	4281	ev_run (EV_A_ 0);
1828		4282
1829	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	4283	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
1830	which is often provided by the following macro.	4284	which is often provided by the following macro.
1831		4285
1832	=item C<EV_P>, C<EV_P_>	4286	=item C<EV_P>, C<EV_P_>
1833		4287
1834	This provides the loop I<parameter> for functions, if one is required ("ev	4288	This provides the loop I<parameter> for functions, if one is required ("ev
1835	loop parameter"). The C<EV_P> form is used when this is the sole parameter,	4289	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
1836	C<EV_P_> is used when other parameters are following. Example:	4290	C<EV_P_> is used when other parameters are following. Example:
1837		4291
1838	// this is how ev_unref is being declared	4292	// this is how ev_unref is being declared
1839	static void ev_unref (EV_P);	4293	static void ev_unref (EV_P);
1840		4294
1841	// this is how you can declare your typical callback	4295	// this is how you can declare your typical callback
1842	static void cb (EV_P_ ev_timer *w, int revents)	4296	static void cb (EV_P_ ev_timer *w, int revents)
1843		4297
1844	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	4298	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
1845	suitable for use with C<EV_A>.	4299	suitable for use with C<EV_A>.
1846		4300
1847	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	4301	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
1848		4302
1849	Similar to the other two macros, this gives you the value of the default	4303	Similar to the other two macros, this gives you the value of the default
1850	loop, if multiple loops are supported ("ev loop default").	4304	loop, if multiple loops are supported ("ev loop default"). The default loop
		4305	will be initialised if it isn't already initialised.
		4306
		4307	For non-multiplicity builds, these macros do nothing, so you always have
		4308	to initialise the loop somewhere.
		4309
		4310	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		4311
		4312	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		4313	default loop has been initialised (C<UC> == unchecked). Their behaviour
		4314	is undefined when the default loop has not been initialised by a previous
		4315	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		4316
		4317	It is often prudent to use C<EV_DEFAULT> when initialising the first
		4318	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
1851		4319
1852	=back	4320	=back
1853		4321
1854	Example: Declare and initialise a check watcher, utilising the above	4322	Example: Declare and initialise a check watcher, utilising the above
1855	macros so it will work regardless of wether multiple loops are supported	4323	macros so it will work regardless of whether multiple loops are supported
1856	or not.	4324	or not.
1857		4325
1858	static void	4326	static void
1859	check_cb (EV_P_ ev_timer *w, int revents)	4327	check_cb (EV_P_ ev_timer *w, int revents)
1860	{	4328	{
1861	ev_check_stop (EV_A_ w);	4329	ev_check_stop (EV_A_ w);
1862	}	4330	}
1863		4331
1864	ev_check check;	4332	ev_check check;
1865	ev_check_init (&check, check_cb);	4333	ev_check_init (&check, check_cb);
1866	ev_check_start (EV_DEFAULT_ &check);	4334	ev_check_start (EV_DEFAULT_ &check);
1867	ev_loop (EV_DEFAULT_ 0);	4335	ev_run (EV_DEFAULT_ 0);
1868		4336
1869	=head1 EMBEDDING	4337	=head1 EMBEDDING
1870		4338
1871	Libev can (and often is) directly embedded into host	4339	Libev can (and often is) directly embedded into host
1872	applications. Examples of applications that embed it include the Deliantra	4340	applications. Examples of applications that embed it include the Deliantra
1873	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)	4341	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
1874	and rxvt-unicode.	4342	and rxvt-unicode.
1875		4343
1876	The goal is to enable you to just copy the neecssary files into your	4344	The goal is to enable you to just copy the necessary files into your
1877	source directory without having to change even a single line in them, so	4345	source directory without having to change even a single line in them, so
1878	you can easily upgrade by simply copying (or having a checked-out copy of	4346	you can easily upgrade by simply copying (or having a checked-out copy of
1879	libev somewhere in your source tree).	4347	libev somewhere in your source tree).
1880		4348
1881	=head2 FILESETS	4349	=head2 FILESETS
1882		4350
1883	Depending on what features you need you need to include one or more sets of files	4351	Depending on what features you need you need to include one or more sets of files
1884	in your app.	4352	in your application.
1885		4353
1886	=head3 CORE EVENT LOOP	4354	=head3 CORE EVENT LOOP
1887		4355
1888	To include only the libev core (all the C<ev_*> functions), with manual	4356	To include only the libev core (all the C<ev_*> functions), with manual
1889	configuration (no autoconf):	4357	configuration (no autoconf):
1890		4358
1891	#define EV_STANDALONE 1	4359	#define EV_STANDALONE 1
1892	#include "ev.c"	4360	#include "ev.c"
1893		4361
1894	This will automatically include F<ev.h>, too, and should be done in a	4362	This will automatically include F<ev.h>, too, and should be done in a
1895	single C source file only to provide the function implementations. To use	4363	single C source file only to provide the function implementations. To use
1896	it, do the same for F<ev.h> in all files wishing to use this API (best	4364	it, do the same for F<ev.h> in all files wishing to use this API (best
1897	done by writing a wrapper around F<ev.h> that you can include instead and	4365	done by writing a wrapper around F<ev.h> that you can include instead and
1898	where you can put other configuration options):	4366	where you can put other configuration options):
1899		4367
1900	#define EV_STANDALONE 1	4368	#define EV_STANDALONE 1
1901	#include "ev.h"	4369	#include "ev.h"
1902		4370
1903	Both header files and implementation files can be compiled with a C++	4371	Both header files and implementation files can be compiled with a C++
1904	compiler (at least, thats a stated goal, and breakage will be treated	4372	compiler (at least, that's a stated goal, and breakage will be treated
1905	as a bug).	4373	as a bug).
1906		4374
1907	You need the following files in your source tree, or in a directory	4375	You need the following files in your source tree, or in a directory
1908	in your include path (e.g. in libev/ when using -Ilibev):	4376	in your include path (e.g. in libev/ when using -Ilibev):
1909		4377
1910	ev.h	4378	ev.h
1911	ev.c	4379	ev.c
1912	ev_vars.h	4380	ev_vars.h
1913	ev_wrap.h	4381	ev_wrap.h
1914		4382
1915	ev_win32.c required on win32 platforms only	4383	ev_win32.c required on win32 platforms only
1916		4384
1917	ev_select.c only when select backend is enabled (which is enabled by default)	4385	ev_select.c only when select backend is enabled (which is enabled by default)
1918	ev_poll.c only when poll backend is enabled (disabled by default)	4386	ev_poll.c only when poll backend is enabled (disabled by default)
1919	ev_epoll.c only when the epoll backend is enabled (disabled by default)	4387	ev_epoll.c only when the epoll backend is enabled (disabled by default)
1920	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	4388	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
1921	ev_port.c only when the solaris port backend is enabled (disabled by default)	4389	ev_port.c only when the solaris port backend is enabled (disabled by default)
1922		4390
1923	F<ev.c> includes the backend files directly when enabled, so you only need	4391	F<ev.c> includes the backend files directly when enabled, so you only need
1924	to compile this single file.	4392	to compile this single file.
1925		4393
1926	=head3 LIBEVENT COMPATIBILITY API	4394	=head3 LIBEVENT COMPATIBILITY API
1927		4395
1928	To include the libevent compatibility API, also include:	4396	To include the libevent compatibility API, also include:
1929		4397
1930	#include "event.c"	4398	#include "event.c"
1931		4399
1932	in the file including F<ev.c>, and:	4400	in the file including F<ev.c>, and:
1933		4401
1934	#include "event.h"	4402	#include "event.h"
1935		4403
1936	in the files that want to use the libevent API. This also includes F<ev.h>.	4404	in the files that want to use the libevent API. This also includes F<ev.h>.
