ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.53 by root, Tue Nov 27 20:15:02 2007 UTC vs.
Revision 1.157 by root, Tue May 20 23:49:41 2008 UTC

…		…
2		2
3	libev - a high performance full-featured event loop written in C	3	libev - a high performance full-featured event loop written in C
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	/* this is the only header you need */
8	#include <ev.h>	7	#include <ev.h>
9		8
10	/* what follows is a fully working example program */	9	=head2 EXAMPLE PROGRAM
		10
		11	// a single header file is required
		12	#include <ev.h>
		13
		14	// every watcher type has its own typedef'd struct
		15	// with the name ev_<type>
11	ev_io stdin_watcher;	16	ev_io stdin_watcher;
12	ev_timer timeout_watcher;	17	ev_timer timeout_watcher;
13		18
		19	// all watcher callbacks have a similar signature
14	/* called when data readable on stdin */	20	// this callback is called when data is readable on stdin
15	static void	21	static void
16	stdin_cb (EV_P_ struct ev_io *w, int revents)	22	stdin_cb (EV_P_ struct ev_io *w, int revents)
17	{	23	{
18	/* puts ("stdin ready"); */	24	puts ("stdin ready");
19	ev_io_stop (EV_A_ w); /* just a syntax example */	25	// for one-shot events, one must manually stop the watcher
20	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */	26	// with its corresponding stop function.
		27	ev_io_stop (EV_A_ w);
		28
		29	// this causes all nested ev_loop's to stop iterating
		30	ev_unloop (EV_A_ EVUNLOOP_ALL);
21	}	31	}
22		32
		33	// another callback, this time for a time-out
23	static void	34	static void
24	timeout_cb (EV_P_ struct ev_timer *w, int revents)	35	timeout_cb (EV_P_ struct ev_timer *w, int revents)
25	{	36	{
26	/* puts ("timeout"); */	37	puts ("timeout");
27	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */	38	// this causes the innermost ev_loop to stop iterating
		39	ev_unloop (EV_A_ EVUNLOOP_ONE);
28	}	40	}
29		41
30	int	42	int
31	main (void)	43	main (void)
32	{	44	{
		45	// use the default event loop unless you have special needs
33	struct ev_loop *loop = ev_default_loop (0);	46	struct ev_loop *loop = ev_default_loop (0);
34		47
35	/* initialise an io watcher, then start it */	48	// initialise an io watcher, then start it
		49	// this one will watch for stdin to become readable
36	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);	50	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
37	ev_io_start (loop, &stdin_watcher);	51	ev_io_start (loop, &stdin_watcher);
38		52
		53	// initialise a timer watcher, then start it
39	/* simple non-repeating 5.5 second timeout */	54	// simple non-repeating 5.5 second timeout
40	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	55	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
41	ev_timer_start (loop, &timeout_watcher);	56	ev_timer_start (loop, &timeout_watcher);
42		57
43	/* loop till timeout or data ready */	58	// now wait for events to arrive
44	ev_loop (loop, 0);	59	ev_loop (loop, 0);
45		60
		61	// unloop was called, so exit
46	return 0;	62	return 0;
47	}	63	}
48		64
49	=head1 DESCRIPTION	65	=head1 DESCRIPTION
50		66
		67	The newest version of this document is also available as an html-formatted
		68	web page you might find easier to navigate when reading it for the first
		69	time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
		70
51	Libev is an event loop: you register interest in certain events (such as a	71	Libev is an event loop: you register interest in certain events (such as a
52	file descriptor being readable or a timeout occuring), and it will manage	72	file descriptor being readable or a timeout occurring), and it will manage
53	these event sources and provide your program with events.	73	these event sources and provide your program with events.
54		74
55	To do this, it must take more or less complete control over your process	75	To do this, it must take more or less complete control over your process
56	(or thread) by executing the I<event loop> handler, and will then	76	(or thread) by executing the I<event loop> handler, and will then
57	communicate events via a callback mechanism.	77	communicate events via a callback mechanism.
…		…
59	You register interest in certain events by registering so-called I<event	79	You register interest in certain events by registering so-called I<event
60	watchers>, which are relatively small C structures you initialise with the	80	watchers>, which are relatively small C structures you initialise with the
61	details of the event, and then hand it over to libev by I<starting> the	81	details of the event, and then hand it over to libev by I<starting> the
62	watcher.	82	watcher.
63		83
64	=head1 FEATURES	84	=head2 FEATURES
65		85
66	Libev supports select, poll, the linux-specific epoll and the bsd-specific	86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
67	kqueue mechanisms for file descriptor events, relative timers, absolute	87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
68	timers with customised rescheduling, signal events, process status change	88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
69	events (related to SIGCHLD), and event watchers dealing with the event	89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
70	loop mechanism itself (idle, prepare and check watchers). It also is quite	90	with customised rescheduling (C<ev_periodic>), synchronous signals
		91	(C<ev_signal>), process status change events (C<ev_child>), and event
		92	watchers dealing with the event loop mechanism itself (C<ev_idle>,
		93	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
		94	file watchers (C<ev_stat>) and even limited support for fork events
		95	(C<ev_fork>).
		96
		97	It also is quite fast (see this
71	fast (see this L<benchmark|http://libev.schmorp.de/bench.html> comparing	98	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
72	it to libevent for example).	99	for example).
73		100
74	=head1 CONVENTIONS	101	=head2 CONVENTIONS
75		102
76	Libev is very configurable. In this manual the default configuration	103	Libev is very configurable. In this manual the default (and most common)
77	will be described, which supports multiple event loops. For more info	104	configuration will be described, which supports multiple event loops. For
78	about various configuration options please have a look at the file	105	more info about various configuration options please have a look at
79	F<README.embed> in the libev distribution. If libev was configured without	106	B<EMBED> section in this manual. If libev was configured without support
80	support for multiple event loops, then all functions taking an initial	107	for multiple event loops, then all functions taking an initial argument of
81	argument of name C<loop> (which is always of type C<struct ev_loop *>)	108	name C<loop> (which is always of type C<struct ev_loop *>) will not have
82	will not have this argument.	109	this argument.
83		110
84	=head1 TIME REPRESENTATION	111	=head2 TIME REPRESENTATION
85		112
86	Libev represents time as a single floating point number, representing the	113	Libev represents time as a single floating point number, representing the
87	(fractional) number of seconds since the (POSIX) epoch (somewhere near	114	(fractional) number of seconds since the (POSIX) epoch (somewhere near
88	the beginning of 1970, details are complicated, don't ask). This type is	115	the beginning of 1970, details are complicated, don't ask). This type is
89	called C<ev_tstamp>, which is what you should use too. It usually aliases	116	called C<ev_tstamp>, which is what you should use too. It usually aliases
90	to the C<double> type in C, and when you need to do any calculations on	117	to the C<double> type in C, and when you need to do any calculations on
91	it, you should treat it as such.	118	it, you should treat it as some floatingpoint value. Unlike the name
		119	component C<stamp> might indicate, it is also used for time differences
		120	throughout libev.
92		121
93	=head1 GLOBAL FUNCTIONS	122	=head1 GLOBAL FUNCTIONS
94		123
95	These functions can be called anytime, even before initialising the	124	These functions can be called anytime, even before initialising the
96	library in any way.	125	library in any way.
…		…
101		130
102	Returns the current time as libev would use it. Please note that the	131	Returns the current time as libev would use it. Please note that the
103	C<ev_now> function is usually faster and also often returns the timestamp	132	C<ev_now> function is usually faster and also often returns the timestamp
104	you actually want to know.	133	you actually want to know.
105		134
		135	=item ev_sleep (ev_tstamp interval)
		136
		137	Sleep for the given interval: The current thread will be blocked until
		138	either it is interrupted or the given time interval has passed. Basically
		139	this is a subsecond-resolution C<sleep ()>.
		140
106	=item int ev_version_major ()	141	=item int ev_version_major ()
107		142
108	=item int ev_version_minor ()	143	=item int ev_version_minor ()
109		144
110	You can find out the major and minor version numbers of the library	145	You can find out the major and minor ABI version numbers of the library
111	you linked against by calling the functions C<ev_version_major> and	146	you linked against by calling the functions C<ev_version_major> and
112	C<ev_version_minor>. If you want, you can compare against the global	147	C<ev_version_minor>. If you want, you can compare against the global
113	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	148	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
114	version of the library your program was compiled against.	149	version of the library your program was compiled against.
115		150
		151	These version numbers refer to the ABI version of the library, not the
		152	release version.
		153
116	Usually, it's a good idea to terminate if the major versions mismatch,	154	Usually, it's a good idea to terminate if the major versions mismatch,
117	as this indicates an incompatible change. Minor versions are usually	155	as this indicates an incompatible change. Minor versions are usually
118	compatible to older versions, so a larger minor version alone is usually	156	compatible to older versions, so a larger minor version alone is usually
119	not a problem.	157	not a problem.
120		158
121	Example: make sure we haven't accidentally been linked against the wrong	159	Example: Make sure we haven't accidentally been linked against the wrong
122	version:	160	version.
123		161
124	assert (("libev version mismatch",	162	assert (("libev version mismatch",
125	ev_version_major () == EV_VERSION_MAJOR	163	ev_version_major () == EV_VERSION_MAJOR
126	&& ev_version_minor () >= EV_VERSION_MINOR));	164	&& ev_version_minor () >= EV_VERSION_MINOR));
127		165
…		…
155	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	193	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
156	recommended ones.	194	recommended ones.
157		195
158	See the description of C<ev_embed> watchers for more info.	196	See the description of C<ev_embed> watchers for more info.
159		197
160	=item ev_set_allocator (void *(*cb)(void *ptr, size_t size))	198	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
161		199
162	Sets the allocation function to use (the prototype and semantics are	200	Sets the allocation function to use (the prototype is similar - the
163	identical to the realloc C function). It is used to allocate and free	201	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
164	memory (no surprises here). If it returns zero when memory needs to be	202	used to allocate and free memory (no surprises here). If it returns zero
165	allocated, the library might abort or take some potentially destructive	203	when memory needs to be allocated (C<size != 0>), the library might abort
166	action. The default is your system realloc function.	204	or take some potentially destructive action.
		205
		206	Since some systems (at least OpenBSD and Darwin) fail to implement
		207	correct C<realloc> semantics, libev will use a wrapper around the system
		208	C<realloc> and C<free> functions by default.
167		209
168	You could override this function in high-availability programs to, say,	210	You could override this function in high-availability programs to, say,
169	free some memory if it cannot allocate memory, to use a special allocator,	211	free some memory if it cannot allocate memory, to use a special allocator,
170	or even to sleep a while and retry until some memory is available.	212	or even to sleep a while and retry until some memory is available.
171		213
172	Example: replace the libev allocator with one that waits a bit and then	214	Example: Replace the libev allocator with one that waits a bit and then
173	retries: better than mine).	215	retries (example requires a standards-compliant C<realloc>).
174		216
175	static void *	217	static void *
176	persistent_realloc (void *ptr, size_t size)	218	persistent_realloc (void *ptr, size_t size)
177	{	219	{
178	for (;;)	220	for (;;)
…		…
197	callback is set, then libev will expect it to remedy the sitution, no	239	callback is set, then libev will expect it to remedy the sitution, no
198	matter what, when it returns. That is, libev will generally retry the	240	matter what, when it returns. That is, libev will generally retry the
199	requested operation, or, if the condition doesn't go away, do bad stuff	241	requested operation, or, if the condition doesn't go away, do bad stuff
200	(such as abort).	242	(such as abort).
201		243
202	Example: do the same thing as libev does internally:	244	Example: This is basically the same thing that libev does internally, too.
203		245
204	static void	246	static void
205	fatal_error (const char *msg)	247	fatal_error (const char *msg)
206	{	248	{
207	perror (msg);	249	perror (msg);
…		…
217		259
218	An event loop is described by a C<struct ev_loop *>. The library knows two	260	An event loop is described by a C<struct ev_loop *>. The library knows two
219	types of such loops, the I<default> loop, which supports signals and child	261	types of such loops, the I<default> loop, which supports signals and child
220	events, and dynamically created loops which do not.	262	events, and dynamically created loops which do not.
221		263
222	If you use threads, a common model is to run the default event loop
223	in your main thread (or in a separate thread) and for each thread you
224	create, you also create another event loop. Libev itself does no locking
225	whatsoever, so if you mix calls to the same event loop in different
226	threads, make sure you lock (this is usually a bad idea, though, even if
227	done correctly, because it's hideous and inefficient).
228
229	=over 4	264	=over 4
230		265
231	=item struct ev_loop *ev_default_loop (unsigned int flags)	266	=item struct ev_loop *ev_default_loop (unsigned int flags)
232		267
233	This will initialise the default event loop if it hasn't been initialised	268	This will initialise the default event loop if it hasn't been initialised
…		…
235	false. If it already was initialised it simply returns it (and ignores the	270	false. If it already was initialised it simply returns it (and ignores the
236	flags. If that is troubling you, check C<ev_backend ()> afterwards).	271	flags. If that is troubling you, check C<ev_backend ()> afterwards).
237		272
238	If you don't know what event loop to use, use the one returned from this	273	If you don't know what event loop to use, use the one returned from this
239	function.	274	function.
		275
		276	Note that this function is I<not> thread-safe, so if you want to use it
		277	from multiple threads, you have to lock (note also that this is unlikely,
		278	as loops cannot bes hared easily between threads anyway).
		279
		280	The default loop is the only loop that can handle C<ev_signal> and
		281	C<ev_child> watchers, and to do this, it always registers a handler
		282	for C<SIGCHLD>. If this is a problem for your app you can either
		283	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
		284	can simply overwrite the C<SIGCHLD> signal handler I<after> calling
		285	C<ev_default_init>.
240		286
241	The flags argument can be used to specify special behaviour or specific	287	The flags argument can be used to specify special behaviour or specific
242	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).	288	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
243		289
244	The following flags are supported:	290	The following flags are supported:
…		…
257	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	303	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
258	override the flags completely if it is found in the environment. This is	304	override the flags completely if it is found in the environment. This is
259	useful to try out specific backends to test their performance, or to work	305	useful to try out specific backends to test their performance, or to work
260	around bugs.	306	around bugs.
261		307
		308	=item C<EVFLAG_FORKCHECK>
		309
		310	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after
		311	a fork, you can also make libev check for a fork in each iteration by
		312	enabling this flag.
		313
		314	This works by calling C<getpid ()> on every iteration of the loop,
		315	and thus this might slow down your event loop if you do a lot of loop
		316	iterations and little real work, but is usually not noticeable (on my
		317	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
		318	without a syscall and thus I<very> fast, but my GNU/Linux system also has
		319	C<pthread_atfork> which is even faster).
		320
		321	The big advantage of this flag is that you can forget about fork (and
		322	forget about forgetting to tell libev about forking) when you use this
		323	flag.
		324
		325	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>
		326	environment variable.
		327
262	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	328	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
263		329
264	This is your standard select(2) backend. Not I<completely> standard, as	330	This is your standard select(2) backend. Not I<completely> standard, as
265	libev tries to roll its own fd_set with no limits on the number of fds,	331	libev tries to roll its own fd_set with no limits on the number of fds,
266	but if that fails, expect a fairly low limit on the number of fds when	332	but if that fails, expect a fairly low limit on the number of fds when
267	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	333	using this backend. It doesn't scale too well (O(highest_fd)), but its
268	the fastest backend for a low number of fds.	334	usually the fastest backend for a low number of (low-numbered :) fds.
		335
		336	To get good performance out of this backend you need a high amount of
		337	parallelity (most of the file descriptors should be busy). If you are
		338	writing a server, you should C<accept ()> in a loop to accept as many
		339	connections as possible during one iteration. You might also want to have
		340	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		341	readiness notifications you get per iteration.
269		342
270	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	343	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
271		344
272	And this is your standard poll(2) backend. It's more complicated than	345	And this is your standard poll(2) backend. It's more complicated
273	select, but handles sparse fds better and has no artificial limit on the	346	than select, but handles sparse fds better and has no artificial
274	number of fds you can use (except it will slow down considerably with a	347	limit on the number of fds you can use (except it will slow down
275	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	348	considerably with a lot of inactive fds). It scales similarly to select,
		349	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		350	performance tips.
276		351
277	=item C<EVBACKEND_EPOLL> (value 4, Linux)	352	=item C<EVBACKEND_EPOLL> (value 4, Linux)
278		353
279	For few fds, this backend is a bit little slower than poll and select,	354	For few fds, this backend is a bit little slower than poll and select,
280	but it scales phenomenally better. While poll and select usually scale like	355	but it scales phenomenally better. While poll and select usually scale
281	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	356	like O(total_fds) where n is the total number of fds (or the highest fd),
282	either O(1) or O(active_fds).	357	epoll scales either O(1) or O(active_fds). The epoll design has a number
		358	of shortcomings, such as silently dropping events in some hard-to-detect
		359	cases and requiring a syscall per fd change, no fork support and bad
		360	support for dup.
283		361
284	While stopping and starting an I/O watcher in the same iteration will	362	While stopping, setting and starting an I/O watcher in the same iteration
285	result in some caching, there is still a syscall per such incident	363	will result in some caching, there is still a syscall per such incident
286	(because the fd could point to a different file description now), so its	364	(because the fd could point to a different file description now), so its
287	best to avoid that. Also, dup()ed file descriptors might not work very	365	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
288	well if you register events for both fds.	366	very well if you register events for both fds.
