ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.2 by root, Mon Nov 12 08:02:55 2007 UTC vs.
Revision 1.61 by root, Thu Nov 29 12:21:05 2007 UTC

…		…
3	libev - a high performance full-featured event loop written in C	3	libev - a high performance full-featured event loop written in C
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
		8
		9	=head1 EXAMPLE PROGRAM
		10
		11	#include <ev.h>
		12
		13	ev_io stdin_watcher;
		14	ev_timer timeout_watcher;
		15
		16	/* called when data readable on stdin */
		17	static void
		18	stdin_cb (EV_P_ struct ev_io *w, int revents)
		19	{
		20	/* puts ("stdin ready"); */
		21	ev_io_stop (EV_A_ w); /* just a syntax example */
		22	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */
		23	}
		24
		25	static void
		26	timeout_cb (EV_P_ struct ev_timer *w, int revents)
		27	{
		28	/* puts ("timeout"); */
		29	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */
		30	}
		31
		32	int
		33	main (void)
		34	{
		35	struct ev_loop *loop = ev_default_loop (0);
		36
		37	/* initialise an io watcher, then start it */
		38	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
		39	ev_io_start (loop, &stdin_watcher);
		40
		41	/* simple non-repeating 5.5 second timeout */
		42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		43	ev_timer_start (loop, &timeout_watcher);
		44
		45	/* loop till timeout or data ready */
		46	ev_loop (loop, 0);
		47
		48	return 0;
		49	}
8		50
9	=head1 DESCRIPTION	51	=head1 DESCRIPTION
10		52
11	Libev is an event loop: you register interest in certain events (such as a	53	Libev is an event loop: you register interest in certain events (such as a
12	file descriptor being readable or a timeout occuring), and it will manage	54	file descriptor being readable or a timeout occuring), and it will manage
13	these event sources and provide your program events.	55	these event sources and provide your program with events.
14		56
15	To do this, it must take more or less complete control over your process	57	To do this, it must take more or less complete control over your process
16	(or thread) by executing the I<event loop> handler, and will then	58	(or thread) by executing the I<event loop> handler, and will then
17	communicate events via a callback mechanism.	59	communicate events via a callback mechanism.
18		60
…		…
21	details of the event, and then hand it over to libev by I<starting> the	63	details of the event, and then hand it over to libev by I<starting> the
22	watcher.	64	watcher.
23		65
24	=head1 FEATURES	66	=head1 FEATURES
25		67
26	Libev supports select, poll, the linux-specific epoll and the bsd-specific	68	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
27	kqueue mechanisms for file descriptor events, relative timers, absolute	69	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
28	timers with customised rescheduling, signal events, process status change	70	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
29	events (related to SIGCHLD), and event watchers dealing with the event	71	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
30	loop mechanism itself (idle, prepare and check watchers).	72	with customised rescheduling (C<ev_periodic>), synchronous signals
		73	(C<ev_signal>), process status change events (C<ev_child>), and event
		74	watchers dealing with the event loop mechanism itself (C<ev_idle>,
		75	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
		76	file watchers (C<ev_stat>) and even limited support for fork events
		77	(C<ev_fork>).
		78
		79	It also is quite fast (see this
		80	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
		81	for example).
31		82
32	=head1 CONVENTIONS	83	=head1 CONVENTIONS
33		84
34	Libev is very configurable. In this manual the default configuration	85	Libev is very configurable. In this manual the default configuration will
35	will be described, which supports multiple event loops. For more info	86	be described, which supports multiple event loops. For more info about
36	about various configuraiton options please have a look at the file	87	various configuration options please have a look at B<EMBED> section in
37	F<README.embed> in the libev distribution. If libev was configured without	88	this manual. If libev was configured without support for multiple event
38	support for multiple event loops, then all functions taking an initial	89	loops, then all functions taking an initial argument of name C<loop>
39	argument of name C<loop> (which is always of type C<struct ev_loop *>)	90	(which is always of type C<struct ev_loop *>) will not have this argument.
40	will not have this argument.
41		91
42	=head1 TIME AND OTHER GLOBAL FUNCTIONS	92	=head1 TIME REPRESENTATION
43		93
44	Libev represents time as a single floating point number, representing the	94	Libev represents time as a single floating point number, representing the
45	(fractional) number of seconds since the (POSIX) epoch (somewhere near	95	(fractional) number of seconds since the (POSIX) epoch (somewhere near
46	the beginning of 1970, details are complicated, don't ask). This type is	96	the beginning of 1970, details are complicated, don't ask). This type is
47	called C<ev_tstamp>, which is what you should use too. It usually aliases	97	called C<ev_tstamp>, which is what you should use too. It usually aliases
48	to the double type in C.	98	to the C<double> type in C, and when you need to do any calculations on
		99	it, you should treat it as such.
		100
		101	=head1 GLOBAL FUNCTIONS
		102
		103	These functions can be called anytime, even before initialising the
		104	library in any way.
49		105
50	=over 4	106	=over 4
51		107
52	=item ev_tstamp ev_time ()	108	=item ev_tstamp ev_time ()
53		109
54	Returns the current time as libev would use it.	110	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
		112	you actually want to know.
55		113
56	=item int ev_version_major ()	114	=item int ev_version_major ()
57		115
58	=item int ev_version_minor ()	116	=item int ev_version_minor ()
59		117
…		…
61	you linked against by calling the functions C<ev_version_major> and	119	you linked against by calling the functions C<ev_version_major> and
62	C<ev_version_minor>. If you want, you can compare against the global	120	C<ev_version_minor>. If you want, you can compare against the global
63	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	121	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
64	version of the library your program was compiled against.	122	version of the library your program was compiled against.
65		123
66	Usually, its a good idea to terminate if the major versions mismatch,	124	Usually, it's a good idea to terminate if the major versions mismatch,
67	as this indicates an incompatible change. Minor versions are usually	125	as this indicates an incompatible change. Minor versions are usually
68	compatible to older versions, so a larger minor version alone is usually	126	compatible to older versions, so a larger minor version alone is usually
69	not a problem.	127	not a problem.
70		128
		129	Example: Make sure we haven't accidentally been linked against the wrong
		130	version.
		131
		132	assert (("libev version mismatch",
		133	ev_version_major () == EV_VERSION_MAJOR
		134	&& ev_version_minor () >= EV_VERSION_MINOR));
		135
		136	=item unsigned int ev_supported_backends ()
		137
		138	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
		139	value) compiled into this binary of libev (independent of their
		140	availability on the system you are running on). See C<ev_default_loop> for
		141	a description of the set values.
		142
		143	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
		145
		146	assert (("sorry, no epoll, no sex",
		147	ev_supported_backends () & EVBACKEND_EPOLL));
		148
		149	=item unsigned int ev_recommended_backends ()
		150
		151	Return the set of all backends compiled into this binary of libev and also
		152	recommended for this platform. This set is often smaller than the one
		153	returned by C<ev_supported_backends>, as for example kqueue is broken on
		154	most BSDs and will not be autodetected unless you explicitly request it
		155	(assuming you know what you are doing). This is the set of backends that
		156	libev will probe for if you specify no backends explicitly.
		157
		158	=item unsigned int ev_embeddable_backends ()
		159
		160	Returns the set of backends that are embeddable in other event loops. This
		161	is the theoretical, all-platform, value. To find which backends
		162	might be supported on the current system, you would need to look at
		163	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
		164	recommended ones.
		165
		166	See the description of C<ev_embed> watchers for more info.
		167
71	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	168	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
72		169
73	Sets the allocation function to use (the prototype is similar to the	170	Sets the allocation function to use (the prototype is similar - the
74	realloc function). It is used to allocate and free memory (no surprises	171	semantics is identical - to the realloc C function). It is used to
75	here). If it returns zero when memory needs to be allocated, the library	172	allocate and free memory (no surprises here). If it returns zero when
76	might abort or take some potentially destructive action. The default is	173	memory needs to be allocated, the library might abort or take some
77	your system realloc function.	174	potentially destructive action. The default is your system realloc
		175	function.
78		176
79	You could override this function in high-availability programs to, say,	177	You could override this function in high-availability programs to, say,
80	free some memory if it cannot allocate memory, to use a special allocator,	178	free some memory if it cannot allocate memory, to use a special allocator,
81	or even to sleep a while and retry until some memory is available.	179	or even to sleep a while and retry until some memory is available.
		180
		181	Example: Replace the libev allocator with one that waits a bit and then
		182	retries).
		183
		184	static void *
		185	persistent_realloc (void *ptr, size_t size)
		186	{
		187	for (;;)
		188	{
		189	void *newptr = realloc (ptr, size);
		190
		191	if (newptr)
		192	return newptr;
		193
		194	sleep (60);
		195	}
		196	}
		197
		198	...
		199	ev_set_allocator (persistent_realloc);
82		200
83	=item ev_set_syserr_cb (void (*cb)(const char *msg));	201	=item ev_set_syserr_cb (void (*cb)(const char *msg));
84		202
85	Set the callback function to call on a retryable syscall error (such	203	Set the callback function to call on a retryable syscall error (such
86	as failed select, poll, epoll_wait). The message is a printable string	204	as failed select, poll, epoll_wait). The message is a printable string
87	indicating the system call or subsystem causing the problem. If this	205	indicating the system call or subsystem causing the problem. If this
88	callback is set, then libev will expect it to remedy the sitution, no	206	callback is set, then libev will expect it to remedy the sitution, no
89	matter what, when it returns. That is, libev will geenrally retry the	207	matter what, when it returns. That is, libev will generally retry the
90	requested operation, or, if the condition doesn't go away, do bad stuff	208	requested operation, or, if the condition doesn't go away, do bad stuff
91	(such as abort).	209	(such as abort).
		210
		211	Example: This is basically the same thing that libev does internally, too.
		212
		213	static void
		214	fatal_error (const char *msg)
		215	{
		216	perror (msg);
		217	abort ();
		218	}
		219
		220	...
		221	ev_set_syserr_cb (fatal_error);
92		222
93	=back	223	=back
94		224
95	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	225	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
96		226
97	An event loop is described by a C<struct ev_loop *>. The library knows two	227	An event loop is described by a C<struct ev_loop *>. The library knows two
98	types of such loops, the I<default> loop, which supports signals and child	228	types of such loops, the I<default> loop, which supports signals and child
99	events, and dynamically created loops which do not.	229	events, and dynamically created loops which do not.
100		230
101	If you use threads, a common model is to run the default event loop	231	If you use threads, a common model is to run the default event loop
102	in your main thread (or in a separate thrad) and for each thread you	232	in your main thread (or in a separate thread) and for each thread you
103	create, you also create another event loop. Libev itself does no lockign	233	create, you also create another event loop. Libev itself does no locking
104	whatsoever, so if you mix calls to different event loops, make sure you	234	whatsoever, so if you mix calls to the same event loop in different
105	lock (this is usually a bad idea, though, even if done right).	235	threads, make sure you lock (this is usually a bad idea, though, even if
		236	done correctly, because it's hideous and inefficient).
106		237
107	=over 4	238	=over 4
108		239
109	=item struct ev_loop *ev_default_loop (unsigned int flags)	240	=item struct ev_loop *ev_default_loop (unsigned int flags)
110		241
111	This will initialise the default event loop if it hasn't been initialised	242	This will initialise the default event loop if it hasn't been initialised
112	yet and return it. If the default loop could not be initialised, returns	243	yet and return it. If the default loop could not be initialised, returns
113	false. If it already was initialised it simply returns it (and ignores the	244	false. If it already was initialised it simply returns it (and ignores the
114	flags).	245	flags. If that is troubling you, check C<ev_backend ()> afterwards).
115		246
116	If you don't know what event loop to use, use the one returned from this	247	If you don't know what event loop to use, use the one returned from this
117	function.	248	function.
118		249
119	The flags argument can be used to specify special behaviour or specific	250	The flags argument can be used to specify special behaviour or specific
120	backends to use, and is usually specified as 0 (or EVFLAG_AUTO)	251	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
121		252
122	It supports the following flags:	253	The following flags are supported:
123		254
124	=over 4	255	=over 4
125		256
126	=item EVFLAG_AUTO	257	=item C<EVFLAG_AUTO>
127		258
128	The default flags value. Use this if you have no clue (its the right	259	The default flags value. Use this if you have no clue (it's the right
129	thing, believe me).	260	thing, believe me).