1937		4405
1938	You need the following additional files for this:	4406	You need the following additional files for this:
1939		4407
1940	event.h	4408	event.h
1941	event.c	4409	event.c
1942		4410
1943	=head3 AUTOCONF SUPPORT	4411	=head3 AUTOCONF SUPPORT
1944		4412
1945	Instead of using C<EV_STANDALONE=1> and providing your config in	4413	Instead of using C<EV_STANDALONE=1> and providing your configuration in
1946	whatever way you want, you can also C<m4_include([libev.m4])> in your	4414	whatever way you want, you can also C<m4_include([libev.m4])> in your
1947	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then	4415	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
1948	include F<config.h> and configure itself accordingly.	4416	include F<config.h> and configure itself accordingly.
1949		4417
1950	For this of course you need the m4 file:	4418	For this of course you need the m4 file:
1951		4419
1952	libev.m4	4420	libev.m4
1953		4421
1954	=head2 PREPROCESSOR SYMBOLS/MACROS	4422	=head2 PREPROCESSOR SYMBOLS/MACROS
1955		4423
1956	Libev can be configured via a variety of preprocessor symbols you have to define	4424	Libev can be configured via a variety of preprocessor symbols you have to
1957	before including any of its files. The default is not to build for multiplicity	4425	define before including (or compiling) any of its files. The default in
1958	and only include the select backend.	4426	the absence of autoconf is documented for every option.
		4427
		4428	Symbols marked with "(h)" do not change the ABI, and can have different
		4429	values when compiling libev vs. including F<ev.h>, so it is permissible
		4430	to redefine them before including F<ev.h> without breaking compatibility
		4431	to a compiled library. All other symbols change the ABI, which means all
		4432	users of libev and the libev code itself must be compiled with compatible
		4433	settings.
1959		4434
1960	=over 4	4435	=over 4
1961		4436
		4437	=item EV_COMPAT3 (h)
		4438
		4439	Backwards compatibility is a major concern for libev. This is why this
		4440	release of libev comes with wrappers for the functions and symbols that
		4441	have been renamed between libev version 3 and 4.
		4442
		4443	You can disable these wrappers (to test compatibility with future
		4444	versions) by defining C<EV_COMPAT3> to C<0> when compiling your
		4445	sources. This has the additional advantage that you can drop the C<struct>
		4446	from C<struct ev_loop> declarations, as libev will provide an C<ev_loop>
		4447	typedef in that case.
		4448
		4449	In some future version, the default for C<EV_COMPAT3> will become C<0>,
		4450	and in some even more future version the compatibility code will be
		4451	removed completely.
		4452
1962	=item EV_STANDALONE	4453	=item EV_STANDALONE (h)
1963		4454
1964	Must always be C<1> if you do not use autoconf configuration, which	4455	Must always be C<1> if you do not use autoconf configuration, which
1965	keeps libev from including F<config.h>, and it also defines dummy	4456	keeps libev from including F<config.h>, and it also defines dummy
1966	implementations for some libevent functions (such as logging, which is not	4457	implementations for some libevent functions (such as logging, which is not
1967	supported). It will also not define any of the structs usually found in	4458	supported). It will also not define any of the structs usually found in
1968	F<event.h> that are not directly supported by the libev core alone.	4459	F<event.h> that are not directly supported by the libev core alone.
1969		4460
		4461	In standalone mode, libev will still try to automatically deduce the
		4462	configuration, but has to be more conservative.
		4463
		4464	=item EV_USE_FLOOR
		4465
		4466	If defined to be C<1>, libev will use the C<floor ()> function for its
		4467	periodic reschedule calculations, otherwise libev will fall back on a
		4468	portable (slower) implementation. If you enable this, you usually have to
		4469	link against libm or something equivalent. Enabling this when the C<floor>
		4470	function is not available will fail, so the safe default is to not enable
		4471	this.
		4472
1970	=item EV_USE_MONOTONIC	4473	=item EV_USE_MONOTONIC
1971		4474
1972	If defined to be C<1>, libev will try to detect the availability of the	4475	If defined to be C<1>, libev will try to detect the availability of the
1973	monotonic clock option at both compiletime and runtime. Otherwise no use	4476	monotonic clock option at both compile time and runtime. Otherwise no
1974	of the monotonic clock option will be attempted. If you enable this, you	4477	use of the monotonic clock option will be attempted. If you enable this,
1975	usually have to link against librt or something similar. Enabling it when	4478	you usually have to link against librt or something similar. Enabling it
1976	the functionality isn't available is safe, though, althoguh you have	4479	when the functionality isn't available is safe, though, although you have
1977	to make sure you link against any libraries where the C<clock_gettime>	4480	to make sure you link against any libraries where the C<clock_gettime>
1978	function is hiding in (often F<-lrt>).	4481	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
1979		4482
1980	=item EV_USE_REALTIME	4483	=item EV_USE_REALTIME
1981		4484
1982	If defined to be C<1>, libev will try to detect the availability of the	4485	If defined to be C<1>, libev will try to detect the availability of the
1983	realtime clock option at compiletime (and assume its availability at	4486	real-time clock option at compile time (and assume its availability
1984	runtime if successful). Otherwise no use of the realtime clock option will	4487	at runtime if successful). Otherwise no use of the real-time clock
1985	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	4488	option will be attempted. This effectively replaces C<gettimeofday>
1986	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See tzhe note about libraries	4489	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
1987	in the description of C<EV_USE_MONOTONIC>, though.	4490	correctness. See the note about libraries in the description of
		4491	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		4492	C<EV_USE_CLOCK_SYSCALL>.
		4493
		4494	=item EV_USE_CLOCK_SYSCALL
		4495
		4496	If defined to be C<1>, libev will try to use a direct syscall instead
		4497	of calling the system-provided C<clock_gettime> function. This option
		4498	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		4499	unconditionally pulls in C<libpthread>, slowing down single-threaded
		4500	programs needlessly. Using a direct syscall is slightly slower (in
		4501	theory), because no optimised vdso implementation can be used, but avoids
		4502	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		4503	higher, as it simplifies linking (no need for C<-lrt>).
		4504
		4505	=item EV_USE_NANOSLEEP
		4506
		4507	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		4508	and will use it for delays. Otherwise it will use C<select ()>.
		4509
		4510	=item EV_USE_EVENTFD
		4511
		4512	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		4513	available and will probe for kernel support at runtime. This will improve
		4514	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		4515	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		4516	2.7 or newer, otherwise disabled.
1988		4517
1989	=item EV_USE_SELECT	4518	=item EV_USE_SELECT
1990		4519
1991	If undefined or defined to be C<1>, libev will compile in support for the	4520	If undefined or defined to be C<1>, libev will compile in support for the
1992	C<select>(2) backend. No attempt at autodetection will be done: if no	4521	C<select>(2) backend. No attempt at auto-detection will be done: if no
1993	other method takes over, select will be it. Otherwise the select backend	4522	other method takes over, select will be it. Otherwise the select backend
1994	will not be compiled in.	4523	will not be compiled in.
1995		4524
1996	=item EV_SELECT_USE_FD_SET	4525	=item EV_SELECT_USE_FD_SET
1997		4526
1998	If defined to C<1>, then the select backend will use the system C<fd_set>	4527	If defined to C<1>, then the select backend will use the system C<fd_set>
1999	structure. This is useful if libev doesn't compile due to a missing	4528	structure. This is useful if libev doesn't compile due to a missing
2000	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on	4529	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
2001	exotic systems. This usually limits the range of file descriptors to some	4530	on exotic systems. This usually limits the range of file descriptors to
2002	low limit such as 1024 or might have other limitations (winsocket only	4531	some low limit such as 1024 or might have other limitations (winsocket
2003	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	4532	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
2004	influence the size of the C<fd_set> used.	4533	configures the maximum size of the C<fd_set>.