289		367
290	Please note that epoll sometimes generates spurious notifications, so you	368	Please note that epoll sometimes generates spurious notifications, so you
291	need to use non-blocking I/O or other means to avoid blocking when no data	369	need to use non-blocking I/O or other means to avoid blocking when no data
292	(or space) is available.	370	(or space) is available.
293		371
		372	Best performance from this backend is achieved by not unregistering all
		373	watchers for a file descriptor until it has been closed, if possible, i.e.
		374	keep at least one watcher active per fd at all times.
		375
		376	While nominally embeddeble in other event loops, this feature is broken in
		377	all kernel versions tested so far.
		378
294	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	379	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
295		380
296	Kqueue deserves special mention, as at the time of this writing, it	381	Kqueue deserves special mention, as at the time of this writing, it
297	was broken on all BSDs except NetBSD (usually it doesn't work with	382	was broken on all BSDs except NetBSD (usually it doesn't work reliably
298	anything but sockets and pipes, except on Darwin, where of course its	383	with anything but sockets and pipes, except on Darwin, where of course
299	completely useless). For this reason its not being "autodetected"	384	it's completely useless). For this reason it's not being "autodetected"
300	unless you explicitly specify it explicitly in the flags (i.e. using	385	unless you explicitly specify it explicitly in the flags (i.e. using
301	C<EVBACKEND_KQUEUE>).	386	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		387	system like NetBSD.
		388
		389	You still can embed kqueue into a normal poll or select backend and use it
		390	only for sockets (after having made sure that sockets work with kqueue on
		391	the target platform). See C<ev_embed> watchers for more info.
302		392
303	It scales in the same way as the epoll backend, but the interface to the	393	It scales in the same way as the epoll backend, but the interface to the
304	kernel is more efficient (which says nothing about its actual speed, of	394	kernel is more efficient (which says nothing about its actual speed, of
305	course). While starting and stopping an I/O watcher does not cause an	395	course). While stopping, setting and starting an I/O watcher does never
306	extra syscall as with epoll, it still adds up to four event changes per	396	cause an extra syscall as with C<EVBACKEND_EPOLL>, it still adds up to
307	incident, so its best to avoid that.	397	two event changes per incident, support for C<fork ()> is very bad and it
		398	drops fds silently in similarly hard-to-detect cases.
		399
		400	This backend usually performs well under most conditions.
		401
		402	While nominally embeddable in other event loops, this doesn't work
		403	everywhere, so you might need to test for this. And since it is broken
		404	almost everywhere, you should only use it when you have a lot of sockets
		405	(for which it usually works), by embedding it into another event loop
		406	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for
		407	sockets.
308		408
309	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	409	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
310		410
311	This is not implemented yet (and might never be).	411	This is not implemented yet (and might never be, unless you send me an
		412	implementation). According to reports, C</dev/poll> only supports sockets
		413	and is not embeddable, which would limit the usefulness of this backend
		414	immensely.
312		415
313	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	416	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
314		417
315	This uses the Solaris 10 port mechanism. As with everything on Solaris,	418	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
316	it's really slow, but it still scales very well (O(active_fds)).	419	it's really slow, but it still scales very well (O(active_fds)).
317		420
318	Please note that solaris ports can result in a lot of spurious	421	Please note that solaris event ports can deliver a lot of spurious
319	notifications, so you need to use non-blocking I/O or other means to avoid	422	notifications, so you need to use non-blocking I/O or other means to avoid
320	blocking when no data (or space) is available.	423	blocking when no data (or space) is available.
		424
		425	While this backend scales well, it requires one system call per active
		426	file descriptor per loop iteration. For small and medium numbers of file
		427	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		428	might perform better.
		429
		430	On the positive side, ignoring the spurious readiness notifications, this
		431	backend actually performed to specification in all tests and is fully
		432	embeddable, which is a rare feat among the OS-specific backends.
321		433
322	=item C<EVBACKEND_ALL>	434	=item C<EVBACKEND_ALL>
323		435
324	Try all backends (even potentially broken ones that wouldn't be tried	436	Try all backends (even potentially broken ones that wouldn't be tried
325	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	437	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
326	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	438	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
327		439
		440	It is definitely not recommended to use this flag.
		441
328	=back	442	=back
329		443
330	If one or more of these are ored into the flags value, then only these	444	If one or more of these are ored into the flags value, then only these
331	backends will be tried (in the reverse order as given here). If none are	445	backends will be tried (in the reverse order as listed here). If none are
332	specified, most compiled-in backend will be tried, usually in reverse	446	specified, all backends in C<ev_recommended_backends ()> will be tried.
333	order of their flag values :)
334		447
335	The most typical usage is like this:	448	The most typical usage is like this:
336		449
337	if (!ev_default_loop (0))	450	if (!ev_default_loop (0))
338	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	451	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
…		…
353	Similar to C<ev_default_loop>, but always creates a new event loop that is	466	Similar to C<ev_default_loop>, but always creates a new event loop that is
354	always distinct from the default loop. Unlike the default loop, it cannot	467	always distinct from the default loop. Unlike the default loop, it cannot
355	handle signal and child watchers, and attempts to do so will be greeted by	468	handle signal and child watchers, and attempts to do so will be greeted by
356	undefined behaviour (or a failed assertion if assertions are enabled).	469	undefined behaviour (or a failed assertion if assertions are enabled).
357		470
		471	Note that this function I<is> thread-safe, and the recommended way to use
		472	libev with threads is indeed to create one loop per thread, and using the
		473	default loop in the "main" or "initial" thread.
		474
358	Example: try to create a event loop that uses epoll and nothing else.	475	Example: Try to create a event loop that uses epoll and nothing else.
359		476
360	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	477	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
361	if (!epoller)	478	if (!epoller)
362	fatal ("no epoll found here, maybe it hides under your chair");	479	fatal ("no epoll found here, maybe it hides under your chair");
363		480
…		…
366	Destroys the default loop again (frees all memory and kernel state	483	Destroys the default loop again (frees all memory and kernel state
367	etc.). None of the active event watchers will be stopped in the normal	484	etc.). None of the active event watchers will be stopped in the normal
368	sense, so e.g. C<ev_is_active> might still return true. It is your	485	sense, so e.g. C<ev_is_active> might still return true. It is your
369	responsibility to either stop all watchers cleanly yoursef I<before>	486	responsibility to either stop all watchers cleanly yoursef I<before>
370	calling this function, or cope with the fact afterwards (which is usually	487	calling this function, or cope with the fact afterwards (which is usually
371	the easiest thing, youc na just ignore the watchers and/or C<free ()> them	488	the easiest thing, you can just ignore the watchers and/or C<free ()> them
372	for example).	489	for example).
		490
		491	Note that certain global state, such as signal state, will not be freed by
		492	this function, and related watchers (such as signal and child watchers)
		493	would need to be stopped manually.
		494
		495	In general it is not advisable to call this function except in the
		496	rare occasion where you really need to free e.g. the signal handling
		497	pipe fds. If you need dynamically allocated loops it is better to use
		498	C<ev_loop_new> and C<ev_loop_destroy>).
373		499
374	=item ev_loop_destroy (loop)	500	=item ev_loop_destroy (loop)
375		501
376	Like C<ev_default_destroy>, but destroys an event loop created by an	502	Like C<ev_default_destroy>, but destroys an event loop created by an
377	earlier call to C<ev_loop_new>.	503	earlier call to C<ev_loop_new>.
378		504
379	=item ev_default_fork ()	505	=item ev_default_fork ()
380		506
		507	This function sets a flag that causes subsequent C<ev_loop> iterations
381	This function reinitialises the kernel state for backends that have	508	to reinitialise the kernel state for backends that have one. Despite the
382	one. Despite the name, you can call it anytime, but it makes most sense	509	name, you can call it anytime, but it makes most sense after forking, in
383	after forking, in either the parent or child process (or both, but that	510	the child process (or both child and parent, but that again makes little
384	again makes little sense).	511	sense). You I<must> call it in the child before using any of the libev
		512	functions, and it will only take effect at the next C<ev_loop> iteration.
385		513
386	You I<must> call this function in the child process after forking if and	514	On the other hand, you only need to call this function in the child
387	only if you want to use the event library in both processes. If you just	515	process if and only if you want to use the event library in the child. If
388	fork+exec, you don't have to call it.	516	you just fork+exec, you don't have to call it at all.
389		517
390	The function itself is quite fast and it's usually not a problem to call	518	The function itself is quite fast and it's usually not a problem to call
391	it just in case after a fork. To make this easy, the function will fit in	519	it just in case after a fork. To make this easy, the function will fit in
392	quite nicely into a call to C<pthread_atfork>:	520	quite nicely into a call to C<pthread_atfork>:
393		521
394	pthread_atfork (0, 0, ev_default_fork);	522	pthread_atfork (0, 0, ev_default_fork);
395		523
396	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
397	without calling this function, so if you force one of those backends you
398	do not need to care.
399
400	=item ev_loop_fork (loop)	524	=item ev_loop_fork (loop)
401		525
402	Like C<ev_default_fork>, but acts on an event loop created by	526	Like C<ev_default_fork>, but acts on an event loop created by
403	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	527	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
404	after fork, and how you do this is entirely your own problem.	528	after fork, and how you do this is entirely your own problem.
		529
		530	=item int ev_is_default_loop (loop)
		531
		532	Returns true when the given loop actually is the default loop, false otherwise.
		533
		534	=item unsigned int ev_loop_count (loop)
		535
		536	Returns the count of loop iterations for the loop, which is identical to
		537	the number of times libev did poll for new events. It starts at C<0> and
		538	happily wraps around with enough iterations.
		539
		540	This value can sometimes be useful as a generation counter of sorts (it
		541	"ticks" the number of loop iterations), as it roughly corresponds with
		542	C<ev_prepare> and C<ev_check> calls.
405		543
406	=item unsigned int ev_backend (loop)	544	=item unsigned int ev_backend (loop)
407		545
408	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	546	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
409	use.	547	use.
…		…
412		550
413	Returns the current "event loop time", which is the time the event loop	551	Returns the current "event loop time", which is the time the event loop
414	received events and started processing them. This timestamp does not	552	received events and started processing them. This timestamp does not
415	change as long as callbacks are being processed, and this is also the base	553	change as long as callbacks are being processed, and this is also the base
416	time used for relative timers. You can treat it as the timestamp of the	554	time used for relative timers. You can treat it as the timestamp of the
417	event occuring (or more correctly, libev finding out about it).	555	event occurring (or more correctly, libev finding out about it).
418		556
419	=item ev_loop (loop, int flags)	557	=item ev_loop (loop, int flags)
420		558
421	Finally, this is it, the event handler. This function usually is called	559	Finally, this is it, the event handler. This function usually is called
422	after you initialised all your watchers and you want to start handling	560	after you initialised all your watchers and you want to start handling
…		…
443	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	581	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is
444	usually a better approach for this kind of thing.	582	usually a better approach for this kind of thing.
445		583
446	Here are the gory details of what C<ev_loop> does:	584	Here are the gory details of what C<ev_loop> does:
447		585
448	* If there are no active watchers (reference count is zero), return.	586	- Before the first iteration, call any pending watchers.
449	- Queue prepare watchers and then call all outstanding watchers.	587	* If EVFLAG_FORKCHECK was used, check for a fork.
		588	- If a fork was detected, queue and call all fork watchers.
		589	- Queue and call all prepare watchers.
450	- If we have been forked, recreate the kernel state.	590	- If we have been forked, recreate the kernel state.
451	- Update the kernel state with all outstanding changes.	591	- Update the kernel state with all outstanding changes.
452	- Update the "event loop time".	592	- Update the "event loop time".
453	- Calculate for how long to block.	593	- Calculate for how long to sleep or block, if at all
		594	(active idle watchers, EVLOOP_NONBLOCK or not having
		595	any active watchers at all will result in not sleeping).
		596	- Sleep if the I/O and timer collect interval say so.
454	- Block the process, waiting for any events.	597	- Block the process, waiting for any events.
455	- Queue all outstanding I/O (fd) events.	598	- Queue all outstanding I/O (fd) events.
456	- Update the "event loop time" and do time jump handling.	599	- Update the "event loop time" and do time jump handling.
457	- Queue all outstanding timers.	600	- Queue all outstanding timers.
458	- Queue all outstanding periodics.	601	- Queue all outstanding periodics.
459	- If no events are pending now, queue all idle watchers.	602	- If no events are pending now, queue all idle watchers.
460	- Queue all check watchers.	603	- Queue all check watchers.
461	- Call all queued watchers in reverse order (i.e. check watchers first).	604	- Call all queued watchers in reverse order (i.e. check watchers first).
462	Signals and child watchers are implemented as I/O watchers, and will	605	Signals and child watchers are implemented as I/O watchers, and will
463	be handled here by queueing them when their watcher gets executed.	606	be handled here by queueing them when their watcher gets executed.
464	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	607	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
465	were used, return, otherwise continue with step *.	608	were used, or there are no active watchers, return, otherwise
		609	continue with step *.
466		610
467	Example: queue some jobs and then loop until no events are outsanding	611	Example: Queue some jobs and then loop until no events are outstanding
468	anymore.	612	anymore.
469		613
470	... queue jobs here, make sure they register event watchers as long	614	... queue jobs here, make sure they register event watchers as long
471	... as they still have work to do (even an idle watcher will do..)	615	... as they still have work to do (even an idle watcher will do..)
472	ev_loop (my_loop, 0);	616	ev_loop (my_loop, 0);
…		…
476		620
477	Can be used to make a call to C<ev_loop> return early (but only after it	621	Can be used to make a call to C<ev_loop> return early (but only after it
478	has processed all outstanding events). The C<how> argument must be either	622	has processed all outstanding events). The C<how> argument must be either
479	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	623	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
480	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	624	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
		625
		626	This "unloop state" will be cleared when entering C<ev_loop> again.
481		627
482	=item ev_ref (loop)	628	=item ev_ref (loop)
483		629
484	=item ev_unref (loop)	630	=item ev_unref (loop)
485		631
…		…
490	returning, ev_unref() after starting, and ev_ref() before stopping it. For	636	returning, ev_unref() after starting, and ev_ref() before stopping it. For
491	example, libev itself uses this for its internal signal pipe: It is not	637	example, libev itself uses this for its internal signal pipe: It is not
492	visible to the libev user and should not keep C<ev_loop> from exiting if	638	visible to the libev user and should not keep C<ev_loop> from exiting if
493	no event watchers registered by it are active. It is also an excellent	639	no event watchers registered by it are active. It is also an excellent
494	way to do this for generic recurring timers or from within third-party	640	way to do this for generic recurring timers or from within third-party
495	libraries. Just remember to I<unref after start> and I<ref before stop>.	641	libraries. Just remember to I<unref after start> and I<ref before stop>
		642	(but only if the watcher wasn't active before, or was active before,
		643	respectively).
496		644
497	Example: create a signal watcher, but keep it from keeping C<ev_loop>	645	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
498	running when nothing else is active.	646	running when nothing else is active.
499		647
500	struct dv_signal exitsig;	648	struct ev_signal exitsig;
501	ev_signal_init (&exitsig, sig_cb, SIGINT);	649	ev_signal_init (&exitsig, sig_cb, SIGINT);
502	ev_signal_start (myloop, &exitsig);	650	ev_signal_start (loop, &exitsig);
503	evf_unref (myloop);	651	evf_unref (loop);
504		652
505	Example: for some weird reason, unregister the above signal handler again.	653	Example: For some weird reason, unregister the above signal handler again.
506		654
507	ev_ref (myloop);	655	ev_ref (loop);
508	ev_signal_stop (myloop, &exitsig);	656	ev_signal_stop (loop, &exitsig);
		657
		658	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		659
		660	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		661
		662	These advanced functions influence the time that libev will spend waiting
		663	for events. Both are by default C<0>, meaning that libev will try to
		664	invoke timer/periodic callbacks and I/O callbacks with minimum latency.
		665
		666	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		667	allows libev to delay invocation of I/O and timer/periodic callbacks to
		668	increase efficiency of loop iterations.
		669
		670	The background is that sometimes your program runs just fast enough to
		671	handle one (or very few) event(s) per loop iteration. While this makes
		672	the program responsive, it also wastes a lot of CPU time to poll for new
		673	events, especially with backends like C<select ()> which have a high
		674	overhead for the actual polling but can deliver many events at once.
		675
		676	By setting a higher I<io collect interval> you allow libev to spend more
		677	time collecting I/O events, so you can handle more events per iteration,
		678	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		679	C<ev_timer>) will be not affected. Setting this to a non-null value will
		680	introduce an additional C<ev_sleep ()> call into most loop iterations.
		681
		682	Likewise, by setting a higher I<timeout collect interval> you allow libev
		683	to spend more time collecting timeouts, at the expense of increased
		684	latency (the watcher callback will be called later). C<ev_io> watchers
		685	will not be affected. Setting this to a non-null value will not introduce
		686	any overhead in libev.
		687
		688	Many (busy) programs can usually benefit by setting the io collect
		689	interval to a value near C<0.1> or so, which is often enough for
		690	interactive servers (of course not for games), likewise for timeouts. It
		691	usually doesn't make much sense to set it to a lower value than C<0.01>,
		692	as this approsaches the timing granularity of most systems.
