ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.69 by root, Fri Dec 7 19:15:39 2007 UTC vs.
Revision 1.176 by root, Mon Sep 8 17:24:39 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	#include <ev.h>	7	#include <ev.h>
8		8
9	=head1 EXAMPLE PROGRAM	9	=head2 EXAMPLE PROGRAM
10		10
		11	// a single header file is required
11	#include <ev.h>	12	#include <ev.h>
12		13
		14	// every watcher type has its own typedef'd struct
		15	// with the name ev_<type>
13	ev_io stdin_watcher;	16	ev_io stdin_watcher;
14	ev_timer timeout_watcher;	17	ev_timer timeout_watcher;
15		18
16	/* called when data readable on stdin */	19	// all watcher callbacks have a similar signature
		20	// this callback is called when data is readable on stdin
17	static void	21	static void
18	stdin_cb (EV_P_ struct ev_io *w, int revents)	22	stdin_cb (EV_P_ struct ev_io *w, int revents)
19	{	23	{
20	/* puts ("stdin ready"); */	24	puts ("stdin ready");
21	ev_io_stop (EV_A_ w); /* just a syntax example */	25	// for one-shot events, one must manually stop the watcher
22	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);
23	}	31	}
24		32
		33	// another callback, this time for a time-out
25	static void	34	static void
26	timeout_cb (EV_P_ struct ev_timer *w, int revents)	35	timeout_cb (EV_P_ struct ev_timer *w, int revents)
27	{	36	{
28	/* puts ("timeout"); */	37	puts ("timeout");
29	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);
30	}	40	}
31		41
32	int	42	int
33	main (void)	43	main (void)
34	{	44	{
		45	// use the default event loop unless you have special needs
35	struct ev_loop *loop = ev_default_loop (0);	46	struct ev_loop *loop = ev_default_loop (0);
36		47
37	/* 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
38	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);
39	ev_io_start (loop, &stdin_watcher);	51	ev_io_start (loop, &stdin_watcher);
40		52
		53	// initialise a timer watcher, then start it
41	/* simple non-repeating 5.5 second timeout */	54	// simple non-repeating 5.5 second timeout
42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	55	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
43	ev_timer_start (loop, &timeout_watcher);	56	ev_timer_start (loop, &timeout_watcher);
44		57
45	/* loop till timeout or data ready */	58	// now wait for events to arrive
46	ev_loop (loop, 0);	59	ev_loop (loop, 0);
47		60
		61	// unloop was called, so exit
48	return 0;	62	return 0;
49	}	63	}
50		64
51	=head1 DESCRIPTION	65	=head1 DESCRIPTION
52		66
53	The newest version of this document is also available as a html-formatted	67	The newest version of this document is also available as an html-formatted
54	web page you might find easier to navigate when reading it for the first	68	web page you might find easier to navigate when reading it for the first
55	time: L<http://cvs.schmorp.de/libev/ev.html>.	69	time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
56		70
57	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
58	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
59	these event sources and provide your program with events.	73	these event sources and provide your program with events.
60		74
61	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
62	(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
63	communicate events via a callback mechanism.	77	communicate events via a callback mechanism.
…		…
65	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
66	watchers>, which are relatively small C structures you initialise with the	80	watchers>, which are relatively small C structures you initialise with the
67	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
68	watcher.	82	watcher.
69		83
70	=head1 FEATURES	84	=head2 FEATURES
71		85
72	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the	86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
73	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms	87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
74	for file descriptor events (C<ev_io>), the Linux C<inotify> interface	88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
75	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers	89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
…		…
82		96
83	It also is quite fast (see this	97	It also is quite fast (see this
84	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent	98	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
85	for example).	99	for example).
86		100
87	=head1 CONVENTIONS	101	=head2 CONVENTIONS
88		102
89	Libev is very configurable. In this manual the default configuration will	103	Libev is very configurable. In this manual the default (and most common)
90	be described, which supports multiple event loops. For more info about	104	configuration will be described, which supports multiple event loops. For
91	various configuration options please have a look at B<EMBED> section in	105	more info about various configuration options please have a look at
92	this manual. If libev was configured without support for multiple event	106	B<EMBED> section in this manual. If libev was configured without support
93	loops, then all functions taking an initial argument of name C<loop>	107	for multiple event loops, then all functions taking an initial argument of
94	(which is always of type C<struct ev_loop *>) will not have this argument.	108	name C<loop> (which is always of type C<struct ev_loop *>) will not have
		109	this argument.
95		110
96	=head1 TIME REPRESENTATION	111	=head2 TIME REPRESENTATION
97		112
98	Libev represents time as a single floating point number, representing the	113	Libev represents time as a single floating point number, representing the
99	(fractional) number of seconds since the (POSIX) epoch (somewhere near	114	(fractional) number of seconds since the (POSIX) epoch (somewhere near
100	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
101	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
102	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
103	it, you should treat it as such.	118	it, you should treat it as some floating point value. Unlike the name
		119	component C<stamp> might indicate, it is also used for time differences
		120	throughout libev.
		121
		122	=head1 ERROR HANDLING
		123
		124	Libev knows three classes of errors: operating system errors, usage errors
		125	and internal errors (bugs).
		126
		127	When libev catches an operating system error it cannot handle (for example
		128	a system call indicating a condition libev cannot fix), it calls the callback
		129	set via C<ev_set_syserr_cb>, which is supposed to fix the problem or
		130	abort. The default is to print a diagnostic message and to call C<abort
		131	()>.
		132
		133	When libev detects a usage error such as a negative timer interval, then
		134	it will print a diagnostic message and abort (via the C<assert> mechanism,
		135	so C<NDEBUG> will disable this checking): these are programming errors in
		136	the libev caller and need to be fixed there.
		137
		138	Libev also has a few internal error-checking C<assert>ions, and also has
		139	extensive consistency checking code. These do not trigger under normal
		140	circumstances, as they indicate either a bug in libev or worse.
		141
104		142
105	=head1 GLOBAL FUNCTIONS	143	=head1 GLOBAL FUNCTIONS
106		144
107	These functions can be called anytime, even before initialising the	145	These functions can be called anytime, even before initialising the
108	library in any way.	146	library in any way.
…		…
113		151
114	Returns the current time as libev would use it. Please note that the	152	Returns the current time as libev would use it. Please note that the
115	C<ev_now> function is usually faster and also often returns the timestamp	153	C<ev_now> function is usually faster and also often returns the timestamp
116	you actually want to know.	154	you actually want to know.
117		155
		156	=item ev_sleep (ev_tstamp interval)
		157
		158	Sleep for the given interval: The current thread will be blocked until
		159	either it is interrupted or the given time interval has passed. Basically
		160	this is a sub-second-resolution C<sleep ()>.
		161
118	=item int ev_version_major ()	162	=item int ev_version_major ()
119		163
120	=item int ev_version_minor ()	164	=item int ev_version_minor ()
121		165
122	You can find out the major and minor version numbers of the library	166	You can find out the major and minor ABI version numbers of the library
123	you linked against by calling the functions C<ev_version_major> and	167	you linked against by calling the functions C<ev_version_major> and
124	C<ev_version_minor>. If you want, you can compare against the global	168	C<ev_version_minor>. If you want, you can compare against the global
125	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	169	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
126	version of the library your program was compiled against.	170	version of the library your program was compiled against.
127		171
		172	These version numbers refer to the ABI version of the library, not the
		173	release version.
		174
128	Usually, it's a good idea to terminate if the major versions mismatch,	175	Usually, it's a good idea to terminate if the major versions mismatch,
129	as this indicates an incompatible change. Minor versions are usually	176	as this indicates an incompatible change. Minor versions are usually
130	compatible to older versions, so a larger minor version alone is usually	177	compatible to older versions, so a larger minor version alone is usually
131	not a problem.	178	not a problem.
132		179
133	Example: Make sure we haven't accidentally been linked against the wrong	180	Example: Make sure we haven't accidentally been linked against the wrong
134	version.	181	version.
135		182
136	assert (("libev version mismatch",	183	assert (("libev version mismatch",
137	ev_version_major () == EV_VERSION_MAJOR	184	ev_version_major () == EV_VERSION_MAJOR
138	&& ev_version_minor () >= EV_VERSION_MINOR));	185	&& ev_version_minor () >= EV_VERSION_MINOR));
139		186
140	=item unsigned int ev_supported_backends ()	187	=item unsigned int ev_supported_backends ()
141		188
142	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>	189	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
143	value) compiled into this binary of libev (independent of their	190	value) compiled into this binary of libev (independent of their
…		…
145	a description of the set values.	192	a description of the set values.
146		193
147	Example: make sure we have the epoll method, because yeah this is cool and	194	Example: make sure we have the epoll method, because yeah this is cool and
148	a must have and can we have a torrent of it please!!!11	195	a must have and can we have a torrent of it please!!!11
149		196
150	assert (("sorry, no epoll, no sex",	197	assert (("sorry, no epoll, no sex",
151	ev_supported_backends () & EVBACKEND_EPOLL));	198	ev_supported_backends () & EVBACKEND_EPOLL));
152		199
153	=item unsigned int ev_recommended_backends ()	200	=item unsigned int ev_recommended_backends ()
154		201
155	Return the set of all backends compiled into this binary of libev and also	202	Return the set of all backends compiled into this binary of libev and also
156	recommended for this platform. This set is often smaller than the one	203	recommended for this platform. This set is often smaller than the one
157	returned by C<ev_supported_backends>, as for example kqueue is broken on	204	returned by C<ev_supported_backends>, as for example kqueue is broken on
158	most BSDs and will not be autodetected unless you explicitly request it	205	most BSDs and will not be auto-detected unless you explicitly request it
159	(assuming you know what you are doing). This is the set of backends that	206	(assuming you know what you are doing). This is the set of backends that
160	libev will probe for if you specify no backends explicitly.	207	libev will probe for if you specify no backends explicitly.
161		208
162	=item unsigned int ev_embeddable_backends ()	209	=item unsigned int ev_embeddable_backends ()
163		210
…		…
170	See the description of C<ev_embed> watchers for more info.	217	See the description of C<ev_embed> watchers for more info.
171		218
172	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	219	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
173		220
174	Sets the allocation function to use (the prototype is similar - the	221	Sets the allocation function to use (the prototype is similar - the
175	semantics is identical - to the realloc C function). It is used to	222	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
176	allocate and free memory (no surprises here). If it returns zero when	223	used to allocate and free memory (no surprises here). If it returns zero
177	memory needs to be allocated, the library might abort or take some	224	when memory needs to be allocated (C<size != 0>), the library might abort
178	potentially destructive action. The default is your system realloc	225	or take some potentially destructive action.