2005		4534
2006	=item EV_SELECT_IS_WINSOCKET	4535	=item EV_SELECT_IS_WINSOCKET
2007		4536
2008	When defined to C<1>, the select backend will assume that	4537	When defined to C<1>, the select backend will assume that
2009	select/socket/connect etc. don't understand file descriptors but	4538	select/socket/connect etc. don't understand file descriptors but
…		…
2011	be used is the winsock select). This means that it will call	4540	be used is the winsock select). This means that it will call
2012	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	4541	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
2013	it is assumed that all these functions actually work on fds, even	4542	it is assumed that all these functions actually work on fds, even
2014	on win32. Should not be defined on non-win32 platforms.	4543	on win32. Should not be defined on non-win32 platforms.
2015		4544
		4545	=item EV_FD_TO_WIN32_HANDLE(fd)
		4546
		4547	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		4548	file descriptors to socket handles. When not defining this symbol (the
		4549	default), then libev will call C<_get_osfhandle>, which is usually
		4550	correct. In some cases, programs use their own file descriptor management,
		4551	in which case they can provide this function to map fds to socket handles.
		4552
		4553	=item EV_WIN32_HANDLE_TO_FD(handle)
		4554
		4555	If C<EV_SELECT_IS_WINSOCKET> then libev maps handles to file descriptors
		4556	using the standard C<_open_osfhandle> function. For programs implementing
		4557	their own fd to handle mapping, overwriting this function makes it easier
		4558	to do so. This can be done by defining this macro to an appropriate value.
		4559
		4560	=item EV_WIN32_CLOSE_FD(fd)
		4561
		4562	If programs implement their own fd to handle mapping on win32, then this
		4563	macro can be used to override the C<close> function, useful to unregister
		4564	file descriptors again. Note that the replacement function has to close
		4565	the underlying OS handle.
		4566
		4567	=item EV_USE_WSASOCKET
		4568
		4569	If defined to be C<1>, libev will use C<WSASocket> to create its internal
		4570	communication socket, which works better in some environments. Otherwise,
		4571	the normal C<socket> function will be used, which works better in other
		4572	environments.
		4573
2016	=item EV_USE_POLL	4574	=item EV_USE_POLL
2017		4575
2018	If defined to be C<1>, libev will compile in support for the C<poll>(2)	4576	If defined to be C<1>, libev will compile in support for the C<poll>(2)
2019	backend. Otherwise it will be enabled on non-win32 platforms. It	4577	backend. Otherwise it will be enabled on non-win32 platforms. It
2020	takes precedence over select.	4578	takes precedence over select.
2021		4579
2022	=item EV_USE_EPOLL	4580	=item EV_USE_EPOLL
2023		4581
2024	If defined to be C<1>, libev will compile in support for the Linux	4582	If defined to be C<1>, libev will compile in support for the Linux
2025	C<epoll>(7) backend. Its availability will be detected at runtime,	4583	C<epoll>(7) backend. Its availability will be detected at runtime,
2026	otherwise another method will be used as fallback. This is the	4584	otherwise another method will be used as fallback. This is the preferred
2027	preferred backend for GNU/Linux systems.	4585	backend for GNU/Linux systems. If undefined, it will be enabled if the
		4586	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2028		4587
2029	=item EV_USE_KQUEUE	4588	=item EV_USE_KQUEUE
2030		4589
2031	If defined to be C<1>, libev will compile in support for the BSD style	4590	If defined to be C<1>, libev will compile in support for the BSD style
2032	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	4591	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
2045	otherwise another method will be used as fallback. This is the preferred	4604	otherwise another method will be used as fallback. This is the preferred
2046	backend for Solaris 10 systems.	4605	backend for Solaris 10 systems.
2047		4606
2048	=item EV_USE_DEVPOLL	4607	=item EV_USE_DEVPOLL
2049		4608
2050	reserved for future expansion, works like the USE symbols above.	4609	Reserved for future expansion, works like the USE symbols above.
2051		4610
2052	=item EV_USE_INOTIFY	4611	=item EV_USE_INOTIFY
2053		4612
2054	If defined to be C<1>, libev will compile in support for the Linux inotify	4613	If defined to be C<1>, libev will compile in support for the Linux inotify
2055	interface to speed up C<ev_stat> watchers. Its actual availability will	4614	interface to speed up C<ev_stat> watchers. Its actual availability will
2056	be detected at runtime.	4615	be detected at runtime. If undefined, it will be enabled if the headers
		4616	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2057		4617
		4618	=item EV_NO_SMP
		4619
		4620	If defined to be C<1>, libev will assume that memory is always coherent
		4621	between threads, that is, threads can be used, but threads never run on
		4622	different cpus (or different cpu cores). This reduces dependencies
		4623	and makes libev faster.
		4624
		4625	=item EV_NO_THREADS
		4626
		4627	If defined to be C<1>, libev will assume that it will never be called from
		4628	different threads (that includes signal handlers), which is a stronger
		4629	assumption than C<EV_NO_SMP>, above. This reduces dependencies and makes
		4630	libev faster.
		4631
		4632	=item EV_ATOMIC_T
		4633
		4634	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		4635	access is atomic with respect to other threads or signal contexts. No
		4636	such type is easily found in the C language, so you can provide your own
		4637	type that you know is safe for your purposes. It is used both for signal
		4638	handler "locking" as well as for signal and thread safety in C<ev_async>
		4639	watchers.
		4640
		4641	In the absence of this define, libev will use C<sig_atomic_t volatile>
		4642	(from F<signal.h>), which is usually good enough on most platforms.
		4643
2058	=item EV_H	4644	=item EV_H (h)
2059		4645
2060	The name of the F<ev.h> header file used to include it. The default if	4646	The name of the F<ev.h> header file used to include it. The default if
2061	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This	4647	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2062	can be used to virtually rename the F<ev.h> header file in case of conflicts.	4648	used to virtually rename the F<ev.h> header file in case of conflicts.
2063		4649
2064	=item EV_CONFIG_H	4650	=item EV_CONFIG_H (h)
2065		4651
2066	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	4652	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2067	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	4653	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2068	C<EV_H>, above.	4654	C<EV_H>, above.
2069		4655
2070	=item EV_EVENT_H	4656	=item EV_EVENT_H (h)
2071		4657
2072	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	4658	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2073	of how the F<event.h> header can be found.	4659	of how the F<event.h> header can be found, the default is C<"event.h">.
2074		4660
2075	=item EV_PROTOTYPES	4661	=item EV_PROTOTYPES (h)
2076		4662
2077	If defined to be C<0>, then F<ev.h> will not define any function	4663	If defined to be C<0>, then F<ev.h> will not define any function
2078	prototypes, but still define all the structs and other symbols. This is	4664	prototypes, but still define all the structs and other symbols. This is
2079	occasionally useful if you want to provide your own wrapper functions	4665	occasionally useful if you want to provide your own wrapper functions
2080	around libev functions.	4666	around libev functions.
…		…
2085	will have the C<struct ev_loop *> as first argument, and you can create	4671	will have the C<struct ev_loop *> as first argument, and you can create
2086	additional independent event loops. Otherwise there will be no support	4672	additional independent event loops. Otherwise there will be no support
2087	for multiple event loops and there is no first event loop pointer	4673	for multiple event loops and there is no first event loop pointer
2088	argument. Instead, all functions act on the single default loop.	4674	argument. Instead, all functions act on the single default loop.
2089		4675
2090	=item EV_PERIODIC_ENABLE	4676	Note that C<EV_DEFAULT> and C<EV_DEFAULT_> will no longer provide a
		4677	default loop when multiplicity is switched off - you always have to
		4678	initialise the loop manually in this case.
2091		4679
2092	If undefined or defined to be C<1>, then periodic timers are supported. If
2093	defined to be C<0>, then they are not. Disabling them saves a few kB of
2094	code.
2095
2096	=item EV_EMBED_ENABLE
2097
2098	If undefined or defined to be C<1>, then embed watchers are supported. If
2099	defined to be C<0>, then they are not.
2100
2101	=item EV_STAT_ENABLE
2102
2103	If undefined or defined to be C<1>, then stat watchers are supported. If
2104	defined to be C<0>, then they are not.