509		693
510	=back	694	=back
511		695
512		696
513	=head1 ANATOMY OF A WATCHER	697	=head1 ANATOMY OF A WATCHER
…		…
613	=item C<EV_FORK>	797	=item C<EV_FORK>
614		798
615	The event loop has been resumed in the child process after fork (see	799	The event loop has been resumed in the child process after fork (see
616	C<ev_fork>).	800	C<ev_fork>).
617		801
		802	=item C<EV_ASYNC>
		803
		804	The given async watcher has been asynchronously notified (see C<ev_async>).
		805
618	=item C<EV_ERROR>	806	=item C<EV_ERROR>
619		807
620	An unspecified error has occured, the watcher has been stopped. This might	808	An unspecified error has occured, the watcher has been stopped. This might
621	happen because the watcher could not be properly started because libev	809	happen because the watcher could not be properly started because libev
622	ran out of memory, a file descriptor was found to be closed or any other	810	ran out of memory, a file descriptor was found to be closed or any other
…		…
693	=item bool ev_is_pending (ev_TYPE *watcher)	881	=item bool ev_is_pending (ev_TYPE *watcher)
694		882
695	Returns a true value iff the watcher is pending, (i.e. it has outstanding	883	Returns a true value iff the watcher is pending, (i.e. it has outstanding
696	events but its callback has not yet been invoked). As long as a watcher	884	events but its callback has not yet been invoked). As long as a watcher
697	is pending (but not active) you must not call an init function on it (but	885	is pending (but not active) you must not call an init function on it (but
698	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to	886	C<ev_TYPE_set> is safe), you must not change its priority, and you must
699	libev (e.g. you cnanot C<free ()> it).	887	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		888	it).
700		889
701	=item callback = ev_cb (ev_TYPE *watcher)	890	=item callback ev_cb (ev_TYPE *watcher)
702		891
703	Returns the callback currently set on the watcher.	892	Returns the callback currently set on the watcher.
704		893
705	=item ev_cb_set (ev_TYPE *watcher, callback)	894	=item ev_cb_set (ev_TYPE *watcher, callback)
706		895
707	Change the callback. You can change the callback at virtually any time	896	Change the callback. You can change the callback at virtually any time
708	(modulo threads).	897	(modulo threads).
		898
		899	=item ev_set_priority (ev_TYPE *watcher, priority)
		900
		901	=item int ev_priority (ev_TYPE *watcher)
		902
		903	Set and query the priority of the watcher. The priority is a small
		904	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		905	(default: C<-2>). Pending watchers with higher priority will be invoked
		906	before watchers with lower priority, but priority will not keep watchers
		907	from being executed (except for C<ev_idle> watchers).
		908
		909	This means that priorities are I<only> used for ordering callback
		910	invocation after new events have been received. This is useful, for
		911	example, to reduce latency after idling, or more often, to bind two
		912	watchers on the same event and make sure one is called first.
		913
		914	If you need to suppress invocation when higher priority events are pending
		915	you need to look at C<ev_idle> watchers, which provide this functionality.
		916
		917	You I<must not> change the priority of a watcher as long as it is active or
		918	pending.
		919
		920	The default priority used by watchers when no priority has been set is
		921	always C<0>, which is supposed to not be too high and not be too low :).
		922
		923	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		924	fine, as long as you do not mind that the priority value you query might
		925	or might not have been adjusted to be within valid range.
		926
		927	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		928
		929	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		930	C<loop> nor C<revents> need to be valid as long as the watcher callback
		931	can deal with that fact.
		932
		933	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		934
		935	If the watcher is pending, this function returns clears its pending status
		936	and returns its C<revents> bitset (as if its callback was invoked). If the
		937	watcher isn't pending it does nothing and returns C<0>.
709		938
710	=back	939	=back
711		940
712		941
713	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	942	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
…		…
734	{	963	{
735	struct my_io *w = (struct my_io *)w_;	964	struct my_io *w = (struct my_io *)w_;
736	...	965	...
737	}	966	}
738		967
739	More interesting and less C-conformant ways of catsing your callback type	968	More interesting and less C-conformant ways of casting your callback type
740	have been omitted....	969	instead have been omitted.
		970
		971	Another common scenario is having some data structure with multiple
		972	watchers:
		973
		974	struct my_biggy
		975	{
		976	int some_data;
		977	ev_timer t1;
		978	ev_timer t2;
		979	}
		980
		981	In this case getting the pointer to C<my_biggy> is a bit more complicated,
		982	you need to use C<offsetof>:
		983
		984	#include <stddef.h>
		985
		986	static void
		987	t1_cb (EV_P_ struct ev_timer *w, int revents)
		988	{
		989	struct my_biggy big = (struct my_biggy *
		990	(((char *)w) - offsetof (struct my_biggy, t1));
		991	}
		992
		993	static void
		994	t2_cb (EV_P_ struct ev_timer *w, int revents)
		995	{
		996	struct my_biggy big = (struct my_biggy *
		997	(((char *)w) - offsetof (struct my_biggy, t2));
		998	}
741		999
742		1000
743	=head1 WATCHER TYPES	1001	=head1 WATCHER TYPES
744		1002
745	This section describes each watcher in detail, but will not repeat	1003	This section describes each watcher in detail, but will not repeat
…		…
769	In general you can register as many read and/or write event watchers per	1027	In general you can register as many read and/or write event watchers per
770	fd as you want (as long as you don't confuse yourself). Setting all file	1028	fd as you want (as long as you don't confuse yourself). Setting all file
771	descriptors to non-blocking mode is also usually a good idea (but not	1029	descriptors to non-blocking mode is also usually a good idea (but not
772	required if you know what you are doing).	1030	required if you know what you are doing).
773		1031
774	You have to be careful with dup'ed file descriptors, though. Some backends
775	(the linux epoll backend is a notable example) cannot handle dup'ed file
776	descriptors correctly if you register interest in two or more fds pointing
777	to the same underlying file/socket/etc. description (that is, they share
778	the same underlying "file open").
779
780	If you must do this, then force the use of a known-to-be-good backend	1032	If you must do this, then force the use of a known-to-be-good backend
781	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1033	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
782	C<EVBACKEND_POLL>).	1034	C<EVBACKEND_POLL>).
783		1035
784	Another thing you have to watch out for is that it is quite easy to	1036	Another thing you have to watch out for is that it is quite easy to
785	receive "spurious" readyness notifications, that is your callback might	1037	receive "spurious" readiness notifications, that is your callback might
786	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1038	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
787	because there is no data. Not only are some backends known to create a	1039	because there is no data. Not only are some backends known to create a
788	lot of those (for example solaris ports), it is very easy to get into	1040	lot of those (for example solaris ports), it is very easy to get into
789	this situation even with a relatively standard program structure. Thus	1041	this situation even with a relatively standard program structure. Thus
790	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1042	it is best to always use non-blocking I/O: An extra C<read>(2) returning
791	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1043	C<EAGAIN> is far preferable to a program hanging until some data arrives.
792		1044
793	If you cannot run the fd in non-blocking mode (for example you should not	1045	If you cannot run the fd in non-blocking mode (for example you should not
794	play around with an Xlib connection), then you have to seperately re-test	1046	play around with an Xlib connection), then you have to seperately re-test
795	wether a file descriptor is really ready with a known-to-be good interface	1047	whether a file descriptor is really ready with a known-to-be good interface
796	such as poll (fortunately in our Xlib example, Xlib already does this on	1048	such as poll (fortunately in our Xlib example, Xlib already does this on
797	its own, so its quite safe to use).	1049	its own, so its quite safe to use).
		1050
		1051	=head3 The special problem of disappearing file descriptors
		1052
		1053	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1054	descriptor (either by calling C<close> explicitly or by any other means,
		1055	such as C<dup>). The reason is that you register interest in some file
		1056	descriptor, but when it goes away, the operating system will silently drop
		1057	this interest. If another file descriptor with the same number then is
		1058	registered with libev, there is no efficient way to see that this is, in
		1059	fact, a different file descriptor.
		1060
		1061	To avoid having to explicitly tell libev about such cases, libev follows
		1062	the following policy: Each time C<ev_io_set> is being called, libev
		1063	will assume that this is potentially a new file descriptor, otherwise
		1064	it is assumed that the file descriptor stays the same. That means that
		1065	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1066	descriptor even if the file descriptor number itself did not change.
		1067
		1068	This is how one would do it normally anyway, the important point is that
		1069	the libev application should not optimise around libev but should leave
		1070	optimisations to libev.
		1071
		1072	=head3 The special problem of dup'ed file descriptors
		1073
		1074	Some backends (e.g. epoll), cannot register events for file descriptors,
		1075	but only events for the underlying file descriptions. That means when you
		1076	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1077	events for them, only one file descriptor might actually receive events.
		1078
		1079	There is no workaround possible except not registering events
		1080	for potentially C<dup ()>'ed file descriptors, or to resort to
		1081	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1082
		1083	=head3 The special problem of fork
		1084
		1085	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1086	useless behaviour. Libev fully supports fork, but needs to be told about
		1087	it in the child.
		1088
		1089	To support fork in your programs, you either have to call
		1090	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1091	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1092	C<EVBACKEND_POLL>.
		1093
		1094	=head3 The special problem of SIGPIPE
		1095
		1096	While not really specific to libev, it is easy to forget about SIGPIPE:
		1097	when reading from a pipe whose other end has been closed, your program
		1098	gets send a SIGPIPE, which, by default, aborts your program. For most
		1099	programs this is sensible behaviour, for daemons, this is usually
		1100	undesirable.
		1101
		1102	So when you encounter spurious, unexplained daemon exits, make sure you
		1103	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1104	somewhere, as that would have given you a big clue).
		1105
		1106
		1107	=head3 Watcher-Specific Functions
798		1108
799	=over 4	1109	=over 4
800		1110
801	=item ev_io_init (ev_io *, callback, int fd, int events)	1111	=item ev_io_init (ev_io *, callback, int fd, int events)
802		1112
…		…
814		1124
815	The events being watched.	1125	The events being watched.
816		1126
817	=back	1127	=back
818		1128
		1129	=head3 Examples
		1130
819	Example: call C<stdin_readable_cb> when STDIN_FILENO has become, well	1131	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
820	readable, but only once. Since it is likely line-buffered, you could	1132	readable, but only once. Since it is likely line-buffered, you could
821	attempt to read a whole line in the callback:	1133	attempt to read a whole line in the callback.
822		1134
823	static void	1135	static void
824	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1136	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
825	{	1137	{
826	ev_io_stop (loop, w);	1138	ev_io_stop (loop, w);
…		…
839		1151
840	Timer watchers are simple relative timers that generate an event after a	1152	Timer watchers are simple relative timers that generate an event after a
841	given time, and optionally repeating in regular intervals after that.	1153	given time, and optionally repeating in regular intervals after that.
842		1154
843	The timers are based on real time, that is, if you register an event that	1155	The timers are based on real time, that is, if you register an event that
844	times out after an hour and you reset your system clock to last years	1156	times out after an hour and you reset your system clock to january last
845	time, it will still time out after (roughly) and hour. "Roughly" because	1157	year, it will still time out after (roughly) and hour. "Roughly" because
846	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1158	detecting time jumps is hard, and some inaccuracies are unavoidable (the
847	monotonic clock option helps a lot here).	1159	monotonic clock option helps a lot here).
848		1160
849	The relative timeouts are calculated relative to the C<ev_now ()>	1161	The relative timeouts are calculated relative to the C<ev_now ()>
850	time. This is usually the right thing as this timestamp refers to the time	1162	time. This is usually the right thing as this timestamp refers to the time
…		…
852	you suspect event processing to be delayed and you I<need> to base the timeout	1164	you suspect event processing to be delayed and you I<need> to base the timeout
853	on the current time, use something like this to adjust for this:	1165	on the current time, use something like this to adjust for this:
854		1166
855	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1167	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
856		1168
857	The callback is guarenteed to be invoked only when its timeout has passed,	1169	The callback is guarenteed to be invoked only after its timeout has passed,
858	but if multiple timers become ready during the same loop iteration then	1170	but if multiple timers become ready during the same loop iteration then
859	order of execution is undefined.	1171	order of execution is undefined.
860		1172
		1173	=head3 Watcher-Specific Functions and Data Members
		1174
861	=over 4	1175	=over 4
862		1176
863	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1177	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
864		1178
865	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1179	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
866		1180
867	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1181	Configure the timer to trigger after C<after> seconds. If C<repeat>
868	C<0.>, then it will automatically be stopped. If it is positive, then the	1182	is C<0.>, then it will automatically be stopped once the timeout is
869	timer will automatically be configured to trigger again C<repeat> seconds	1183	reached. If it is positive, then the timer will automatically be
870	later, again, and again, until stopped manually.	1184	configured to trigger again C<repeat> seconds later, again, and again,
		1185	until stopped manually.
871		1186
872	The timer itself will do a best-effort at avoiding drift, that is, if you	1187	The timer itself will do a best-effort at avoiding drift, that is, if
873	configure a timer to trigger every 10 seconds, then it will trigger at	1188	you configure a timer to trigger every 10 seconds, then it will normally
874	exactly 10 second intervals. If, however, your program cannot keep up with	1189	trigger at exactly 10 second intervals. If, however, your program cannot
875	the timer (because it takes longer than those 10 seconds to do stuff) the	1190	keep up with the timer (because it takes longer than those 10 seconds to
876	timer will not fire more than once per event loop iteration.	1191	do stuff) the timer will not fire more than once per event loop iteration.
877		1192
878	=item ev_timer_again (loop)	1193	=item ev_timer_again (loop, ev_timer *)
879		1194
880	This will act as if the timer timed out and restart it again if it is	1195	This will act as if the timer timed out and restart it again if it is
881	repeating. The exact semantics are:	1196	repeating. The exact semantics are:
882		1197
		1198	If the timer is pending, its pending status is cleared.
		1199
883	If the timer is started but nonrepeating, stop it.	1200	If the timer is started but nonrepeating, stop it (as if it timed out).
884		1201
885	If the timer is repeating, either start it if necessary (with the repeat	1202	If the timer is repeating, either start it if necessary (with the
886	value), or reset the running timer to the repeat value.	1203	C<repeat> value), or reset the running timer to the C<repeat> value.
887		1204
888	This sounds a bit complicated, but here is a useful and typical	1205	This sounds a bit complicated, but here is a useful and typical
889	example: Imagine you have a tcp connection and you want a so-called	1206	example: Imagine you have a tcp connection and you want a so-called idle
890	idle timeout, that is, you want to be called when there have been,	1207	timeout, that is, you want to be called when there have been, say, 60
891	say, 60 seconds of inactivity on the socket. The easiest way to do	1208	seconds of inactivity on the socket. The easiest way to do this is to
892	this is to configure an C<ev_timer> with C<after>=C<repeat>=C<60> and calling	1209	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
893	C<ev_timer_again> each time you successfully read or write some data. If	1210	C<ev_timer_again> each time you successfully read or write some data. If
894	you go into an idle state where you do not expect data to travel on the	1211	you go into an idle state where you do not expect data to travel on the
895	socket, you can stop the timer, and again will automatically restart it if	1212	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
896	need be.	1213	automatically restart it if need be.
897		1214
898	You can also ignore the C<after> value and C<ev_timer_start> altogether	1215	That means you can ignore the C<after> value and C<ev_timer_start>
899	and only ever use the C<repeat> value:	1216	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
900		1217
901	ev_timer_init (timer, callback, 0., 5.);	1218	ev_timer_init (timer, callback, 0., 5.);
902	ev_timer_again (loop, timer);	1219	ev_timer_again (loop, timer);
903	...	1220	...
904	timer->again = 17.;	1221	timer->again = 17.;
905	ev_timer_again (loop, timer);	1222	ev_timer_again (loop, timer);
906	...	1223	...
907	timer->again = 10.;	1224	timer->again = 10.;
908	ev_timer_again (loop, timer);	1225	ev_timer_again (loop, timer);
909		1226
910	This is more efficient then stopping/starting the timer eahc time you want	1227	This is more slightly efficient then stopping/starting the timer each time
911	to modify its timeout value.	1228	you want to modify its timeout value.
912		1229
913	=item ev_tstamp repeat [read-write]	1230	=item ev_tstamp repeat [read-write]
914		1231
915	The current C<repeat> value. Will be used each time the watcher times out	1232	The current C<repeat> value. Will be used each time the watcher times out
916	or C<ev_timer_again> is called and determines the next timeout (if any),	1233	or C<ev_timer_again> is called and determines the next timeout (if any),
917	which is also when any modifications are taken into account.	1234	which is also when any modifications are taken into account.
918		1235
919	=back	1236	=back
920		1237
		1238	=head3 Examples
		1239
921	Example: create a timer that fires after 60 seconds.	1240	Example: Create a timer that fires after 60 seconds.
922		1241
923	static void	1242	static void
924	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1243	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
925	{	1244	{
926	.. one minute over, w is actually stopped right here	1245	.. one minute over, w is actually stopped right here
…		…
928		1247
929	struct ev_timer mytimer;	1248	struct ev_timer mytimer;
930	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1249	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
931	ev_timer_start (loop, &mytimer);	1250	ev_timer_start (loop, &mytimer);
932		1251
933	Example: create a timeout timer that times out after 10 seconds of	1252	Example: Create a timeout timer that times out after 10 seconds of
934	inactivity.	1253	inactivity.