179	function.	226
		227	Since some systems (at least OpenBSD and Darwin) fail to implement
		228	correct C<realloc> semantics, libev will use a wrapper around the system
		229	C<realloc> and C<free> functions by default.
180		230
181	You could override this function in high-availability programs to, say,	231	You could override this function in high-availability programs to, say,
182	free some memory if it cannot allocate memory, to use a special allocator,	232	free some memory if it cannot allocate memory, to use a special allocator,
183	or even to sleep a while and retry until some memory is available.	233	or even to sleep a while and retry until some memory is available.
184		234
185	Example: Replace the libev allocator with one that waits a bit and then	235	Example: Replace the libev allocator with one that waits a bit and then
186	retries).	236	retries (example requires a standards-compliant C<realloc>).
187		237
188	static void *	238	static void *
189	persistent_realloc (void *ptr, size_t size)	239	persistent_realloc (void *ptr, size_t size)
190	{	240	{
191	for (;;)	241	for (;;)
…		…
202	...	252	...
203	ev_set_allocator (persistent_realloc);	253	ev_set_allocator (persistent_realloc);
204		254
205	=item ev_set_syserr_cb (void (*cb)(const char *msg));	255	=item ev_set_syserr_cb (void (*cb)(const char *msg));
206		256
207	Set the callback function to call on a retryable syscall error (such	257	Set the callback function to call on a retryable system call error (such
208	as failed select, poll, epoll_wait). The message is a printable string	258	as failed select, poll, epoll_wait). The message is a printable string
209	indicating the system call or subsystem causing the problem. If this	259	indicating the system call or subsystem causing the problem. If this
210	callback is set, then libev will expect it to remedy the sitution, no	260	callback is set, then libev will expect it to remedy the situation, no
211	matter what, when it returns. That is, libev will generally retry the	261	matter what, when it returns. That is, libev will generally retry the
212	requested operation, or, if the condition doesn't go away, do bad stuff	262	requested operation, or, if the condition doesn't go away, do bad stuff
213	(such as abort).	263	(such as abort).
214		264
215	Example: This is basically the same thing that libev does internally, too.	265	Example: This is basically the same thing that libev does internally, too.
…		…
229	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	279	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
230		280
231	An event loop is described by a C<struct ev_loop *>. The library knows two	281	An event loop is described by a C<struct ev_loop *>. The library knows two
232	types of such loops, the I<default> loop, which supports signals and child	282	types of such loops, the I<default> loop, which supports signals and child
233	events, and dynamically created loops which do not.	283	events, and dynamically created loops which do not.
234
235	If you use threads, a common model is to run the default event loop
236	in your main thread (or in a separate thread) and for each thread you
237	create, you also create another event loop. Libev itself does no locking
238	whatsoever, so if you mix calls to the same event loop in different
239	threads, make sure you lock (this is usually a bad idea, though, even if
240	done correctly, because it's hideous and inefficient).
241		284
242	=over 4	285	=over 4
243		286
244	=item struct ev_loop *ev_default_loop (unsigned int flags)	287	=item struct ev_loop *ev_default_loop (unsigned int flags)
245		288
…		…
249	flags. If that is troubling you, check C<ev_backend ()> afterwards).	292	flags. If that is troubling you, check C<ev_backend ()> afterwards).
250		293
251	If you don't know what event loop to use, use the one returned from this	294	If you don't know what event loop to use, use the one returned from this
252	function.	295	function.
253		296
		297	Note that this function is I<not> thread-safe, so if you want to use it
		298	from multiple threads, you have to lock (note also that this is unlikely,
		299	as loops cannot bes hared easily between threads anyway).
		300
		301	The default loop is the only loop that can handle C<ev_signal> and
		302	C<ev_child> watchers, and to do this, it always registers a handler
		303	for C<SIGCHLD>. If this is a problem for your application you can either
		304	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
		305	can simply overwrite the C<SIGCHLD> signal handler I<after> calling
		306	C<ev_default_init>.
		307
254	The flags argument can be used to specify special behaviour or specific	308	The flags argument can be used to specify special behaviour or specific
255	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).	309	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
256		310
257	The following flags are supported:	311	The following flags are supported:
258		312
…		…
263	The default flags value. Use this if you have no clue (it's the right	317	The default flags value. Use this if you have no clue (it's the right
264	thing, believe me).	318	thing, believe me).
265		319
266	=item C<EVFLAG_NOENV>	320	=item C<EVFLAG_NOENV>
267		321
268	If this flag bit is ored into the flag value (or the program runs setuid	322	If this flag bit is or'ed into the flag value (or the program runs setuid
269	or setgid) then libev will I<not> look at the environment variable	323	or setgid) then libev will I<not> look at the environment variable
270	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	324	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
271	override the flags completely if it is found in the environment. This is	325	override the flags completely if it is found in the environment. This is
272	useful to try out specific backends to test their performance, or to work	326	useful to try out specific backends to test their performance, or to work
273	around bugs.	327	around bugs.
…		…
279	enabling this flag.	333	enabling this flag.
280		334
281	This works by calling C<getpid ()> on every iteration of the loop,	335	This works by calling C<getpid ()> on every iteration of the loop,
282	and thus this might slow down your event loop if you do a lot of loop	336	and thus this might slow down your event loop if you do a lot of loop
283	iterations and little real work, but is usually not noticeable (on my	337	iterations and little real work, but is usually not noticeable (on my
284	Linux system for example, C<getpid> is actually a simple 5-insn sequence	338	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
285	without a syscall and thus I<very> fast, but my Linux system also has	339	without a system call and thus I<very> fast, but my GNU/Linux system also has
286	C<pthread_atfork> which is even faster).	340	C<pthread_atfork> which is even faster).
287		341
288	The big advantage of this flag is that you can forget about fork (and	342	The big advantage of this flag is that you can forget about fork (and
289	forget about forgetting to tell libev about forking) when you use this	343	forget about forgetting to tell libev about forking) when you use this
290	flag.	344	flag.
291		345
292	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>	346	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
293	environment variable.	347	environment variable.
294		348
295	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	349	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
296		350
297	This is your standard select(2) backend. Not I<completely> standard, as	351	This is your standard select(2) backend. Not I<completely> standard, as
298	libev tries to roll its own fd_set with no limits on the number of fds,	352	libev tries to roll its own fd_set with no limits on the number of fds,
299	but if that fails, expect a fairly low limit on the number of fds when	353	but if that fails, expect a fairly low limit on the number of fds when
300	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	354	using this backend. It doesn't scale too well (O(highest_fd)), but its
301	the fastest backend for a low number of fds.	355	usually the fastest backend for a low number of (low-numbered :) fds.
		356
		357	To get good performance out of this backend you need a high amount of
		358	parallelism (most of the file descriptors should be busy). If you are
		359	writing a server, you should C<accept ()> in a loop to accept as many
		360	connections as possible during one iteration. You might also want to have
		361	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		362	readiness notifications you get per iteration.
302		363
303	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	364	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
304		365
305	And this is your standard poll(2) backend. It's more complicated than	366	And this is your standard poll(2) backend. It's more complicated
306	select, but handles sparse fds better and has no artificial limit on the	367	than select, but handles sparse fds better and has no artificial
307	number of fds you can use (except it will slow down considerably with a	368	limit on the number of fds you can use (except it will slow down
308	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	369	considerably with a lot of inactive fds). It scales similarly to select,
		370	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		371	performance tips.
309		372
310	=item C<EVBACKEND_EPOLL> (value 4, Linux)	373	=item C<EVBACKEND_EPOLL> (value 4, Linux)
311		374
312	For few fds, this backend is a bit little slower than poll and select,	375	For few fds, this backend is a bit little slower than poll and select,
313	but it scales phenomenally better. While poll and select usually scale like	376	but it scales phenomenally better. While poll and select usually scale
314	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	377	like O(total_fds) where n is the total number of fds (or the highest fd),
315	either O(1) or O(active_fds).	378	epoll scales either O(1) or O(active_fds). The epoll design has a number
		379	of shortcomings, such as silently dropping events in some hard-to-detect
		380	cases and requiring a system call per fd change, no fork support and bad
		381	support for dup.
316		382
317	While stopping and starting an I/O watcher in the same iteration will	383	While stopping, setting and starting an I/O watcher in the same iteration
318	result in some caching, there is still a syscall per such incident	384	will result in some caching, there is still a system call per such incident
319	(because the fd could point to a different file description now), so its	385	(because the fd could point to a different file description now), so its
320	best to avoid that. Also, dup()ed file descriptors might not work very	386	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
321	well if you register events for both fds.	387	very well if you register events for both fds.
322		388
323	Please note that epoll sometimes generates spurious notifications, so you	389	Please note that epoll sometimes generates spurious notifications, so you
324	need to use non-blocking I/O or other means to avoid blocking when no data	390	need to use non-blocking I/O or other means to avoid blocking when no data
325	(or space) is available.	391	(or space) is available.
326		392
		393	Best performance from this backend is achieved by not unregistering all
		394	watchers for a file descriptor until it has been closed, if possible, i.e.
		395	keep at least one watcher active per fd at all times.
		396
		397	While nominally embeddable in other event loops, this feature is broken in
		398	all kernel versions tested so far.
		399
327	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	400	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
328		401
329	Kqueue deserves special mention, as at the time of this writing, it	402	Kqueue deserves special mention, as at the time of this writing, it
330	was broken on all BSDs except NetBSD (usually it doesn't work with	403	was broken on all BSDs except NetBSD (usually it doesn't work reliably
331	anything but sockets and pipes, except on Darwin, where of course its	404	with anything but sockets and pipes, except on Darwin, where of course
332	completely useless). For this reason its not being "autodetected"	405	it's completely useless). For this reason it's not being "auto-detected"
333	unless you explicitly specify it explicitly in the flags (i.e. using	406	unless you explicitly specify it explicitly in the flags (i.e. using
334	C<EVBACKEND_KQUEUE>).	407	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		408	system like NetBSD.
		409
		410	You still can embed kqueue into a normal poll or select backend and use it
		411	only for sockets (after having made sure that sockets work with kqueue on
		412	the target platform). See C<ev_embed> watchers for more info.
335		413
336	It scales in the same way as the epoll backend, but the interface to the	414	It scales in the same way as the epoll backend, but the interface to the
337	kernel is more efficient (which says nothing about its actual speed, of	415	kernel is more efficient (which says nothing about its actual speed, of
338	course). While starting and stopping an I/O watcher does not cause an	416	course). While stopping, setting and starting an I/O watcher does never
339	extra syscall as with epoll, it still adds up to four event changes per	417	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
340	incident, so its best to avoid that.	418	two event changes per incident, support for C<fork ()> is very bad and it
		419	drops fds silently in similarly hard-to-detect cases.