2105
2106	=item EV_FORK_ENABLE
2107
2108	If undefined or defined to be C<1>, then fork watchers are supported. If
2109	defined to be C<0>, then they are not.
2110
2111	=item EV_MINIMAL	4680	=item EV_MINPRI
		4681
		4682	=item EV_MAXPRI
		4683
		4684	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		4685	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		4686	provide for more priorities by overriding those symbols (usually defined
		4687	to be C<-2> and C<2>, respectively).
		4688
		4689	When doing priority-based operations, libev usually has to linearly search
		4690	all the priorities, so having many of them (hundreds) uses a lot of space
		4691	and time, so using the defaults of five priorities (-2 .. +2) is usually
		4692	fine.
		4693
		4694	If your embedding application does not need any priorities, defining these
		4695	both to C<0> will save some memory and CPU.
		4696
		4697	=item EV_PERIODIC_ENABLE, EV_IDLE_ENABLE, EV_EMBED_ENABLE, EV_STAT_ENABLE,
		4698	EV_PREPARE_ENABLE, EV_CHECK_ENABLE, EV_FORK_ENABLE, EV_SIGNAL_ENABLE,
		4699	EV_ASYNC_ENABLE, EV_CHILD_ENABLE.
		4700
		4701	If undefined or defined to be C<1> (and the platform supports it), then
		4702	the respective watcher type is supported. If defined to be C<0>, then it
		4703	is not. Disabling watcher types mainly saves code size.
		4704
		4705	=item EV_FEATURES
2112		4706
2113	If you need to shave off some kilobytes of code at the expense of some	4707	If you need to shave off some kilobytes of code at the expense of some
2114	speed, define this symbol to C<1>. Currently only used for gcc to override	4708	speed (but with the full API), you can define this symbol to request
2115	some inlining decisions, saves roughly 30% codesize of amd64.	4709	certain subsets of functionality. The default is to enable all features
		4710	that can be enabled on the platform.
		4711
		4712	A typical way to use this symbol is to define it to C<0> (or to a bitset
		4713	with some broad features you want) and then selectively re-enable
		4714	additional parts you want, for example if you want everything minimal,
		4715	but multiple event loop support, async and child watchers and the poll
		4716	backend, use this:
		4717
		4718	#define EV_FEATURES 0
		4719	#define EV_MULTIPLICITY 1
		4720	#define EV_USE_POLL 1
		4721	#define EV_CHILD_ENABLE 1
		4722	#define EV_ASYNC_ENABLE 1
		4723
		4724	The actual value is a bitset, it can be a combination of the following
		4725	values (by default, all of these are enabled):
		4726
		4727	=over 4
		4728
		4729	=item C<1> - faster/larger code
		4730
		4731	Use larger code to speed up some operations.
		4732
		4733	Currently this is used to override some inlining decisions (enlarging the
		4734	code size by roughly 30% on amd64).
		4735
		4736	When optimising for size, use of compiler flags such as C<-Os> with
		4737	gcc is recommended, as well as C<-DNDEBUG>, as libev contains a number of
		4738	assertions.
		4739
		4740	The default is off when C<__OPTIMIZE_SIZE__> is defined by your compiler
		4741	(e.g. gcc with C<-Os>).
		4742
		4743	=item C<2> - faster/larger data structures
		4744
		4745	Replaces the small 2-heap for timer management by a faster 4-heap, larger
		4746	hash table sizes and so on. This will usually further increase code size
		4747	and can additionally have an effect on the size of data structures at
		4748	runtime.
		4749
		4750	The default is off when C<__OPTIMIZE_SIZE__> is defined by your compiler
		4751	(e.g. gcc with C<-Os>).
		4752
		4753	=item C<4> - full API configuration
		4754
		4755	This enables priorities (sets C<EV_MAXPRI>=2 and C<EV_MINPRI>=-2), and
		4756	enables multiplicity (C<EV_MULTIPLICITY>=1).
		4757
		4758	=item C<8> - full API
		4759
		4760	This enables a lot of the "lesser used" API functions. See C<ev.h> for
		4761	details on which parts of the API are still available without this
		4762	feature, and do not complain if this subset changes over time.
		4763
		4764	=item C<16> - enable all optional watcher types
		4765
		4766	Enables all optional watcher types. If you want to selectively enable
		4767	only some watcher types other than I/O and timers (e.g. prepare,
		4768	embed, async, child...) you can enable them manually by defining
		4769	C<EV_watchertype_ENABLE> to C<1> instead.
		4770
		4771	=item C<32> - enable all backends
		4772
		4773	This enables all backends - without this feature, you need to enable at
		4774	least one backend manually (C<EV_USE_SELECT> is a good choice).
		4775
		4776	=item C<64> - enable OS-specific "helper" APIs
		4777
		4778	Enable inotify, eventfd, signalfd and similar OS-specific helper APIs by
		4779	default.
		4780
		4781	=back
		4782
		4783	Compiling with C<gcc -Os -DEV_STANDALONE -DEV_USE_EPOLL=1 -DEV_FEATURES=0>
		4784	reduces the compiled size of libev from 24.7Kb code/2.8Kb data to 6.5Kb
		4785	code/0.3Kb data on my GNU/Linux amd64 system, while still giving you I/O
		4786	watchers, timers and monotonic clock support.
		4787
		4788	With an intelligent-enough linker (gcc+binutils are intelligent enough
		4789	when you use C<-Wl,--gc-sections -ffunction-sections>) functions unused by
		4790	your program might be left out as well - a binary starting a timer and an
		4791	I/O watcher then might come out at only 5Kb.
		4792
		4793	=item EV_API_STATIC
		4794
		4795	If this symbol is defined (by default it is not), then all identifiers
		4796	will have static linkage. This means that libev will not export any
		4797	identifiers, and you cannot link against libev anymore. This can be useful
		4798	when you embed libev, only want to use libev functions in a single file,
		4799	and do not want its identifiers to be visible.
		4800
		4801	To use this, define C<EV_API_STATIC> and include F<ev.c> in the file that
		4802	wants to use libev.
		4803
		4804	This option only works when libev is compiled with a C compiler, as C++
		4805	doesn't support the required declaration syntax.
		4806
		4807	=item EV_AVOID_STDIO
		4808
		4809	If this is set to C<1> at compiletime, then libev will avoid using stdio
		4810	functions (printf, scanf, perror etc.). This will increase the code size
		4811	somewhat, but if your program doesn't otherwise depend on stdio and your
		4812	libc allows it, this avoids linking in the stdio library which is quite
		4813	big.
		4814
		4815	Note that error messages might become less precise when this option is
		4816	enabled.
		4817
		4818	=item EV_NSIG
		4819
		4820	The highest supported signal number, +1 (or, the number of
		4821	signals): Normally, libev tries to deduce the maximum number of signals
		4822	automatically, but sometimes this fails, in which case it can be
		4823	specified. Also, using a lower number than detected (C<32> should be
		4824	good for about any system in existence) can save some memory, as libev
		4825	statically allocates some 12-24 bytes per signal number.
2116		4826
2117	=item EV_PID_HASHSIZE	4827	=item EV_PID_HASHSIZE
2118		4828
2119	C<ev_child> watchers use a small hash table to distribute workload by	4829	C<ev_child> watchers use a small hash table to distribute workload by
2120	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	4830	pid. The default size is C<16> (or C<1> with C<EV_FEATURES> disabled),
2121	than enough. If you need to manage thousands of children you might want to	4831	usually more than enough. If you need to manage thousands of children you
2122	increase this value (I<must> be a power of two).	4832	might want to increase this value (I<must> be a power of two).