935		1254
936	static void	1255	static void
937	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1256	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
938	{	1257	{
…		…
954	Periodic watchers are also timers of a kind, but they are very versatile	1273	Periodic watchers are also timers of a kind, but they are very versatile
955	(and unfortunately a bit complex).	1274	(and unfortunately a bit complex).
956		1275
957	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1276	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
958	but on wallclock time (absolute time). You can tell a periodic watcher	1277	but on wallclock time (absolute time). You can tell a periodic watcher
959	to trigger "at" some specific point in time. For example, if you tell a	1278	to trigger after some specific point in time. For example, if you tell a
960	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	1279	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()
961	+ 10.>) and then reset your system clock to the last year, then it will	1280	+ 10.>, that is, an absolute time not a delay) and then reset your system
		1281	clock to january of the previous year, then it will take more than year
962	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1282	to trigger the event (unlike an C<ev_timer>, which would still trigger
963	roughly 10 seconds later and of course not if you reset your system time	1283	roughly 10 seconds later as it uses a relative timeout).
964	again).
965		1284
966	They can also be used to implement vastly more complex timers, such as	1285	C<ev_periodic>s can also be used to implement vastly more complex timers,
967	triggering an event on eahc midnight, local time.	1286	such as triggering an event on each "midnight, local time", or other
		1287	complicated, rules.
968		1288
969	As with timers, the callback is guarenteed to be invoked only when the	1289	As with timers, the callback is guarenteed to be invoked only when the
970	time (C<at>) has been passed, but if multiple periodic timers become ready	1290	time (C<at>) has passed, but if multiple periodic timers become ready
971	during the same loop iteration then order of execution is undefined.	1291	during the same loop iteration then order of execution is undefined.
		1292
		1293	=head3 Watcher-Specific Functions and Data Members
972		1294
973	=over 4	1295	=over 4
974		1296
975	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1297	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
976		1298
…		…
979	Lots of arguments, lets sort it out... There are basically three modes of	1301	Lots of arguments, lets sort it out... There are basically three modes of
980	operation, and we will explain them from simplest to complex:	1302	operation, and we will explain them from simplest to complex:
981		1303
982	=over 4	1304	=over 4
983		1305
984	=item * absolute timer (interval = reschedule_cb = 0)	1306	=item * absolute timer (at = time, interval = reschedule_cb = 0)
985		1307
986	In this configuration the watcher triggers an event at the wallclock time	1308	In this configuration the watcher triggers an event after the wallclock
987	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1309	time C<at> has passed and doesn't repeat. It will not adjust when a time
988	that is, if it is to be run at January 1st 2011 then it will run when the	1310	jump occurs, that is, if it is to be run at January 1st 2011 then it will
989	system time reaches or surpasses this time.	1311	run when the system time reaches or surpasses this time.
990		1312
991	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1313	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
992		1314
993	In this mode the watcher will always be scheduled to time out at the next	1315	In this mode the watcher will always be scheduled to time out at the next
994	C<at + N * interval> time (for some integer N) and then repeat, regardless	1316	C<at + N * interval> time (for some integer N, which can also be negative)
995	of any time jumps.	1317	and then repeat, regardless of any time jumps.
996		1318
997	This can be used to create timers that do not drift with respect to system	1319	This can be used to create timers that do not drift with respect to system
998	time:	1320	time, for example, here is a C<ev_periodic> that triggers each hour, on
		1321	the hour:
999		1322
1000	ev_periodic_set (&periodic, 0., 3600., 0);	1323	ev_periodic_set (&periodic, 0., 3600., 0);
1001		1324
1002	This doesn't mean there will always be 3600 seconds in between triggers,	1325	This doesn't mean there will always be 3600 seconds in between triggers,
1003	but only that the the callback will be called when the system time shows a	1326	but only that the the callback will be called when the system time shows a
…		…
1006		1329
1007	Another way to think about it (for the mathematically inclined) is that	1330	Another way to think about it (for the mathematically inclined) is that
1008	C<ev_periodic> will try to run the callback in this mode at the next possible	1331	C<ev_periodic> will try to run the callback in this mode at the next possible
1009	time where C<time = at (mod interval)>, regardless of any time jumps.	1332	time where C<time = at (mod interval)>, regardless of any time jumps.
1010		1333
		1334	For numerical stability it is preferable that the C<at> value is near
		1335	C<ev_now ()> (the current time), but there is no range requirement for
		1336	this value, and in fact is often specified as zero.
		1337
1011	=item * manual reschedule mode (reschedule_cb = callback)	1338	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
1012		1339
1013	In this mode the values for C<interval> and C<at> are both being	1340	In this mode the values for C<interval> and C<at> are both being
1014	ignored. Instead, each time the periodic watcher gets scheduled, the	1341	ignored. Instead, each time the periodic watcher gets scheduled, the
1015	reschedule callback will be called with the watcher as first, and the	1342	reschedule callback will be called with the watcher as first, and the
1016	current time as second argument.	1343	current time as second argument.
1017		1344
1018	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1345	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
1019	ever, or make any event loop modifications>. If you need to stop it,	1346	ever, or make ANY event loop modifications whatsoever>.
1020	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
1021	starting a prepare watcher).
1022		1347
		1348	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1349	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1350	only event loop modification you are allowed to do).
		1351
1023	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1352	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic
1024	ev_tstamp now)>, e.g.:	1353	*w, ev_tstamp now)>, e.g.:
1025		1354
1026	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1355	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
1027	{	1356	{
1028	return now + 60.;	1357	return now + 60.;
1029	}	1358	}
…		…
1031	It must return the next time to trigger, based on the passed time value	1360	It must return the next time to trigger, based on the passed time value
1032	(that is, the lowest time value larger than to the second argument). It	1361	(that is, the lowest time value larger than to the second argument). It
1033	will usually be called just before the callback will be triggered, but	1362	will usually be called just before the callback will be triggered, but
1034	might be called at other times, too.	1363	might be called at other times, too.
1035		1364
1036	NOTE: I<< This callback must always return a time that is later than the	1365	NOTE: I<< This callback must always return a time that is higher than or
1037	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	1366	equal to the passed C<now> value >>.
1038		1367
1039	This can be used to create very complex timers, such as a timer that	1368	This can be used to create very complex timers, such as a timer that
1040	triggers on each midnight, local time. To do this, you would calculate the	1369	triggers on "next midnight, local time". To do this, you would calculate the
1041	next midnight after C<now> and return the timestamp value for this. How	1370	next midnight after C<now> and return the timestamp value for this. How
1042	you do this is, again, up to you (but it is not trivial, which is the main	1371	you do this is, again, up to you (but it is not trivial, which is the main
1043	reason I omitted it as an example).	1372	reason I omitted it as an example).
1044		1373
1045	=back	1374	=back
…		…
1049	Simply stops and restarts the periodic watcher again. This is only useful	1378	Simply stops and restarts the periodic watcher again. This is only useful
1050	when you changed some parameters or the reschedule callback would return	1379	when you changed some parameters or the reschedule callback would return
1051	a different time than the last time it was called (e.g. in a crond like	1380	a different time than the last time it was called (e.g. in a crond like
1052	program when the crontabs have changed).	1381	program when the crontabs have changed).
1053		1382
		1383	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1384
		1385	When active, returns the absolute time that the watcher is supposed to
		1386	trigger next.
		1387
		1388	=item ev_tstamp offset [read-write]
		1389
		1390	When repeating, this contains the offset value, otherwise this is the
		1391	absolute point in time (the C<at> value passed to C<ev_periodic_set>).
		1392
		1393	Can be modified any time, but changes only take effect when the periodic
		1394	timer fires or C<ev_periodic_again> is being called.
		1395
1054	=item ev_tstamp interval [read-write]	1396	=item ev_tstamp interval [read-write]
1055		1397
1056	The current interval value. Can be modified any time, but changes only	1398	The current interval value. Can be modified any time, but changes only
1057	take effect when the periodic timer fires or C<ev_periodic_again> is being	1399	take effect when the periodic timer fires or C<ev_periodic_again> is being
1058	called.	1400	called.
…		…
1063	switched off. Can be changed any time, but changes only take effect when	1405	switched off. Can be changed any time, but changes only take effect when
1064	the periodic timer fires or C<ev_periodic_again> is being called.	1406	the periodic timer fires or C<ev_periodic_again> is being called.
1065		1407
1066	=back	1408	=back
1067		1409
		1410	=head3 Examples
		1411
1068	Example: call a callback every hour, or, more precisely, whenever the	1412	Example: Call a callback every hour, or, more precisely, whenever the
1069	system clock is divisible by 3600. The callback invocation times have	1413	system clock is divisible by 3600. The callback invocation times have
1070	potentially a lot of jittering, but good long-term stability.	1414	potentially a lot of jittering, but good long-term stability.
1071		1415
1072	static void	1416	static void
1073	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1417	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
…		…
1077		1421
1078	struct ev_periodic hourly_tick;	1422	struct ev_periodic hourly_tick;
1079	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1423	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1080	ev_periodic_start (loop, &hourly_tick);	1424	ev_periodic_start (loop, &hourly_tick);
1081		1425
1082	Example: the same as above, but use a reschedule callback to do it:	1426	Example: The same as above, but use a reschedule callback to do it:
1083		1427
1084	#include <math.h>	1428	#include <math.h>
1085		1429
1086	static ev_tstamp	1430	static ev_tstamp
1087	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1431	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
…		…
1089	return fmod (now, 3600.) + 3600.;	1433	return fmod (now, 3600.) + 3600.;
1090	}	1434	}
1091		1435
1092	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1436	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1093		1437
1094	Example: call a callback every hour, starting now:	1438	Example: Call a callback every hour, starting now:
1095		1439
1096	struct ev_periodic hourly_tick;	1440	struct ev_periodic hourly_tick;
1097	ev_periodic_init (&hourly_tick, clock_cb,	1441	ev_periodic_init (&hourly_tick, clock_cb,
1098	fmod (ev_now (loop), 3600.), 3600., 0);	1442	fmod (ev_now (loop), 3600.), 3600., 0);
1099	ev_periodic_start (loop, &hourly_tick);	1443	ev_periodic_start (loop, &hourly_tick);
…		…
1111	with the kernel (thus it coexists with your own signal handlers as long	1455	with the kernel (thus it coexists with your own signal handlers as long
1112	as you don't register any with libev). Similarly, when the last signal	1456	as you don't register any with libev). Similarly, when the last signal
1113	watcher for a signal is stopped libev will reset the signal handler to	1457	watcher for a signal is stopped libev will reset the signal handler to
1114	SIG_DFL (regardless of what it was set to before).	1458	SIG_DFL (regardless of what it was set to before).
1115		1459
		1460	If possible and supported, libev will install its handlers with
		1461	C<SA_RESTART> behaviour enabled, so syscalls should not be unduly
		1462	interrupted. If you have a problem with syscalls getting interrupted by
		1463	signals you can block all signals in an C<ev_check> watcher and unblock
		1464	them in an C<ev_prepare> watcher.
		1465
		1466	=head3 Watcher-Specific Functions and Data Members
		1467
1116	=over 4	1468	=over 4
1117		1469
1118	=item ev_signal_init (ev_signal *, callback, int signum)	1470	=item ev_signal_init (ev_signal *, callback, int signum)
1119		1471
1120	=item ev_signal_set (ev_signal *, int signum)	1472	=item ev_signal_set (ev_signal *, int signum)
…		…
1126		1478
1127	The signal the watcher watches out for.	1479	The signal the watcher watches out for.
1128		1480
1129	=back	1481	=back
1130		1482
		1483	=head3 Examples
		1484
		1485	Example: Try to exit cleanly on SIGINT and SIGTERM.
		1486
		1487	static void
		1488	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
		1489	{
		1490	ev_unloop (loop, EVUNLOOP_ALL);
		1491	}
		1492
		1493	struct ev_signal signal_watcher;
		1494	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1495	ev_signal_start (loop, &sigint_cb);
		1496
1131		1497
1132	=head2 C<ev_child> - watch out for process status changes	1498	=head2 C<ev_child> - watch out for process status changes
1133		1499
1134	Child watchers trigger when your process receives a SIGCHLD in response to	1500	Child watchers trigger when your process receives a SIGCHLD in response to
1135	some child status changes (most typically when a child of yours dies).	1501	some child status changes (most typically when a child of yours dies). It
		1502	is permissible to install a child watcher I<after> the child has been
		1503	forked (which implies it might have already exited), as long as the event
		1504	loop isn't entered (or is continued from a watcher).
		1505
		1506	Only the default event loop is capable of handling signals, and therefore
		1507	you can only rgeister child watchers in the default event loop.
		1508
		1509	=head3 Process Interaction
		1510
		1511	Libev grabs C<SIGCHLD> as soon as the default event loop is
		1512	initialised. This is necessary to guarantee proper behaviour even if
		1513	the first child watcher is started after the child exits. The occurance
		1514	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		1515	synchronously as part of the event loop processing. Libev always reaps all
		1516	children, even ones not watched.
		1517
		1518	=head3 Overriding the Built-In Processing
		1519
		1520	Libev offers no special support for overriding the built-in child
		1521	processing, but if your application collides with libev's default child
		1522	handler, you can override it easily by installing your own handler for
		1523	C<SIGCHLD> after initialising the default loop, and making sure the
		1524	default loop never gets destroyed. You are encouraged, however, to use an
		1525	event-based approach to child reaping and thus use libev's support for
		1526	that, so other libev users can use C<ev_child> watchers freely.
		1527
		1528	=head3 Watcher-Specific Functions and Data Members
1136		1529
1137	=over 4	1530	=over 4
1138		1531
1139	=item ev_child_init (ev_child *, callback, int pid)	1532	=item ev_child_init (ev_child *, callback, int pid, int trace)
1140		1533
1141	=item ev_child_set (ev_child *, int pid)	1534	=item ev_child_set (ev_child *, int pid, int trace)
1142		1535
1143	Configures the watcher to wait for status changes of process C<pid> (or	1536	Configures the watcher to wait for status changes of process C<pid> (or
1144	I<any> process if C<pid> is specified as C<0>). The callback can look	1537	I<any> process if C<pid> is specified as C<0>). The callback can look
1145	at the C<rstatus> member of the C<ev_child> watcher structure to see	1538	at the C<rstatus> member of the C<ev_child> watcher structure to see
1146	the status word (use the macros from C<sys/wait.h> and see your systems	1539	the status word (use the macros from C<sys/wait.h> and see your systems
1147	C<waitpid> documentation). The C<rpid> member contains the pid of the	1540	C<waitpid> documentation). The C<rpid> member contains the pid of the
1148	process causing the status change.	1541	process causing the status change. C<trace> must be either C<0> (only
		1542	activate the watcher when the process terminates) or C<1> (additionally
		1543	activate the watcher when the process is stopped or continued).
1149		1544
1150	=item int pid [read-only]	1545	=item int pid [read-only]
1151		1546
1152	The process id this watcher watches out for, or C<0>, meaning any process id.	1547	The process id this watcher watches out for, or C<0>, meaning any process id.
1153		1548
…		…
1160	The process exit/trace status caused by C<rpid> (see your systems	1555	The process exit/trace status caused by C<rpid> (see your systems
1161	C<waitpid> and C<sys/wait.h> documentation for details).	1556	C<waitpid> and C<sys/wait.h> documentation for details).
1162		1557
1163	=back	1558	=back
1164		1559
1165	Example: try to exit cleanly on SIGINT and SIGTERM.	1560	=head3 Examples
		1561
		1562	Example: C<fork()> a new process and install a child handler to wait for
		1563	its completion.
		1564
		1565	ev_child cw;
1166		1566
1167	static void	1567	static void
1168	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1568	child_cb (EV_P_ struct ev_child *w, int revents)
1169	{	1569	{
1170	ev_unloop (loop, EVUNLOOP_ALL);	1570	ev_child_stop (EV_A_ w);
		1571	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1171	}	1572	}
1172		1573
1173	struct ev_signal signal_watcher;	1574	pid_t pid = fork ();
1174	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1575
1175	ev_signal_start (loop, &sigint_cb);	1576	if (pid < 0)
		1577	// error
		1578	else if (pid == 0)
		1579	{
		1580	// the forked child executes here
		1581	exit (1);
		1582	}
		1583	else
		1584	{
		1585	ev_child_init (&cw, child_cb, pid, 0);
		1586	ev_child_start (EV_DEFAULT_ &cw);
		1587	}
1176		1588
1177		1589
1178	=head2 C<ev_stat> - did the file attributes just change?	1590	=head2 C<ev_stat> - did the file attributes just change?
1179		1591
1180	This watches a filesystem path for attribute changes. That is, it calls	1592	This watches a filesystem path for attribute changes. That is, it calls
…		…
1185	not exist" is a status change like any other. The condition "path does	1597	not exist" is a status change like any other. The condition "path does
1186	not exist" is signified by the C<st_nlink> field being zero (which is	1598	not exist" is signified by the C<st_nlink> field being zero (which is
1187	otherwise always forced to be at least one) and all the other fields of	1599	otherwise always forced to be at least one) and all the other fields of
1188	the stat buffer having unspecified contents.	1600	the stat buffer having unspecified contents.
1189		1601
		1602	The path I<should> be absolute and I<must not> end in a slash. If it is
		1603	relative and your working directory changes, the behaviour is undefined.