130		261
131	=item EVFLAG_NOENV	262	=item C<EVFLAG_NOENV>
132		263
133	If this flag bit is ored into the flag value then libev will I<not> look	264	If this flag bit is ored into the flag value (or the program runs setuid
134	at the environment variable C<LIBEV_FLAGS>. Otherwise (the default), this	265	or setgid) then libev will I<not> look at the environment variable
135	environment variable will override the flags completely. This is useful	266	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
		267	override the flags completely if it is found in the environment. This is
136	to try out specific backends to tets their performance, or to work around	268	useful to try out specific backends to test their performance, or to work
137	bugs.	269	around bugs.
138		270
139	=item EVMETHOD_SELECT portable select backend	271	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
140		272
141	=item EVMETHOD_POLL poll backend (everywhere except windows)	273	This is your standard select(2) backend. Not I<completely> standard, as
		274	libev tries to roll its own fd_set with no limits on the number of fds,
		275	but if that fails, expect a fairly low limit on the number of fds when
		276	using this backend. It doesn't scale too well (O(highest_fd)), but its usually
		277	the fastest backend for a low number of fds.
142		278
143	=item EVMETHOD_EPOLL linux only	279	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
144		280
145	=item EVMETHOD_KQUEUE some bsds only	281	And this is your standard poll(2) backend. It's more complicated than
		282	select, but handles sparse fds better and has no artificial limit on the
		283	number of fds you can use (except it will slow down considerably with a
		284	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).
146		285
147	=item EVMETHOD_DEVPOLL solaris 8 only	286	=item C<EVBACKEND_EPOLL> (value 4, Linux)
148		287
149	=item EVMETHOD_PORT solaris 10 only	288	For few fds, this backend is a bit little slower than poll and select,
		289	but it scales phenomenally better. While poll and select usually scale like
		290	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales
		291	either O(1) or O(active_fds).
		292
		293	While stopping and starting an I/O watcher in the same iteration will
		294	result in some caching, there is still a syscall per such incident
		295	(because the fd could point to a different file description now), so its
		296	best to avoid that. Also, dup()ed file descriptors might not work very
		297	well if you register events for both fds.
		298
		299	Please note that epoll sometimes generates spurious notifications, so you
		300	need to use non-blocking I/O or other means to avoid blocking when no data
		301	(or space) is available.
		302
		303	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
		304
		305	Kqueue deserves special mention, as at the time of this writing, it
		306	was broken on all BSDs except NetBSD (usually it doesn't work with
		307	anything but sockets and pipes, except on Darwin, where of course its
		308	completely useless). For this reason its not being "autodetected"
		309	unless you explicitly specify it explicitly in the flags (i.e. using
		310	C<EVBACKEND_KQUEUE>).
		311
		312	It scales in the same way as the epoll backend, but the interface to the
		313	kernel is more efficient (which says nothing about its actual speed, of
		314	course). While starting and stopping an I/O watcher does not cause an
		315	extra syscall as with epoll, it still adds up to four event changes per
		316	incident, so its best to avoid that.
		317
		318	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
		319
		320	This is not implemented yet (and might never be).
		321
		322	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
		323
		324	This uses the Solaris 10 port mechanism. As with everything on Solaris,
		325	it's really slow, but it still scales very well (O(active_fds)).
		326
		327	Please note that solaris ports can result in a lot of spurious
		328	notifications, so you need to use non-blocking I/O or other means to avoid
		329	blocking when no data (or space) is available.
		330
		331	=item C<EVBACKEND_ALL>
		332
		333	Try all backends (even potentially broken ones that wouldn't be tried
		334	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
		335	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
		336
		337	=back
150		338
151	If one or more of these are ored into the flags value, then only these	339	If one or more of these are ored into the flags value, then only these
152	backends will be tried (in the reverse order as given here). If one are	340	backends will be tried (in the reverse order as given here). If none are
153	specified, any backend will do.	341	specified, most compiled-in backend will be tried, usually in reverse
		342	order of their flag values :)
154		343
155	=back	344	The most typical usage is like this:
		345
		346	if (!ev_default_loop (0))
		347	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
		348
		349	Restrict libev to the select and poll backends, and do not allow
		350	environment settings to be taken into account:
		351
		352	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
		353
		354	Use whatever libev has to offer, but make sure that kqueue is used if
		355	available (warning, breaks stuff, best use only with your own private
		356	event loop and only if you know the OS supports your types of fds):
		357
		358	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
156		359
157	=item struct ev_loop *ev_loop_new (unsigned int flags)	360	=item struct ev_loop *ev_loop_new (unsigned int flags)
158		361
159	Similar to C<ev_default_loop>, but always creates a new event loop that is	362	Similar to C<ev_default_loop>, but always creates a new event loop that is
160	always distinct from the default loop. Unlike the default loop, it cannot	363	always distinct from the default loop. Unlike the default loop, it cannot
161	handle signal and child watchers, and attempts to do so will be greeted by	364	handle signal and child watchers, and attempts to do so will be greeted by
162	undefined behaviour (or a failed assertion if assertions are enabled).	365	undefined behaviour (or a failed assertion if assertions are enabled).
163		366
		367	Example: Try to create a event loop that uses epoll and nothing else.
		368
		369	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
		370	if (!epoller)
		371	fatal ("no epoll found here, maybe it hides under your chair");
		372
164	=item ev_default_destroy ()	373	=item ev_default_destroy ()
165		374
166	Destroys the default loop again (frees all memory and kernel state	375	Destroys the default loop again (frees all memory and kernel state
167	etc.). This stops all registered event watchers (by not touching them in	376	etc.). None of the active event watchers will be stopped in the normal
168	any way whatsoever, although you cnanot rely on this :).	377	sense, so e.g. C<ev_is_active> might still return true. It is your
		378	responsibility to either stop all watchers cleanly yoursef I<before>
		379	calling this function, or cope with the fact afterwards (which is usually
		380	the easiest thing, youc na just ignore the watchers and/or C<free ()> them
		381	for example).
169		382
170	=item ev_loop_destroy (loop)	383	=item ev_loop_destroy (loop)
171		384
172	Like C<ev_default_destroy>, but destroys an event loop created by an	385	Like C<ev_default_destroy>, but destroys an event loop created by an
173	earlier call to C<ev_loop_new>.	386	earlier call to C<ev_loop_new>.
…		…
177	This function reinitialises the kernel state for backends that have	390	This function reinitialises the kernel state for backends that have
178	one. Despite the name, you can call it anytime, but it makes most sense	391	one. Despite the name, you can call it anytime, but it makes most sense
179	after forking, in either the parent or child process (or both, but that	392	after forking, in either the parent or child process (or both, but that
180	again makes little sense).	393	again makes little sense).
181		394
182	You I<must> call this function after forking if and only if you want to	395	You I<must> call this function in the child process after forking if and
183	use the event library in both processes. If you just fork+exec, you don't	396	only if you want to use the event library in both processes. If you just
184	have to call it.	397	fork+exec, you don't have to call it.
185		398
186	The function itself is quite fast and its usually not a problem to call	399	The function itself is quite fast and it's usually not a problem to call
187	it just in case after a fork. To make this easy, the function will fit in	400	it just in case after a fork. To make this easy, the function will fit in
188	quite nicely into a call to C<pthread_atfork>:	401	quite nicely into a call to C<pthread_atfork>:
189		402
190	pthread_atfork (0, 0, ev_default_fork);	403	pthread_atfork (0, 0, ev_default_fork);
		404
		405	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
		406	without calling this function, so if you force one of those backends you
		407	do not need to care.
191		408
192	=item ev_loop_fork (loop)	409	=item ev_loop_fork (loop)
193		410
194	Like C<ev_default_fork>, but acts on an event loop created by	411	Like C<ev_default_fork>, but acts on an event loop created by
195	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	412	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
196	after fork, and how you do this is entirely your own problem.	413	after fork, and how you do this is entirely your own problem.
197		414
198	=item unsigned int ev_method (loop)	415	=item unsigned int ev_backend (loop)
199		416
200	Returns one of the C<EVMETHOD_*> flags indicating the event backend in	417	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
201	use.	418	use.
202		419
203	=item ev_tstamp = ev_now (loop)	420	=item ev_tstamp ev_now (loop)
204		421
205	Returns the current "event loop time", which is the time the event loop	422	Returns the current "event loop time", which is the time the event loop
206	got events and started processing them. This timestamp does not change	423	received events and started processing them. This timestamp does not
207	as long as callbacks are being processed, and this is also the base time	424	change as long as callbacks are being processed, and this is also the base
208	used for relative timers. You can treat it as the timestamp of the event	425	time used for relative timers. You can treat it as the timestamp of the
209	occuring (or more correctly, the mainloop finding out about it).	426	event occuring (or more correctly, libev finding out about it).
210		427
211	=item ev_loop (loop, int flags)	428	=item ev_loop (loop, int flags)
212		429
213	Finally, this is it, the event handler. This function usually is called	430	Finally, this is it, the event handler. This function usually is called
214	after you initialised all your watchers and you want to start handling	431	after you initialised all your watchers and you want to start handling
215	events.	432	events.
216		433
217	If the flags argument is specified as 0, it will not return until either	434	If the flags argument is specified as C<0>, it will not return until
218	no event watchers are active anymore or C<ev_unloop> was called.	435	either no event watchers are active anymore or C<ev_unloop> was called.
		436
		437	Please note that an explicit C<ev_unloop> is usually better than
		438	relying on all watchers to be stopped when deciding when a program has
		439	finished (especially in interactive programs), but having a program that
		440	automatically loops as long as it has to and no longer by virtue of
		441	relying on its watchers stopping correctly is a thing of beauty.
219		442
220	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	443	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
221	those events and any outstanding ones, but will not block your process in	444	those events and any outstanding ones, but will not block your process in
222	case there are no events.	445	case there are no events and will return after one iteration of the loop.
223		446
224	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	447	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
225	neccessary) and will handle those and any outstanding ones. It will block	448	neccessary) and will handle those and any outstanding ones. It will block
226	your process until at least one new event arrives.	449	your process until at least one new event arrives, and will return after
		450	one iteration of the loop. This is useful if you are waiting for some
		451	external event in conjunction with something not expressible using other
		452	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is
		453	usually a better approach for this kind of thing.
227		454
228	This flags value could be used to implement alternative looping	455	Here are the gory details of what C<ev_loop> does:
229	constructs, but the C<prepare> and C<check> watchers provide a better and	456
230	more generic mechanism.	457	* If there are no active watchers (reference count is zero), return.
		458	- Queue prepare watchers and then call all outstanding watchers.
		459	- If we have been forked, recreate the kernel state.
		460	- Update the kernel state with all outstanding changes.
		461	- Update the "event loop time".
		462	- Calculate for how long to block.
		463	- Block the process, waiting for any events.
		464	- Queue all outstanding I/O (fd) events.
		465	- Update the "event loop time" and do time jump handling.
		466	- Queue all outstanding timers.
		467	- Queue all outstanding periodics.
		468	- If no events are pending now, queue all idle watchers.
		469	- Queue all check watchers.
		470	- Call all queued watchers in reverse order (i.e. check watchers first).
		471	Signals and child watchers are implemented as I/O watchers, and will
		472	be handled here by queueing them when their watcher gets executed.
		473	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
		474	were used, return, otherwise continue with step *.
		475
		476	Example: Queue some jobs and then loop until no events are outsanding
		477	anymore.
		478
		479	... queue jobs here, make sure they register event watchers as long
		480	... as they still have work to do (even an idle watcher will do..)
		481	ev_loop (my_loop, 0);
		482	... jobs done. yeah!