2123		4833
2124	=item EV_INOTIFY_HASHSIZE	4834	=item EV_INOTIFY_HASHSIZE
2125		4835
2126	C<ev_staz> watchers use a small hash table to distribute workload by	4836	C<ev_stat> watchers use a small hash table to distribute workload by
2127	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	4837	inotify watch id. The default size is C<16> (or C<1> with C<EV_FEATURES>
2128	usually more than enough. If you need to manage thousands of C<ev_stat>	4838	disabled), usually more than enough. If you need to manage thousands of
2129	watchers you might want to increase this value (I<must> be a power of	4839	C<ev_stat> watchers you might want to increase this value (I<must> be a
2130	two).	4840	power of two).
		4841
		4842	=item EV_USE_4HEAP
		4843
		4844	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4845	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		4846	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		4847	faster performance with many (thousands) of watchers.
		4848
		4849	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4850	will be C<0>.
		4851
		4852	=item EV_HEAP_CACHE_AT
		4853
		4854	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		4855	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		4856	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		4857	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		4858	but avoids random read accesses on heap changes. This improves performance
		4859	noticeably with many (hundreds) of watchers.
		4860
		4861	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4862	will be C<0>.
		4863
		4864	=item EV_VERIFY
		4865
		4866	Controls how much internal verification (see C<ev_verify ()>) will
		4867	be done: If set to C<0>, no internal verification code will be compiled
		4868	in. If set to C<1>, then verification code will be compiled in, but not
		4869	called. If set to C<2>, then the internal verification code will be
		4870	called once per loop, which can slow down libev. If set to C<3>, then the
		4871	verification code will be called very frequently, which will slow down
		4872	libev considerably.
		4873
		4874	The default is C<1>, unless C<EV_FEATURES> overrides it, in which case it
		4875	will be C<0>.
2131		4876
2132	=item EV_COMMON	4877	=item EV_COMMON
2133		4878
2134	By default, all watchers have a C<void *data> member. By redefining	4879	By default, all watchers have a C<void *data> member. By redefining
2135	this macro to a something else you can include more and other types of	4880	this macro to something else you can include more and other types of
2136	members. You have to define it each time you include one of the files,	4881	members. You have to define it each time you include one of the files,
2137	though, and it must be identical each time.	4882	though, and it must be identical each time.
2138		4883
2139	For example, the perl EV module uses something like this:	4884	For example, the perl EV module uses something like this:
2140		4885
2141	#define EV_COMMON \	4886	#define EV_COMMON \
2142	SV *self; /* contains this struct */ \	4887	SV *self; /* contains this struct */ \
2143	SV *cb_sv, *fh /* note no trailing ";" */	4888	SV *cb_sv, *fh /* note no trailing ";" */
2144		4889
2145	=item EV_CB_DECLARE (type)	4890	=item EV_CB_DECLARE (type)
2146		4891
2147	=item EV_CB_INVOKE (watcher, revents)	4892	=item EV_CB_INVOKE (watcher, revents)
2148		4893
2149	=item ev_set_cb (ev, cb)	4894	=item ev_set_cb (ev, cb)
2150		4895
2151	Can be used to change the callback member declaration in each watcher,	4896	Can be used to change the callback member declaration in each watcher,
2152	and the way callbacks are invoked and set. Must expand to a struct member	4897	and the way callbacks are invoked and set. Must expand to a struct member
2153	definition and a statement, respectively. See the F<ev.v> header file for	4898	definition and a statement, respectively. See the F<ev.h> header file for
2154	their default definitions. One possible use for overriding these is to	4899	their default definitions. One possible use for overriding these is to
2155	avoid the C<struct ev_loop *> as first argument in all cases, or to use	4900	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2156	method calls instead of plain function calls in C++.	4901	method calls instead of plain function calls in C++.
		4902
		4903	=back
		4904
		4905	=head2 EXPORTED API SYMBOLS
		4906
		4907	If you need to re-export the API (e.g. via a DLL) and you need a list of
		4908	exported symbols, you can use the provided F<Symbol.*> files which list
		4909	all public symbols, one per line:
		4910
		4911	Symbols.ev for libev proper
		4912	Symbols.event for the libevent emulation
		4913
		4914	This can also be used to rename all public symbols to avoid clashes with
		4915	multiple versions of libev linked together (which is obviously bad in
		4916	itself, but sometimes it is inconvenient to avoid this).
		4917
		4918	A sed command like this will create wrapper C<#define>'s that you need to
		4919	include before including F<ev.h>:
		4920
		4921	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		4922
		4923	This would create a file F<wrap.h> which essentially looks like this:
		4924
		4925	#define ev_backend myprefix_ev_backend
		4926	#define ev_check_start myprefix_ev_check_start
		4927	#define ev_check_stop myprefix_ev_check_stop
		4928	...
2157		4929
2158	=head2 EXAMPLES	4930	=head2 EXAMPLES
2159		4931
2160	For a real-world example of a program the includes libev	4932	For a real-world example of a program the includes libev
2161	verbatim, you can have a look at the EV perl module	4933	verbatim, you can have a look at the EV perl module
…		…
2166	file.	4938	file.
2167		4939
2168	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	4940	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2169	that everybody includes and which overrides some configure choices:	4941	that everybody includes and which overrides some configure choices:
2170		4942
2171	#define EV_MINIMAL 1	4943	#define EV_FEATURES 8
2172	#define EV_USE_POLL 0	4944	#define EV_USE_SELECT 1
2173	#define EV_MULTIPLICITY 0
2174	#define EV_PERIODIC_ENABLE 0	4945	#define EV_PREPARE_ENABLE 1
		4946	#define EV_IDLE_ENABLE 1
2175	#define EV_STAT_ENABLE 0	4947	#define EV_SIGNAL_ENABLE 1
2176	#define EV_FORK_ENABLE 0	4948	#define EV_CHILD_ENABLE 1
		4949	#define EV_USE_STDEXCEPT 0
2177	#define EV_CONFIG_H <config.h>	4950	#define EV_CONFIG_H <config.h>
2178	#define EV_MINPRI 0
2179	#define EV_MAXPRI 0
2180		4951
2181	#include "ev++.h"	4952	#include "ev++.h"
2182		4953
2183	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	4954	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
2184		4955
2185	#include "ev_cpp.h"	4956	#include "ev_cpp.h"
2186	#include "ev.c"	4957	#include "ev.c"
2187		4958
		4959	=head1 INTERACTION WITH OTHER PROGRAMS, LIBRARIES OR THE ENVIRONMENT
2188		4960
		4961	=head2 THREADS AND COROUTINES
		4962
		4963	=head3 THREADS
		4964
		4965	All libev functions are reentrant and thread-safe unless explicitly
		4966	documented otherwise, but libev implements no locking itself. This means
		4967	that you can use as many loops as you want in parallel, as long as there
		4968	are no concurrent calls into any libev function with the same loop
		4969	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		4970	of course): libev guarantees that different event loops share no data
		4971	structures that need any locking.
		4972
		4973	Or to put it differently: calls with different loop parameters can be done
		4974	concurrently from multiple threads, calls with the same loop parameter
		4975	must be done serially (but can be done from different threads, as long as
		4976	only one thread ever is inside a call at any point in time, e.g. by using
		4977	a mutex per loop).
		4978
		4979	Specifically to support threads (and signal handlers), libev implements
		4980	so-called C<ev_async> watchers, which allow some limited form of
		4981	concurrency on the same event loop, namely waking it up "from the
		4982	outside".
		4983
		4984	If you want to know which design (one loop, locking, or multiple loops
		4985	without or something else still) is best for your problem, then I cannot
		4986	help you, but here is some generic advice:
		4987
		4988	=over 4
		4989
		4990	=item * most applications have a main thread: use the default libev loop
		4991	in that thread, or create a separate thread running only the default loop.
		4992
		4993	This helps integrating other libraries or software modules that use libev
		4994	themselves and don't care/know about threading.
		4995
		4996	=item * one loop per thread is usually a good model.
		4997
		4998	Doing this is almost never wrong, sometimes a better-performance model
		4999	exists, but it is always a good start.