		1604
1190	Since there is no standard to do this, the portable implementation simply	1605	Since there is no standard to do this, the portable implementation simply
1191	calls C<stat (2)> regulalry on the path to see if it changed somehow. You	1606	calls C<stat (2)> regularly on the path to see if it changed somehow. You
1192	can specify a recommended polling interval for this case. If you specify	1607	can specify a recommended polling interval for this case. If you specify
1193	a polling interval of C<0> (highly recommended!) then a I<suitable,	1608	a polling interval of C<0> (highly recommended!) then a I<suitable,
1194	unspecified default> value will be used (which you can expect to be around	1609	unspecified default> value will be used (which you can expect to be around
1195	five seconds, although this might change dynamically). Libev will also	1610	five seconds, although this might change dynamically). Libev will also
1196	impose a minimum interval which is currently around C<0.1>, but thats	1611	impose a minimum interval which is currently around C<0.1>, but thats
…		…
1198		1613
1199	This watcher type is not meant for massive numbers of stat watchers,	1614	This watcher type is not meant for massive numbers of stat watchers,
1200	as even with OS-supported change notifications, this can be	1615	as even with OS-supported change notifications, this can be
1201	resource-intensive.	1616	resource-intensive.
1202		1617
1203	At the time of this writing, no specific OS backends are implemented, but	1618	At the time of this writing, only the Linux inotify interface is
1204	if demand increases, at least a kqueue and inotify backend will be added.	1619	implemented (implementing kqueue support is left as an exercise for the
		1620	reader, note, however, that the author sees no way of implementing ev_stat
		1621	semantics with kqueue). Inotify will be used to give hints only and should
		1622	not change the semantics of C<ev_stat> watchers, which means that libev
		1623	sometimes needs to fall back to regular polling again even with inotify,
		1624	but changes are usually detected immediately, and if the file exists there
		1625	will be no polling.
		1626
		1627	=head3 ABI Issues (Largefile Support)
		1628
		1629	Libev by default (unless the user overrides this) uses the default
		1630	compilation environment, which means that on systems with optionally
		1631	disabled large file support, you get the 32 bit version of the stat
		1632	structure. When using the library from programs that change the ABI to
		1633	use 64 bit file offsets the programs will fail. In that case you have to
		1634	compile libev with the same flags to get binary compatibility. This is
		1635	obviously the case with any flags that change the ABI, but the problem is
		1636	most noticably with ev_stat and largefile support.
		1637
		1638	=head3 Inotify
		1639
		1640	When C<inotify (7)> support has been compiled into libev (generally only
		1641	available on Linux) and present at runtime, it will be used to speed up
		1642	change detection where possible. The inotify descriptor will be created lazily
		1643	when the first C<ev_stat> watcher is being started.
		1644
		1645	Inotify presence does not change the semantics of C<ev_stat> watchers
		1646	except that changes might be detected earlier, and in some cases, to avoid
		1647	making regular C<stat> calls. Even in the presence of inotify support
		1648	there are many cases where libev has to resort to regular C<stat> polling.
		1649
		1650	(There is no support for kqueue, as apparently it cannot be used to
		1651	implement this functionality, due to the requirement of having a file
		1652	descriptor open on the object at all times).
		1653
		1654	=head3 The special problem of stat time resolution
		1655
		1656	The C<stat ()> syscall only supports full-second resolution portably, and
		1657	even on systems where the resolution is higher, many filesystems still
		1658	only support whole seconds.
		1659
		1660	That means that, if the time is the only thing that changes, you can
		1661	easily miss updates: on the first update, C<ev_stat> detects a change and
		1662	calls your callback, which does something. When there is another update
		1663	within the same second, C<ev_stat> will be unable to detect it as the stat
		1664	data does not change.
		1665
		1666	The solution to this is to delay acting on a change for slightly more
		1667	than a second (or till slightly after the next full second boundary), using
		1668	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		1669	ev_timer_again (loop, w)>).
		1670
		1671	The C<.02> offset is added to work around small timing inconsistencies
		1672	of some operating systems (where the second counter of the current time
		1673	might be be delayed. One such system is the Linux kernel, where a call to
		1674	C<gettimeofday> might return a timestamp with a full second later than
		1675	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		1676	update file times then there will be a small window where the kernel uses
		1677	the previous second to update file times but libev might already execute
		1678	the timer callback).
		1679
		1680	=head3 Watcher-Specific Functions and Data Members
1205		1681
1206	=over 4	1682	=over 4
1207		1683
1208	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)	1684	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
1209		1685
…		…
1213	C<path>. The C<interval> is a hint on how quickly a change is expected to	1689	C<path>. The C<interval> is a hint on how quickly a change is expected to
1214	be detected and should normally be specified as C<0> to let libev choose	1690	be detected and should normally be specified as C<0> to let libev choose
1215	a suitable value. The memory pointed to by C<path> must point to the same	1691	a suitable value. The memory pointed to by C<path> must point to the same
1216	path for as long as the watcher is active.	1692	path for as long as the watcher is active.
1217		1693
1218	The callback will be receive C<EV_STAT> when a change was detected,	1694	The callback will receive C<EV_STAT> when a change was detected, relative
1219	relative to the attributes at the time the watcher was started (or the	1695	to the attributes at the time the watcher was started (or the last change
1220	last change was detected).	1696	was detected).
1221		1697
1222	=item ev_stat_stat (ev_stat *)	1698	=item ev_stat_stat (loop, ev_stat *)
1223		1699
1224	Updates the stat buffer immediately with new values. If you change the	1700	Updates the stat buffer immediately with new values. If you change the
1225	watched path in your callback, you could call this fucntion to avoid	1701	watched path in your callback, you could call this function to avoid
1226	detecting this change (while introducing a race condition). Can also be	1702	detecting this change (while introducing a race condition if you are not
1227	useful simply to find out the new values.	1703	the only one changing the path). Can also be useful simply to find out the
		1704	new values.
1228		1705
1229	=item ev_statdata attr [read-only]	1706	=item ev_statdata attr [read-only]
1230		1707
1231	The most-recently detected attributes of the file. Although the type is of	1708	The most-recently detected attributes of the file. Although the type is
1232	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	1709	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1233	suitable for your system. If the C<st_nlink> member is C<0>, then there	1710	suitable for your system, but you can only rely on the POSIX-standardised
		1711	members to be present. If the C<st_nlink> member is C<0>, then there was
1234	was some error while C<stat>ing the file.	1712	some error while C<stat>ing the file.
1235		1713
1236	=item ev_statdata prev [read-only]	1714	=item ev_statdata prev [read-only]
1237		1715
1238	The previous attributes of the file. The callback gets invoked whenever	1716	The previous attributes of the file. The callback gets invoked whenever
1239	C<prev> != C<attr>.	1717	C<prev> != C<attr>, or, more precisely, one or more of these members
		1718	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		1719	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1240		1720
1241	=item ev_tstamp interval [read-only]	1721	=item ev_tstamp interval [read-only]
1242		1722
1243	The specified interval.	1723	The specified interval.
1244		1724
1245	=item const char *path [read-only]	1725	=item const char *path [read-only]
1246		1726
1247	The filesystem path that is being watched.	1727	The filesystem path that is being watched.
1248		1728
1249	=back	1729	=back
		1730
		1731	=head3 Examples
1250		1732
1251	Example: Watch C</etc/passwd> for attribute changes.	1733	Example: Watch C</etc/passwd> for attribute changes.
1252		1734
1253	static void	1735	static void
1254	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	1736	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
…		…
1267	}	1749	}
1268		1750
1269	...	1751	...
1270	ev_stat passwd;	1752	ev_stat passwd;
1271		1753
1272	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");	1754	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1273	ev_stat_start (loop, &passwd);	1755	ev_stat_start (loop, &passwd);
1274		1756
		1757	Example: Like above, but additionally use a one-second delay so we do not
		1758	miss updates (however, frequent updates will delay processing, too, so
		1759	one might do the work both on C<ev_stat> callback invocation I<and> on
		1760	C<ev_timer> callback invocation).
		1761
		1762	static ev_stat passwd;
		1763	static ev_timer timer;
		1764
		1765	static void
		1766	timer_cb (EV_P_ ev_timer *w, int revents)
		1767	{
		1768	ev_timer_stop (EV_A_ w);
		1769
		1770	/* now it's one second after the most recent passwd change */
		1771	}
		1772
		1773	static void
		1774	stat_cb (EV_P_ ev_stat *w, int revents)
		1775	{
		1776	/* reset the one-second timer */
		1777	ev_timer_again (EV_A_ &timer);
		1778	}
		1779
		1780	...
		1781	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		1782	ev_stat_start (loop, &passwd);
		1783	ev_timer_init (&timer, timer_cb, 0., 1.02);
		1784
1275		1785
1276	=head2 C<ev_idle> - when you've got nothing better to do...	1786	=head2 C<ev_idle> - when you've got nothing better to do...
1277		1787
1278	Idle watchers trigger events when there are no other events are pending	1788	Idle watchers trigger events when no other events of the same or higher
1279	(prepare, check and other idle watchers do not count). That is, as long	1789	priority are pending (prepare, check and other idle watchers do not
1280	as your process is busy handling sockets or timeouts (or even signals,	1790	count).
1281	imagine) it will not be triggered. But when your process is idle all idle	1791
1282	watchers are being called again and again, once per event loop iteration -	1792	That is, as long as your process is busy handling sockets or timeouts
		1793	(or even signals, imagine) of the same or higher priority it will not be
		1794	triggered. But when your process is idle (or only lower-priority watchers
		1795	are pending), the idle watchers are being called once per event loop
1283	until stopped, that is, or your process receives more events and becomes	1796	iteration - until stopped, that is, or your process receives more events
1284	busy.	1797	and becomes busy again with higher priority stuff.
1285		1798
1286	The most noteworthy effect is that as long as any idle watchers are	1799	The most noteworthy effect is that as long as any idle watchers are
1287	active, the process will not block when waiting for new events.	1800	active, the process will not block when waiting for new events.
1288		1801
1289	Apart from keeping your process non-blocking (which is a useful	1802	Apart from keeping your process non-blocking (which is a useful
1290	effect on its own sometimes), idle watchers are a good place to do	1803	effect on its own sometimes), idle watchers are a good place to do
1291	"pseudo-background processing", or delay processing stuff to after the	1804	"pseudo-background processing", or delay processing stuff to after the
1292	event loop has handled all outstanding events.	1805	event loop has handled all outstanding events.
1293		1806
		1807	=head3 Watcher-Specific Functions and Data Members
		1808
1294	=over 4	1809	=over 4
1295		1810
1296	=item ev_idle_init (ev_signal *, callback)	1811	=item ev_idle_init (ev_signal *, callback)
1297		1812
1298	Initialises and configures the idle watcher - it has no parameters of any	1813	Initialises and configures the idle watcher - it has no parameters of any
1299	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1814	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1300	believe me.	1815	believe me.
1301		1816
1302	=back	1817	=back
1303		1818
		1819	=head3 Examples
		1820
1304	Example: dynamically allocate an C<ev_idle>, start it, and in the	1821	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1305	callback, free it. Alos, use no error checking, as usual.	1822	callback, free it. Also, use no error checking, as usual.
1306		1823
1307	static void	1824	static void
1308	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	1825	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1309	{	1826	{
1310	free (w);	1827	free (w);
1311	// now do something you wanted to do when the program has	1828	// now do something you wanted to do when the program has
1312	// no longer asnything immediate to do.	1829	// no longer anything immediate to do.
1313	}	1830	}
1314		1831
1315	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1832	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1316	ev_idle_init (idle_watcher, idle_cb);	1833	ev_idle_init (idle_watcher, idle_cb);
1317	ev_idle_start (loop, idle_cb);	1834	ev_idle_start (loop, idle_cb);
…		…
1355	with priority higher than or equal to the event loop and one coroutine	1872	with priority higher than or equal to the event loop and one coroutine
1356	of lower priority, but only once, using idle watchers to keep the event	1873	of lower priority, but only once, using idle watchers to keep the event
1357	loop from blocking if lower-priority coroutines are active, thus mapping	1874	loop from blocking if lower-priority coroutines are active, thus mapping
1358	low-priority coroutines to idle/background tasks).	1875	low-priority coroutines to idle/background tasks).
1359		1876
		1877	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		1878	priority, to ensure that they are being run before any other watchers
		1879	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,
		1880	too) should not activate ("feed") events into libev. While libev fully
		1881	supports this, they might get executed before other C<ev_check> watchers
		1882	did their job. As C<ev_check> watchers are often used to embed other
		1883	(non-libev) event loops those other event loops might be in an unusable
		1884	state until their C<ev_check> watcher ran (always remind yourself to
		1885	coexist peacefully with others).
		1886
		1887	=head3 Watcher-Specific Functions and Data Members
		1888
1360	=over 4	1889	=over 4
1361		1890
1362	=item ev_prepare_init (ev_prepare *, callback)	1891	=item ev_prepare_init (ev_prepare *, callback)
1363		1892
1364	=item ev_check_init (ev_check *, callback)	1893	=item ev_check_init (ev_check *, callback)
…		…
1367	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1896	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1368	macros, but using them is utterly, utterly and completely pointless.	1897	macros, but using them is utterly, utterly and completely pointless.
1369		1898
1370	=back	1899	=back
1371		1900
1372	Example: To include a library such as adns, you would add IO watchers	1901	=head3 Examples
1373	and a timeout watcher in a prepare handler, as required by libadns, and	1902
		1903	There are a number of principal ways to embed other event loops or modules
		1904	into libev. Here are some ideas on how to include libadns into libev
		1905	(there is a Perl module named C<EV::ADNS> that does this, which you could
		1906	use as a working example. Another Perl module named C<EV::Glib> embeds a
		1907	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		1908	Glib event loop).
		1909
		1910	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1374	in a check watcher, destroy them and call into libadns. What follows is	1911	and in a check watcher, destroy them and call into libadns. What follows
1375	pseudo-code only of course:	1912	is pseudo-code only of course. This requires you to either use a low
		1913	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		1914	the callbacks for the IO/timeout watchers might not have been called yet.
1376		1915
1377	static ev_io iow [nfd];	1916	static ev_io iow [nfd];
1378	static ev_timer tw;	1917	static ev_timer tw;
1379		1918
1380	static void	1919	static void
1381	io_cb (ev_loop *loop, ev_io *w, int revents)	1920	io_cb (ev_loop *loop, ev_io *w, int revents)
1382	{	1921	{
1383	// set the relevant poll flags
1384	// could also call adns_processreadable etc. here
1385	struct pollfd *fd = (struct pollfd *)w->data;
1386	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1387	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1388	}	1922	}
1389		1923
1390	// create io watchers for each fd and a timer before blocking	1924	// create io watchers for each fd and a timer before blocking
1391	static void	1925	static void
1392	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	1926	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1393	{	1927	{
1394	int timeout = 3600000;truct pollfd fds [nfd];	1928	int timeout = 3600000;
		1929	struct pollfd fds [nfd];
1395	// actual code will need to loop here and realloc etc.	1930	// actual code will need to loop here and realloc etc.
1396	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	1931	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1397		1932
1398	/* the callback is illegal, but won't be called as we stop during check */	1933	/* the callback is illegal, but won't be called as we stop during check */
1399	ev_timer_init (&tw, 0, timeout * 1e-3);	1934	ev_timer_init (&tw, 0, timeout * 1e-3);
1400	ev_timer_start (loop, &tw);	1935	ev_timer_start (loop, &tw);
1401		1936
1402	// create on ev_io per pollfd	1937	// create one ev_io per pollfd
1403	for (int i = 0; i < nfd; ++i)	1938	for (int i = 0; i < nfd; ++i)
1404	{	1939	{
1405	ev_io_init (iow + i, io_cb, fds [i].fd,	1940	ev_io_init (iow + i, io_cb, fds [i].fd,
1406	((fds [i].events & POLLIN ? EV_READ : 0)	1941	((fds [i].events & POLLIN ? EV_READ : 0)
1407	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	1942	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1408		1943
1409	fds [i].revents = 0;	1944	fds [i].revents = 0;
1410	iow [i].data = fds + i;
1411	ev_io_start (loop, iow + i);	1945	ev_io_start (loop, iow + i);
1412	}	1946	}
1413	}	1947	}
1414		1948
1415	// stop all watchers after blocking	1949	// stop all watchers after blocking
…		…
1417	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	1951	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1418	{	1952	{
1419	ev_timer_stop (loop, &tw);	1953	ev_timer_stop (loop, &tw);
1420		1954
1421	for (int i = 0; i < nfd; ++i)	1955	for (int i = 0; i < nfd; ++i)
		1956	{
		1957	// set the relevant poll flags
		1958	// could also call adns_processreadable etc. here
		1959	struct pollfd *fd = fds + i;
		1960	int revents = ev_clear_pending (iow + i);
		1961	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		1962	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		1963
		1964	// now stop the watcher
1422	ev_io_stop (loop, iow + i);	1965	ev_io_stop (loop, iow + i);
		1966	}
1423		1967
1424	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	1968	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
		1969	}
		1970
		1971	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		1972	in the prepare watcher and would dispose of the check watcher.
		1973
		1974	Method 3: If the module to be embedded supports explicit event
		1975	notification (adns does), you can also make use of the actual watcher
		1976	callbacks, and only destroy/create the watchers in the prepare watcher.