231		483
232	=item ev_unloop (loop, how)	484	=item ev_unloop (loop, how)
233		485
234	Can be used to make a call to C<ev_loop> return early. The C<how> argument	486	Can be used to make a call to C<ev_loop> return early (but only after it
		487	has processed all outstanding events). The C<how> argument must be either
235	must be either C<EVUNLOOP_ONCE>, which will make the innermost C<ev_loop>	488	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
236	call return, or C<EVUNLOOP_ALL>, which will make all nested C<ev_loop>	489	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
237	calls return.
238		490
239	=item ev_ref (loop)	491	=item ev_ref (loop)
240		492
241	=item ev_unref (loop)	493	=item ev_unref (loop)
242		494
243	Ref/unref can be used to add or remove a refcount on the event loop: Every	495	Ref/unref can be used to add or remove a reference count on the event
244	watcher keeps one reference. If you have a long-runing watcher you never	496	loop: Every watcher keeps one reference, and as long as the reference
245	unregister that should not keep ev_loop from running, ev_unref() after	497	count is nonzero, C<ev_loop> will not return on its own. If you have
246	starting, and ev_ref() before stopping it. Libev itself uses this for	498	a watcher you never unregister that should not keep C<ev_loop> from
247	example for its internal signal pipe: It is not visible to you as a user	499	returning, ev_unref() after starting, and ev_ref() before stopping it. For
248	and should not keep C<ev_loop> from exiting if the work is done. It is	500	example, libev itself uses this for its internal signal pipe: It is not
249	also an excellent way to do this for generic recurring timers or from	501	visible to the libev user and should not keep C<ev_loop> from exiting if
250	within third-party libraries. Just remember to unref after start and ref	502	no event watchers registered by it are active. It is also an excellent
251	before stop.	503	way to do this for generic recurring timers or from within third-party
		504	libraries. Just remember to I<unref after start> and I<ref before stop>.
		505
		506	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
		507	running when nothing else is active.
		508
		509	struct ev_signal exitsig;
		510	ev_signal_init (&exitsig, sig_cb, SIGINT);
		511	ev_signal_start (loop, &exitsig);
		512	evf_unref (loop);
		513
		514	Example: For some weird reason, unregister the above signal handler again.
		515
		516	ev_ref (loop);
		517	ev_signal_stop (loop, &exitsig);
252		518
253	=back	519	=back
		520
254		521
255	=head1 ANATOMY OF A WATCHER	522	=head1 ANATOMY OF A WATCHER
256		523
257	A watcher is a structure that you create and register to record your	524	A watcher is a structure that you create and register to record your
258	interest in some event. For instance, if you want to wait for STDIN to	525	interest in some event. For instance, if you want to wait for STDIN to
259	become readable, you would create an ev_io watcher for that:	526	become readable, you would create an C<ev_io> watcher for that:
260		527
261	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	528	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)
262	{	529	{
263	ev_io_stop (w);	530	ev_io_stop (w);
264	ev_unloop (loop, EVUNLOOP_ALL);	531	ev_unloop (loop, EVUNLOOP_ALL);
…		…
291	*) >>), and you can stop watching for events at any time by calling the	558	*) >>), and you can stop watching for events at any time by calling the
292	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	559	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.
293		560
294	As long as your watcher is active (has been started but not stopped) you	561	As long as your watcher is active (has been started but not stopped) you
295	must not touch the values stored in it. Most specifically you must never	562	must not touch the values stored in it. Most specifically you must never
296	reinitialise it or call its set method.	563	reinitialise it or call its C<set> macro.
297
298	You cna check wether an event is active by calling the C<ev_is_active
299	(watcher *)> macro. To see wether an event is outstanding (but the
300	callback for it has not been called yet) you cna use the C<ev_is_pending
301	(watcher *)> macro.
302		564
303	Each and every callback receives the event loop pointer as first, the	565	Each and every callback receives the event loop pointer as first, the
304	registered watcher structure as second, and a bitset of received events as	566	registered watcher structure as second, and a bitset of received events as
305	third argument.	567	third argument.
306		568
307	The rceeived events usually include a single bit per event type received	569	The received events usually include a single bit per event type received
308	(you can receive multiple events at the same time). The possible bit masks	570	(you can receive multiple events at the same time). The possible bit masks
309	are:	571	are:
310		572
311	=over 4	573	=over 4
312		574
313	=item EV_READ	575	=item C<EV_READ>
314		576
315	=item EV_WRITE	577	=item C<EV_WRITE>
316		578
317	The file descriptor in the ev_io watcher has become readable and/or	579	The file descriptor in the C<ev_io> watcher has become readable and/or
318	writable.	580	writable.
319		581
320	=item EV_TIMEOUT	582	=item C<EV_TIMEOUT>
321		583
322	The ev_timer watcher has timed out.	584	The C<ev_timer> watcher has timed out.
323		585
324	=item EV_PERIODIC	586	=item C<EV_PERIODIC>
325		587
326	The ev_periodic watcher has timed out.	588	The C<ev_periodic> watcher has timed out.
327		589
328	=item EV_SIGNAL	590	=item C<EV_SIGNAL>
329		591
330	The signal specified in the ev_signal watcher has been received by a thread.	592	The signal specified in the C<ev_signal> watcher has been received by a thread.
331		593
332	=item EV_CHILD	594	=item C<EV_CHILD>
333		595
334	The pid specified in the ev_child watcher has received a status change.	596	The pid specified in the C<ev_child> watcher has received a status change.
335		597
		598	=item C<EV_STAT>
		599
		600	The path specified in the C<ev_stat> watcher changed its attributes somehow.
		601
336	=item EV_IDLE	602	=item C<EV_IDLE>
337		603
338	The ev_idle watcher has determined that you have nothing better to do.	604	The C<ev_idle> watcher has determined that you have nothing better to do.
339		605
340	=item EV_PREPARE	606	=item C<EV_PREPARE>
341		607
342	=item EV_CHECK	608	=item C<EV_CHECK>
343		609
344	All ev_prepare watchers are invoked just I<before> C<ev_loop> starts	610	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts
345	to gather new events, and all ev_check watchers are invoked just after	611	to gather new events, and all C<ev_check> watchers are invoked just after
346	C<ev_loop> has gathered them, but before it invokes any callbacks for any	612	C<ev_loop> has gathered them, but before it invokes any callbacks for any
347	received events. Callbacks of both watcher types can start and stop as	613	received events. Callbacks of both watcher types can start and stop as
348	many watchers as they want, and all of them will be taken into account	614	many watchers as they want, and all of them will be taken into account
349	(for example, a ev_prepare watcher might start an idle watcher to keep	615	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
350	C<ev_loop> from blocking).	616	C<ev_loop> from blocking).
351		617
		618	=item C<EV_EMBED>
		619
		620	The embedded event loop specified in the C<ev_embed> watcher needs attention.
		621
		622	=item C<EV_FORK>
		623
		624	The event loop has been resumed in the child process after fork (see
		625	C<ev_fork>).
		626
352	=item EV_ERROR	627	=item C<EV_ERROR>
353		628
354	An unspecified error has occured, the watcher has been stopped. This might	629	An unspecified error has occured, the watcher has been stopped. This might
355	happen because the watcher could not be properly started because libev	630	happen because the watcher could not be properly started because libev
356	ran out of memory, a file descriptor was found to be closed or any other	631	ran out of memory, a file descriptor was found to be closed or any other
357	problem. You best act on it by reporting the problem and somehow coping	632	problem. You best act on it by reporting the problem and somehow coping
…		…
363	with the error from read() or write(). This will not work in multithreaded	638	with the error from read() or write(). This will not work in multithreaded
364	programs, though, so beware.	639	programs, though, so beware.
365		640
366	=back	641	=back
367		642
		643	=head2 GENERIC WATCHER FUNCTIONS
		644
		645	In the following description, C<TYPE> stands for the watcher type,
		646	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
		647
		648	=over 4
		649
		650	=item C<ev_init> (ev_TYPE *watcher, callback)
		651
		652	This macro initialises the generic portion of a watcher. The contents
		653	of the watcher object can be arbitrary (so C<malloc> will do). Only
		654	the generic parts of the watcher are initialised, you I<need> to call
		655	the type-specific C<ev_TYPE_set> macro afterwards to initialise the
		656	type-specific parts. For each type there is also a C<ev_TYPE_init> macro
		657	which rolls both calls into one.
		658
		659	You can reinitialise a watcher at any time as long as it has been stopped
		660	(or never started) and there are no pending events outstanding.
		661
		662	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,
		663	int revents)>.
		664
		665	=item C<ev_TYPE_set> (ev_TYPE *, [args])
		666
		667	This macro initialises the type-specific parts of a watcher. You need to
		668	call C<ev_init> at least once before you call this macro, but you can
		669	call C<ev_TYPE_set> any number of times. You must not, however, call this
		670	macro on a watcher that is active (it can be pending, however, which is a
		671	difference to the C<ev_init> macro).
		672
		673	Although some watcher types do not have type-specific arguments
		674	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
		675
		676	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
		677
		678	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
		679	calls into a single call. This is the most convinient method to initialise
		680	a watcher. The same limitations apply, of course.
		681
		682	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
		683
		684	Starts (activates) the given watcher. Only active watchers will receive
		685	events. If the watcher is already active nothing will happen.
		686
		687	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
		688
		689	Stops the given watcher again (if active) and clears the pending
		690	status. It is possible that stopped watchers are pending (for example,
		691	non-repeating timers are being stopped when they become pending), but
		692	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If
		693	you want to free or reuse the memory used by the watcher it is therefore a
		694	good idea to always call its C<ev_TYPE_stop> function.
		695
		696	=item bool ev_is_active (ev_TYPE *watcher)
		697
		698	Returns a true value iff the watcher is active (i.e. it has been started
		699	and not yet been stopped). As long as a watcher is active you must not modify
		700	it.
		701
		702	=item bool ev_is_pending (ev_TYPE *watcher)
		703
		704	Returns a true value iff the watcher is pending, (i.e. it has outstanding
		705	events but its callback has not yet been invoked). As long as a watcher
		706	is pending (but not active) you must not call an init function on it (but
		707	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to
		708	libev (e.g. you cnanot C<free ()> it).
		709
		710	=item callback ev_cb (ev_TYPE *watcher)
		711
		712	Returns the callback currently set on the watcher.
		713
		714	=item ev_cb_set (ev_TYPE *watcher, callback)
		715
		716	Change the callback. You can change the callback at virtually any time
		717	(modulo threads).
		718
		719	=back
		720
		721
368	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	722	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
369		723
370	Each watcher has, by default, a member C<void *data> that you can change	724	Each watcher has, by default, a member C<void *data> that you can change
371	and read at any time, libev will completely ignore it. This cna be used	725	and read at any time, libev will completely ignore it. This can be used
372	to associate arbitrary data with your watcher. If you need more data and	726	to associate arbitrary data with your watcher. If you need more data and
373	don't want to allocate memory and store a pointer to it in that data	727	don't want to allocate memory and store a pointer to it in that data
374	member, you can also "subclass" the watcher type and provide your own	728	member, you can also "subclass" the watcher type and provide your own
375	data:	729	data:
376		730
…		…
389	{	743	{
390	struct my_io *w = (struct my_io *)w_;	744	struct my_io *w = (struct my_io *)w_;
391	...	745	...
392	}	746	}
393		747
394	More interesting and less C-conformant ways of catsing your callback type	748	More interesting and less C-conformant ways of casting your callback type
395	have been omitted....	749	instead have been omitted.
		750
		751	Another common scenario is having some data structure with multiple
		752	watchers:
		753
		754	struct my_biggy
		755	{
		756	int some_data;
		757	ev_timer t1;
		758	ev_timer t2;
		759	}
		760
		761	In this case getting the pointer to C<my_biggy> is a bit more complicated,
		762	you need to use C<offsetof>:
		763
		764	#include <stddef.h>
		765
		766	static void
		767	t1_cb (EV_P_ struct ev_timer *w, int revents)
		768	{
		769	struct my_biggy big = (struct my_biggy *
		770	(((char *)w) - offsetof (struct my_biggy, t1));
		771	}
		772
		773	static void
		774	t2_cb (EV_P_ struct ev_timer *w, int revents)
		775	{
		776	struct my_biggy big = (struct my_biggy *
		777	(((char *)w) - offsetof (struct my_biggy, t2));
		778	}
396		779
397		780
398	=head1 WATCHER TYPES	781	=head1 WATCHER TYPES
399		782
400	This section describes each watcher in detail, but will not repeat	783	This section describes each watcher in detail, but will not repeat
401	information given in the last section.	784	information given in the last section. Any initialisation/set macros,
		785	functions and members specific to the watcher type are explained.