		420
		421	This backend usually performs well under most conditions.
		422
		423	While nominally embeddable in other event loops, this doesn't work
		424	everywhere, so you might need to test for this. And since it is broken
		425	almost everywhere, you should only use it when you have a lot of sockets
		426	(for which it usually works), by embedding it into another event loop
		427	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for
		428	sockets.
341		429
342	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	430	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
343		431
344	This is not implemented yet (and might never be).	432	This is not implemented yet (and might never be, unless you send me an
		433	implementation). According to reports, C</dev/poll> only supports sockets
		434	and is not embeddable, which would limit the usefulness of this backend
		435	immensely.
345		436
346	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	437	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
347		438
348	This uses the Solaris 10 port mechanism. As with everything on Solaris,	439	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
349	it's really slow, but it still scales very well (O(active_fds)).	440	it's really slow, but it still scales very well (O(active_fds)).
350		441
351	Please note that solaris ports can result in a lot of spurious	442	Please note that Solaris event ports can deliver a lot of spurious
352	notifications, so you need to use non-blocking I/O or other means to avoid	443	notifications, so you need to use non-blocking I/O or other means to avoid
353	blocking when no data (or space) is available.	444	blocking when no data (or space) is available.
		445
		446	While this backend scales well, it requires one system call per active
		447	file descriptor per loop iteration. For small and medium numbers of file
		448	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		449	might perform better.
		450
		451	On the positive side, ignoring the spurious readiness notifications, this
		452	backend actually performed to specification in all tests and is fully
		453	embeddable, which is a rare feat among the OS-specific backends.
354		454
355	=item C<EVBACKEND_ALL>	455	=item C<EVBACKEND_ALL>
356		456
357	Try all backends (even potentially broken ones that wouldn't be tried	457	Try all backends (even potentially broken ones that wouldn't be tried
358	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	458	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
359	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	459	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
360		460
		461	It is definitely not recommended to use this flag.
		462
361	=back	463	=back
362		464
363	If one or more of these are ored into the flags value, then only these	465	If one or more of these are or'ed into the flags value, then only these
364	backends will be tried (in the reverse order as given here). If none are	466	backends will be tried (in the reverse order as listed here). If none are
365	specified, most compiled-in backend will be tried, usually in reverse	467	specified, all backends in C<ev_recommended_backends ()> will be tried.
366	order of their flag values :)
367		468
368	The most typical usage is like this:	469	The most typical usage is like this:
369		470
370	if (!ev_default_loop (0))	471	if (!ev_default_loop (0))
371	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	472	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
372		473
373	Restrict libev to the select and poll backends, and do not allow	474	Restrict libev to the select and poll backends, and do not allow
374	environment settings to be taken into account:	475	environment settings to be taken into account:
375		476
376	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);	477	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
377		478
378	Use whatever libev has to offer, but make sure that kqueue is used if	479	Use whatever libev has to offer, but make sure that kqueue is used if
379	available (warning, breaks stuff, best use only with your own private	480	available (warning, breaks stuff, best use only with your own private
380	event loop and only if you know the OS supports your types of fds):	481	event loop and only if you know the OS supports your types of fds):
381		482
382	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);	483	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
383		484
384	=item struct ev_loop *ev_loop_new (unsigned int flags)	485	=item struct ev_loop *ev_loop_new (unsigned int flags)
385		486
386	Similar to C<ev_default_loop>, but always creates a new event loop that is	487	Similar to C<ev_default_loop>, but always creates a new event loop that is
387	always distinct from the default loop. Unlike the default loop, it cannot	488	always distinct from the default loop. Unlike the default loop, it cannot
388	handle signal and child watchers, and attempts to do so will be greeted by	489	handle signal and child watchers, and attempts to do so will be greeted by
389	undefined behaviour (or a failed assertion if assertions are enabled).	490	undefined behaviour (or a failed assertion if assertions are enabled).
390		491
		492	Note that this function I<is> thread-safe, and the recommended way to use
		493	libev with threads is indeed to create one loop per thread, and using the
		494	default loop in the "main" or "initial" thread.
		495
391	Example: Try to create a event loop that uses epoll and nothing else.	496	Example: Try to create a event loop that uses epoll and nothing else.
392		497
393	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	498	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
394	if (!epoller)	499	if (!epoller)
395	fatal ("no epoll found here, maybe it hides under your chair");	500	fatal ("no epoll found here, maybe it hides under your chair");
396		501
397	=item ev_default_destroy ()	502	=item ev_default_destroy ()
398		503
399	Destroys the default loop again (frees all memory and kernel state	504	Destroys the default loop again (frees all memory and kernel state
400	etc.). None of the active event watchers will be stopped in the normal	505	etc.). None of the active event watchers will be stopped in the normal
401	sense, so e.g. C<ev_is_active> might still return true. It is your	506	sense, so e.g. C<ev_is_active> might still return true. It is your
402	responsibility to either stop all watchers cleanly yoursef I<before>	507	responsibility to either stop all watchers cleanly yourself I<before>
403	calling this function, or cope with the fact afterwards (which is usually	508	calling this function, or cope with the fact afterwards (which is usually
404	the easiest thing, youc na just ignore the watchers and/or C<free ()> them	509	the easiest thing, you can just ignore the watchers and/or C<free ()> them
405	for example).	510	for example).
		511
		512	Note that certain global state, such as signal state, will not be freed by
		513	this function, and related watchers (such as signal and child watchers)
		514	would need to be stopped manually.
		515
		516	In general it is not advisable to call this function except in the
		517	rare occasion where you really need to free e.g. the signal handling
		518	pipe fds. If you need dynamically allocated loops it is better to use
		519	C<ev_loop_new> and C<ev_loop_destroy>).
406		520
407	=item ev_loop_destroy (loop)	521	=item ev_loop_destroy (loop)
408		522
409	Like C<ev_default_destroy>, but destroys an event loop created by an	523	Like C<ev_default_destroy>, but destroys an event loop created by an
410	earlier call to C<ev_loop_new>.	524	earlier call to C<ev_loop_new>.
411		525
412	=item ev_default_fork ()	526	=item ev_default_fork ()
413		527
		528	This function sets a flag that causes subsequent C<ev_loop> iterations
414	This function reinitialises the kernel state for backends that have	529	to reinitialise the kernel state for backends that have one. Despite the
415	one. Despite the name, you can call it anytime, but it makes most sense	530	name, you can call it anytime, but it makes most sense after forking, in
416	after forking, in either the parent or child process (or both, but that	531	the child process (or both child and parent, but that again makes little
417	again makes little sense).	532	sense). You I<must> call it in the child before using any of the libev
		533	functions, and it will only take effect at the next C<ev_loop> iteration.
418		534
419	You I<must> call this function in the child process after forking if and	535	On the other hand, you only need to call this function in the child
420	only if you want to use the event library in both processes. If you just	536	process if and only if you want to use the event library in the child. If
421	fork+exec, you don't have to call it.	537	you just fork+exec, you don't have to call it at all.
422		538
423	The function itself is quite fast and it's usually not a problem to call	539	The function itself is quite fast and it's usually not a problem to call
424	it just in case after a fork. To make this easy, the function will fit in	540	it just in case after a fork. To make this easy, the function will fit in
425	quite nicely into a call to C<pthread_atfork>:	541	quite nicely into a call to C<pthread_atfork>:
426		542
427	pthread_atfork (0, 0, ev_default_fork);	543	pthread_atfork (0, 0, ev_default_fork);
428		544
429	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
430	without calling this function, so if you force one of those backends you
431	do not need to care.
432
433	=item ev_loop_fork (loop)	545	=item ev_loop_fork (loop)
434		546
435	Like C<ev_default_fork>, but acts on an event loop created by	547	Like C<ev_default_fork>, but acts on an event loop created by
436	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	548	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
437	after fork, and how you do this is entirely your own problem.	549	after fork, and how you do this is entirely your own problem.
		550
		551	=item int ev_is_default_loop (loop)
		552
		553	Returns true when the given loop actually is the default loop, false otherwise.
438		554
439	=item unsigned int ev_loop_count (loop)	555	=item unsigned int ev_loop_count (loop)
440		556
441	Returns the count of loop iterations for the loop, which is identical to	557	Returns the count of loop iterations for the loop, which is identical to
442	the number of times libev did poll for new events. It starts at C<0> and	558	the number of times libev did poll for new events. It starts at C<0> and
…		…
455		571
456	Returns the current "event loop time", which is the time the event loop	572	Returns the current "event loop time", which is the time the event loop
457	received events and started processing them. This timestamp does not	573	received events and started processing them. This timestamp does not
458	change as long as callbacks are being processed, and this is also the base	574	change as long as callbacks are being processed, and this is also the base
459	time used for relative timers. You can treat it as the timestamp of the	575	time used for relative timers. You can treat it as the timestamp of the
460	event occuring (or more correctly, libev finding out about it).	576	event occurring (or more correctly, libev finding out about it).
		577
		578	=item ev_now_update (loop)
		579
		580	Establishes the current time by querying the kernel, updating the time
		581	returned by C<ev_now ()> in the progress. This is a costly operation and
		582	is usually done automatically within C<ev_loop ()>.
		583
		584	This function is rarely useful, but when some event callback runs for a
		585	very long time without entering the event loop, updating libev's idea of
		586	the current time is a good idea.
		587
		588	See also "The special problem of time updates" in the C<ev_timer> section.
461		589
462	=item ev_loop (loop, int flags)	590	=item ev_loop (loop, int flags)
463		591
464	Finally, this is it, the event handler. This function usually is called	592	Finally, this is it, the event handler. This function usually is called
465	after you initialised all your watchers and you want to start handling	593	after you initialised all your watchers and you want to start handling
…		…
477	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	605	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
478	those events and any outstanding ones, but will not block your process in	606	those events and any outstanding ones, but will not block your process in
479	case there are no events and will return after one iteration of the loop.	607	case there are no events and will return after one iteration of the loop.
480		608
481	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	609	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
482	neccessary) and will handle those and any outstanding ones. It will block	610	necessary) and will handle those and any outstanding ones. It will block
483	your process until at least one new event arrives, and will return after	611	your process until at least one new event arrives, and will return after
484	one iteration of the loop. This is useful if you are waiting for some	612	one iteration of the loop. This is useful if you are waiting for some
485	external event in conjunction with something not expressible using other	613	external event in conjunction with something not expressible using other
486	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	614	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is
487	usually a better approach for this kind of thing.	615	usually a better approach for this kind of thing.
488		616
489	Here are the gory details of what C<ev_loop> does:	617	Here are the gory details of what C<ev_loop> does:
490		618
491	* If there are no active watchers (reference count is zero), return.	619	- Before the first iteration, call any pending watchers.
492	- Queue prepare watchers and then call all outstanding watchers.	620	* If EVFLAG_FORKCHECK was used, check for a fork.
		621	- If a fork was detected (by any means), queue and call all fork watchers.