		1977
		1978	static void
		1979	timer_cb (EV_P_ ev_timer *w, int revents)
		1980	{
		1981	adns_state ads = (adns_state)w->data;
		1982	update_now (EV_A);
		1983
		1984	adns_processtimeouts (ads, &tv_now);
		1985	}
		1986
		1987	static void
		1988	io_cb (EV_P_ ev_io *w, int revents)
		1989	{
		1990	adns_state ads = (adns_state)w->data;
		1991	update_now (EV_A);
		1992
		1993	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		1994	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		1995	}
		1996
		1997	// do not ever call adns_afterpoll
		1998
		1999	Method 4: Do not use a prepare or check watcher because the module you
		2000	want to embed is too inflexible to support it. Instead, youc na override
		2001	their poll function. The drawback with this solution is that the main
		2002	loop is now no longer controllable by EV. The C<Glib::EV> module does
		2003	this.
		2004
		2005	static gint
		2006	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		2007	{
		2008	int got_events = 0;
		2009
		2010	for (n = 0; n < nfds; ++n)
		2011	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		2012
		2013	if (timeout >= 0)
		2014	// create/start timer
		2015
		2016	// poll
		2017	ev_loop (EV_A_ 0);
		2018
		2019	// stop timer again
		2020	if (timeout >= 0)
		2021	ev_timer_stop (EV_A_ &to);
		2022
		2023	// stop io watchers again - their callbacks should have set
		2024	for (n = 0; n < nfds; ++n)
		2025	ev_io_stop (EV_A_ iow [n]);
		2026
		2027	return got_events;
1425	}	2028	}
1426		2029
1427		2030
1428	=head2 C<ev_embed> - when one backend isn't enough...	2031	=head2 C<ev_embed> - when one backend isn't enough...
1429		2032
…		…
1472	portable one.	2075	portable one.
1473		2076
1474	So when you want to use this feature you will always have to be prepared	2077	So when you want to use this feature you will always have to be prepared
1475	that you cannot get an embeddable loop. The recommended way to get around	2078	that you cannot get an embeddable loop. The recommended way to get around
1476	this is to have a separate variables for your embeddable loop, try to	2079	this is to have a separate variables for your embeddable loop, try to
1477	create it, and if that fails, use the normal loop for everything:	2080	create it, and if that fails, use the normal loop for everything.
		2081
		2082	=head3 Watcher-Specific Functions and Data Members
		2083
		2084	=over 4
		2085
		2086	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
		2087
		2088	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)
		2089
		2090	Configures the watcher to embed the given loop, which must be
		2091	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		2092	invoked automatically, otherwise it is the responsibility of the callback
		2093	to invoke it (it will continue to be called until the sweep has been done,
		2094	if you do not want thta, you need to temporarily stop the embed watcher).
		2095
		2096	=item ev_embed_sweep (loop, ev_embed *)
		2097
		2098	Make a single, non-blocking sweep over the embedded loop. This works
		2099	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		2100	apropriate way for embedded loops.
		2101
		2102	=item struct ev_loop *other [read-only]
		2103
		2104	The embedded event loop.
		2105
		2106	=back
		2107
		2108	=head3 Examples
		2109
		2110	Example: Try to get an embeddable event loop and embed it into the default
		2111	event loop. If that is not possible, use the default loop. The default
		2112	loop is stored in C<loop_hi>, while the mebeddable loop is stored in
		2113	C<loop_lo> (which is C<loop_hi> in the acse no embeddable loop can be
		2114	used).
1478		2115
1479	struct ev_loop *loop_hi = ev_default_init (0);	2116	struct ev_loop *loop_hi = ev_default_init (0);
1480	struct ev_loop *loop_lo = 0;	2117	struct ev_loop *loop_lo = 0;
1481	struct ev_embed embed;	2118	struct ev_embed embed;
1482		2119
…		…
1493	ev_embed_start (loop_hi, &embed);	2130	ev_embed_start (loop_hi, &embed);
1494	}	2131	}
1495	else	2132	else
1496	loop_lo = loop_hi;	2133	loop_lo = loop_hi;
1497		2134
1498	=over 4	2135	Example: Check if kqueue is available but not recommended and create
		2136	a kqueue backend for use with sockets (which usually work with any
		2137	kqueue implementation). Store the kqueue/socket-only event loop in
		2138	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
1499		2139
1500	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	2140	struct ev_loop *loop = ev_default_init (0);
		2141	struct ev_loop *loop_socket = 0;
		2142	struct ev_embed embed;
		2143
		2144	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2145	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2146	{
		2147	ev_embed_init (&embed, 0, loop_socket);
		2148	ev_embed_start (loop, &embed);
		2149	}
1501		2150
1502	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)	2151	if (!loop_socket)
		2152	loop_socket = loop;
1503		2153
1504	Configures the watcher to embed the given loop, which must be	2154	// now use loop_socket for all sockets, and loop for everything else
1505	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1506	invoked automatically, otherwise it is the responsibility of the callback
1507	to invoke it (it will continue to be called until the sweep has been done,
1508	if you do not want thta, you need to temporarily stop the embed watcher).
1509
1510	=item ev_embed_sweep (loop, ev_embed *)
1511
1512	Make a single, non-blocking sweep over the embedded loop. This works
1513	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
1514	apropriate way for embedded loops.
1515
1516	=item struct ev_loop *loop [read-only]
1517
1518	The embedded event loop.
1519
1520	=back
1521		2155
1522		2156
1523	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	2157	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
1524		2158
1525	Fork watchers are called when a C<fork ()> was detected (usually because	2159	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
1528	event loop blocks next and before C<ev_check> watchers are being called,	2162	event loop blocks next and before C<ev_check> watchers are being called,
1529	and only in the child after the fork. If whoever good citizen calling	2163	and only in the child after the fork. If whoever good citizen calling
1530	C<ev_default_fork> cheats and calls it in the wrong process, the fork	2164	C<ev_default_fork> cheats and calls it in the wrong process, the fork
1531	handlers will be invoked, too, of course.	2165	handlers will be invoked, too, of course.
1532		2166
		2167	=head3 Watcher-Specific Functions and Data Members
		2168
1533	=over 4	2169	=over 4
1534		2170
1535	=item ev_fork_init (ev_signal *, callback)	2171	=item ev_fork_init (ev_signal *, callback)
1536		2172
1537	Initialises and configures the fork watcher - it has no parameters of any	2173	Initialises and configures the fork watcher - it has no parameters of any
1538	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,	2174	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
1539	believe me.	2175	believe me.
		2176
		2177	=back
		2178
		2179
		2180	=head2 C<ev_async> - how to wake up another event loop
		2181
		2182	In general, you cannot use an C<ev_loop> from multiple threads or other
		2183	asynchronous sources such as signal handlers (as opposed to multiple event
		2184	loops - those are of course safe to use in different threads).
		2185
		2186	Sometimes, however, you need to wake up another event loop you do not
		2187	control, for example because it belongs to another thread. This is what
		2188	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2189	can signal it by calling C<ev_async_send>, which is thread- and signal
		2190	safe.
		2191
		2192	This functionality is very similar to C<ev_signal> watchers, as signals,
		2193	too, are asynchronous in nature, and signals, too, will be compressed
		2194	(i.e. the number of callback invocations may be less than the number of
		2195	C<ev_async_sent> calls).
		2196
		2197	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2198	just the default loop.
		2199
		2200	=head3 Queueing
		2201
		2202	C<ev_async> does not support queueing of data in any way. The reason
		2203	is that the author does not know of a simple (or any) algorithm for a
		2204	multiple-writer-single-reader queue that works in all cases and doesn't
		2205	need elaborate support such as pthreads.
		2206
		2207	That means that if you want to queue data, you have to provide your own
		2208	queue. But at least I can tell you would implement locking around your
		2209	queue:
		2210
		2211	=over 4
		2212
		2213	=item queueing from a signal handler context
		2214
		2215	To implement race-free queueing, you simply add to the queue in the signal
		2216	handler but you block the signal handler in the watcher callback. Here is an example that does that for
		2217	some fictitiuous SIGUSR1 handler:
		2218
		2219	static ev_async mysig;
		2220
		2221	static void
		2222	sigusr1_handler (void)
		2223	{
		2224	sometype data;
		2225
		2226	// no locking etc.
		2227	queue_put (data);
		2228	ev_async_send (EV_DEFAULT_ &mysig);
		2229	}
		2230
		2231	static void
		2232	mysig_cb (EV_P_ ev_async *w, int revents)
		2233	{
		2234	sometype data;
		2235	sigset_t block, prev;
		2236
		2237	sigemptyset (&block);
		2238	sigaddset (&block, SIGUSR1);
		2239	sigprocmask (SIG_BLOCK, &block, &prev);
		2240
		2241	while (queue_get (&data))
		2242	process (data);
		2243
		2244	if (sigismember (&prev, SIGUSR1)
		2245	sigprocmask (SIG_UNBLOCK, &block, 0);
		2246	}
		2247
		2248	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2249	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2250	either...).
		2251
		2252	=item queueing from a thread context
		2253
		2254	The strategy for threads is different, as you cannot (easily) block
		2255	threads but you can easily preempt them, so to queue safely you need to
		2256	employ a traditional mutex lock, such as in this pthread example:
		2257
		2258	static ev_async mysig;
		2259	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2260
		2261	static void
		2262	otherthread (void)
		2263	{
		2264	// only need to lock the actual queueing operation
		2265	pthread_mutex_lock (&mymutex);
		2266	queue_put (data);
		2267	pthread_mutex_unlock (&mymutex);
		2268
		2269	ev_async_send (EV_DEFAULT_ &mysig);
		2270	}
		2271
		2272	static void
		2273	mysig_cb (EV_P_ ev_async *w, int revents)
		2274	{
		2275	pthread_mutex_lock (&mymutex);
		2276
		2277	while (queue_get (&data))
		2278	process (data);
		2279
		2280	pthread_mutex_unlock (&mymutex);
		2281	}
		2282
		2283	=back
		2284
		2285
		2286	=head3 Watcher-Specific Functions and Data Members
		2287
		2288	=over 4
		2289
		2290	=item ev_async_init (ev_async *, callback)
		2291
		2292	Initialises and configures the async watcher - it has no parameters of any
		2293	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,
		2294	believe me.
		2295
		2296	=item ev_async_send (loop, ev_async *)
		2297
		2298	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2299	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2300	C<ev_feed_event>, this call is safe to do in other threads, signal or
		2301	similar contexts (see the dicusssion of C<EV_ATOMIC_T> in the embedding
		2302	section below on what exactly this means).
		2303
		2304	This call incurs the overhead of a syscall only once per loop iteration,
		2305	so while the overhead might be noticable, it doesn't apply to repeated
		2306	calls to C<ev_async_send>.
		2307
		2308	=item bool = ev_async_pending (ev_async *)
		2309
		2310	Returns a non-zero value when C<ev_async_send> has been called on the
		2311	watcher but the event has not yet been processed (or even noted) by the
		2312	event loop.
		2313
		2314	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2315	the loop iterates next and checks for the watcher to have become active,
		2316	it will reset the flag again. C<ev_async_pending> can be used to very
		2317	quickly check wether invoking the loop might be a good idea.
		2318
		2319	Not that this does I<not> check wether the watcher itself is pending, only
		2320	wether it has been requested to make this watcher pending.
1540		2321
1541	=back	2322	=back
1542		2323
1543		2324
1544	=head1 OTHER FUNCTIONS	2325	=head1 OTHER FUNCTIONS
…		…
1616		2397
1617	=item * Priorities are not currently supported. Initialising priorities	2398	=item * Priorities are not currently supported. Initialising priorities
1618	will fail and all watchers will have the same priority, even though there	2399	will fail and all watchers will have the same priority, even though there
1619	is an ev_pri field.	2400	is an ev_pri field.
1620		2401
		2402	=item * In libevent, the last base created gets the signals, in libev, the
		2403	first base created (== the default loop) gets the signals.
		2404
1621	=item * Other members are not supported.	2405	=item * Other members are not supported.
1622		2406
1623	=item * The libev emulation is I<not> ABI compatible to libevent, you need	2407	=item * The libev emulation is I<not> ABI compatible to libevent, you need
1624	to use the libev header file and library.	2408	to use the libev header file and library.
1625		2409
…		…
1633		2417
1634	To use it,	2418	To use it,
1635		2419
1636	#include <ev++.h>	2420	#include <ev++.h>
1637		2421
1638	(it is not installed by default). This automatically includes F<ev.h>	2422	This automatically includes F<ev.h> and puts all of its definitions (many
1639	and puts all of its definitions (many of them macros) into the global	2423	of them macros) into the global namespace. All C++ specific things are
1640	namespace. All C++ specific things are put into the C<ev> namespace.	2424	put into the C<ev> namespace. It should support all the same embedding
		2425	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
1641		2426
1642	It should support all the same embedding options as F<ev.h>, most notably	2427	Care has been taken to keep the overhead low. The only data member the C++
1643	C<EV_MULTIPLICITY>.	2428	classes add (compared to plain C-style watchers) is the event loop pointer
		2429	that the watcher is associated with (or no additional members at all if
		2430	you disable C<EV_MULTIPLICITY> when embedding libev).
		2431
		2432	Currently, functions, and static and non-static member functions can be
		2433	used as callbacks. Other types should be easy to add as long as they only
		2434	need one additional pointer for context. If you need support for other
		2435	types of functors please contact the author (preferably after implementing
		2436	it).
1644		2437
1645	Here is a list of things available in the C<ev> namespace:	2438	Here is a list of things available in the C<ev> namespace:
1646		2439
1647	=over 4	2440	=over 4
1648		2441
…		…
1664		2457
1665	All of those classes have these methods:	2458	All of those classes have these methods:
1666		2459
1667	=over 4	2460	=over 4
1668		2461
1669	=item ev::TYPE::TYPE (object *, object::method *)	2462	=item ev::TYPE::TYPE ()
1670		2463
1671	=item ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)	2464	=item ev::TYPE::TYPE (struct ev_loop *)
1672		2465
1673	=item ev::TYPE::~TYPE	2466	=item ev::TYPE::~TYPE
1674		2467
1675	The constructor takes a pointer to an object and a method pointer to	2468	The constructor (optionally) takes an event loop to associate the watcher
1676	the event handler callback to call in this class. The constructor calls	2469	with. If it is omitted, it will use C<EV_DEFAULT>.
1677	C<ev_init> for you, which means you have to call the C<set> method	2470
1678	before starting it. If you do not specify a loop then the constructor	2471	The constructor calls C<ev_init> for you, which means you have to call the
1679	automatically associates the default loop with this watcher.	2472	C<set> method before starting it.
		2473
		2474	It will not set a callback, however: You have to call the templated C<set>
		2475	method to set a callback before you can start the watcher.
		2476
		2477	(The reason why you have to use a method is a limitation in C++ which does
		2478	not allow explicit template arguments for constructors).
1680		2479
1681	The destructor automatically stops the watcher if it is active.	2480	The destructor automatically stops the watcher if it is active.
		2481
		2482	=item w->set<class, &class::method> (object *)
		2483
		2484	This method sets the callback method to call. The method has to have a
		2485	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		2486	first argument and the C<revents> as second. The object must be given as
		2487	parameter and is stored in the C<data> member of the watcher.
		2488
		2489	This method synthesizes efficient thunking code to call your method from
		2490	the C callback that libev requires. If your compiler can inline your
		2491	callback (i.e. it is visible to it at the place of the C<set> call and
		2492	your compiler is good :), then the method will be fully inlined into the
		2493	thunking function, making it as fast as a direct C callback.
		2494
		2495	Example: simple class declaration and watcher initialisation
		2496
		2497	struct myclass
		2498	{
		2499	void io_cb (ev::io &w, int revents) { }
		2500	}
		2501
		2502	myclass obj;
		2503	ev::io iow;
		2504	iow.set <myclass, &myclass::io_cb> (&obj);
		2505
		2506	=item w->set<function> (void *data = 0)
		2507
		2508	Also sets a callback, but uses a static method or plain function as
		2509	callback. The optional C<data> argument will be stored in the watcher's
		2510	C<data> member and is free for you to use.
		2511
		2512	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		2513
		2514	See the method-C<set> above for more details.
		2515
		2516	Example:
		2517
		2518	static void io_cb (ev::io &w, int revents) { }
		2519	iow.set <io_cb> ();
1682		2520
1683	=item w->set (struct ev_loop *)	2521	=item w->set (struct ev_loop *)
1684		2522
1685	Associates a different C<struct ev_loop> with this watcher. You can only	2523	Associates a different C<struct ev_loop> with this watcher. You can only
1686	do this when the watcher is inactive (and not pending either).	2524	do this when the watcher is inactive (and not pending either).
1687		2525
1688	=item w->set ([args])	2526	=item w->set ([args])
1689		2527
1690	Basically the same as C<ev_TYPE_set>, with the same args. Must be	2528	Basically the same as C<ev_TYPE_set>, with the same args. Must be
1691	called at least once. Unlike the C counterpart, an active watcher gets	2529	called at least once. Unlike the C counterpart, an active watcher gets
1692	automatically stopped and restarted.	2530	automatically stopped and restarted when reconfiguring it with this
		2531	method.
1693		2532
1694	=item w->start ()	2533	=item w->start ()
1695		2534
1696	Starts the watcher. Note that there is no C<loop> argument as the	2535	Starts the watcher. Note that there is no C<loop> argument, as the
1697	constructor already takes the loop.	2536	constructor already stores the event loop.
1698		2537
1699	=item w->stop ()	2538	=item w->stop ()
1700		2539
1701	Stops the watcher if it is active. Again, no C<loop> argument.	2540	Stops the watcher if it is active. Again, no C<loop> argument.