402		786
		787	Members are additionally marked with either I<[read-only]>, meaning that,
		788	while the watcher is active, you can look at the member and expect some
		789	sensible content, but you must not modify it (you can modify it while the
		790	watcher is stopped to your hearts content), or I<[read-write]>, which
		791	means you can expect it to have some sensible content while the watcher
		792	is active, but you can also modify it. Modifying it may not do something
		793	sensible or take immediate effect (or do anything at all), but libev will
		794	not crash or malfunction in any way.
		795
		796
403	=head2 struct ev_io - is my file descriptor readable or writable	797	=head2 C<ev_io> - is this file descriptor readable or writable?
404		798
405	I/O watchers check wether a file descriptor is readable or writable	799	I/O watchers check whether a file descriptor is readable or writable
406	in each iteration of the event loop (This behaviour is called	800	in each iteration of the event loop, or, more precisely, when reading
407	level-triggering because you keep receiving events as long as the	801	would not block the process and writing would at least be able to write
408	condition persists. Remember you cna stop the watcher if you don't want to	802	some data. This behaviour is called level-triggering because you keep
409	act on the event and neither want to receive future events).	803	receiving events as long as the condition persists. Remember you can stop
		804	the watcher if you don't want to act on the event and neither want to
		805	receive future events.
		806
		807	In general you can register as many read and/or write event watchers per
		808	fd as you want (as long as you don't confuse yourself). Setting all file
		809	descriptors to non-blocking mode is also usually a good idea (but not
		810	required if you know what you are doing).
		811
		812	You have to be careful with dup'ed file descriptors, though. Some backends
		813	(the linux epoll backend is a notable example) cannot handle dup'ed file
		814	descriptors correctly if you register interest in two or more fds pointing
		815	to the same underlying file/socket/etc. description (that is, they share
		816	the same underlying "file open").
		817
		818	If you must do this, then force the use of a known-to-be-good backend
		819	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
		820	C<EVBACKEND_POLL>).
		821
		822	Another thing you have to watch out for is that it is quite easy to
		823	receive "spurious" readyness notifications, that is your callback might
		824	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
		825	because there is no data. Not only are some backends known to create a
		826	lot of those (for example solaris ports), it is very easy to get into
		827	this situation even with a relatively standard program structure. Thus
		828	it is best to always use non-blocking I/O: An extra C<read>(2) returning
		829	C<EAGAIN> is far preferable to a program hanging until some data arrives.
		830
		831	If you cannot run the fd in non-blocking mode (for example you should not
		832	play around with an Xlib connection), then you have to seperately re-test
		833	wether a file descriptor is really ready with a known-to-be good interface
		834	such as poll (fortunately in our Xlib example, Xlib already does this on
		835	its own, so its quite safe to use).
410		836
411	=over 4	837	=over 4
412		838
413	=item ev_io_init (ev_io *, callback, int fd, int events)	839	=item ev_io_init (ev_io *, callback, int fd, int events)
414		840
415	=item ev_io_set (ev_io *, int fd, int events)	841	=item ev_io_set (ev_io *, int fd, int events)
416		842
417	Configures an ev_io watcher. The fd is the file descriptor to rceeive	843	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
418	events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ |	844	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or
419	EV_WRITE> to receive the given events.	845	C<EV_READ | EV_WRITE> to receive the given events.
		846
		847	=item int fd [read-only]
		848
		849	The file descriptor being watched.
		850
		851	=item int events [read-only]
		852
		853	The events being watched.
420		854
421	=back	855	=back
422		856
		857	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
		858	readable, but only once. Since it is likely line-buffered, you could
		859	attempt to read a whole line in the callback.
		860
		861	static void
		862	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
		863	{
		864	ev_io_stop (loop, w);
		865	.. read from stdin here (or from w->fd) and haqndle any I/O errors
		866	}
		867
		868	...
		869	struct ev_loop *loop = ev_default_init (0);
		870	struct ev_io stdin_readable;
		871	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
		872	ev_io_start (loop, &stdin_readable);
		873	ev_loop (loop, 0);
		874
		875
423	=head2 struct ev_timer - relative and optionally recurring timeouts	876	=head2 C<ev_timer> - relative and optionally repeating timeouts
424		877
425	Timer watchers are simple relative timers that generate an event after a	878	Timer watchers are simple relative timers that generate an event after a
426	given time, and optionally repeating in regular intervals after that.	879	given time, and optionally repeating in regular intervals after that.
427		880
428	The timers are based on real time, that is, if you register an event that	881	The timers are based on real time, that is, if you register an event that
429	times out after an hour and youreset your system clock to last years	882	times out after an hour and you reset your system clock to last years
430	time, it will still time out after (roughly) and hour. "Roughly" because	883	time, it will still time out after (roughly) and hour. "Roughly" because
431	detecting time jumps is hard, and soem inaccuracies are unavoidable (the	884	detecting time jumps is hard, and some inaccuracies are unavoidable (the
432	monotonic clock option helps a lot here).	885	monotonic clock option helps a lot here).
		886
		887	The relative timeouts are calculated relative to the C<ev_now ()>
		888	time. This is usually the right thing as this timestamp refers to the time
		889	of the event triggering whatever timeout you are modifying/starting. If
		890	you suspect event processing to be delayed and you I<need> to base the timeout
		891	on the current time, use something like this to adjust for this:
		892
		893	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
		894
		895	The callback is guarenteed to be invoked only when its timeout has passed,
		896	but if multiple timers become ready during the same loop iteration then
		897	order of execution is undefined.
433		898
434	=over 4	899	=over 4
435		900
436	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	901	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
437		902
…		…
443	later, again, and again, until stopped manually.	908	later, again, and again, until stopped manually.
444		909
445	The timer itself will do a best-effort at avoiding drift, that is, if you	910	The timer itself will do a best-effort at avoiding drift, that is, if you
446	configure a timer to trigger every 10 seconds, then it will trigger at	911	configure a timer to trigger every 10 seconds, then it will trigger at
447	exactly 10 second intervals. If, however, your program cannot keep up with	912	exactly 10 second intervals. If, however, your program cannot keep up with
448	the timer (ecause it takes longer than those 10 seconds to do stuff) the	913	the timer (because it takes longer than those 10 seconds to do stuff) the
449	timer will not fire more than once per event loop iteration.	914	timer will not fire more than once per event loop iteration.
450		915
451	=item ev_timer_again (loop)	916	=item ev_timer_again (loop)
452		917
453	This will act as if the timer timed out and restart it again if it is	918	This will act as if the timer timed out and restart it again if it is
454	repeating. The exact semantics are:	919	repeating. The exact semantics are:
455		920
		921	If the timer is pending, its pending status is cleared.
		922
456	If the timer is started but nonrepeating, stop it.	923	If the timer is started but nonrepeating, stop it (as if it timed out).
457		924
458	If the timer is repeating, either start it if necessary (with the repeat	925	If the timer is repeating, either start it if necessary (with the
459	value), or reset the running timer to the repeat value.	926	C<repeat> value), or reset the running timer to the C<repeat> value.
460		927
461	This sounds a bit complicated, but here is a useful and typical	928	This sounds a bit complicated, but here is a useful and typical
462	example: Imagine you have a tcp connection and you want a so-called idle	929	example: Imagine you have a tcp connection and you want a so-called idle
463	timeout, that is, you want to be called when there have been, say, 60	930	timeout, that is, you want to be called when there have been, say, 60
464	seconds of inactivity on the socket. The easiest way to do this is to	931	seconds of inactivity on the socket. The easiest way to do this is to
465	configure an ev_timer with after=repeat=60 and calling ev_timer_again each	932	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
466	time you successfully read or write some data. If you go into an idle	933	C<ev_timer_again> each time you successfully read or write some data. If
467	state where you do not expect data to travel on the socket, you can stop	934	you go into an idle state where you do not expect data to travel on the
468	the timer, and again will automatically restart it if need be.	935	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
		936	automatically restart it if need be.
		937
		938	That means you can ignore the C<after> value and C<ev_timer_start>
		939	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
		940
		941	ev_timer_init (timer, callback, 0., 5.);
		942	ev_timer_again (loop, timer);
		943	...
		944	timer->again = 17.;
		945	ev_timer_again (loop, timer);
		946	...
		947	timer->again = 10.;
		948	ev_timer_again (loop, timer);
		949
		950	This is more slightly efficient then stopping/starting the timer each time
		951	you want to modify its timeout value.
		952
		953	=item ev_tstamp repeat [read-write]
		954
		955	The current C<repeat> value. Will be used each time the watcher times out
		956	or C<ev_timer_again> is called and determines the next timeout (if any),
		957	which is also when any modifications are taken into account.
469		958
470	=back	959	=back
471		960
472	=head2 ev_periodic	961	Example: Create a timer that fires after 60 seconds.
		962
		963	static void
		964	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
		965	{
		966	.. one minute over, w is actually stopped right here
		967	}
		968
		969	struct ev_timer mytimer;
		970	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
		971	ev_timer_start (loop, &mytimer);
		972
		973	Example: Create a timeout timer that times out after 10 seconds of
		974	inactivity.
		975
		976	static void
		977	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
		978	{
		979	.. ten seconds without any activity
		980	}
		981
		982	struct ev_timer mytimer;
		983	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
		984	ev_timer_again (&mytimer); /* start timer */
		985	ev_loop (loop, 0);
		986
		987	// and in some piece of code that gets executed on any "activity":
		988	// reset the timeout to start ticking again at 10 seconds
		989	ev_timer_again (&mytimer);
		990
		991
		992	=head2 C<ev_periodic> - to cron or not to cron?
473		993
474	Periodic watchers are also timers of a kind, but they are very versatile	994	Periodic watchers are also timers of a kind, but they are very versatile
475	(and unfortunately a bit complex).	995	(and unfortunately a bit complex).
476		996
477	Unlike ev_timer's, they are not based on real time (or relative time)	997	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
478	but on wallclock time (absolute time). You can tell a periodic watcher	998	but on wallclock time (absolute time). You can tell a periodic watcher
479	to trigger "at" some specific point in time. For example, if you tell a	999	to trigger "at" some specific point in time. For example, if you tell a
480	periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now ()	1000	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()
481	+ 10.>) and then reset your system clock to the last year, then it will	1001	+ 10.>) and then reset your system clock to the last year, then it will
482	take a year to trigger the event (unlike an ev_timer, which would trigger	1002	take a year to trigger the event (unlike an C<ev_timer>, which would trigger
483	roughly 10 seconds later and of course not if you reset your system time	1003	roughly 10 seconds later and of course not if you reset your system time
484	again).	1004	again).
485		1005
486	They can also be used to implement vastly more complex timers, such as	1006	They can also be used to implement vastly more complex timers, such as
487	triggering an event on eahc midnight, local time.	1007	triggering an event on eahc midnight, local time.
488		1008
		1009	As with timers, the callback is guarenteed to be invoked only when the
		1010	time (C<at>) has been passed, but if multiple periodic timers become ready
		1011	during the same loop iteration then order of execution is undefined.
		1012
489	=over 4	1013	=over 4
490		1014
491	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1015	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
492		1016
493	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1017	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)
494		1018
495	Lots of arguments, lets sort it out... There are basically three modes of	1019	Lots of arguments, lets sort it out... There are basically three modes of
496	operation, and we will explain them from simplest to complex:	1020	operation, and we will explain them from simplest to complex:
497
498		1021
499	=over 4	1022	=over 4
500		1023
501	=item * absolute timer (interval = reschedule_cb = 0)	1024	=item * absolute timer (interval = reschedule_cb = 0)
502		1025
…		…
516		1039
517	ev_periodic_set (&periodic, 0., 3600., 0);	1040	ev_periodic_set (&periodic, 0., 3600., 0);
518		1041
519	This doesn't mean there will always be 3600 seconds in between triggers,	1042	This doesn't mean there will always be 3600 seconds in between triggers,
520	but only that the the callback will be called when the system time shows a	1043	but only that the the callback will be called when the system time shows a
521	full hour (UTC), or more correct, when the system time is evenly divisible	1044	full hour (UTC), or more correctly, when the system time is evenly divisible
522	by 3600.	1045	by 3600.