		622	- Queue and call all prepare watchers.
493	- If we have been forked, recreate the kernel state.	623	- If we have been forked, detach and recreate the kernel state
		624	as to not disturb the other process.
494	- Update the kernel state with all outstanding changes.	625	- Update the kernel state with all outstanding changes.
495	- Update the "event loop time".	626	- Update the "event loop time" (ev_now ()).
496	- Calculate for how long to block.	627	- Calculate for how long to sleep or block, if at all
		628	(active idle watchers, EVLOOP_NONBLOCK or not having
		629	any active watchers at all will result in not sleeping).
		630	- Sleep if the I/O and timer collect interval say so.
497	- Block the process, waiting for any events.	631	- Block the process, waiting for any events.
498	- Queue all outstanding I/O (fd) events.	632	- Queue all outstanding I/O (fd) events.
499	- Update the "event loop time" and do time jump handling.	633	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
500	- Queue all outstanding timers.	634	- Queue all outstanding timers.
501	- Queue all outstanding periodics.	635	- Queue all outstanding periodics.
502	- If no events are pending now, queue all idle watchers.	636	- Unless any events are pending now, queue all idle watchers.
503	- Queue all check watchers.	637	- Queue all check watchers.
504	- Call all queued watchers in reverse order (i.e. check watchers first).	638	- Call all queued watchers in reverse order (i.e. check watchers first).
505	Signals and child watchers are implemented as I/O watchers, and will	639	Signals and child watchers are implemented as I/O watchers, and will
506	be handled here by queueing them when their watcher gets executed.	640	be handled here by queueing them when their watcher gets executed.
507	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	641	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
508	were used, return, otherwise continue with step *.	642	were used, or there are no active watchers, return, otherwise
		643	continue with step *.
509		644
510	Example: Queue some jobs and then loop until no events are outsanding	645	Example: Queue some jobs and then loop until no events are outstanding
511	anymore.	646	anymore.
512		647
513	... queue jobs here, make sure they register event watchers as long	648	... queue jobs here, make sure they register event watchers as long
514	... as they still have work to do (even an idle watcher will do..)	649	... as they still have work to do (even an idle watcher will do..)
515	ev_loop (my_loop, 0);	650	ev_loop (my_loop, 0);
516	... jobs done. yeah!	651	... jobs done or somebody called unloop. yeah!
517		652
518	=item ev_unloop (loop, how)	653	=item ev_unloop (loop, how)
519		654
520	Can be used to make a call to C<ev_loop> return early (but only after it	655	Can be used to make a call to C<ev_loop> return early (but only after it
521	has processed all outstanding events). The C<how> argument must be either	656	has processed all outstanding events). The C<how> argument must be either
522	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	657	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
523	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	658	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
		659
		660	This "unloop state" will be cleared when entering C<ev_loop> again.
524		661
525	=item ev_ref (loop)	662	=item ev_ref (loop)
526		663
527	=item ev_unref (loop)	664	=item ev_unref (loop)
528		665
…		…
533	returning, ev_unref() after starting, and ev_ref() before stopping it. For	670	returning, ev_unref() after starting, and ev_ref() before stopping it. For
534	example, libev itself uses this for its internal signal pipe: It is not	671	example, libev itself uses this for its internal signal pipe: It is not
535	visible to the libev user and should not keep C<ev_loop> from exiting if	672	visible to the libev user and should not keep C<ev_loop> from exiting if
536	no event watchers registered by it are active. It is also an excellent	673	no event watchers registered by it are active. It is also an excellent
537	way to do this for generic recurring timers or from within third-party	674	way to do this for generic recurring timers or from within third-party
538	libraries. Just remember to I<unref after start> and I<ref before stop>.	675	libraries. Just remember to I<unref after start> and I<ref before stop>
		676	(but only if the watcher wasn't active before, or was active before,
		677	respectively).
539		678
540	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	679	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
541	running when nothing else is active.	680	running when nothing else is active.
542		681
543	struct ev_signal exitsig;	682	struct ev_signal exitsig;
544	ev_signal_init (&exitsig, sig_cb, SIGINT);	683	ev_signal_init (&exitsig, sig_cb, SIGINT);
545	ev_signal_start (loop, &exitsig);	684	ev_signal_start (loop, &exitsig);
546	evf_unref (loop);	685	evf_unref (loop);
547		686
548	Example: For some weird reason, unregister the above signal handler again.	687	Example: For some weird reason, unregister the above signal handler again.
549		688
550	ev_ref (loop);	689	ev_ref (loop);
551	ev_signal_stop (loop, &exitsig);	690	ev_signal_stop (loop, &exitsig);
		691
		692	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		693
		694	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		695
		696	These advanced functions influence the time that libev will spend waiting
		697	for events. Both time intervals are by default C<0>, meaning that libev
		698	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		699	latency.
		700
		701	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		702	allows libev to delay invocation of I/O and timer/periodic callbacks
		703	to increase efficiency of loop iterations (or to increase power-saving
		704	opportunities).
		705
		706	The background is that sometimes your program runs just fast enough to
		707	handle one (or very few) event(s) per loop iteration. While this makes
		708	the program responsive, it also wastes a lot of CPU time to poll for new
		709	events, especially with backends like C<select ()> which have a high
		710	overhead for the actual polling but can deliver many events at once.
		711
		712	By setting a higher I<io collect interval> you allow libev to spend more
		713	time collecting I/O events, so you can handle more events per iteration,
		714	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		715	C<ev_timer>) will be not affected. Setting this to a non-null value will
		716	introduce an additional C<ev_sleep ()> call into most loop iterations.
		717
		718	Likewise, by setting a higher I<timeout collect interval> you allow libev
		719	to spend more time collecting timeouts, at the expense of increased
		720	latency (the watcher callback will be called later). C<ev_io> watchers
		721	will not be affected. Setting this to a non-null value will not introduce
		722	any overhead in libev.
		723
		724	Many (busy) programs can usually benefit by setting the I/O collect
		725	interval to a value near C<0.1> or so, which is often enough for
		726	interactive servers (of course not for games), likewise for timeouts. It
		727	usually doesn't make much sense to set it to a lower value than C<0.01>,
		728	as this approaches the timing granularity of most systems.
		729
		730	Setting the I<timeout collect interval> can improve the opportunity for
		731	saving power, as the program will "bundle" timer callback invocations that
		732	are "near" in time together, by delaying some, thus reducing the number of
		733	times the process sleeps and wakes up again. Another useful technique to
		734	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		735	they fire on, say, one-second boundaries only.
		736
		737	=item ev_loop_verify (loop)
		738
		739	This function only does something when C<EV_VERIFY> support has been
		740	compiled in. It tries to go through all internal structures and checks
		741	them for validity. If anything is found to be inconsistent, it will print
		742	an error message to standard error and call C<abort ()>.
		743
		744	This can be used to catch bugs inside libev itself: under normal
		745	circumstances, this function will never abort as of course libev keeps its
		746	data structures consistent.
552		747
553	=back	748	=back
554		749
555		750
556	=head1 ANATOMY OF A WATCHER	751	=head1 ANATOMY OF A WATCHER
557		752
558	A watcher is a structure that you create and register to record your	753	A watcher is a structure that you create and register to record your
559	interest in some event. For instance, if you want to wait for STDIN to	754	interest in some event. For instance, if you want to wait for STDIN to
560	become readable, you would create an C<ev_io> watcher for that:	755	become readable, you would create an C<ev_io> watcher for that:
561		756
562	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	757	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)
563	{	758	{
564	ev_io_stop (w);	759	ev_io_stop (w);
565	ev_unloop (loop, EVUNLOOP_ALL);	760	ev_unloop (loop, EVUNLOOP_ALL);
566	}	761	}
567		762
568	struct ev_loop *loop = ev_default_loop (0);	763	struct ev_loop *loop = ev_default_loop (0);
569	struct ev_io stdin_watcher;	764	struct ev_io stdin_watcher;
570	ev_init (&stdin_watcher, my_cb);	765	ev_init (&stdin_watcher, my_cb);
571	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	766	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
572	ev_io_start (loop, &stdin_watcher);	767	ev_io_start (loop, &stdin_watcher);
573	ev_loop (loop, 0);	768	ev_loop (loop, 0);
574		769
575	As you can see, you are responsible for allocating the memory for your	770	As you can see, you are responsible for allocating the memory for your
576	watcher structures (and it is usually a bad idea to do this on the stack,	771	watcher structures (and it is usually a bad idea to do this on the stack,
577	although this can sometimes be quite valid).	772	although this can sometimes be quite valid).
578		773
579	Each watcher structure must be initialised by a call to C<ev_init	774	Each watcher structure must be initialised by a call to C<ev_init
580	(watcher *, callback)>, which expects a callback to be provided. This	775	(watcher *, callback)>, which expects a callback to be provided. This
581	callback gets invoked each time the event occurs (or, in the case of io	776	callback gets invoked each time the event occurs (or, in the case of I/O
582	watchers, each time the event loop detects that the file descriptor given	777	watchers, each time the event loop detects that the file descriptor given
583	is readable and/or writable).	778	is readable and/or writable).
584		779
585	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	780	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro
586	with arguments specific to this watcher type. There is also a macro	781	with arguments specific to this watcher type. There is also a macro
…		…
656	=item C<EV_FORK>	851	=item C<EV_FORK>
657		852
658	The event loop has been resumed in the child process after fork (see	853	The event loop has been resumed in the child process after fork (see
659	C<ev_fork>).	854	C<ev_fork>).
660		855
		856	=item C<EV_ASYNC>
		857
		858	The given async watcher has been asynchronously notified (see C<ev_async>).
		859
661	=item C<EV_ERROR>	860	=item C<EV_ERROR>
662		861
663	An unspecified error has occured, the watcher has been stopped. This might	862	An unspecified error has occurred, the watcher has been stopped. This might
664	happen because the watcher could not be properly started because libev	863	happen because the watcher could not be properly started because libev
665	ran out of memory, a file descriptor was found to be closed or any other	864	ran out of memory, a file descriptor was found to be closed or any other
666	problem. You best act on it by reporting the problem and somehow coping	865	problem. You best act on it by reporting the problem and somehow coping
667	with the watcher being stopped.	866	with the watcher being stopped.
668		867
669	Libev will usually signal a few "dummy" events together with an error,	868	Libev will usually signal a few "dummy" events together with an error,
670	for example it might indicate that a fd is readable or writable, and if	869	for example it might indicate that a fd is readable or writable, and if
671	your callbacks is well-written it can just attempt the operation and cope	870	your callbacks is well-written it can just attempt the operation and cope
672	with the error from read() or write(). This will not work in multithreaded	871	with the error from read() or write(). This will not work in multi-threaded
673	programs, though, so beware.	872	programs, though, so beware.