1702		2541
1703	=item w->again () C<ev::timer>, C<ev::periodic> only	2542	=item w->again () (C<ev::timer>, C<ev::periodic> only)
1704		2543
1705	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding	2544	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
1706	C<ev_TYPE_again> function.	2545	C<ev_TYPE_again> function.
1707		2546
1708	=item w->sweep () C<ev::embed> only	2547	=item w->sweep () (C<ev::embed> only)
1709		2548
1710	Invokes C<ev_embed_sweep>.	2549	Invokes C<ev_embed_sweep>.
1711		2550
1712	=item w->update () C<ev::stat> only	2551	=item w->update () (C<ev::stat> only)
1713		2552
1714	Invokes C<ev_stat_stat>.	2553	Invokes C<ev_stat_stat>.
1715		2554
1716	=back	2555	=back
1717		2556
…		…
1720	Example: Define a class with an IO and idle watcher, start one of them in	2559	Example: Define a class with an IO and idle watcher, start one of them in
1721	the constructor.	2560	the constructor.
1722		2561
1723	class myclass	2562	class myclass
1724	{	2563	{
1725	ev_io io; void io_cb (ev::io &w, int revents);	2564	ev::io io; void io_cb (ev::io &w, int revents);
1726	ev_idle idle void idle_cb (ev::idle &w, int revents);	2565	ev:idle idle void idle_cb (ev::idle &w, int revents);
1727		2566
1728	myclass ();	2567	myclass (int fd)
1729	}
1730
1731	myclass::myclass (int fd)
1732	: io (this, &myclass::io_cb),
1733	idle (this, &myclass::idle_cb)
1734	{	2568	{
		2569	io .set <myclass, &myclass::io_cb > (this);
		2570	idle.set <myclass, &myclass::idle_cb> (this);
		2571
1735	io.start (fd, ev::READ);	2572	io.start (fd, ev::READ);
		2573	}
1736	}	2574	};
		2575
		2576
		2577	=head1 OTHER LANGUAGE BINDINGS
		2578
		2579	Libev does not offer other language bindings itself, but bindings for a
		2580	numbe rof languages exist in the form of third-party packages. If you know
		2581	any interesting language binding in addition to the ones listed here, drop
		2582	me a note.
		2583
		2584	=over 4
		2585
		2586	=item Perl
		2587
		2588	The EV module implements the full libev API and is actually used to test
		2589	libev. EV is developed together with libev. Apart from the EV core module,
		2590	there are additional modules that implement libev-compatible interfaces
		2591	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the
		2592	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).
		2593
		2594	It can be found and installed via CPAN, its homepage is found at
		2595	L<http://software.schmorp.de/pkg/EV>.
		2596
		2597	=item Ruby
		2598
		2599	Tony Arcieri has written a ruby extension that offers access to a subset
		2600	of the libev API and adds filehandle abstractions, asynchronous DNS and
		2601	more on top of it. It can be found via gem servers. Its homepage is at
		2602	L<http://rev.rubyforge.org/>.
		2603
		2604	=item D
		2605
		2606	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		2607	be found at L<http://git.llucax.com.ar/?p=software/ev.d.git;a=summary>.
		2608
		2609	=back
1737		2610
1738		2611
1739	=head1 MACRO MAGIC	2612	=head1 MACRO MAGIC
1740		2613
1741	Libev can be compiled with a variety of options, the most fundemantal is	2614	Libev can be compiled with a variety of options, the most fundamantal
1742	C<EV_MULTIPLICITY>. This option determines wether (most) functions and	2615	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
1743	callbacks have an initial C<struct ev_loop *> argument.	2616	functions and callbacks have an initial C<struct ev_loop *> argument.
1744		2617
1745	To make it easier to write programs that cope with either variant, the	2618	To make it easier to write programs that cope with either variant, the
1746	following macros are defined:	2619	following macros are defined:
1747		2620
1748	=over 4	2621	=over 4
…		…
1778	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	2651	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
1779		2652
1780	Similar to the other two macros, this gives you the value of the default	2653	Similar to the other two macros, this gives you the value of the default
1781	loop, if multiple loops are supported ("ev loop default").	2654	loop, if multiple loops are supported ("ev loop default").
1782		2655
		2656	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		2657
		2658	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		2659	default loop has been initialised (C<UC> == unchecked). Their behaviour
		2660	is undefined when the default loop has not been initialised by a previous
		2661	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		2662
		2663	It is often prudent to use C<EV_DEFAULT> when initialising the first
		2664	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
		2665
1783	=back	2666	=back
1784		2667
1785	Example: Declare and initialise a check watcher, working regardless of	2668	Example: Declare and initialise a check watcher, utilising the above
1786	wether multiple loops are supported or not.	2669	macros so it will work regardless of whether multiple loops are supported
		2670	or not.
1787		2671
1788	static void	2672	static void
1789	check_cb (EV_P_ ev_timer *w, int revents)	2673	check_cb (EV_P_ ev_timer *w, int revents)
1790	{	2674	{
1791	ev_check_stop (EV_A_ w);	2675	ev_check_stop (EV_A_ w);
…		…
1794	ev_check check;	2678	ev_check check;
1795	ev_check_init (&check, check_cb);	2679	ev_check_init (&check, check_cb);
1796	ev_check_start (EV_DEFAULT_ &check);	2680	ev_check_start (EV_DEFAULT_ &check);
1797	ev_loop (EV_DEFAULT_ 0);	2681	ev_loop (EV_DEFAULT_ 0);
1798		2682
1799
1800	=head1 EMBEDDING	2683	=head1 EMBEDDING
1801		2684
1802	Libev can (and often is) directly embedded into host	2685	Libev can (and often is) directly embedded into host
1803	applications. Examples of applications that embed it include the Deliantra	2686	applications. Examples of applications that embed it include the Deliantra
1804	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)	2687	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
1805	and rxvt-unicode.	2688	and rxvt-unicode.
1806		2689
1807	The goal is to enable you to just copy the neecssary files into your	2690	The goal is to enable you to just copy the necessary files into your
1808	source directory without having to change even a single line in them, so	2691	source directory without having to change even a single line in them, so
1809	you can easily upgrade by simply copying (or having a checked-out copy of	2692	you can easily upgrade by simply copying (or having a checked-out copy of
1810	libev somewhere in your source tree).	2693	libev somewhere in your source tree).
1811		2694
1812	=head2 FILESETS	2695	=head2 FILESETS
…		…
1843	ev_vars.h	2726	ev_vars.h
1844	ev_wrap.h	2727	ev_wrap.h
1845		2728
1846	ev_win32.c required on win32 platforms only	2729	ev_win32.c required on win32 platforms only
1847		2730
1848	ev_select.c only when select backend is enabled (which is by default)	2731	ev_select.c only when select backend is enabled (which is enabled by default)
1849	ev_poll.c only when poll backend is enabled (disabled by default)	2732	ev_poll.c only when poll backend is enabled (disabled by default)
1850	ev_epoll.c only when the epoll backend is enabled (disabled by default)	2733	ev_epoll.c only when the epoll backend is enabled (disabled by default)
1851	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	2734	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
1852	ev_port.c only when the solaris port backend is enabled (disabled by default)	2735	ev_port.c only when the solaris port backend is enabled (disabled by default)
1853		2736
…		…
1882		2765
1883	libev.m4	2766	libev.m4
1884		2767
1885	=head2 PREPROCESSOR SYMBOLS/MACROS	2768	=head2 PREPROCESSOR SYMBOLS/MACROS
1886		2769
1887	Libev can be configured via a variety of preprocessor symbols you have to define	2770	Libev can be configured via a variety of preprocessor symbols you have to
1888	before including any of its files. The default is not to build for multiplicity	2771	define before including any of its files. The default in the absense of
1889	and only include the select backend.	2772	autoconf is noted for every option.
1890		2773
1891	=over 4	2774	=over 4
1892		2775
1893	=item EV_STANDALONE	2776	=item EV_STANDALONE
1894		2777
…		…
1902		2785
1903	If defined to be C<1>, libev will try to detect the availability of the	2786	If defined to be C<1>, libev will try to detect the availability of the
1904	monotonic clock option at both compiletime and runtime. Otherwise no use	2787	monotonic clock option at both compiletime and runtime. Otherwise no use
1905	of the monotonic clock option will be attempted. If you enable this, you	2788	of the monotonic clock option will be attempted. If you enable this, you
1906	usually have to link against librt or something similar. Enabling it when	2789	usually have to link against librt or something similar. Enabling it when
1907	the functionality isn't available is safe, though, althoguh you have	2790	the functionality isn't available is safe, though, although you have
1908	to make sure you link against any libraries where the C<clock_gettime>	2791	to make sure you link against any libraries where the C<clock_gettime>
1909	function is hiding in (often F<-lrt>).	2792	function is hiding in (often F<-lrt>).
1910		2793
1911	=item EV_USE_REALTIME	2794	=item EV_USE_REALTIME
1912		2795
1913	If defined to be C<1>, libev will try to detect the availability of the	2796	If defined to be C<1>, libev will try to detect the availability of the
1914	realtime clock option at compiletime (and assume its availability at	2797	realtime clock option at compiletime (and assume its availability at
1915	runtime if successful). Otherwise no use of the realtime clock option will	2798	runtime if successful). Otherwise no use of the realtime clock option will
1916	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	2799	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
1917	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See tzhe note about libraries	2800	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
1918	in the description of C<EV_USE_MONOTONIC>, though.	2801	note about libraries in the description of C<EV_USE_MONOTONIC>, though.
		2802
		2803	=item EV_USE_NANOSLEEP
		2804
		2805	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		2806	and will use it for delays. Otherwise it will use C<select ()>.
		2807
		2808	=item EV_USE_EVENTFD
		2809
		2810	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		2811	available and will probe for kernel support at runtime. This will improve
		2812	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		2813	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		2814	2.7 or newer, otherwise disabled.
1919		2815
1920	=item EV_USE_SELECT	2816	=item EV_USE_SELECT
1921		2817
1922	If undefined or defined to be C<1>, libev will compile in support for the	2818	If undefined or defined to be C<1>, libev will compile in support for the
1923	C<select>(2) backend. No attempt at autodetection will be done: if no	2819	C<select>(2) backend. No attempt at autodetection will be done: if no
…		…
1942	be used is the winsock select). This means that it will call	2838	be used is the winsock select). This means that it will call
1943	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	2839	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
1944	it is assumed that all these functions actually work on fds, even	2840	it is assumed that all these functions actually work on fds, even
1945	on win32. Should not be defined on non-win32 platforms.	2841	on win32. Should not be defined on non-win32 platforms.
1946		2842
		2843	=item EV_FD_TO_WIN32_HANDLE
		2844
		2845	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		2846	file descriptors to socket handles. When not defining this symbol (the
		2847	default), then libev will call C<_get_osfhandle>, which is usually
		2848	correct. In some cases, programs use their own file descriptor management,
		2849	in which case they can provide this function to map fds to socket handles.
		2850
1947	=item EV_USE_POLL	2851	=item EV_USE_POLL
1948		2852
1949	If defined to be C<1>, libev will compile in support for the C<poll>(2)	2853	If defined to be C<1>, libev will compile in support for the C<poll>(2)
1950	backend. Otherwise it will be enabled on non-win32 platforms. It	2854	backend. Otherwise it will be enabled on non-win32 platforms. It
1951	takes precedence over select.	2855	takes precedence over select.
1952		2856
1953	=item EV_USE_EPOLL	2857	=item EV_USE_EPOLL
1954		2858
1955	If defined to be C<1>, libev will compile in support for the Linux	2859	If defined to be C<1>, libev will compile in support for the Linux
1956	C<epoll>(7) backend. Its availability will be detected at runtime,	2860	C<epoll>(7) backend. Its availability will be detected at runtime,
1957	otherwise another method will be used as fallback. This is the	2861	otherwise another method will be used as fallback. This is the preferred
1958	preferred backend for GNU/Linux systems.	2862	backend for GNU/Linux systems. If undefined, it will be enabled if the
		2863	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
1959		2864
1960	=item EV_USE_KQUEUE	2865	=item EV_USE_KQUEUE
1961		2866
1962	If defined to be C<1>, libev will compile in support for the BSD style	2867	If defined to be C<1>, libev will compile in support for the BSD style
1963	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	2868	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
1978		2883
1979	=item EV_USE_DEVPOLL	2884	=item EV_USE_DEVPOLL
1980		2885
1981	reserved for future expansion, works like the USE symbols above.	2886	reserved for future expansion, works like the USE symbols above.
1982		2887
		2888	=item EV_USE_INOTIFY
		2889
		2890	If defined to be C<1>, libev will compile in support for the Linux inotify
		2891	interface to speed up C<ev_stat> watchers. Its actual availability will
		2892	be detected at runtime. If undefined, it will be enabled if the headers
		2893	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		2894
		2895	=item EV_ATOMIC_T
		2896
		2897	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		2898	access is atomic with respect to other threads or signal contexts. No such
		2899	type is easily found in the C language, so you can provide your own type
		2900	that you know is safe for your purposes. It is used both for signal handler "locking"
		2901	as well as for signal and thread safety in C<ev_async> watchers.
		2902
		2903	In the absense of this define, libev will use C<sig_atomic_t volatile>
		2904	(from F<signal.h>), which is usually good enough on most platforms.
		2905
1983	=item EV_H	2906	=item EV_H
1984		2907
1985	The name of the F<ev.h> header file used to include it. The default if	2908	The name of the F<ev.h> header file used to include it. The default if
1986	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This	2909	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
1987	can be used to virtually rename the F<ev.h> header file in case of conflicts.	2910	used to virtually rename the F<ev.h> header file in case of conflicts.
1988		2911
1989	=item EV_CONFIG_H	2912	=item EV_CONFIG_H
1990		2913
1991	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	2914	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
1992	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	2915	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
1993	C<EV_H>, above.	2916	C<EV_H>, above.
1994		2917
1995	=item EV_EVENT_H	2918	=item EV_EVENT_H
1996		2919
1997	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	2920	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
1998	of how the F<event.h> header can be found.	2921	of how the F<event.h> header can be found, the default is C<"event.h">.
1999		2922
2000	=item EV_PROTOTYPES	2923	=item EV_PROTOTYPES
2001		2924
2002	If defined to be C<0>, then F<ev.h> will not define any function	2925	If defined to be C<0>, then F<ev.h> will not define any function
2003	prototypes, but still define all the structs and other symbols. This is	2926	prototypes, but still define all the structs and other symbols. This is
…		…
2010	will have the C<struct ev_loop *> as first argument, and you can create	2933	will have the C<struct ev_loop *> as first argument, and you can create
2011	additional independent event loops. Otherwise there will be no support	2934	additional independent event loops. Otherwise there will be no support
2012	for multiple event loops and there is no first event loop pointer	2935	for multiple event loops and there is no first event loop pointer
2013	argument. Instead, all functions act on the single default loop.	2936	argument. Instead, all functions act on the single default loop.
2014		2937
		2938	=item EV_MINPRI
		2939
		2940	=item EV_MAXPRI
		2941
		2942	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		2943	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		2944	provide for more priorities by overriding those symbols (usually defined
		2945	to be C<-2> and C<2>, respectively).
		2946
		2947	When doing priority-based operations, libev usually has to linearly search
		2948	all the priorities, so having many of them (hundreds) uses a lot of space
		2949	and time, so using the defaults of five priorities (-2 .. +2) is usually
		2950	fine.
		2951
		2952	If your embedding app does not need any priorities, defining these both to
		2953	C<0> will save some memory and cpu.
		2954
2015	=item EV_PERIODIC_ENABLE	2955	=item EV_PERIODIC_ENABLE
2016		2956
2017	If undefined or defined to be C<1>, then periodic timers are supported. If	2957	If undefined or defined to be C<1>, then periodic timers are supported. If
2018	defined to be C<0>, then they are not. Disabling them saves a few kB of	2958	defined to be C<0>, then they are not. Disabling them saves a few kB of
2019	code.	2959	code.
2020		2960
		2961	=item EV_IDLE_ENABLE
		2962
		2963	If undefined or defined to be C<1>, then idle watchers are supported. If
		2964	defined to be C<0>, then they are not. Disabling them saves a few kB of
		2965	code.
		2966
2021	=item EV_EMBED_ENABLE	2967	=item EV_EMBED_ENABLE
2022		2968
2023	If undefined or defined to be C<1>, then embed watchers are supported. If	2969	If undefined or defined to be C<1>, then embed watchers are supported. If
2024	defined to be C<0>, then they are not.	2970	defined to be C<0>, then they are not.
2025		2971
…		…
2031	=item EV_FORK_ENABLE	2977	=item EV_FORK_ENABLE
2032		2978
2033	If undefined or defined to be C<1>, then fork watchers are supported. If	2979	If undefined or defined to be C<1>, then fork watchers are supported. If
2034	defined to be C<0>, then they are not.	2980	defined to be C<0>, then they are not.
2035		2981
		2982	=item EV_ASYNC_ENABLE
		2983
		2984	If undefined or defined to be C<1>, then async watchers are supported. If
		2985	defined to be C<0>, then they are not.