		5000
		5001	=item * other models exist, such as the leader/follower pattern, where one
		5002	loop is handed through multiple threads in a kind of round-robin fashion.
		5003
		5004	Choosing a model is hard - look around, learn, know that usually you can do
		5005	better than you currently do :-)
		5006
		5007	=item * often you need to talk to some other thread which blocks in the
		5008	event loop.
		5009
		5010	C<ev_async> watchers can be used to wake them up from other threads safely
		5011	(or from signal contexts...).
		5012
		5013	An example use would be to communicate signals or other events that only
		5014	work in the default loop by registering the signal watcher with the
		5015	default loop and triggering an C<ev_async> watcher from the default loop
		5016	watcher callback into the event loop interested in the signal.
		5017
		5018	=back
		5019
		5020	See also L</THREAD LOCKING EXAMPLE>.
		5021
		5022	=head3 COROUTINES
		5023
		5024	Libev is very accommodating to coroutines ("cooperative threads"):
		5025	libev fully supports nesting calls to its functions from different
		5026	coroutines (e.g. you can call C<ev_run> on the same loop from two
		5027	different coroutines, and switch freely between both coroutines running
		5028	the loop, as long as you don't confuse yourself). The only exception is
		5029	that you must not do this from C<ev_periodic> reschedule callbacks.
		5030
		5031	Care has been taken to ensure that libev does not keep local state inside
		5032	C<ev_run>, and other calls do not usually allow for coroutine switches as
		5033	they do not call any callbacks.
		5034
		5035	=head2 COMPILER WARNINGS
		5036
		5037	Depending on your compiler and compiler settings, you might get no or a
		5038	lot of warnings when compiling libev code. Some people are apparently
		5039	scared by this.
		5040
		5041	However, these are unavoidable for many reasons. For one, each compiler
		5042	has different warnings, and each user has different tastes regarding
		5043	warning options. "Warn-free" code therefore cannot be a goal except when
		5044	targeting a specific compiler and compiler-version.
		5045
		5046	Another reason is that some compiler warnings require elaborate
		5047	workarounds, or other changes to the code that make it less clear and less
		5048	maintainable.
		5049
		5050	And of course, some compiler warnings are just plain stupid, or simply
		5051	wrong (because they don't actually warn about the condition their message
		5052	seems to warn about). For example, certain older gcc versions had some
		5053	warnings that resulted in an extreme number of false positives. These have
		5054	been fixed, but some people still insist on making code warn-free with
		5055	such buggy versions.
		5056
		5057	While libev is written to generate as few warnings as possible,
		5058	"warn-free" code is not a goal, and it is recommended not to build libev
		5059	with any compiler warnings enabled unless you are prepared to cope with
		5060	them (e.g. by ignoring them). Remember that warnings are just that:
		5061	warnings, not errors, or proof of bugs.
		5062
		5063
		5064	=head2 VALGRIND
		5065
		5066	Valgrind has a special section here because it is a popular tool that is
		5067	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		5068
		5069	If you think you found a bug (memory leak, uninitialised data access etc.)
		5070	in libev, then check twice: If valgrind reports something like:
		5071
		5072	==2274== definitely lost: 0 bytes in 0 blocks.
		5073	==2274== possibly lost: 0 bytes in 0 blocks.
		5074	==2274== still reachable: 256 bytes in 1 blocks.
		5075
		5076	Then there is no memory leak, just as memory accounted to global variables
		5077	is not a memleak - the memory is still being referenced, and didn't leak.
		5078
		5079	Similarly, under some circumstances, valgrind might report kernel bugs
		5080	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		5081	although an acceptable workaround has been found here), or it might be
		5082	confused.
		5083
		5084	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		5085	make it into some kind of religion.
		5086
		5087	If you are unsure about something, feel free to contact the mailing list
		5088	with the full valgrind report and an explanation on why you think this
		5089	is a bug in libev (best check the archives, too :). However, don't be
		5090	annoyed when you get a brisk "this is no bug" answer and take the chance
		5091	of learning how to interpret valgrind properly.
		5092
		5093	If you need, for some reason, empty reports from valgrind for your project
		5094	I suggest using suppression lists.
		5095
		5096
		5097	=head1 PORTABILITY NOTES
		5098
		5099	=head2 GNU/LINUX 32 BIT LIMITATIONS
		5100
		5101	GNU/Linux is the only common platform that supports 64 bit file/large file
		5102	interfaces but I<disables> them by default.
		5103
		5104	That means that libev compiled in the default environment doesn't support
		5105	files larger than 2GiB or so, which mainly affects C<ev_stat> watchers.
		5106
		5107	Unfortunately, many programs try to work around this GNU/Linux issue
		5108	by enabling the large file API, which makes them incompatible with the
		5109	standard libev compiled for their system.
		5110
		5111	Likewise, libev cannot enable the large file API itself as this would
		5112	suddenly make it incompatible to the default compile time environment,
		5113	i.e. all programs not using special compile switches.
		5114
		5115	=head2 OS/X AND DARWIN BUGS
		5116
		5117	The whole thing is a bug if you ask me - basically any system interface
		5118	you touch is broken, whether it is locales, poll, kqueue or even the
		5119	OpenGL drivers.
		5120
		5121	=head3 C<kqueue> is buggy
		5122
		5123	The kqueue syscall is broken in all known versions - most versions support
		5124	only sockets, many support pipes.
		5125
		5126	Libev tries to work around this by not using C<kqueue> by default on this
		5127	rotten platform, but of course you can still ask for it when creating a
		5128	loop - embedding a socket-only kqueue loop into a select-based one is
		5129	probably going to work well.
		5130
		5131	=head3 C<poll> is buggy
		5132
		5133	Instead of fixing C<kqueue>, Apple replaced their (working) C<poll>
		5134	implementation by something calling C<kqueue> internally around the 10.5.6
		5135	release, so now C<kqueue> I<and> C<poll> are broken.
		5136
		5137	Libev tries to work around this by not using C<poll> by default on
		5138	this rotten platform, but of course you can still ask for it when creating
		5139	a loop.
		5140
		5141	=head3 C<select> is buggy
		5142
		5143	All that's left is C<select>, and of course Apple found a way to fuck this
		5144	one up as well: On OS/X, C<select> actively limits the number of file
		5145	descriptors you can pass in to 1024 - your program suddenly crashes when
		5146	you use more.
		5147
		5148	There is an undocumented "workaround" for this - defining
		5149	C<_DARWIN_UNLIMITED_SELECT>, which libev tries to use, so select I<should>
		5150	work on OS/X.
		5151
		5152	=head2 SOLARIS PROBLEMS AND WORKAROUNDS
		5153
		5154	=head3 C<errno> reentrancy
		5155
		5156	The default compile environment on Solaris is unfortunately so
		5157	thread-unsafe that you can't even use components/libraries compiled
		5158	without C<-D_REENTRANT> in a threaded program, which, of course, isn't
		5159	defined by default. A valid, if stupid, implementation choice.
		5160
		5161	If you want to use libev in threaded environments you have to make sure
		5162	it's compiled with C<_REENTRANT> defined.
		5163
		5164	=head3 Event port backend
		5165
		5166	The scalable event interface for Solaris is called "event
		5167	ports". Unfortunately, this mechanism is very buggy in all major
		5168	releases. If you run into high CPU usage, your program freezes or you get
		5169	a large number of spurious wakeups, make sure you have all the relevant
		5170	and latest kernel patches applied. No, I don't know which ones, but there
		5171	are multiple ones to apply, and afterwards, event ports actually work
		5172	great.
		5173
		5174	If you can't get it to work, you can try running the program by setting
		5175	the environment variable C<LIBEV_FLAGS=3> to only allow C<poll> and
		5176	C<select> backends.
		5177
		5178	=head2 AIX POLL BUG
		5179
		5180	AIX unfortunately has a broken C<poll.h> header. Libev works around
		5181	this by trying to avoid the poll backend altogether (i.e. it's not even
		5182	compiled in), which normally isn't a big problem as C<select> works fine
		5183	with large bitsets on AIX, and AIX is dead anyway.