674		873
675	=back	874	=back
676		875
677	=head2 GENERIC WATCHER FUNCTIONS	876	=head2 GENERIC WATCHER FUNCTIONS
…		…
707	Although some watcher types do not have type-specific arguments	906	Although some watcher types do not have type-specific arguments
708	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	907	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
709		908
710	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	909	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
711		910
712	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	911	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
713	calls into a single call. This is the most convinient method to initialise	912	calls into a single call. This is the most convenient method to initialise
714	a watcher. The same limitations apply, of course.	913	a watcher. The same limitations apply, of course.
715		914
716	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	915	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
717		916
718	Starts (activates) the given watcher. Only active watchers will receive	917	Starts (activates) the given watcher. Only active watchers will receive
…		…
736	=item bool ev_is_pending (ev_TYPE *watcher)	935	=item bool ev_is_pending (ev_TYPE *watcher)
737		936
738	Returns a true value iff the watcher is pending, (i.e. it has outstanding	937	Returns a true value iff the watcher is pending, (i.e. it has outstanding
739	events but its callback has not yet been invoked). As long as a watcher	938	events but its callback has not yet been invoked). As long as a watcher
740	is pending (but not active) you must not call an init function on it (but	939	is pending (but not active) you must not call an init function on it (but
741	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to	940	C<ev_TYPE_set> is safe), you must not change its priority, and you must
742	libev (e.g. you cnanot C<free ()> it).	941	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		942	it).
743		943
744	=item callback ev_cb (ev_TYPE *watcher)	944	=item callback ev_cb (ev_TYPE *watcher)
745		945
746	Returns the callback currently set on the watcher.	946	Returns the callback currently set on the watcher.
747		947
…		…
766	watchers on the same event and make sure one is called first.	966	watchers on the same event and make sure one is called first.
767		967
768	If you need to suppress invocation when higher priority events are pending	968	If you need to suppress invocation when higher priority events are pending
769	you need to look at C<ev_idle> watchers, which provide this functionality.	969	you need to look at C<ev_idle> watchers, which provide this functionality.
770		970
		971	You I<must not> change the priority of a watcher as long as it is active or
		972	pending.
		973
771	The default priority used by watchers when no priority has been set is	974	The default priority used by watchers when no priority has been set is
772	always C<0>, which is supposed to not be too high and not be too low :).	975	always C<0>, which is supposed to not be too high and not be too low :).
773		976
774	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is	977	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
775	fine, as long as you do not mind that the priority value you query might	978	fine, as long as you do not mind that the priority value you query might
776	or might not have been adjusted to be within valid range.	979	or might not have been adjusted to be within valid range.
		980
		981	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		982
		983	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		984	C<loop> nor C<revents> need to be valid as long as the watcher callback
		985	can deal with that fact.
		986
		987	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		988
		989	If the watcher is pending, this function returns clears its pending status
		990	and returns its C<revents> bitset (as if its callback was invoked). If the
		991	watcher isn't pending it does nothing and returns C<0>.
777		992
778	=back	993	=back
779		994
780		995
781	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	996	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
…		…
785	to associate arbitrary data with your watcher. If you need more data and	1000	to associate arbitrary data with your watcher. If you need more data and
786	don't want to allocate memory and store a pointer to it in that data	1001	don't want to allocate memory and store a pointer to it in that data
787	member, you can also "subclass" the watcher type and provide your own	1002	member, you can also "subclass" the watcher type and provide your own
788	data:	1003	data:
789		1004
790	struct my_io	1005	struct my_io
791	{	1006	{
792	struct ev_io io;	1007	struct ev_io io;
793	int otherfd;	1008	int otherfd;
794	void *somedata;	1009	void *somedata;
795	struct whatever *mostinteresting;	1010	struct whatever *mostinteresting;
796	}	1011	}
797		1012
798	And since your callback will be called with a pointer to the watcher, you	1013	And since your callback will be called with a pointer to the watcher, you
799	can cast it back to your own type:	1014	can cast it back to your own type:
800		1015
801	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1016	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)
802	{	1017	{
803	struct my_io *w = (struct my_io *)w_;	1018	struct my_io *w = (struct my_io *)w_;
804	...	1019	...
805	}	1020	}
806		1021
807	More interesting and less C-conformant ways of casting your callback type	1022	More interesting and less C-conformant ways of casting your callback type
808	instead have been omitted.	1023	instead have been omitted.
809		1024
810	Another common scenario is having some data structure with multiple	1025	Another common scenario is having some data structure with multiple
811	watchers:	1026	watchers:
812		1027
813	struct my_biggy	1028	struct my_biggy
814	{	1029	{
815	int some_data;	1030	int some_data;
816	ev_timer t1;	1031	ev_timer t1;
817	ev_timer t2;	1032	ev_timer t2;
818	}	1033	}
819		1034
820	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1035	In this case getting the pointer to C<my_biggy> is a bit more complicated,
821	you need to use C<offsetof>:	1036	you need to use C<offsetof>:
822		1037
823	#include <stddef.h>	1038	#include <stddef.h>
824		1039
825	static void	1040	static void
826	t1_cb (EV_P_ struct ev_timer *w, int revents)	1041	t1_cb (EV_P_ struct ev_timer *w, int revents)
827	{	1042	{
828	struct my_biggy big = (struct my_biggy *	1043	struct my_biggy big = (struct my_biggy *
829	(((char *)w) - offsetof (struct my_biggy, t1));	1044	(((char *)w) - offsetof (struct my_biggy, t1));
830	}	1045	}
831		1046
832	static void	1047	static void
833	t2_cb (EV_P_ struct ev_timer *w, int revents)	1048	t2_cb (EV_P_ struct ev_timer *w, int revents)
834	{	1049	{
835	struct my_biggy big = (struct my_biggy *	1050	struct my_biggy big = (struct my_biggy *
836	(((char *)w) - offsetof (struct my_biggy, t2));	1051	(((char *)w) - offsetof (struct my_biggy, t2));
837	}	1052	}
838		1053
839		1054
840	=head1 WATCHER TYPES	1055	=head1 WATCHER TYPES
841		1056
842	This section describes each watcher in detail, but will not repeat	1057	This section describes each watcher in detail, but will not repeat
…		…
866	In general you can register as many read and/or write event watchers per	1081	In general you can register as many read and/or write event watchers per
867	fd as you want (as long as you don't confuse yourself). Setting all file	1082	fd as you want (as long as you don't confuse yourself). Setting all file
868	descriptors to non-blocking mode is also usually a good idea (but not	1083	descriptors to non-blocking mode is also usually a good idea (but not
869	required if you know what you are doing).	1084	required if you know what you are doing).
870		1085
871	You have to be careful with dup'ed file descriptors, though. Some backends
872	(the linux epoll backend is a notable example) cannot handle dup'ed file
873	descriptors correctly if you register interest in two or more fds pointing
874	to the same underlying file/socket/etc. description (that is, they share
875	the same underlying "file open").
876
877	If you must do this, then force the use of a known-to-be-good backend	1086	If you must do this, then force the use of a known-to-be-good backend
878	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1087	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
879	C<EVBACKEND_POLL>).	1088	C<EVBACKEND_POLL>).
880		1089
881	Another thing you have to watch out for is that it is quite easy to	1090	Another thing you have to watch out for is that it is quite easy to
882	receive "spurious" readyness notifications, that is your callback might	1091	receive "spurious" readiness notifications, that is your callback might
883	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1092	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
884	because there is no data. Not only are some backends known to create a	1093	because there is no data. Not only are some backends known to create a
885	lot of those (for example solaris ports), it is very easy to get into	1094	lot of those (for example Solaris ports), it is very easy to get into
886	this situation even with a relatively standard program structure. Thus	1095	this situation even with a relatively standard program structure. Thus
887	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1096	it is best to always use non-blocking I/O: An extra C<read>(2) returning
888	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1097	C<EAGAIN> is far preferable to a program hanging until some data arrives.
889		1098
890	If you cannot run the fd in non-blocking mode (for example you should not	1099	If you cannot run the fd in non-blocking mode (for example you should not
891	play around with an Xlib connection), then you have to seperately re-test	1100	play around with an Xlib connection), then you have to separately re-test
892	whether a file descriptor is really ready with a known-to-be good interface	1101	whether a file descriptor is really ready with a known-to-be good interface
893	such as poll (fortunately in our Xlib example, Xlib already does this on	1102	such as poll (fortunately in our Xlib example, Xlib already does this on
894	its own, so its quite safe to use).	1103	its own, so its quite safe to use).
895		1104
		1105	=head3 The special problem of disappearing file descriptors
		1106
		1107	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1108	descriptor (either by calling C<close> explicitly or by any other means,
		1109	such as C<dup>). The reason is that you register interest in some file
		1110	descriptor, but when it goes away, the operating system will silently drop
		1111	this interest. If another file descriptor with the same number then is
		1112	registered with libev, there is no efficient way to see that this is, in
		1113	fact, a different file descriptor.
		1114
		1115	To avoid having to explicitly tell libev about such cases, libev follows
		1116	the following policy: Each time C<ev_io_set> is being called, libev
		1117	will assume that this is potentially a new file descriptor, otherwise
		1118	it is assumed that the file descriptor stays the same. That means that
		1119	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1120	descriptor even if the file descriptor number itself did not change.
		1121
		1122	This is how one would do it normally anyway, the important point is that
		1123	the libev application should not optimise around libev but should leave
		1124	optimisations to libev.
		1125
		1126	=head3 The special problem of dup'ed file descriptors
		1127
		1128	Some backends (e.g. epoll), cannot register events for file descriptors,
		1129	but only events for the underlying file descriptions. That means when you
		1130	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1131	events for them, only one file descriptor might actually receive events.
		1132
		1133	There is no workaround possible except not registering events
		1134	for potentially C<dup ()>'ed file descriptors, or to resort to
		1135	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1136
		1137	=head3 The special problem of fork
		1138
		1139	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1140	useless behaviour. Libev fully supports fork, but needs to be told about
		1141	it in the child.
		1142
		1143	To support fork in your programs, you either have to call
		1144	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1145	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1146	C<EVBACKEND_POLL>.
		1147
		1148	=head3 The special problem of SIGPIPE
		1149
		1150	While not really specific to libev, it is easy to forget about SIGPIPE:
		1151	when writing to a pipe whose other end has been closed, your program gets
		1152	send a SIGPIPE, which, by default, aborts your program. For most programs
		1153	this is sensible behaviour, for daemons, this is usually undesirable.
		1154
		1155	So when you encounter spurious, unexplained daemon exits, make sure you
		1156	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1157	somewhere, as that would have given you a big clue).