523		1046
524	Another way to think about it (for the mathematically inclined) is that	1047	Another way to think about it (for the mathematically inclined) is that
525	ev_periodic will try to run the callback in this mode at the next possible	1048	C<ev_periodic> will try to run the callback in this mode at the next possible
526	time where C<time = at (mod interval)>, regardless of any time jumps.	1049	time where C<time = at (mod interval)>, regardless of any time jumps.
527		1050
528	=item * manual reschedule mode (reschedule_cb = callback)	1051	=item * manual reschedule mode (reschedule_cb = callback)
529		1052
530	In this mode the values for C<interval> and C<at> are both being	1053	In this mode the values for C<interval> and C<at> are both being
531	ignored. Instead, each time the periodic watcher gets scheduled, the	1054	ignored. Instead, each time the periodic watcher gets scheduled, the
532	reschedule callback will be called with the watcher as first, and the	1055	reschedule callback will be called with the watcher as first, and the
533	current time as second argument.	1056	current time as second argument.
534		1057
535	NOTE: I<This callback MUST NOT stop or destroy the periodic or any other	1058	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
536	periodic watcher, ever, or make any event loop modificstions>. If you need	1059	ever, or make any event loop modifications>. If you need to stop it,
537	to stop it, return 1e30 (or so, fudge fudge) and stop it afterwards.	1060	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
		1061	starting a prepare watcher).
538		1062
539	Its prototype is c<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1063	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,
540	ev_tstamp now)>, e.g.:	1064	ev_tstamp now)>, e.g.:
541		1065
542	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1066	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
543	{	1067	{
544	return now + 60.;	1068	return now + 60.;
…		…
547	It must return the next time to trigger, based on the passed time value	1071	It must return the next time to trigger, based on the passed time value
548	(that is, the lowest time value larger than to the second argument). It	1072	(that is, the lowest time value larger than to the second argument). It
549	will usually be called just before the callback will be triggered, but	1073	will usually be called just before the callback will be triggered, but
550	might be called at other times, too.	1074	might be called at other times, too.
551		1075
		1076	NOTE: I<< This callback must always return a time that is later than the
		1077	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.
		1078
552	This can be used to create very complex timers, such as a timer that	1079	This can be used to create very complex timers, such as a timer that
553	triggers on each midnight, local time. To do this, you would calculate the	1080	triggers on each midnight, local time. To do this, you would calculate the
554	next midnight after C<now> and return the timestamp value for this. How you do this	1081	next midnight after C<now> and return the timestamp value for this. How
555	is, again, up to you (but it is not trivial).	1082	you do this is, again, up to you (but it is not trivial, which is the main
		1083	reason I omitted it as an example).
556		1084
557	=back	1085	=back
558		1086
559	=item ev_periodic_again (loop, ev_periodic *)	1087	=item ev_periodic_again (loop, ev_periodic *)
560		1088
561	Simply stops and restarts the periodic watcher again. This is only useful	1089	Simply stops and restarts the periodic watcher again. This is only useful
562	when you changed some parameters or the reschedule callback would return	1090	when you changed some parameters or the reschedule callback would return
563	a different time than the last time it was called (e.g. in a crond like	1091	a different time than the last time it was called (e.g. in a crond like
564	program when the crontabs have changed).	1092	program when the crontabs have changed).
565		1093
		1094	=item ev_tstamp interval [read-write]
		1095
		1096	The current interval value. Can be modified any time, but changes only
		1097	take effect when the periodic timer fires or C<ev_periodic_again> is being
		1098	called.
		1099
		1100	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]
		1101
		1102	The current reschedule callback, or C<0>, if this functionality is
		1103	switched off. Can be changed any time, but changes only take effect when
		1104	the periodic timer fires or C<ev_periodic_again> is being called.
		1105
566	=back	1106	=back
567		1107
		1108	Example: Call a callback every hour, or, more precisely, whenever the
		1109	system clock is divisible by 3600. The callback invocation times have
		1110	potentially a lot of jittering, but good long-term stability.
		1111
		1112	static void
		1113	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
		1114	{
		1115	... its now a full hour (UTC, or TAI or whatever your clock follows)
		1116	}
		1117
		1118	struct ev_periodic hourly_tick;
		1119	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
		1120	ev_periodic_start (loop, &hourly_tick);
		1121
		1122	Example: The same as above, but use a reschedule callback to do it:
		1123
		1124	#include <math.h>
		1125
		1126	static ev_tstamp
		1127	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
		1128	{
		1129	return fmod (now, 3600.) + 3600.;
		1130	}
		1131
		1132	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
		1133
		1134	Example: Call a callback every hour, starting now:
		1135
		1136	struct ev_periodic hourly_tick;
		1137	ev_periodic_init (&hourly_tick, clock_cb,
		1138	fmod (ev_now (loop), 3600.), 3600., 0);
		1139	ev_periodic_start (loop, &hourly_tick);
		1140
		1141
568	=head2 ev_signal - signal me when a signal gets signalled	1142	=head2 C<ev_signal> - signal me when a signal gets signalled!
569		1143
570	Signal watchers will trigger an event when the process receives a specific	1144	Signal watchers will trigger an event when the process receives a specific
571	signal one or more times. Even though signals are very asynchronous, libev	1145	signal one or more times. Even though signals are very asynchronous, libev
572	will try its best to deliver signals synchronously, i.e. as part of the	1146	will try it's best to deliver signals synchronously, i.e. as part of the
573	normal event processing, like any other event.	1147	normal event processing, like any other event.
574		1148
575	You cna configure as many watchers as you like per signal. Only when the	1149	You can configure as many watchers as you like per signal. Only when the
576	first watcher gets started will libev actually register a signal watcher	1150	first watcher gets started will libev actually register a signal watcher
577	with the kernel (thus it coexists with your own signal handlers as long	1151	with the kernel (thus it coexists with your own signal handlers as long
578	as you don't register any with libev). Similarly, when the last signal	1152	as you don't register any with libev). Similarly, when the last signal
579	watcher for a signal is stopped libev will reset the signal handler to	1153	watcher for a signal is stopped libev will reset the signal handler to
580	SIG_DFL (regardless of what it was set to before).	1154	SIG_DFL (regardless of what it was set to before).
…		…
586	=item ev_signal_set (ev_signal *, int signum)	1160	=item ev_signal_set (ev_signal *, int signum)
587		1161
588	Configures the watcher to trigger on the given signal number (usually one	1162	Configures the watcher to trigger on the given signal number (usually one
589	of the C<SIGxxx> constants).	1163	of the C<SIGxxx> constants).
590		1164
		1165	=item int signum [read-only]
		1166
		1167	The signal the watcher watches out for.
		1168
591	=back	1169	=back
592		1170
		1171
593	=head2 ev_child - wait for pid status changes	1172	=head2 C<ev_child> - watch out for process status changes
594		1173
595	Child watchers trigger when your process receives a SIGCHLD in response to	1174	Child watchers trigger when your process receives a SIGCHLD in response to
596	some child status changes (most typically when a child of yours dies).	1175	some child status changes (most typically when a child of yours dies).
597		1176
598	=over 4	1177	=over 4
…		…
602	=item ev_child_set (ev_child *, int pid)	1181	=item ev_child_set (ev_child *, int pid)
603		1182
604	Configures the watcher to wait for status changes of process C<pid> (or	1183	Configures the watcher to wait for status changes of process C<pid> (or
605	I<any> process if C<pid> is specified as C<0>). The callback can look	1184	I<any> process if C<pid> is specified as C<0>). The callback can look
606	at the C<rstatus> member of the C<ev_child> watcher structure to see	1185	at the C<rstatus> member of the C<ev_child> watcher structure to see
607	the status word (use the macros from C<sys/wait.h>). The C<rpid> member	1186	the status word (use the macros from C<sys/wait.h> and see your systems
608	contains the pid of the process causing the status change.	1187	C<waitpid> documentation). The C<rpid> member contains the pid of the
		1188	process causing the status change.
		1189
		1190	=item int pid [read-only]
		1191
		1192	The process id this watcher watches out for, or C<0>, meaning any process id.
		1193
		1194	=item int rpid [read-write]
		1195
		1196	The process id that detected a status change.
		1197
		1198	=item int rstatus [read-write]
		1199
		1200	The process exit/trace status caused by C<rpid> (see your systems
		1201	C<waitpid> and C<sys/wait.h> documentation for details).
609		1202
610	=back	1203	=back
611		1204
		1205	Example: Try to exit cleanly on SIGINT and SIGTERM.
		1206
		1207	static void
		1208	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
		1209	{
		1210	ev_unloop (loop, EVUNLOOP_ALL);
		1211	}
		1212
		1213	struct ev_signal signal_watcher;
		1214	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1215	ev_signal_start (loop, &sigint_cb);
		1216
		1217
		1218	=head2 C<ev_stat> - did the file attributes just change?
		1219
		1220	This watches a filesystem path for attribute changes. That is, it calls
		1221	C<stat> regularly (or when the OS says it changed) and sees if it changed
		1222	compared to the last time, invoking the callback if it did.
		1223
		1224	The path does not need to exist: changing from "path exists" to "path does
		1225	not exist" is a status change like any other. The condition "path does
		1226	not exist" is signified by the C<st_nlink> field being zero (which is
		1227	otherwise always forced to be at least one) and all the other fields of
		1228	the stat buffer having unspecified contents.
		1229
		1230	The path I<should> be absolute and I<must not> end in a slash. If it is
		1231	relative and your working directory changes, the behaviour is undefined.
		1232
		1233	Since there is no standard to do this, the portable implementation simply
		1234	calls C<stat (2)> regularly on the path to see if it changed somehow. You
		1235	can specify a recommended polling interval for this case. If you specify
		1236	a polling interval of C<0> (highly recommended!) then a I<suitable,
		1237	unspecified default> value will be used (which you can expect to be around
		1238	five seconds, although this might change dynamically). Libev will also
		1239	impose a minimum interval which is currently around C<0.1>, but thats
		1240	usually overkill.
		1241
		1242	This watcher type is not meant for massive numbers of stat watchers,
		1243	as even with OS-supported change notifications, this can be
		1244	resource-intensive.
		1245
		1246	At the time of this writing, only the Linux inotify interface is
		1247	implemented (implementing kqueue support is left as an exercise for the
		1248	reader). Inotify will be used to give hints only and should not change the
		1249	semantics of C<ev_stat> watchers, which means that libev sometimes needs
		1250	to fall back to regular polling again even with inotify, but changes are
		1251	usually detected immediately, and if the file exists there will be no
		1252	polling.
		1253
		1254	=over 4
		1255
		1256	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
		1257
		1258	=item ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)
		1259
		1260	Configures the watcher to wait for status changes of the given
		1261	C<path>. The C<interval> is a hint on how quickly a change is expected to
		1262	be detected and should normally be specified as C<0> to let libev choose
		1263	a suitable value. The memory pointed to by C<path> must point to the same
		1264	path for as long as the watcher is active.
		1265
		1266	The callback will be receive C<EV_STAT> when a change was detected,
		1267	relative to the attributes at the time the watcher was started (or the
		1268	last change was detected).
		1269
		1270	=item ev_stat_stat (ev_stat *)
		1271
		1272	Updates the stat buffer immediately with new values. If you change the
		1273	watched path in your callback, you could call this fucntion to avoid
		1274	detecting this change (while introducing a race condition). Can also be
		1275	useful simply to find out the new values.
		1276
		1277	=item ev_statdata attr [read-only]
		1278
		1279	The most-recently detected attributes of the file. Although the type is of
		1280	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
		1281	suitable for your system. If the C<st_nlink> member is C<0>, then there
		1282	was some error while C<stat>ing the file.