		5184
		5185	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		5186
		5187	=head3 General issues
		5188
		5189	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		5190	requires, and its I/O model is fundamentally incompatible with the POSIX
		5191	model. Libev still offers limited functionality on this platform in
		5192	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		5193	descriptors. This only applies when using Win32 natively, not when using
		5194	e.g. cygwin. Actually, it only applies to the microsofts own compilers,
		5195	as every compiler comes with a slightly differently broken/incompatible
		5196	environment.
		5197
		5198	Lifting these limitations would basically require the full
		5199	re-implementation of the I/O system. If you are into this kind of thing,
		5200	then note that glib does exactly that for you in a very portable way (note
		5201	also that glib is the slowest event library known to man).
		5202
		5203	There is no supported compilation method available on windows except
		5204	embedding it into other applications.
		5205
		5206	Sensible signal handling is officially unsupported by Microsoft - libev
		5207	tries its best, but under most conditions, signals will simply not work.
		5208
		5209	Not a libev limitation but worth mentioning: windows apparently doesn't
		5210	accept large writes: instead of resulting in a partial write, windows will
		5211	either accept everything or return C<ENOBUFS> if the buffer is too large,
		5212	so make sure you only write small amounts into your sockets (less than a
		5213	megabyte seems safe, but this apparently depends on the amount of memory
		5214	available).
		5215
		5216	Due to the many, low, and arbitrary limits on the win32 platform and
		5217	the abysmal performance of winsockets, using a large number of sockets
		5218	is not recommended (and not reasonable). If your program needs to use
		5219	more than a hundred or so sockets, then likely it needs to use a totally
		5220	different implementation for windows, as libev offers the POSIX readiness
		5221	notification model, which cannot be implemented efficiently on windows
		5222	(due to Microsoft monopoly games).
		5223
		5224	A typical way to use libev under windows is to embed it (see the embedding
		5225	section for details) and use the following F<evwrap.h> header file instead
		5226	of F<ev.h>:
		5227
		5228	#define EV_STANDALONE /* keeps ev from requiring config.h */
		5229	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		5230
		5231	#include "ev.h"
		5232
		5233	And compile the following F<evwrap.c> file into your project (make sure
		5234	you do I<not> compile the F<ev.c> or any other embedded source files!):
		5235
		5236	#include "evwrap.h"
		5237	#include "ev.c"
		5238
		5239	=head3 The winsocket C<select> function
		5240
		5241	The winsocket C<select> function doesn't follow POSIX in that it
		5242	requires socket I<handles> and not socket I<file descriptors> (it is
		5243	also extremely buggy). This makes select very inefficient, and also
		5244	requires a mapping from file descriptors to socket handles (the Microsoft
		5245	C runtime provides the function C<_open_osfhandle> for this). See the
		5246	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		5247	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
		5248
		5249	The configuration for a "naked" win32 using the Microsoft runtime
		5250	libraries and raw winsocket select is:
		5251
		5252	#define EV_USE_SELECT 1
		5253	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		5254
		5255	Note that winsockets handling of fd sets is O(n), so you can easily get a
		5256	complexity in the O(n²) range when using win32.
		5257
		5258	=head3 Limited number of file descriptors
		5259
		5260	Windows has numerous arbitrary (and low) limits on things.
		5261
		5262	Early versions of winsocket's select only supported waiting for a maximum
		5263	of C<64> handles (probably owning to the fact that all windows kernels
		5264	can only wait for C<64> things at the same time internally; Microsoft
		5265	recommends spawning a chain of threads and wait for 63 handles and the
		5266	previous thread in each. Sounds great!).
		5267
		5268	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		5269	to some high number (e.g. C<2048>) before compiling the winsocket select
		5270	call (which might be in libev or elsewhere, for example, perl and many
		5271	other interpreters do their own select emulation on windows).
		5272
		5273	Another limit is the number of file descriptors in the Microsoft runtime
		5274	libraries, which by default is C<64> (there must be a hidden I<64>
		5275	fetish or something like this inside Microsoft). You can increase this
		5276	by calling C<_setmaxstdio>, which can increase this limit to C<2048>
		5277	(another arbitrary limit), but is broken in many versions of the Microsoft
		5278	runtime libraries. This might get you to about C<512> or C<2048> sockets
		5279	(depending on windows version and/or the phase of the moon). To get more,
		5280	you need to wrap all I/O functions and provide your own fd management, but
		5281	the cost of calling select (O(n²)) will likely make this unworkable.
		5282
		5283	=head2 PORTABILITY REQUIREMENTS
		5284
		5285	In addition to a working ISO-C implementation and of course the
		5286	backend-specific APIs, libev relies on a few additional extensions:
		5287
		5288	=over 4
		5289
		5290	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		5291	calling conventions regardless of C<ev_watcher_type *>.
		5292
		5293	Libev assumes not only that all watcher pointers have the same internal
		5294	structure (guaranteed by POSIX but not by ISO C for example), but it also
		5295	assumes that the same (machine) code can be used to call any watcher
		5296	callback: The watcher callbacks have different type signatures, but libev
		5297	calls them using an C<ev_watcher *> internally.
		5298
		5299	=item pointer accesses must be thread-atomic
		5300
		5301	Accessing a pointer value must be atomic, it must both be readable and
		5302	writable in one piece - this is the case on all current architectures.
		5303
		5304	=item C<sig_atomic_t volatile> must be thread-atomic as well
		5305
		5306	The type C<sig_atomic_t volatile> (or whatever is defined as
		5307	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		5308	threads. This is not part of the specification for C<sig_atomic_t>, but is
		5309	believed to be sufficiently portable.
		5310
		5311	=item C<sigprocmask> must work in a threaded environment
		5312
		5313	Libev uses C<sigprocmask> to temporarily block signals. This is not
		5314	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		5315	pthread implementations will either allow C<sigprocmask> in the "main
		5316	thread" or will block signals process-wide, both behaviours would
		5317	be compatible with libev. Interaction between C<sigprocmask> and
		5318	C<pthread_sigmask> could complicate things, however.
		5319
		5320	The most portable way to handle signals is to block signals in all threads
		5321	except the initial one, and run the signal handling loop in the initial
		5322	thread as well.
		5323
		5324	=item C<long> must be large enough for common memory allocation sizes
		5325
		5326	To improve portability and simplify its API, libev uses C<long> internally
		5327	instead of C<size_t> when allocating its data structures. On non-POSIX
		5328	systems (Microsoft...) this might be unexpectedly low, but is still at
		5329	least 31 bits everywhere, which is enough for hundreds of millions of
		5330	watchers.
		5331
		5332	=item C<double> must hold a time value in seconds with enough accuracy
		5333
		5334	The type C<double> is used to represent timestamps. It is required to
		5335	have at least 51 bits of mantissa (and 9 bits of exponent), which is
		5336	good enough for at least into the year 4000 with millisecond accuracy
		5337	(the design goal for libev). This requirement is overfulfilled by
		5338	implementations using IEEE 754, which is basically all existing ones.
		5339
		5340	With IEEE 754 doubles, you get microsecond accuracy until at least the
		5341	year 2255 (and millisecond accuracy till the year 287396 - by then, libev
		5342	is either obsolete or somebody patched it to use C<long double> or
		5343	something like that, just kidding).
		5344
		5345	=back
		5346
		5347	If you know of other additional requirements drop me a note.
		5348
		5349
2189	=head1 COMPLEXITIES	5350	=head1 ALGORITHMIC COMPLEXITIES
2190		5351
2191	In this section the complexities of (many of) the algorithms used inside	5352	In this section the complexities of (many of) the algorithms used inside
2192	libev will be explained. For complexity discussions about backends see the	5353	libev will be documented. For complexity discussions about backends see
2193	documentation for C<ev_default_init>.	5354	the documentation for C<ev_default_init>.