		1158
		1159
		1160	=head3 Watcher-Specific Functions
		1161
896	=over 4	1162	=over 4
897		1163
898	=item ev_io_init (ev_io *, callback, int fd, int events)	1164	=item ev_io_init (ev_io *, callback, int fd, int events)
899		1165
900	=item ev_io_set (ev_io *, int fd, int events)	1166	=item ev_io_set (ev_io *, int fd, int events)
901		1167
902	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1168	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
903	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or	1169	receive events for and events is either C<EV_READ>, C<EV_WRITE> or
904	C<EV_READ | EV_WRITE> to receive the given events.	1170	C<EV_READ | EV_WRITE> to receive the given events.
905		1171
906	=item int fd [read-only]	1172	=item int fd [read-only]
907		1173
908	The file descriptor being watched.	1174	The file descriptor being watched.
…		…
910	=item int events [read-only]	1176	=item int events [read-only]
911		1177
912	The events being watched.	1178	The events being watched.
913		1179
914	=back	1180	=back
		1181
		1182	=head3 Examples
915		1183
916	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1184	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
917	readable, but only once. Since it is likely line-buffered, you could	1185	readable, but only once. Since it is likely line-buffered, you could
918	attempt to read a whole line in the callback.	1186	attempt to read a whole line in the callback.
919		1187
920	static void	1188	static void
921	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1189	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
922	{	1190	{
923	ev_io_stop (loop, w);	1191	ev_io_stop (loop, w);
924	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1192	.. read from stdin here (or from w->fd) and haqndle any I/O errors
925	}	1193	}
926		1194
927	...	1195	...
928	struct ev_loop *loop = ev_default_init (0);	1196	struct ev_loop *loop = ev_default_init (0);
929	struct ev_io stdin_readable;	1197	struct ev_io stdin_readable;
930	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1198	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
931	ev_io_start (loop, &stdin_readable);	1199	ev_io_start (loop, &stdin_readable);
932	ev_loop (loop, 0);	1200	ev_loop (loop, 0);
933		1201
934		1202
935	=head2 C<ev_timer> - relative and optionally repeating timeouts	1203	=head2 C<ev_timer> - relative and optionally repeating timeouts
936		1204
937	Timer watchers are simple relative timers that generate an event after a	1205	Timer watchers are simple relative timers that generate an event after a
938	given time, and optionally repeating in regular intervals after that.	1206	given time, and optionally repeating in regular intervals after that.
939		1207
940	The timers are based on real time, that is, if you register an event that	1208	The timers are based on real time, that is, if you register an event that
941	times out after an hour and you reset your system clock to last years	1209	times out after an hour and you reset your system clock to January last
942	time, it will still time out after (roughly) and hour. "Roughly" because	1210	year, it will still time out after (roughly) and hour. "Roughly" because
943	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1211	detecting time jumps is hard, and some inaccuracies are unavoidable (the
944	monotonic clock option helps a lot here).	1212	monotonic clock option helps a lot here).
		1213
		1214	The callback is guaranteed to be invoked only after its timeout has passed,
		1215	but if multiple timers become ready during the same loop iteration then
		1216	order of execution is undefined.
		1217
		1218	=head3 The special problem of time updates
		1219
		1220	Establishing the current time is a costly operation (it usually takes at
		1221	least two system calls): EV therefore updates its idea of the current
		1222	time only before and after C<ev_loop> polls for new events, which causes
		1223	a growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1224	lots of events.
945		1225
946	The relative timeouts are calculated relative to the C<ev_now ()>	1226	The relative timeouts are calculated relative to the C<ev_now ()>
947	time. This is usually the right thing as this timestamp refers to the time	1227	time. This is usually the right thing as this timestamp refers to the time
948	of the event triggering whatever timeout you are modifying/starting. If	1228	of the event triggering whatever timeout you are modifying/starting. If
949	you suspect event processing to be delayed and you I<need> to base the timeout	1229	you suspect event processing to be delayed and you I<need> to base the
950	on the current time, use something like this to adjust for this:	1230	timeout on the current time, use something like this to adjust for this:
951		1231
952	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1232	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
953		1233
954	The callback is guarenteed to be invoked only when its timeout has passed,	1234	If the event loop is suspended for a long time, one can also force an
955	but if multiple timers become ready during the same loop iteration then	1235	update of the time returned by C<ev_now ()> by calling C<ev_now_update
956	order of execution is undefined.	1236	()>.
		1237
		1238	=head3 Watcher-Specific Functions and Data Members
957		1239
958	=over 4	1240	=over 4
959		1241
960	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1242	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
961		1243
962	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1244	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
963		1245
964	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1246	Configure the timer to trigger after C<after> seconds. If C<repeat>
965	C<0.>, then it will automatically be stopped. If it is positive, then the	1247	is C<0.>, then it will automatically be stopped once the timeout is
966	timer will automatically be configured to trigger again C<repeat> seconds	1248	reached. If it is positive, then the timer will automatically be
967	later, again, and again, until stopped manually.	1249	configured to trigger again C<repeat> seconds later, again, and again,
		1250	until stopped manually.
968		1251
969	The timer itself will do a best-effort at avoiding drift, that is, if you	1252	The timer itself will do a best-effort at avoiding drift, that is, if
970	configure a timer to trigger every 10 seconds, then it will trigger at	1253	you configure a timer to trigger every 10 seconds, then it will normally
971	exactly 10 second intervals. If, however, your program cannot keep up with	1254	trigger at exactly 10 second intervals. If, however, your program cannot
972	the timer (because it takes longer than those 10 seconds to do stuff) the	1255	keep up with the timer (because it takes longer than those 10 seconds to
973	timer will not fire more than once per event loop iteration.	1256	do stuff) the timer will not fire more than once per event loop iteration.
974		1257
975	=item ev_timer_again (loop)	1258	=item ev_timer_again (loop, ev_timer *)
976		1259
977	This will act as if the timer timed out and restart it again if it is	1260	This will act as if the timer timed out and restart it again if it is
978	repeating. The exact semantics are:	1261	repeating. The exact semantics are:
979		1262
980	If the timer is pending, its pending status is cleared.	1263	If the timer is pending, its pending status is cleared.
981		1264
982	If the timer is started but nonrepeating, stop it (as if it timed out).	1265	If the timer is started but non-repeating, stop it (as if it timed out).
983		1266
984	If the timer is repeating, either start it if necessary (with the	1267	If the timer is repeating, either start it if necessary (with the
985	C<repeat> value), or reset the running timer to the C<repeat> value.	1268	C<repeat> value), or reset the running timer to the C<repeat> value.
986		1269
987	This sounds a bit complicated, but here is a useful and typical	1270	This sounds a bit complicated, but here is a useful and typical
988	example: Imagine you have a tcp connection and you want a so-called idle	1271	example: Imagine you have a TCP connection and you want a so-called idle
989	timeout, that is, you want to be called when there have been, say, 60	1272	timeout, that is, you want to be called when there have been, say, 60
990	seconds of inactivity on the socket. The easiest way to do this is to	1273	seconds of inactivity on the socket. The easiest way to do this is to
991	configure an C<ev_timer> with a C<repeat> value of C<60> and then call	1274	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
992	C<ev_timer_again> each time you successfully read or write some data. If	1275	C<ev_timer_again> each time you successfully read or write some data. If
993	you go into an idle state where you do not expect data to travel on the	1276	you go into an idle state where you do not expect data to travel on the
…		…
1015	or C<ev_timer_again> is called and determines the next timeout (if any),	1298	or C<ev_timer_again> is called and determines the next timeout (if any),
1016	which is also when any modifications are taken into account.	1299	which is also when any modifications are taken into account.
1017		1300
1018	=back	1301	=back
1019		1302
		1303	=head3 Examples
		1304
1020	Example: Create a timer that fires after 60 seconds.	1305	Example: Create a timer that fires after 60 seconds.
1021		1306
1022	static void	1307	static void
1023	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1308	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1024	{	1309	{
1025	.. one minute over, w is actually stopped right here	1310	.. one minute over, w is actually stopped right here
1026	}	1311	}
1027		1312
1028	struct ev_timer mytimer;	1313	struct ev_timer mytimer;
1029	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1314	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1030	ev_timer_start (loop, &mytimer);	1315	ev_timer_start (loop, &mytimer);
1031		1316
1032	Example: Create a timeout timer that times out after 10 seconds of	1317	Example: Create a timeout timer that times out after 10 seconds of
1033	inactivity.	1318	inactivity.
1034		1319
1035	static void	1320	static void
1036	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1321	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1037	{	1322	{
1038	.. ten seconds without any activity	1323	.. ten seconds without any activity
1039	}	1324	}
1040		1325
1041	struct ev_timer mytimer;	1326	struct ev_timer mytimer;
1042	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	1327	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1043	ev_timer_again (&mytimer); /* start timer */	1328	ev_timer_again (&mytimer); /* start timer */
1044	ev_loop (loop, 0);	1329	ev_loop (loop, 0);
1045		1330
1046	// and in some piece of code that gets executed on any "activity":	1331	// and in some piece of code that gets executed on any "activity":
1047	// reset the timeout to start ticking again at 10 seconds	1332	// reset the timeout to start ticking again at 10 seconds
1048	ev_timer_again (&mytimer);	1333	ev_timer_again (&mytimer);
1049		1334
1050		1335
1051	=head2 C<ev_periodic> - to cron or not to cron?	1336	=head2 C<ev_periodic> - to cron or not to cron?
1052		1337
1053	Periodic watchers are also timers of a kind, but they are very versatile	1338	Periodic watchers are also timers of a kind, but they are very versatile
1054	(and unfortunately a bit complex).	1339	(and unfortunately a bit complex).
1055		1340
1056	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1341	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
1057	but on wallclock time (absolute time). You can tell a periodic watcher	1342	but on wall clock time (absolute time). You can tell a periodic watcher
1058	to trigger "at" some specific point in time. For example, if you tell a	1343	to trigger after some specific point in time. For example, if you tell a
1059	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	1344	periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()
1060	+ 10.>) and then reset your system clock to the last year, then it will	1345	+ 10.>, that is, an absolute time not a delay) and then reset your system
		1346	clock to January of the previous year, then it will take more than year
1061	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1347	to trigger the event (unlike an C<ev_timer>, which would still trigger
1062	roughly 10 seconds later and of course not if you reset your system time	1348	roughly 10 seconds later as it uses a relative timeout).
1063	again).
1064		1349
1065	They can also be used to implement vastly more complex timers, such as	1350	C<ev_periodic>s can also be used to implement vastly more complex timers,
1066	triggering an event on eahc midnight, local time.	1351	such as triggering an event on each "midnight, local time", or other
		1352	complicated, rules.
1067		1353
1068	As with timers, the callback is guarenteed to be invoked only when the	1354	As with timers, the callback is guaranteed to be invoked only when the
1069	time (C<at>) has been passed, but if multiple periodic timers become ready	1355	time (C<at>) has passed, but if multiple periodic timers become ready
1070	during the same loop iteration then order of execution is undefined.	1356	during the same loop iteration then order of execution is undefined.