		2986
2036	=item EV_MINIMAL	2987	=item EV_MINIMAL
2037		2988
2038	If you need to shave off some kilobytes of code at the expense of some	2989	If you need to shave off some kilobytes of code at the expense of some
2039	speed, define this symbol to C<1>. Currently only used for gcc to override	2990	speed, define this symbol to C<1>. Currently this is used to override some
2040	some inlining decisions, saves roughly 30% codesize of amd64.	2991	inlining decisions, saves roughly 30% codesize of amd64. It also selects a
		2992	much smaller 2-heap for timer management over the default 4-heap.
2041		2993
2042	=item EV_PID_HASHSIZE	2994	=item EV_PID_HASHSIZE
2043		2995
2044	C<ev_child> watchers use a small hash table to distribute workload by	2996	C<ev_child> watchers use a small hash table to distribute workload by
2045	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	2997	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
2046	than enough. If you need to manage thousands of children you might want to	2998	than enough. If you need to manage thousands of children you might want to
2047	increase this value.	2999	increase this value (I<must> be a power of two).
		3000
		3001	=item EV_INOTIFY_HASHSIZE
		3002
		3003	C<ev_stat> watchers use a small hash table to distribute workload by
		3004	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
		3005	usually more than enough. If you need to manage thousands of C<ev_stat>
		3006	watchers you might want to increase this value (I<must> be a power of
		3007	two).
		3008
		3009	=item EV_USE_4HEAP
		3010
		3011	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3012	timer and periodics heap, libev uses a 4-heap when this symbol is defined
		3013	to C<1>. The 4-heap uses more complicated (longer) code but has
		3014	noticably faster performance with many (thousands) of watchers.
		3015
		3016	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3017	(disabled).
		3018
		3019	=item EV_HEAP_CACHE_AT
		3020
		3021	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3022	timer and periodics heap, libev can cache the timestamp (I<at>) within
		3023	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3024	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3025	but avoids random read accesses on heap changes. This improves performance
		3026	noticably with with many (hundreds) of watchers.
		3027
		3028	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3029	(disabled).
2048		3030
2049	=item EV_COMMON	3031	=item EV_COMMON
2050		3032
2051	By default, all watchers have a C<void *data> member. By redefining	3033	By default, all watchers have a C<void *data> member. By redefining
2052	this macro to a something else you can include more and other types of	3034	this macro to a something else you can include more and other types of
…		…
2065		3047
2066	=item ev_set_cb (ev, cb)	3048	=item ev_set_cb (ev, cb)
2067		3049
2068	Can be used to change the callback member declaration in each watcher,	3050	Can be used to change the callback member declaration in each watcher,
2069	and the way callbacks are invoked and set. Must expand to a struct member	3051	and the way callbacks are invoked and set. Must expand to a struct member
2070	definition and a statement, respectively. See the F<ev.v> header file for	3052	definition and a statement, respectively. See the F<ev.h> header file for
2071	their default definitions. One possible use for overriding these is to	3053	their default definitions. One possible use for overriding these is to
2072	avoid the C<struct ev_loop *> as first argument in all cases, or to use	3054	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2073	method calls instead of plain function calls in C++.	3055	method calls instead of plain function calls in C++.
		3056
		3057	=head2 EXPORTED API SYMBOLS
		3058
		3059	If you need to re-export the API (e.g. via a dll) and you need a list of
		3060	exported symbols, you can use the provided F<Symbol.*> files which list
		3061	all public symbols, one per line:
		3062
		3063	Symbols.ev for libev proper
		3064	Symbols.event for the libevent emulation
		3065
		3066	This can also be used to rename all public symbols to avoid clashes with
		3067	multiple versions of libev linked together (which is obviously bad in
		3068	itself, but sometimes it is inconvinient to avoid this).
		3069
		3070	A sed command like this will create wrapper C<#define>'s that you need to
		3071	include before including F<ev.h>:
		3072
		3073	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		3074
		3075	This would create a file F<wrap.h> which essentially looks like this:
		3076
		3077	#define ev_backend myprefix_ev_backend
		3078	#define ev_check_start myprefix_ev_check_start
		3079	#define ev_check_stop myprefix_ev_check_stop
		3080	...
2074		3081
2075	=head2 EXAMPLES	3082	=head2 EXAMPLES
2076		3083
2077	For a real-world example of a program the includes libev	3084	For a real-world example of a program the includes libev
2078	verbatim, you can have a look at the EV perl module	3085	verbatim, you can have a look at the EV perl module
…		…
2081	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file	3088	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
2082	will be compiled. It is pretty complex because it provides its own header	3089	will be compiled. It is pretty complex because it provides its own header
2083	file.	3090	file.
2084		3091
2085	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	3092	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2086	that everybody includes and which overrides some autoconf choices:	3093	that everybody includes and which overrides some configure choices:
2087		3094
		3095	#define EV_MINIMAL 1
2088	#define EV_USE_POLL 0	3096	#define EV_USE_POLL 0
2089	#define EV_MULTIPLICITY 0	3097	#define EV_MULTIPLICITY 0
2090	#define EV_PERIODICS 0	3098	#define EV_PERIODIC_ENABLE 0
		3099	#define EV_STAT_ENABLE 0
		3100	#define EV_FORK_ENABLE 0
2091	#define EV_CONFIG_H <config.h>	3101	#define EV_CONFIG_H <config.h>
		3102	#define EV_MINPRI 0
		3103	#define EV_MAXPRI 0
2092		3104
2093	#include "ev++.h"	3105	#include "ev++.h"
2094		3106
2095	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3107	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
2096		3108
2097	#include "ev_cpp.h"	3109	#include "ev_cpp.h"
2098	#include "ev.c"	3110	#include "ev.c"
		3111
		3112
		3113	=head1 THREADS AND COROUTINES
		3114
		3115	=head2 THREADS
		3116
		3117	Libev itself is completely threadsafe, but it uses no locking. This
		3118	means that you can use as many loops as you want in parallel, as long as
		3119	only one thread ever calls into one libev function with the same loop
		3120	parameter.
		3121
		3122	Or put differently: calls with different loop parameters can be done in
		3123	parallel from multiple threads, calls with the same loop parameter must be
		3124	done serially (but can be done from different threads, as long as only one
		3125	thread ever is inside a call at any point in time, e.g. by using a mutex
		3126	per loop).
		3127
		3128	If you want to know which design is best for your problem, then I cannot
		3129	help you but by giving some generic advice:
		3130
		3131	=over 4
		3132
		3133	=item * most applications have a main thread: use the default libev loop
		3134	in that thread, or create a seperate thread running only the default loop.
		3135
		3136	This helps integrating other libraries or software modules that use libev
		3137	themselves and don't care/know about threading.
		3138
		3139	=item * one loop per thread is usually a good model.
		3140
		3141	Doing this is almost never wrong, sometimes a better-performance model
		3142	exists, but it is always a good start.
		3143
		3144	=item * other models exist, such as the leader/follower pattern, where one
		3145	loop is handed through multiple threads in a kind of round-robbin fashion.
		3146
		3147	Chosing a model is hard - look around, learn, know that usually you cna do
		3148	better than you currently do :-)
		3149
		3150	=item * often you need to talk to some other thread which blocks in the
		3151	event loop - C<ev_async> watchers can be used to wake them up from other
		3152	threads safely (or from signal contexts...).
		3153
		3154	=back
		3155
		3156	=head2 COROUTINES
		3157
		3158	Libev is much more accomodating to coroutines ("cooperative threads"):
		3159	libev fully supports nesting calls to it's functions from different
		3160	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		3161	different coroutines and switch freely between both coroutines running the
		3162	loop, as long as you don't confuse yourself). The only exception is that
		3163	you must not do this from C<ev_periodic> reschedule callbacks.
		3164
		3165	Care has been invested into making sure that libev does not keep local
		3166	state inside C<ev_loop>, and other calls do not usually allow coroutine
		3167	switches.
2099		3168
2100		3169
2101	=head1 COMPLEXITIES	3170	=head1 COMPLEXITIES
2102		3171
2103	In this section the complexities of (many of) the algorithms used inside	3172	In this section the complexities of (many of) the algorithms used inside
2104	libev will be explained. For complexity discussions about backends see the	3173	libev will be explained. For complexity discussions about backends see the
2105	documentation for C<ev_default_init>.	3174	documentation for C<ev_default_init>.
2106		3175
		3176	All of the following are about amortised time: If an array needs to be
		3177	extended, libev needs to realloc and move the whole array, but this
		3178	happens asymptotically never with higher number of elements, so O(1) might
		3179	mean it might do a lengthy realloc operation in rare cases, but on average
		3180	it is much faster and asymptotically approaches constant time.
		3181
2107	=over 4	3182	=over 4
2108		3183
2109	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3184	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2110		3185
		3186	This means that, when you have a watcher that triggers in one hour and
		3187	there are 100 watchers that would trigger before that then inserting will
		3188	have to skip roughly seven (C<ld 100>) of these watchers.
		3189
2111	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	3190	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
2112		3191
		3192	That means that changing a timer costs less than removing/adding them
		3193	as only the relative motion in the event queue has to be paid for.
		3194
2113	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	3195	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
2114		3196
		3197	These just add the watcher into an array or at the head of a list.
		3198
2115	=item Stopping check/prepare/idle watchers: O(1)	3199	=item Stopping check/prepare/idle/fork/async watchers: O(1)
2116		3200
2117	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % 16))	3201	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2118		3202
		3203	These watchers are stored in lists then need to be walked to find the
		3204	correct watcher to remove. The lists are usually short (you don't usually
		3205	have many watchers waiting for the same fd or signal).
		3206
2119	=item Finding the next timer per loop iteration: O(1)	3207	=item Finding the next timer in each loop iteration: O(1)
		3208
		3209	By virtue of using a binary or 4-heap, the next timer is always found at a
		3210	fixed position in the storage array.
2120		3211
2121	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	3212	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2122		3213
2123	=item Activating one watcher: O(1)	3214	A change means an I/O watcher gets started or stopped, which requires
		3215	libev to recalculate its status (and possibly tell the kernel, depending
		3216	on backend and wether C<ev_io_set> was used).
		3217
		3218	=item Activating one watcher (putting it into the pending state): O(1)
		3219
		3220	=item Priority handling: O(number_of_priorities)
		3221
		3222	Priorities are implemented by allocating some space for each
		3223	priority. When doing priority-based operations, libev usually has to
		3224	linearly search all the priorities, but starting/stopping and activating
		3225	watchers becomes O(1) w.r.t. priority handling.
		3226
		3227	=item Sending an ev_async: O(1)
		3228
		3229	=item Processing ev_async_send: O(number_of_async_watchers)
		3230
		3231	=item Processing signals: O(max_signal_number)
		3232
		3233	Sending involves a syscall I<iff> there were no other C<ev_async_send>
		3234	calls in the current loop iteration. Checking for async and signal events
		3235	involves iterating over all running async watchers or all signal numbers.
2124		3236
2125	=back	3237	=back
2126		3238
2127		3239
		3240	=head1 Win32 platform limitations and workarounds
		3241
		3242	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		3243	requires, and its I/O model is fundamentally incompatible with the POSIX
		3244	model. Libev still offers limited functionality on this platform in
		3245	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		3246	descriptors. This only applies when using Win32 natively, not when using
		3247	e.g. cygwin.
		3248
		3249	Lifting these limitations would basically require the full
		3250	re-implementation of the I/O system. If you are into these kinds of
		3251	things, then note that glib does exactly that for you in a very portable
		3252	way (note also that glib is the slowest event library known to man).
		3253
		3254	There is no supported compilation method available on windows except
		3255	embedding it into other applications.
		3256
		3257	Due to the many, low, and arbitrary limits on the win32 platform and
		3258	the abysmal performance of winsockets, using a large number of sockets
		3259	is not recommended (and not reasonable). If your program needs to use
		3260	more than a hundred or so sockets, then likely it needs to use a totally
		3261	different implementation for windows, as libev offers the POSIX readiness
		3262	notification model, which cannot be implemented efficiently on windows
		3263	(microsoft monopoly games).
		3264
		3265	=over 4
		3266
		3267	=item The winsocket select function
		3268
		3269	The winsocket C<select> function doesn't follow POSIX in that it requires
		3270	socket I<handles> and not socket I<file descriptors>. This makes select
		3271	very inefficient, and also requires a mapping from file descriptors
		3272	to socket handles. See the discussion of the C<EV_SELECT_USE_FD_SET>,
		3273	C<EV_SELECT_IS_WINSOCKET> and C<EV_FD_TO_WIN32_HANDLE> preprocessor
		3274	symbols for more info.
		3275
		3276	The configuration for a "naked" win32 using the microsoft runtime
		3277	libraries and raw winsocket select is:
		3278
		3279	#define EV_USE_SELECT 1
		3280	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		3281
		3282	Note that winsockets handling of fd sets is O(n), so you can easily get a
		3283	complexity in the O(n²) range when using win32.
		3284
		3285	=item Limited number of file descriptors
		3286
		3287	Windows has numerous arbitrary (and low) limits on things.
		3288
		3289	Early versions of winsocket's select only supported waiting for a maximum
		3290	of C<64> handles (probably owning to the fact that all windows kernels
		3291	can only wait for C<64> things at the same time internally; microsoft
		3292	recommends spawning a chain of threads and wait for 63 handles and the
		3293	previous thread in each. Great).
		3294
		3295	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		3296	to some high number (e.g. C<2048>) before compiling the winsocket select
		3297	call (which might be in libev or elsewhere, for example, perl does its own
		3298	select emulation on windows).
		3299
		3300	Another limit is the number of file descriptors in the microsoft runtime
		3301	libraries, which by default is C<64> (there must be a hidden I<64> fetish
		3302	or something like this inside microsoft). You can increase this by calling
		3303	C<_setmaxstdio>, which can increase this limit to C<2048> (another
		3304	arbitrary limit), but is broken in many versions of the microsoft runtime
		3305	libraries.
		3306
		3307	This might get you to about C<512> or C<2048> sockets (depending on
		3308	windows version and/or the phase of the moon). To get more, you need to
		3309	wrap all I/O functions and provide your own fd management, but the cost of
		3310	calling select (O(n²)) will likely make this unworkable.
		3311
		3312	=back
		3313
		3314
		3315	=head1 PORTABILITY REQUIREMENTS
		3316
		3317	In addition to a working ISO-C implementation, libev relies on a few
		3318	additional extensions:
		3319
		3320	=over 4
		3321
		3322	=item C<sig_atomic_t volatile> must be thread-atomic as well
		3323
		3324	The type C<sig_atomic_t volatile> (or whatever is defined as
		3325	C<EV_ATOMIC_T>) must be atomic w.r.t. accesses from different
		3326	threads. This is not part of the specification for C<sig_atomic_t>, but is
		3327	believed to be sufficiently portable.
		3328
		3329	=item C<sigprocmask> must work in a threaded environment
		3330
		3331	Libev uses C<sigprocmask> to temporarily block signals. This is not
		3332	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		3333	pthread implementations will either allow C<sigprocmask> in the "main
		3334	thread" or will block signals process-wide, both behaviours would
		3335	be compatible with libev. Interaction between C<sigprocmask> and
		3336	C<pthread_sigmask> could complicate things, however.
		3337
		3338	The most portable way to handle signals is to block signals in all threads
		3339	except the initial one, and run the default loop in the initial thread as
		3340	well.
		3341
		3342	=item C<long> must be large enough for common memory allocation sizes
		3343
		3344	To improve portability and simplify using libev, libev uses C<long>
		3345	internally instead of C<size_t> when allocating its data structures. On
		3346	non-POSIX systems (Microsoft...) this might be unexpectedly low, but
		3347	is still at least 31 bits everywhere, which is enough for hundreds of
		3348	millions of watchers.
		3349
		3350	=item C<double> must hold a time value in seconds with enough accuracy
		3351
		3352	The type C<double> is used to represent timestamps. It is required to
		3353	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		3354	enough for at least into the year 4000. This requirement is fulfilled by
		3355	implementations implementing IEEE 754 (basically all existing ones).
		3356
		3357	=back
		3358
		3359	If you know of other additional requirements drop me a note.
		3360
		3361
		3362	=head1 VALGRIND
		3363
		3364	Valgrind has a special section here because it is a popular tool that is
		3365	highly useful, but valgrind reports are very hard to interpret.
		3366
		3367	If you think you found a bug (memory leak, uninitialised data access etc.)
		3368	in libev, then check twice: If valgrind reports something like:
		3369
		3370	==2274== definitely lost: 0 bytes in 0 blocks.
		3371	==2274== possibly lost: 0 bytes in 0 blocks.
		3372	==2274== still reachable: 256 bytes in 1 blocks.
		3373
		3374	then there is no memory leak. Similarly, under some circumstances,
		3375	valgrind might report kernel bugs as if it were a bug in libev, or it
		3376	might be confused (it is a very good tool, but only a tool).
		3377
		3378	If you are unsure about something, feel free to contact the mailing list
		3379	with the full valgrind report and an explanation on why you think this is
		3380	a bug in libev. However, don't be annoyed when you get a brisk "this is
		3381	no bug" answer and take the chance of learning how to interpret valgrind
		3382	properly.
		3383
		3384	If you need, for some reason, empty reports from valgrind for your project
		3385	I suggest using suppression lists.
		3386
		3387
2128	=head1 AUTHOR	3388	=head1 AUTHOR
2129		3389
2130	Marc Lehmann <libev@schmorp.de>.	3390	Marc Lehmann <libev@schmorp.de>.
2131		3391

Diff Legend

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