		1283
		1284	=item ev_statdata prev [read-only]
		1285
		1286	The previous attributes of the file. The callback gets invoked whenever
		1287	C<prev> != C<attr>.
		1288
		1289	=item ev_tstamp interval [read-only]
		1290
		1291	The specified interval.
		1292
		1293	=item const char *path [read-only]
		1294
		1295	The filesystem path that is being watched.
		1296
		1297	=back
		1298
		1299	Example: Watch C</etc/passwd> for attribute changes.
		1300
		1301	static void
		1302	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
		1303	{
		1304	/* /etc/passwd changed in some way */
		1305	if (w->attr.st_nlink)
		1306	{
		1307	printf ("passwd current size %ld\n", (long)w->attr.st_size);
		1308	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
		1309	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
		1310	}
		1311	else
		1312	/* you shalt not abuse printf for puts */
		1313	puts ("wow, /etc/passwd is not there, expect problems. "
		1314	"if this is windows, they already arrived\n");
		1315	}
		1316
		1317	...
		1318	ev_stat passwd;
		1319
		1320	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");
		1321	ev_stat_start (loop, &passwd);
		1322
		1323
612	=head2 ev_idle - when you've got nothing better to do	1324	=head2 C<ev_idle> - when you've got nothing better to do...
613		1325
614	Idle watchers trigger events when there are no other I/O or timer (or	1326	Idle watchers trigger events when there are no other events are pending
615	periodic) events pending. That is, as long as your process is busy	1327	(prepare, check and other idle watchers do not count). That is, as long
616	handling sockets or timeouts it will not be called. But when your process	1328	as your process is busy handling sockets or timeouts (or even signals,
617	is idle all idle watchers are being called again and again - until	1329	imagine) it will not be triggered. But when your process is idle all idle
		1330	watchers are being called again and again, once per event loop iteration -
618	stopped, that is, or your process receives more events.	1331	until stopped, that is, or your process receives more events and becomes
		1332	busy.
619		1333
620	The most noteworthy effect is that as long as any idle watchers are	1334	The most noteworthy effect is that as long as any idle watchers are
621	active, the process will not block when waiting for new events.	1335	active, the process will not block when waiting for new events.
622		1336
623	Apart from keeping your process non-blocking (which is a useful	1337	Apart from keeping your process non-blocking (which is a useful
…		…
633	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1347	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
634	believe me.	1348	believe me.
635		1349
636	=back	1350	=back
637		1351
638	=head2 prepare and check - your hooks into the event loop	1352	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
		1353	callback, free it. Also, use no error checking, as usual.
639		1354
		1355	static void
		1356	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
		1357	{
		1358	free (w);
		1359	// now do something you wanted to do when the program has
		1360	// no longer asnything immediate to do.
		1361	}
		1362
		1363	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
		1364	ev_idle_init (idle_watcher, idle_cb);
		1365	ev_idle_start (loop, idle_cb);
		1366
		1367
		1368	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
		1369
640	Prepare and check watchers usually (but not always) are used in	1370	Prepare and check watchers are usually (but not always) used in tandem:
641	tandom. Prepare watchers get invoked before the process blocks and check	1371	prepare watchers get invoked before the process blocks and check watchers
642	watchers afterwards.	1372	afterwards.
643		1373
		1374	You I<must not> call C<ev_loop> or similar functions that enter
		1375	the current event loop from either C<ev_prepare> or C<ev_check>
		1376	watchers. Other loops than the current one are fine, however. The
		1377	rationale behind this is that you do not need to check for recursion in
		1378	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
		1379	C<ev_check> so if you have one watcher of each kind they will always be
		1380	called in pairs bracketing the blocking call.
		1381
644	Their main purpose is to integrate other event mechanisms into libev. This	1382	Their main purpose is to integrate other event mechanisms into libev and
645	could be used, for example, to track variable changes, implement your own	1383	their use is somewhat advanced. This could be used, for example, to track
646	watchers, integrate net-snmp or a coroutine library and lots more.	1384	variable changes, implement your own watchers, integrate net-snmp or a
		1385	coroutine library and lots more. They are also occasionally useful if
		1386	you cache some data and want to flush it before blocking (for example,
		1387	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
		1388	watcher).
647		1389
648	This is done by examining in each prepare call which file descriptors need	1390	This is done by examining in each prepare call which file descriptors need
649	to be watched by the other library, registering ev_io watchers for them	1391	to be watched by the other library, registering C<ev_io> watchers for
650	and starting an ev_timer watcher for any timeouts (many libraries provide	1392	them and starting an C<ev_timer> watcher for any timeouts (many libraries
651	just this functionality). Then, in the check watcher you check for any	1393	provide just this functionality). Then, in the check watcher you check for
652	events that occured (by making your callbacks set soem flags for example)	1394	any events that occured (by checking the pending status of all watchers
653	and call back into the library.	1395	and stopping them) and call back into the library. The I/O and timer
		1396	callbacks will never actually be called (but must be valid nevertheless,
		1397	because you never know, you know?).
654		1398
655	As another example, the perl Coro module uses these hooks to integrate	1399	As another example, the Perl Coro module uses these hooks to integrate
656	coroutines into libev programs, by yielding to other active coroutines	1400	coroutines into libev programs, by yielding to other active coroutines
657	during each prepare and only letting the process block if no coroutines	1401	during each prepare and only letting the process block if no coroutines
658	are ready to run.	1402	are ready to run (it's actually more complicated: it only runs coroutines
		1403	with priority higher than or equal to the event loop and one coroutine
		1404	of lower priority, but only once, using idle watchers to keep the event
		1405	loop from blocking if lower-priority coroutines are active, thus mapping
		1406	low-priority coroutines to idle/background tasks).
659		1407
660	=over 4	1408	=over 4
661		1409
662	=item ev_prepare_init (ev_prepare *, callback)	1410	=item ev_prepare_init (ev_prepare *, callback)
663		1411
664	=item ev_check_init (ev_check *, callback)	1412	=item ev_check_init (ev_check *, callback)
665		1413
666	Initialises and configures the prepare or check watcher - they have no	1414	Initialises and configures the prepare or check watcher - they have no
667	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1415	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
668	macros, but using them is utterly, utterly pointless.	1416	macros, but using them is utterly, utterly and completely pointless.
669		1417
670	=back	1418	=back
671		1419
		1420	Example: To include a library such as adns, you would add IO watchers
		1421	and a timeout watcher in a prepare handler, as required by libadns, and
		1422	in a check watcher, destroy them and call into libadns. What follows is
		1423	pseudo-code only of course:
		1424
		1425	static ev_io iow [nfd];
		1426	static ev_timer tw;
		1427
		1428	static void
		1429	io_cb (ev_loop *loop, ev_io *w, int revents)
		1430	{
		1431	// set the relevant poll flags
		1432	// could also call adns_processreadable etc. here
		1433	struct pollfd *fd = (struct pollfd *)w->data;
		1434	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		1435	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		1436	}
		1437
		1438	// create io watchers for each fd and a timer before blocking
		1439	static void
		1440	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
		1441	{
		1442	int timeout = 3600000;truct pollfd fds [nfd];
		1443	// actual code will need to loop here and realloc etc.
		1444	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
		1445
		1446	/* the callback is illegal, but won't be called as we stop during check */
		1447	ev_timer_init (&tw, 0, timeout * 1e-3);
		1448	ev_timer_start (loop, &tw);
		1449
		1450	// create on ev_io per pollfd
		1451	for (int i = 0; i < nfd; ++i)
		1452	{
		1453	ev_io_init (iow + i, io_cb, fds [i].fd,
		1454	((fds [i].events & POLLIN ? EV_READ : 0)
		1455	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
		1456
		1457	fds [i].revents = 0;
		1458	iow [i].data = fds + i;
		1459	ev_io_start (loop, iow + i);
		1460	}
		1461	}
		1462
		1463	// stop all watchers after blocking
		1464	static void
		1465	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
		1466	{
		1467	ev_timer_stop (loop, &tw);
		1468
		1469	for (int i = 0; i < nfd; ++i)
		1470	ev_io_stop (loop, iow + i);
		1471
		1472	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
		1473	}
		1474
		1475
		1476	=head2 C<ev_embed> - when one backend isn't enough...
		1477
		1478	This is a rather advanced watcher type that lets you embed one event loop
		1479	into another (currently only C<ev_io> events are supported in the embedded
		1480	loop, other types of watchers might be handled in a delayed or incorrect
		1481	fashion and must not be used).
		1482
		1483	There are primarily two reasons you would want that: work around bugs and
		1484	prioritise I/O.
		1485
		1486	As an example for a bug workaround, the kqueue backend might only support
		1487	sockets on some platform, so it is unusable as generic backend, but you
		1488	still want to make use of it because you have many sockets and it scales
		1489	so nicely. In this case, you would create a kqueue-based loop and embed it
		1490	into your default loop (which might use e.g. poll). Overall operation will
		1491	be a bit slower because first libev has to poll and then call kevent, but
		1492	at least you can use both at what they are best.
		1493
		1494	As for prioritising I/O: rarely you have the case where some fds have
		1495	to be watched and handled very quickly (with low latency), and even
		1496	priorities and idle watchers might have too much overhead. In this case
		1497	you would put all the high priority stuff in one loop and all the rest in
		1498	a second one, and embed the second one in the first.
		1499
		1500	As long as the watcher is active, the callback will be invoked every time
		1501	there might be events pending in the embedded loop. The callback must then
		1502	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke
		1503	their callbacks (you could also start an idle watcher to give the embedded
		1504	loop strictly lower priority for example). You can also set the callback
		1505	to C<0>, in which case the embed watcher will automatically execute the
		1506	embedded loop sweep.
		1507
		1508	As long as the watcher is started it will automatically handle events. The
		1509	callback will be invoked whenever some events have been handled. You can
		1510	set the callback to C<0> to avoid having to specify one if you are not
		1511	interested in that.
		1512
		1513	Also, there have not currently been made special provisions for forking:
		1514	when you fork, you not only have to call C<ev_loop_fork> on both loops,
		1515	but you will also have to stop and restart any C<ev_embed> watchers
		1516	yourself.
		1517
		1518	Unfortunately, not all backends are embeddable, only the ones returned by
		1519	C<ev_embeddable_backends> are, which, unfortunately, does not include any
		1520	portable one.
		1521
		1522	So when you want to use this feature you will always have to be prepared
		1523	that you cannot get an embeddable loop. The recommended way to get around
		1524	this is to have a separate variables for your embeddable loop, try to
		1525	create it, and if that fails, use the normal loop for everything:
		1526
		1527	struct ev_loop *loop_hi = ev_default_init (0);
		1528	struct ev_loop *loop_lo = 0;
		1529	struct ev_embed embed;
		1530
		1531	// see if there is a chance of getting one that works
		1532	// (remember that a flags value of 0 means autodetection)
		1533	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		1534	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		1535	: 0;
		1536
		1537	// if we got one, then embed it, otherwise default to loop_hi
		1538	if (loop_lo)
		1539	{
		1540	ev_embed_init (&embed, 0, loop_lo);
		1541	ev_embed_start (loop_hi, &embed);
		1542	}
		1543	else
		1544	loop_lo = loop_hi;
		1545
		1546	=over 4
		1547
		1548	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
		1549
		1550	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)
		1551
		1552	Configures the watcher to embed the given loop, which must be
		1553	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		1554	invoked automatically, otherwise it is the responsibility of the callback
		1555	to invoke it (it will continue to be called until the sweep has been done,
		1556	if you do not want thta, you need to temporarily stop the embed watcher).
		1557
		1558	=item ev_embed_sweep (loop, ev_embed *)
		1559
		1560	Make a single, non-blocking sweep over the embedded loop. This works
		1561	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		1562	apropriate way for embedded loops.
		1563
		1564	=item struct ev_loop *loop [read-only]
		1565
		1566	The embedded event loop.
		1567
		1568	=back
		1569
		1570
		1571	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
		1572
		1573	Fork watchers are called when a C<fork ()> was detected (usually because
		1574	whoever is a good citizen cared to tell libev about it by calling
		1575	C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the
		1576	event loop blocks next and before C<ev_check> watchers are being called,
		1577	and only in the child after the fork. If whoever good citizen calling
		1578	C<ev_default_fork> cheats and calls it in the wrong process, the fork
		1579	handlers will be invoked, too, of course.