		1357
		1358	=head3 Watcher-Specific Functions and Data Members
1071		1359
1072	=over 4	1360	=over 4
1073		1361
1074	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1362	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
1075		1363
…		…
1078	Lots of arguments, lets sort it out... There are basically three modes of	1366	Lots of arguments, lets sort it out... There are basically three modes of
1079	operation, and we will explain them from simplest to complex:	1367	operation, and we will explain them from simplest to complex:
1080		1368
1081	=over 4	1369	=over 4
1082		1370
1083	=item * absolute timer (interval = reschedule_cb = 0)	1371	=item * absolute timer (at = time, interval = reschedule_cb = 0)
1084		1372
1085	In this configuration the watcher triggers an event at the wallclock time	1373	In this configuration the watcher triggers an event after the wall clock
1086	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1374	time C<at> has passed and doesn't repeat. It will not adjust when a time
1087	that is, if it is to be run at January 1st 2011 then it will run when the	1375	jump occurs, that is, if it is to be run at January 1st 2011 then it will
1088	system time reaches or surpasses this time.	1376	run when the system time reaches or surpasses this time.
1089		1377
1090	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1378	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
1091		1379
1092	In this mode the watcher will always be scheduled to time out at the next	1380	In this mode the watcher will always be scheduled to time out at the next
1093	C<at + N * interval> time (for some integer N) and then repeat, regardless	1381	C<at + N * interval> time (for some integer N, which can also be negative)
1094	of any time jumps.	1382	and then repeat, regardless of any time jumps.
1095		1383
1096	This can be used to create timers that do not drift with respect to system	1384	This can be used to create timers that do not drift with respect to system
1097	time:	1385	time, for example, here is a C<ev_periodic> that triggers each hour, on
		1386	the hour:
1098		1387
1099	ev_periodic_set (&periodic, 0., 3600., 0);	1388	ev_periodic_set (&periodic, 0., 3600., 0);
1100		1389
1101	This doesn't mean there will always be 3600 seconds in between triggers,	1390	This doesn't mean there will always be 3600 seconds in between triggers,
1102	but only that the the callback will be called when the system time shows a	1391	but only that the callback will be called when the system time shows a
1103	full hour (UTC), or more correctly, when the system time is evenly divisible	1392	full hour (UTC), or more correctly, when the system time is evenly divisible
1104	by 3600.	1393	by 3600.
1105		1394
1106	Another way to think about it (for the mathematically inclined) is that	1395	Another way to think about it (for the mathematically inclined) is that
1107	C<ev_periodic> will try to run the callback in this mode at the next possible	1396	C<ev_periodic> will try to run the callback in this mode at the next possible
1108	time where C<time = at (mod interval)>, regardless of any time jumps.	1397	time where C<time = at (mod interval)>, regardless of any time jumps.
1109		1398
		1399	For numerical stability it is preferable that the C<at> value is near
		1400	C<ev_now ()> (the current time), but there is no range requirement for
		1401	this value, and in fact is often specified as zero.
		1402
		1403	Note also that there is an upper limit to how often a timer can fire (CPU
		1404	speed for example), so if C<interval> is very small then timing stability
		1405	will of course deteriorate. Libev itself tries to be exact to be about one
		1406	millisecond (if the OS supports it and the machine is fast enough).
		1407
1110	=item * manual reschedule mode (reschedule_cb = callback)	1408	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
1111		1409
1112	In this mode the values for C<interval> and C<at> are both being	1410	In this mode the values for C<interval> and C<at> are both being
1113	ignored. Instead, each time the periodic watcher gets scheduled, the	1411	ignored. Instead, each time the periodic watcher gets scheduled, the
1114	reschedule callback will be called with the watcher as first, and the	1412	reschedule callback will be called with the watcher as first, and the
1115	current time as second argument.	1413	current time as second argument.
1116		1414
1117	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1415	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
1118	ever, or make any event loop modifications>. If you need to stop it,	1416	ever, or make ANY event loop modifications whatsoever>.
1119	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
1120	starting a prepare watcher).
1121		1417
		1418	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1419	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1420	only event loop modification you are allowed to do).
		1421
1122	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1422	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic
1123	ev_tstamp now)>, e.g.:	1423	*w, ev_tstamp now)>, e.g.:
1124		1424
1125	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1425	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
1126	{	1426	{
1127	return now + 60.;	1427	return now + 60.;
1128	}	1428	}
…		…
1130	It must return the next time to trigger, based on the passed time value	1430	It must return the next time to trigger, based on the passed time value
1131	(that is, the lowest time value larger than to the second argument). It	1431	(that is, the lowest time value larger than to the second argument). It
1132	will usually be called just before the callback will be triggered, but	1432	will usually be called just before the callback will be triggered, but
1133	might be called at other times, too.	1433	might be called at other times, too.
1134		1434
1135	NOTE: I<< This callback must always return a time that is later than the	1435	NOTE: I<< This callback must always return a time that is higher than or
1136	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	1436	equal to the passed C<now> value >>.
1137		1437
1138	This can be used to create very complex timers, such as a timer that	1438	This can be used to create very complex timers, such as a timer that
1139	triggers on each midnight, local time. To do this, you would calculate the	1439	triggers on "next midnight, local time". To do this, you would calculate the
1140	next midnight after C<now> and return the timestamp value for this. How	1440	next midnight after C<now> and return the timestamp value for this. How
1141	you do this is, again, up to you (but it is not trivial, which is the main	1441	you do this is, again, up to you (but it is not trivial, which is the main
1142	reason I omitted it as an example).	1442	reason I omitted it as an example).
1143		1443
1144	=back	1444	=back
…		…
1148	Simply stops and restarts the periodic watcher again. This is only useful	1448	Simply stops and restarts the periodic watcher again. This is only useful
1149	when you changed some parameters or the reschedule callback would return	1449	when you changed some parameters or the reschedule callback would return
1150	a different time than the last time it was called (e.g. in a crond like	1450	a different time than the last time it was called (e.g. in a crond like
1151	program when the crontabs have changed).	1451	program when the crontabs have changed).
1152		1452
		1453	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1454
		1455	When active, returns the absolute time that the watcher is supposed to
		1456	trigger next.
		1457
		1458	=item ev_tstamp offset [read-write]
		1459
		1460	When repeating, this contains the offset value, otherwise this is the
		1461	absolute point in time (the C<at> value passed to C<ev_periodic_set>).
		1462
		1463	Can be modified any time, but changes only take effect when the periodic
		1464	timer fires or C<ev_periodic_again> is being called.
		1465
1153	=item ev_tstamp interval [read-write]	1466	=item ev_tstamp interval [read-write]
1154		1467
1155	The current interval value. Can be modified any time, but changes only	1468	The current interval value. Can be modified any time, but changes only
1156	take effect when the periodic timer fires or C<ev_periodic_again> is being	1469	take effect when the periodic timer fires or C<ev_periodic_again> is being
1157	called.	1470	called.
…		…
1162	switched off. Can be changed any time, but changes only take effect when	1475	switched off. Can be changed any time, but changes only take effect when
1163	the periodic timer fires or C<ev_periodic_again> is being called.	1476	the periodic timer fires or C<ev_periodic_again> is being called.
1164		1477
1165	=back	1478	=back
1166		1479
		1480	=head3 Examples
		1481
1167	Example: Call a callback every hour, or, more precisely, whenever the	1482	Example: Call a callback every hour, or, more precisely, whenever the
1168	system clock is divisible by 3600. The callback invocation times have	1483	system clock is divisible by 3600. The callback invocation times have
1169	potentially a lot of jittering, but good long-term stability.	1484	potentially a lot of jitter, but good long-term stability.
1170		1485
1171	static void	1486	static void
1172	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1487	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1173	{	1488	{
1174	... its now a full hour (UTC, or TAI or whatever your clock follows)	1489	... its now a full hour (UTC, or TAI or whatever your clock follows)
1175	}	1490	}
1176		1491
1177	struct ev_periodic hourly_tick;	1492	struct ev_periodic hourly_tick;
1178	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1493	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1179	ev_periodic_start (loop, &hourly_tick);	1494	ev_periodic_start (loop, &hourly_tick);
1180		1495
1181	Example: The same as above, but use a reschedule callback to do it:	1496	Example: The same as above, but use a reschedule callback to do it:
1182		1497
1183	#include <math.h>	1498	#include <math.h>
1184		1499
1185	static ev_tstamp	1500	static ev_tstamp
1186	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1501	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1187	{	1502	{
1188	return fmod (now, 3600.) + 3600.;	1503	return fmod (now, 3600.) + 3600.;
1189	}	1504	}
1190		1505
1191	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1506	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1192		1507
1193	Example: Call a callback every hour, starting now:	1508	Example: Call a callback every hour, starting now:
1194		1509
1195	struct ev_periodic hourly_tick;	1510	struct ev_periodic hourly_tick;
1196	ev_periodic_init (&hourly_tick, clock_cb,	1511	ev_periodic_init (&hourly_tick, clock_cb,
1197	fmod (ev_now (loop), 3600.), 3600., 0);	1512	fmod (ev_now (loop), 3600.), 3600., 0);
1198	ev_periodic_start (loop, &hourly_tick);	1513	ev_periodic_start (loop, &hourly_tick);
1199		1514
1200		1515
1201	=head2 C<ev_signal> - signal me when a signal gets signalled!	1516	=head2 C<ev_signal> - signal me when a signal gets signalled!
1202		1517
1203	Signal watchers will trigger an event when the process receives a specific	1518	Signal watchers will trigger an event when the process receives a specific
…		…
1210	with the kernel (thus it coexists with your own signal handlers as long	1525	with the kernel (thus it coexists with your own signal handlers as long
1211	as you don't register any with libev). Similarly, when the last signal	1526	as you don't register any with libev). Similarly, when the last signal
1212	watcher for a signal is stopped libev will reset the signal handler to	1527	watcher for a signal is stopped libev will reset the signal handler to
1213	SIG_DFL (regardless of what it was set to before).	1528	SIG_DFL (regardless of what it was set to before).
1214		1529
		1530	If possible and supported, libev will install its handlers with
		1531	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
		1532	interrupted. If you have a problem with system calls getting interrupted by
		1533	signals you can block all signals in an C<ev_check> watcher and unblock
		1534	them in an C<ev_prepare> watcher.