		5355
		5356	All of the following are about amortised time: If an array needs to be
		5357	extended, libev needs to realloc and move the whole array, but this
		5358	happens asymptotically rarer with higher number of elements, so O(1) might
		5359	mean that libev does a lengthy realloc operation in rare cases, but on
		5360	average it is much faster and asymptotically approaches constant time.
2194		5361
2195	=over 4	5362	=over 4
2196		5363
2197	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	5364	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2198		5365
		5366	This means that, when you have a watcher that triggers in one hour and
		5367	there are 100 watchers that would trigger before that, then inserting will
		5368	have to skip roughly seven (C<ld 100>) of these watchers.
		5369
2199	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	5370	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
2200		5371
		5372	That means that changing a timer costs less than removing/adding them,
		5373	as only the relative motion in the event queue has to be paid for.
		5374
2201	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	5375	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
2202		5376
		5377	These just add the watcher into an array or at the head of a list.
		5378
2203	=item Stopping check/prepare/idle watchers: O(1)	5379	=item Stopping check/prepare/idle/fork/async watchers: O(1)
2204		5380
2205	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	5381	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2206		5382
		5383	These watchers are stored in lists, so they need to be walked to find the
		5384	correct watcher to remove. The lists are usually short (you don't usually
		5385	have many watchers waiting for the same fd or signal: one is typical, two
		5386	is rare).
		5387
2207	=item Finding the next timer per loop iteration: O(1)	5388	=item Finding the next timer in each loop iteration: O(1)
		5389
		5390	By virtue of using a binary or 4-heap, the next timer is always found at a
		5391	fixed position in the storage array.
2208		5392
2209	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	5393	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2210		5394
2211	=item Activating one watcher: O(1)	5395	A change means an I/O watcher gets started or stopped, which requires
		5396	libev to recalculate its status (and possibly tell the kernel, depending
		5397	on backend and whether C<ev_io_set> was used).
		5398
		5399	=item Activating one watcher (putting it into the pending state): O(1)
		5400
		5401	=item Priority handling: O(number_of_priorities)
		5402
		5403	Priorities are implemented by allocating some space for each
		5404	priority. When doing priority-based operations, libev usually has to
		5405	linearly search all the priorities, but starting/stopping and activating
		5406	watchers becomes O(1) with respect to priority handling.
		5407
		5408	=item Sending an ev_async: O(1)
		5409
		5410	=item Processing ev_async_send: O(number_of_async_watchers)
		5411
		5412	=item Processing signals: O(max_signal_number)
		5413
		5414	Sending involves a system call I<iff> there were no other C<ev_async_send>
		5415	calls in the current loop iteration and the loop is currently
		5416	blocked. Checking for async and signal events involves iterating over all
		5417	running async watchers or all signal numbers.
2212		5418
2213	=back	5419	=back
2214		5420
2215		5421
		5422	=head1 PORTING FROM LIBEV 3.X TO 4.X
		5423
		5424	The major version 4 introduced some incompatible changes to the API.
		5425
		5426	At the moment, the C<ev.h> header file provides compatibility definitions
		5427	for all changes, so most programs should still compile. The compatibility
		5428	layer might be removed in later versions of libev, so better update to the
		5429	new API early than late.
		5430
		5431	=over 4
		5432
		5433	=item C<EV_COMPAT3> backwards compatibility mechanism
		5434
		5435	The backward compatibility mechanism can be controlled by
		5436	C<EV_COMPAT3>. See L</"PREPROCESSOR SYMBOLS/MACROS"> in the L</EMBEDDING>
		5437	section.
		5438
		5439	=item C<ev_default_destroy> and C<ev_default_fork> have been removed
		5440
		5441	These calls can be replaced easily by their C<ev_loop_xxx> counterparts:
		5442
		5443	ev_loop_destroy (EV_DEFAULT_UC);
		5444	ev_loop_fork (EV_DEFAULT);
		5445
		5446	=item function/symbol renames
		5447
		5448	A number of functions and symbols have been renamed:
		5449
		5450	ev_loop => ev_run
		5451	EVLOOP_NONBLOCK => EVRUN_NOWAIT
		5452	EVLOOP_ONESHOT => EVRUN_ONCE
		5453
		5454	ev_unloop => ev_break
		5455	EVUNLOOP_CANCEL => EVBREAK_CANCEL
		5456	EVUNLOOP_ONE => EVBREAK_ONE
		5457	EVUNLOOP_ALL => EVBREAK_ALL
		5458
		5459	EV_TIMEOUT => EV_TIMER
		5460
		5461	ev_loop_count => ev_iteration
		5462	ev_loop_depth => ev_depth
		5463	ev_loop_verify => ev_verify
		5464
		5465	Most functions working on C<struct ev_loop> objects don't have an
		5466	C<ev_loop_> prefix, so it was removed; C<ev_loop>, C<ev_unloop> and
		5467	associated constants have been renamed to not collide with the C<struct
		5468	ev_loop> anymore and C<EV_TIMER> now follows the same naming scheme
		5469	as all other watcher types. Note that C<ev_loop_fork> is still called
		5470	C<ev_loop_fork> because it would otherwise clash with the C<ev_fork>
		5471	typedef.
		5472
		5473	=item C<EV_MINIMAL> mechanism replaced by C<EV_FEATURES>
		5474
		5475	The preprocessor symbol C<EV_MINIMAL> has been replaced by a different
		5476	mechanism, C<EV_FEATURES>. Programs using C<EV_MINIMAL> usually compile
		5477	and work, but the library code will of course be larger.
		5478
		5479	=back
		5480
		5481
		5482	=head1 GLOSSARY
		5483
		5484	=over 4
		5485
		5486	=item active
		5487
		5488	A watcher is active as long as it has been started and not yet stopped.
		5489	See L</WATCHER STATES> for details.
		5490
		5491	=item application
		5492
		5493	In this document, an application is whatever is using libev.
		5494
		5495	=item backend
		5496
		5497	The part of the code dealing with the operating system interfaces.
		5498
		5499	=item callback
		5500
		5501	The address of a function that is called when some event has been
		5502	detected. Callbacks are being passed the event loop, the watcher that
		5503	received the event, and the actual event bitset.
		5504
		5505	=item callback/watcher invocation
		5506
		5507	The act of calling the callback associated with a watcher.
		5508
		5509	=item event
		5510
		5511	A change of state of some external event, such as data now being available
		5512	for reading on a file descriptor, time having passed or simply not having
		5513	any other events happening anymore.
		5514
		5515	In libev, events are represented as single bits (such as C<EV_READ> or
		5516	C<EV_TIMER>).
		5517
		5518	=item event library
		5519
		5520	A software package implementing an event model and loop.
		5521
		5522	=item event loop
		5523
		5524	An entity that handles and processes external events and converts them
		5525	into callback invocations.
		5526
		5527	=item event model
		5528
		5529	The model used to describe how an event loop handles and processes
		5530	watchers and events.
		5531
		5532	=item pending
		5533
		5534	A watcher is pending as soon as the corresponding event has been
		5535	detected. See L</WATCHER STATES> for details.
		5536
		5537	=item real time
		5538
		5539	The physical time that is observed. It is apparently strictly monotonic :)
		5540
		5541	=item wall-clock time
		5542
		5543	The time and date as shown on clocks. Unlike real time, it can actually
		5544	be wrong and jump forwards and backwards, e.g. when you adjust your
		5545	clock.
		5546
		5547	=item watcher
		5548
		5549	A data structure that describes interest in certain events. Watchers need
		5550	to be started (attached to an event loop) before they can receive events.
		5551
		5552	=back
		5553
2216	=head1 AUTHOR	5554	=head1 AUTHOR
2217		5555
2218	Marc Lehmann <libev@schmorp.de>.	5556	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael
		5557	Magnusson and Emanuele Giaquinta, and minor corrections by many others.
2219		5558

Diff Legend

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