		1580
		1581	=over 4
		1582
		1583	=item ev_fork_init (ev_signal *, callback)
		1584
		1585	Initialises and configures the fork watcher - it has no parameters of any
		1586	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
		1587	believe me.
		1588
		1589	=back
		1590
		1591
672	=head1 OTHER FUNCTIONS	1592	=head1 OTHER FUNCTIONS
673		1593
674	There are some other fucntions of possible interest. Described. Here. Now.	1594	There are some other functions of possible interest. Described. Here. Now.
675		1595
676	=over 4	1596	=over 4
677		1597
678	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	1598	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
679		1599
680	This function combines a simple timer and an I/O watcher, calls your	1600	This function combines a simple timer and an I/O watcher, calls your
681	callback on whichever event happens first and automatically stop both	1601	callback on whichever event happens first and automatically stop both
682	watchers. This is useful if you want to wait for a single event on an fd	1602	watchers. This is useful if you want to wait for a single event on an fd
683	or timeout without havign to allocate/configure/start/stop/free one or	1603	or timeout without having to allocate/configure/start/stop/free one or
684	more watchers yourself.	1604	more watchers yourself.
685		1605
686	If C<fd> is less than 0, then no I/O watcher will be started and events is	1606	If C<fd> is less than 0, then no I/O watcher will be started and events
687	ignored. Otherwise, an ev_io watcher for the given C<fd> and C<events> set	1607	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and
688	will be craeted and started.	1608	C<events> set will be craeted and started.
689		1609
690	If C<timeout> is less than 0, then no timeout watcher will be	1610	If C<timeout> is less than 0, then no timeout watcher will be
691	started. Otherwise an ev_timer watcher with after = C<timeout> (and repeat	1611	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
692	= 0) will be started.	1612	repeat = 0) will be started. While C<0> is a valid timeout, it is of
		1613	dubious value.
693		1614
694	The callback has the type C<void (*cb)(int revents, void *arg)> and	1615	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
695	gets passed an events set (normally a combination of EV_ERROR, EV_READ,	1616	passed an C<revents> set like normal event callbacks (a combination of
696	EV_WRITE or EV_TIMEOUT) and the C<arg> value passed to C<ev_once>:	1617	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
		1618	value passed to C<ev_once>:
697		1619
698	static void stdin_ready (int revents, void *arg)	1620	static void stdin_ready (int revents, void *arg)
699	{	1621	{
700	if (revents & EV_TIMEOUT)	1622	if (revents & EV_TIMEOUT)
701	/* doh, nothing entered */	1623	/* doh, nothing entered */;
702	else if (revents & EV_READ)	1624	else if (revents & EV_READ)
703	/* stdin might have data for us, joy! */	1625	/* stdin might have data for us, joy! */;
704	}	1626	}
705		1627
706	ev_once (STDIN_FILENO, EV_READm 10., stdin_ready, 0);	1628	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
707		1629
708	=item ev_feed_event (loop, watcher, int events)	1630	=item ev_feed_event (ev_loop *, watcher *, int revents)
709		1631
710	Feeds the given event set into the event loop, as if the specified event	1632	Feeds the given event set into the event loop, as if the specified event
711	has happened for the specified watcher (which must be a pointer to an	1633	had happened for the specified watcher (which must be a pointer to an
712	initialised but not necessarily active event watcher).	1634	initialised but not necessarily started event watcher).
713		1635
714	=item ev_feed_fd_event (loop, int fd, int revents)	1636	=item ev_feed_fd_event (ev_loop *, int fd, int revents)
715		1637
716	Feed an event on the given fd, as if a file descriptor backend detected it.	1638	Feed an event on the given fd, as if a file descriptor backend detected
		1639	the given events it.
717		1640
718	=item ev_feed_signal_event (loop, int signum)	1641	=item ev_feed_signal_event (ev_loop *loop, int signum)
719		1642
720	Feed an event as if the given signal occured (loop must be the default loop!).	1643	Feed an event as if the given signal occured (C<loop> must be the default
		1644	loop!).
721		1645
722	=back	1646	=back
723		1647
		1648
		1649	=head1 LIBEVENT EMULATION
		1650
		1651	Libev offers a compatibility emulation layer for libevent. It cannot
		1652	emulate the internals of libevent, so here are some usage hints:
		1653
		1654	=over 4
		1655
		1656	=item * Use it by including <event.h>, as usual.
		1657
		1658	=item * The following members are fully supported: ev_base, ev_callback,
		1659	ev_arg, ev_fd, ev_res, ev_events.
		1660
		1661	=item * Avoid using ev_flags and the EVLIST_*-macros, while it is
		1662	maintained by libev, it does not work exactly the same way as in libevent (consider
		1663	it a private API).
		1664
		1665	=item * Priorities are not currently supported. Initialising priorities
		1666	will fail and all watchers will have the same priority, even though there
		1667	is an ev_pri field.
		1668
		1669	=item * Other members are not supported.
		1670
		1671	=item * The libev emulation is I<not> ABI compatible to libevent, you need
		1672	to use the libev header file and library.
		1673
		1674	=back
		1675
		1676	=head1 C++ SUPPORT
		1677
		1678	Libev comes with some simplistic wrapper classes for C++ that mainly allow
		1679	you to use some convinience methods to start/stop watchers and also change
		1680	the callback model to a model using method callbacks on objects.
		1681
		1682	To use it,
		1683
		1684	#include <ev++.h>
		1685
		1686	(it is not installed by default). This automatically includes F<ev.h>
		1687	and puts all of its definitions (many of them macros) into the global
		1688	namespace. All C++ specific things are put into the C<ev> namespace.
		1689
		1690	It should support all the same embedding options as F<ev.h>, most notably
		1691	C<EV_MULTIPLICITY>.
		1692
		1693	Here is a list of things available in the C<ev> namespace:
		1694
		1695	=over 4
		1696
		1697	=item C<ev::READ>, C<ev::WRITE> etc.
		1698
		1699	These are just enum values with the same values as the C<EV_READ> etc.
		1700	macros from F<ev.h>.
		1701
		1702	=item C<ev::tstamp>, C<ev::now>
		1703
		1704	Aliases to the same types/functions as with the C<ev_> prefix.
		1705
		1706	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.
		1707
		1708	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of
		1709	the same name in the C<ev> namespace, with the exception of C<ev_signal>
		1710	which is called C<ev::sig> to avoid clashes with the C<signal> macro
		1711	defines by many implementations.
		1712
		1713	All of those classes have these methods:
		1714
		1715	=over 4
		1716
		1717	=item ev::TYPE::TYPE (object *, object::method *)
		1718
		1719	=item ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)
		1720
		1721	=item ev::TYPE::~TYPE
		1722
		1723	The constructor takes a pointer to an object and a method pointer to
		1724	the event handler callback to call in this class. The constructor calls
		1725	C<ev_init> for you, which means you have to call the C<set> method
		1726	before starting it. If you do not specify a loop then the constructor
		1727	automatically associates the default loop with this watcher.
		1728
		1729	The destructor automatically stops the watcher if it is active.
		1730
		1731	=item w->set (struct ev_loop *)
		1732
		1733	Associates a different C<struct ev_loop> with this watcher. You can only
		1734	do this when the watcher is inactive (and not pending either).
		1735
		1736	=item w->set ([args])
		1737
		1738	Basically the same as C<ev_TYPE_set>, with the same args. Must be
		1739	called at least once. Unlike the C counterpart, an active watcher gets
		1740	automatically stopped and restarted.
		1741
		1742	=item w->start ()
		1743
		1744	Starts the watcher. Note that there is no C<loop> argument as the
		1745	constructor already takes the loop.
		1746
		1747	=item w->stop ()
		1748
		1749	Stops the watcher if it is active. Again, no C<loop> argument.
		1750
		1751	=item w->again () C<ev::timer>, C<ev::periodic> only
		1752
		1753	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
		1754	C<ev_TYPE_again> function.
		1755
		1756	=item w->sweep () C<ev::embed> only
		1757
		1758	Invokes C<ev_embed_sweep>.
		1759
		1760	=item w->update () C<ev::stat> only
		1761
		1762	Invokes C<ev_stat_stat>.
		1763
		1764	=back
		1765
		1766	=back
		1767
		1768	Example: Define a class with an IO and idle watcher, start one of them in
		1769	the constructor.
		1770
		1771	class myclass
		1772	{
		1773	ev_io io; void io_cb (ev::io &w, int revents);
		1774	ev_idle idle void idle_cb (ev::idle &w, int revents);
		1775
		1776	myclass ();
		1777	}
		1778
		1779	myclass::myclass (int fd)
		1780	: io (this, &myclass::io_cb),
		1781	idle (this, &myclass::idle_cb)
		1782	{
		1783	io.start (fd, ev::READ);
		1784	}
		1785
		1786
		1787	=head1 MACRO MAGIC
		1788
		1789	Libev can be compiled with a variety of options, the most fundemantal is
		1790	C<EV_MULTIPLICITY>. This option determines wether (most) functions and
		1791	callbacks have an initial C<struct ev_loop *> argument.
		1792
		1793	To make it easier to write programs that cope with either variant, the
		1794	following macros are defined:
		1795
		1796	=over 4
		1797
		1798	=item C<EV_A>, C<EV_A_>
		1799
		1800	This provides the loop I<argument> for functions, if one is required ("ev
		1801	loop argument"). The C<EV_A> form is used when this is the sole argument,
		1802	C<EV_A_> is used when other arguments are following. Example:
		1803
		1804	ev_unref (EV_A);
		1805	ev_timer_add (EV_A_ watcher);
		1806	ev_loop (EV_A_ 0);
		1807
		1808	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
		1809	which is often provided by the following macro.
		1810
		1811	=item C<EV_P>, C<EV_P_>
		1812
		1813	This provides the loop I<parameter> for functions, if one is required ("ev
		1814	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
		1815	C<EV_P_> is used when other parameters are following. Example:
		1816
		1817	// this is how ev_unref is being declared
		1818	static void ev_unref (EV_P);
		1819
		1820	// this is how you can declare your typical callback
		1821	static void cb (EV_P_ ev_timer *w, int revents)
		1822
		1823	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
		1824	suitable for use with C<EV_A>.
		1825
		1826	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
		1827
		1828	Similar to the other two macros, this gives you the value of the default
		1829	loop, if multiple loops are supported ("ev loop default").
		1830
		1831	=back
		1832
		1833	Example: Declare and initialise a check watcher, working regardless of
		1834	wether multiple loops are supported or not.
		1835
		1836	static void
		1837	check_cb (EV_P_ ev_timer *w, int revents)
		1838	{
		1839	ev_check_stop (EV_A_ w);
		1840	}
		1841
		1842	ev_check check;
		1843	ev_check_init (&check, check_cb);
		1844	ev_check_start (EV_DEFAULT_ &check);
		1845	ev_loop (EV_DEFAULT_ 0);
		1846
		1847
		1848	=head1 EMBEDDING
		1849
		1850	Libev can (and often is) directly embedded into host
		1851	applications. Examples of applications that embed it include the Deliantra
		1852	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
		1853	and rxvt-unicode.
		1854
		1855	The goal is to enable you to just copy the neecssary files into your
		1856	source directory without having to change even a single line in them, so
		1857	you can easily upgrade by simply copying (or having a checked-out copy of
		1858	libev somewhere in your source tree).
		1859
		1860	=head2 FILESETS
		1861
		1862	Depending on what features you need you need to include one or more sets of files
		1863	in your app.
		1864
		1865	=head3 CORE EVENT LOOP
		1866
		1867	To include only the libev core (all the C<ev_*> functions), with manual
		1868	configuration (no autoconf):
		1869
		1870	#define EV_STANDALONE 1
		1871	#include "ev.c"
		1872
		1873	This will automatically include F<ev.h>, too, and should be done in a
		1874	single C source file only to provide the function implementations. To use
		1875	it, do the same for F<ev.h> in all files wishing to use this API (best
		1876	done by writing a wrapper around F<ev.h> that you can include instead and
		1877	where you can put other configuration options):
		1878
		1879	#define EV_STANDALONE 1
		1880	#include "ev.h"
		1881
		1882	Both header files and implementation files can be compiled with a C++
		1883	compiler (at least, thats a stated goal, and breakage will be treated
		1884	as a bug).
		1885
		1886	You need the following files in your source tree, or in a directory
		1887	in your include path (e.g. in libev/ when using -Ilibev):
		1888
		1889	ev.h
		1890	ev.c
		1891	ev_vars.h
		1892	ev_wrap.h
		1893
		1894	ev_win32.c required on win32 platforms only
		1895
		1896	ev_select.c only when select backend is enabled (which is by default)
		1897	ev_poll.c only when poll backend is enabled (disabled by default)
		1898	ev_epoll.c only when the epoll backend is enabled (disabled by default)
		1899	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
		1900	ev_port.c only when the solaris port backend is enabled (disabled by default)
		1901
		1902	F<ev.c> includes the backend files directly when enabled, so you only need
		1903	to compile this single file.
		1904
		1905	=head3 LIBEVENT COMPATIBILITY API
		1906
		1907	To include the libevent compatibility API, also include:
		1908
		1909	#include "event.c"
		1910
		1911	in the file including F<ev.c>, and:
		1912
		1913	#include "event.h"
		1914
		1915	in the files that want to use the libevent API. This also includes F<ev.h>.
		1916
		1917	You need the following additional files for this:
		1918
		1919	event.h
		1920	event.c
		1921
		1922	=head3 AUTOCONF SUPPORT
		1923
		1924	Instead of using C<EV_STANDALONE=1> and providing your config in
		1925	whatever way you want, you can also C<m4_include([libev.m4])> in your
		1926	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
		1927	include F<config.h> and configure itself accordingly.
		1928
		1929	For this of course you need the m4 file:
		1930
		1931	libev.m4
		1932
		1933	=head2 PREPROCESSOR SYMBOLS/MACROS
		1934
		1935	Libev can be configured via a variety of preprocessor symbols you have to define
		1936	before including any of its files. The default is not to build for multiplicity
		1937	and only include the select backend.
		1938
		1939	=over 4
		1940
		1941	=item EV_STANDALONE
		1942
		1943	Must always be C<1> if you do not use autoconf configuration, which
		1944	keeps libev from including F<config.h>, and it also defines dummy
		1945	implementations for some libevent functions (such as logging, which is not
		1946	supported). It will also not define any of the structs usually found in
		1947	F<event.h> that are not directly supported by the libev core alone.
		1948
		1949	=item EV_USE_MONOTONIC
		1950
		1951	If defined to be C<1>, libev will try to detect the availability of the
		1952	monotonic clock option at both compiletime and runtime. Otherwise no use
		1953	of the monotonic clock option will be attempted. If you enable this, you
		1954	usually have to link against librt or something similar. Enabling it when
		1955	the functionality isn't available is safe, though, althoguh you have
		1956	to make sure you link against any libraries where the C<clock_gettime>
		1957	function is hiding in (often F<-lrt>).
		1958
		1959	=item EV_USE_REALTIME
		1960
		1961	If defined to be C<1>, libev will try to detect the availability of the
		1962	realtime clock option at compiletime (and assume its availability at
		1963	runtime if successful). Otherwise no use of the realtime clock option will
		1964	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
		1965	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See tzhe note about libraries
		1966	in the description of C<EV_USE_MONOTONIC>, though.
		1967
		1968	=item EV_USE_SELECT
		1969
		1970	If undefined or defined to be C<1>, libev will compile in support for the
		1971	C<select>(2) backend. No attempt at autodetection will be done: if no
		1972	other method takes over, select will be it. Otherwise the select backend
		1973	will not be compiled in.
		1974
		1975	=item EV_SELECT_USE_FD_SET
		1976
		1977	If defined to C<1>, then the select backend will use the system C<fd_set>
		1978	structure. This is useful if libev doesn't compile due to a missing
		1979	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on
		1980	exotic systems. This usually limits the range of file descriptors to some
		1981	low limit such as 1024 or might have other limitations (winsocket only
		1982	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
		1983	influence the size of the C<fd_set> used.
		1984
		1985	=item EV_SELECT_IS_WINSOCKET
		1986
		1987	When defined to C<1>, the select backend will assume that
		1988	select/socket/connect etc. don't understand file descriptors but
		1989	wants osf handles on win32 (this is the case when the select to
		1990	be used is the winsock select). This means that it will call
		1991	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
		1992	it is assumed that all these functions actually work on fds, even
		1993	on win32. Should not be defined on non-win32 platforms.
		1994
		1995	=item EV_USE_POLL
		1996
		1997	If defined to be C<1>, libev will compile in support for the C<poll>(2)
		1998	backend. Otherwise it will be enabled on non-win32 platforms. It
		1999	takes precedence over select.
		2000
		2001	=item EV_USE_EPOLL
		2002
		2003	If defined to be C<1>, libev will compile in support for the Linux
		2004	C<epoll>(7) backend. Its availability will be detected at runtime,
		2005	otherwise another method will be used as fallback. This is the
		2006	preferred backend for GNU/Linux systems.
		2007
		2008	=item EV_USE_KQUEUE
		2009
		2010	If defined to be C<1>, libev will compile in support for the BSD style
		2011	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
		2012	otherwise another method will be used as fallback. This is the preferred
		2013	backend for BSD and BSD-like systems, although on most BSDs kqueue only
		2014	supports some types of fds correctly (the only platform we found that
		2015	supports ptys for example was NetBSD), so kqueue might be compiled in, but
		2016	not be used unless explicitly requested. The best way to use it is to find
		2017	out whether kqueue supports your type of fd properly and use an embedded
		2018	kqueue loop.
		2019
		2020	=item EV_USE_PORT
		2021
		2022	If defined to be C<1>, libev will compile in support for the Solaris
		2023	10 port style backend. Its availability will be detected at runtime,
		2024	otherwise another method will be used as fallback. This is the preferred
		2025	backend for Solaris 10 systems.
		2026
		2027	=item EV_USE_DEVPOLL
		2028
		2029	reserved for future expansion, works like the USE symbols above.
		2030
		2031	=item EV_USE_INOTIFY
		2032
		2033	If defined to be C<1>, libev will compile in support for the Linux inotify
		2034	interface to speed up C<ev_stat> watchers. Its actual availability will
		2035	be detected at runtime.
		2036
		2037	=item EV_H
		2038
		2039	The name of the F<ev.h> header file used to include it. The default if
		2040	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This
		2041	can be used to virtually rename the F<ev.h> header file in case of conflicts.
		2042
		2043	=item EV_CONFIG_H
		2044
		2045	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
		2046	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
		2047	C<EV_H>, above.
		2048
		2049	=item EV_EVENT_H
		2050
		2051	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
		2052	of how the F<event.h> header can be found.
		2053
		2054	=item EV_PROTOTYPES
		2055
		2056	If defined to be C<0>, then F<ev.h> will not define any function
		2057	prototypes, but still define all the structs and other symbols. This is
		2058	occasionally useful if you want to provide your own wrapper functions
		2059	around libev functions.
		2060
		2061	=item EV_MULTIPLICITY
		2062
		2063	If undefined or defined to C<1>, then all event-loop-specific functions
		2064	will have the C<struct ev_loop *> as first argument, and you can create
		2065	additional independent event loops. Otherwise there will be no support
		2066	for multiple event loops and there is no first event loop pointer
		2067	argument. Instead, all functions act on the single default loop.
		2068
		2069	=item EV_PERIODIC_ENABLE
		2070
		2071	If undefined or defined to be C<1>, then periodic timers are supported. If
		2072	defined to be C<0>, then they are not. Disabling them saves a few kB of
		2073	code.
		2074
		2075	=item EV_EMBED_ENABLE
		2076
		2077	If undefined or defined to be C<1>, then embed watchers are supported. If
		2078	defined to be C<0>, then they are not.
		2079
		2080	=item EV_STAT_ENABLE
		2081
		2082	If undefined or defined to be C<1>, then stat watchers are supported. If
		2083	defined to be C<0>, then they are not.
		2084
		2085	=item EV_FORK_ENABLE
		2086
		2087	If undefined or defined to be C<1>, then fork watchers are supported. If
		2088	defined to be C<0>, then they are not.
		2089
		2090	=item EV_MINIMAL
		2091
		2092	If you need to shave off some kilobytes of code at the expense of some
		2093	speed, define this symbol to C<1>. Currently only used for gcc to override
		2094	some inlining decisions, saves roughly 30% codesize of amd64.
		2095
		2096	=item EV_PID_HASHSIZE
		2097
		2098	C<ev_child> watchers use a small hash table to distribute workload by
		2099	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
		2100	than enough. If you need to manage thousands of children you might want to
		2101	increase this value (I<must> be a power of two).
		2102
		2103	=item EV_INOTIFY_HASHSIZE
		2104
		2105	C<ev_staz> watchers use a small hash table to distribute workload by
		2106	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
		2107	usually more than enough. If you need to manage thousands of C<ev_stat>
		2108	watchers you might want to increase this value (I<must> be a power of
		2109	two).
		2110
		2111	=item EV_COMMON
		2112
		2113	By default, all watchers have a C<void *data> member. By redefining
		2114	this macro to a something else you can include more and other types of
		2115	members. You have to define it each time you include one of the files,
		2116	though, and it must be identical each time.
		2117
		2118	For example, the perl EV module uses something like this:
		2119
		2120	#define EV_COMMON \
		2121	SV *self; /* contains this struct */ \
		2122	SV *cb_sv, *fh /* note no trailing ";" */
		2123
		2124	=item EV_CB_DECLARE (type)
		2125
		2126	=item EV_CB_INVOKE (watcher, revents)
		2127
		2128	=item ev_set_cb (ev, cb)
		2129
		2130	Can be used to change the callback member declaration in each watcher,
		2131	and the way callbacks are invoked and set. Must expand to a struct member
		2132	definition and a statement, respectively. See the F<ev.v> header file for
		2133	their default definitions. One possible use for overriding these is to
		2134	avoid the C<struct ev_loop *> as first argument in all cases, or to use
		2135	method calls instead of plain function calls in C++.
		2136
		2137	=head2 EXAMPLES
		2138
		2139	For a real-world example of a program the includes libev
		2140	verbatim, you can have a look at the EV perl module
		2141	(L<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
		2142	the F<libev/> subdirectory and includes them in the F<EV/EVAPI.h> (public
		2143	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
		2144	will be compiled. It is pretty complex because it provides its own header
		2145	file.
		2146
		2147	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
		2148	that everybody includes and which overrides some autoconf choices:
		2149
		2150	#define EV_USE_POLL 0
		2151	#define EV_MULTIPLICITY 0
		2152	#define EV_PERIODICS 0
		2153	#define EV_CONFIG_H <config.h>
		2154
		2155	#include "ev++.h"
		2156
		2157	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
		2158
		2159	#include "ev_cpp.h"
		2160	#include "ev.c"
		2161
		2162
		2163	=head1 COMPLEXITIES
		2164
		2165	In this section the complexities of (many of) the algorithms used inside
		2166	libev will be explained. For complexity discussions about backends see the
		2167	documentation for C<ev_default_init>.
		2168
		2169	=over 4
		2170
		2171	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		2172
		2173	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)
		2174
		2175	=item Starting io/check/prepare/idle/signal/child watchers: O(1)
		2176
		2177	=item Stopping check/prepare/idle watchers: O(1)
		2178
		2179	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		2180
		2181	=item Finding the next timer per loop iteration: O(1)
		2182
		2183	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		2184
		2185	=item Activating one watcher: O(1)
		2186
		2187	=back
		2188
		2189
724	=head1 AUTHOR	2190	=head1 AUTHOR
725		2191
726	Marc Lehmann <libev@schmorp.de>.	2192	Marc Lehmann <libev@schmorp.de>.
727		2193

Diff Legend

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