		1535
		1536	=head3 Watcher-Specific Functions and Data Members
		1537
1215	=over 4	1538	=over 4
1216		1539
1217	=item ev_signal_init (ev_signal *, callback, int signum)	1540	=item ev_signal_init (ev_signal *, callback, int signum)
1218		1541
1219	=item ev_signal_set (ev_signal *, int signum)	1542	=item ev_signal_set (ev_signal *, int signum)
…		…
1225		1548
1226	The signal the watcher watches out for.	1549	The signal the watcher watches out for.
1227		1550
1228	=back	1551	=back
1229		1552
		1553	=head3 Examples
		1554
		1555	Example: Try to exit cleanly on SIGINT and SIGTERM.
		1556
		1557	static void
		1558	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
		1559	{
		1560	ev_unloop (loop, EVUNLOOP_ALL);
		1561	}
		1562
		1563	struct ev_signal signal_watcher;
		1564	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1565	ev_signal_start (loop, &sigint_cb);
		1566
1230		1567
1231	=head2 C<ev_child> - watch out for process status changes	1568	=head2 C<ev_child> - watch out for process status changes
1232		1569
1233	Child watchers trigger when your process receives a SIGCHLD in response to	1570	Child watchers trigger when your process receives a SIGCHLD in response to
1234	some child status changes (most typically when a child of yours dies).	1571	some child status changes (most typically when a child of yours dies). It
		1572	is permissible to install a child watcher I<after> the child has been
		1573	forked (which implies it might have already exited), as long as the event
		1574	loop isn't entered (or is continued from a watcher).
		1575
		1576	Only the default event loop is capable of handling signals, and therefore
		1577	you can only register child watchers in the default event loop.
		1578
		1579	=head3 Process Interaction
		1580
		1581	Libev grabs C<SIGCHLD> as soon as the default event loop is
		1582	initialised. This is necessary to guarantee proper behaviour even if
		1583	the first child watcher is started after the child exits. The occurrence
		1584	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		1585	synchronously as part of the event loop processing. Libev always reaps all
		1586	children, even ones not watched.
		1587
		1588	=head3 Overriding the Built-In Processing
		1589
		1590	Libev offers no special support for overriding the built-in child
		1591	processing, but if your application collides with libev's default child
		1592	handler, you can override it easily by installing your own handler for
		1593	C<SIGCHLD> after initialising the default loop, and making sure the
		1594	default loop never gets destroyed. You are encouraged, however, to use an
		1595	event-based approach to child reaping and thus use libev's support for
		1596	that, so other libev users can use C<ev_child> watchers freely.
		1597
		1598	=head3 Stopping the Child Watcher
		1599
		1600	Currently, the child watcher never gets stopped, even when the
		1601	child terminates, so normally one needs to stop the watcher in the
		1602	callback. Future versions of libev might stop the watcher automatically
		1603	when a child exit is detected.
		1604
		1605	=head3 Watcher-Specific Functions and Data Members
1235		1606
1236	=over 4	1607	=over 4
1237		1608
1238	=item ev_child_init (ev_child *, callback, int pid)	1609	=item ev_child_init (ev_child *, callback, int pid, int trace)
1239		1610
1240	=item ev_child_set (ev_child *, int pid)	1611	=item ev_child_set (ev_child *, int pid, int trace)
1241		1612
1242	Configures the watcher to wait for status changes of process C<pid> (or	1613	Configures the watcher to wait for status changes of process C<pid> (or
1243	I<any> process if C<pid> is specified as C<0>). The callback can look	1614	I<any> process if C<pid> is specified as C<0>). The callback can look
1244	at the C<rstatus> member of the C<ev_child> watcher structure to see	1615	at the C<rstatus> member of the C<ev_child> watcher structure to see
1245	the status word (use the macros from C<sys/wait.h> and see your systems	1616	the status word (use the macros from C<sys/wait.h> and see your systems
1246	C<waitpid> documentation). The C<rpid> member contains the pid of the	1617	C<waitpid> documentation). The C<rpid> member contains the pid of the
1247	process causing the status change.	1618	process causing the status change. C<trace> must be either C<0> (only
		1619	activate the watcher when the process terminates) or C<1> (additionally
		1620	activate the watcher when the process is stopped or continued).
1248		1621
1249	=item int pid [read-only]	1622	=item int pid [read-only]
1250		1623
1251	The process id this watcher watches out for, or C<0>, meaning any process id.	1624	The process id this watcher watches out for, or C<0>, meaning any process id.
1252		1625
…		…
1259	The process exit/trace status caused by C<rpid> (see your systems	1632	The process exit/trace status caused by C<rpid> (see your systems
1260	C<waitpid> and C<sys/wait.h> documentation for details).	1633	C<waitpid> and C<sys/wait.h> documentation for details).
1261		1634
1262	=back	1635	=back
1263		1636
1264	Example: Try to exit cleanly on SIGINT and SIGTERM.	1637	=head3 Examples
1265		1638
		1639	Example: C<fork()> a new process and install a child handler to wait for
		1640	its completion.
		1641
		1642	ev_child cw;
		1643
1266	static void	1644	static void
1267	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1645	child_cb (EV_P_ struct ev_child *w, int revents)
1268	{	1646	{
1269	ev_unloop (loop, EVUNLOOP_ALL);	1647	ev_child_stop (EV_A_ w);
		1648	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1270	}	1649	}
1271		1650
1272	struct ev_signal signal_watcher;	1651	pid_t pid = fork ();
1273	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1652
1274	ev_signal_start (loop, &sigint_cb);	1653	if (pid < 0)
		1654	// error
		1655	else if (pid == 0)
		1656	{
		1657	// the forked child executes here
		1658	exit (1);
		1659	}
		1660	else
		1661	{
		1662	ev_child_init (&cw, child_cb, pid, 0);
		1663	ev_child_start (EV_DEFAULT_ &cw);
		1664	}
1275		1665
1276		1666
1277	=head2 C<ev_stat> - did the file attributes just change?	1667	=head2 C<ev_stat> - did the file attributes just change?
1278		1668
1279	This watches a filesystem path for attribute changes. That is, it calls	1669	This watches a file system path for attribute changes. That is, it calls
1280	C<stat> regularly (or when the OS says it changed) and sees if it changed	1670	C<stat> regularly (or when the OS says it changed) and sees if it changed
1281	compared to the last time, invoking the callback if it did.	1671	compared to the last time, invoking the callback if it did.
1282		1672
1283	The path does not need to exist: changing from "path exists" to "path does	1673	The path does not need to exist: changing from "path exists" to "path does
1284	not exist" is a status change like any other. The condition "path does	1674	not exist" is a status change like any other. The condition "path does
…		…
1302	as even with OS-supported change notifications, this can be	1692	as even with OS-supported change notifications, this can be
1303	resource-intensive.	1693	resource-intensive.
1304		1694
1305	At the time of this writing, only the Linux inotify interface is	1695	At the time of this writing, only the Linux inotify interface is
1306	implemented (implementing kqueue support is left as an exercise for the	1696	implemented (implementing kqueue support is left as an exercise for the
		1697	reader, note, however, that the author sees no way of implementing ev_stat
1307	reader). Inotify will be used to give hints only and should not change the	1698	semantics with kqueue). Inotify will be used to give hints only and should
1308	semantics of C<ev_stat> watchers, which means that libev sometimes needs	1699	not change the semantics of C<ev_stat> watchers, which means that libev
1309	to fall back to regular polling again even with inotify, but changes are	1700	sometimes needs to fall back to regular polling again even with inotify,
1310	usually detected immediately, and if the file exists there will be no	1701	but changes are usually detected immediately, and if the file exists there
1311	polling.	1702	will be no polling.
		1703
		1704	=head3 ABI Issues (Largefile Support)
		1705
		1706	Libev by default (unless the user overrides this) uses the default
		1707	compilation environment, which means that on systems with large file
		1708	support disabled by default, you get the 32 bit version of the stat
		1709	structure. When using the library from programs that change the ABI to
		1710	use 64 bit file offsets the programs will fail. In that case you have to
		1711	compile libev with the same flags to get binary compatibility. This is
		1712	obviously the case with any flags that change the ABI, but the problem is
		1713	most noticeably disabled with ev_stat and large file support.
		1714
		1715	The solution for this is to lobby your distribution maker to make large
		1716	file interfaces available by default (as e.g. FreeBSD does) and not
		1717	optional. Libev cannot simply switch on large file support because it has
		1718	to exchange stat structures with application programs compiled using the
		1719	default compilation environment.
		1720
		1721	=head3 Inotify
		1722
		1723	When C<inotify (7)> support has been compiled into libev (generally only
		1724	available on Linux) and present at runtime, it will be used to speed up
		1725	change detection where possible. The inotify descriptor will be created lazily
		1726	when the first C<ev_stat> watcher is being started.
		1727
		1728	Inotify presence does not change the semantics of C<ev_stat> watchers
		1729	except that changes might be detected earlier, and in some cases, to avoid
		1730	making regular C<stat> calls. Even in the presence of inotify support
		1731	there are many cases where libev has to resort to regular C<stat> polling.
		1732
		1733	(There is no support for kqueue, as apparently it cannot be used to
		1734	implement this functionality, due to the requirement of having a file
		1735	descriptor open on the object at all times).
		1736
		1737	=head3 The special problem of stat time resolution
		1738
		1739	The C<stat ()> system call only supports full-second resolution portably, and
		1740	even on systems where the resolution is higher, many file systems still
		1741	only support whole seconds.
		1742
		1743	That means that, if the time is the only thing that changes, you can
		1744	easily miss updates: on the first update, C<ev_stat> detects a change and
		1745	calls your callback, which does something. When there is another update
		1746	within the same second, C<ev_stat> will be unable to detect it as the stat
		1747	data does not change.
		1748
		1749	The solution to this is to delay acting on a change for slightly more
		1750	than a second (or till slightly after the next full second boundary), using
		1751	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		1752	ev_timer_again (loop, w)>).
		1753
		1754	The C<.02> offset is added to work around small timing inconsistencies
		1755	of some operating systems (where the second counter of the current time
		1756	might be be delayed. One such system is the Linux kernel, where a call to
		1757	C<gettimeofday> might return a timestamp with a full second later than
		1758	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		1759	update file times then there will be a small window where the kernel uses
		1760	the previous second to update file times but libev might already execute
		1761	the timer callback).
		1762
		1763	=head3 Watcher-Specific Functions and Data Members
1312		1764
1313	=over 4	1765	=over 4
1314		1766
1315	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)	1767	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
1316		1768
…		…
1320	C<path>. The C<interval> is a hint on how quickly a change is expected to	1772	C<path>. The C<interval> is a hint on how quickly a change is expected to
1321	be detected and should normally be specified as C<0> to let libev choose	1773	be detected and should normally be specified as C<0> to let libev choose
1322	a suitable value. The memory pointed to by C<path> must point to the same	1774	a suitable value. The memory pointed to by C<path> must point to the same
1323	path for as long as the watcher is active.	1775	path for as long as the watcher is active.
1324		1776
1325	The callback will be receive C<EV_STAT> when a change was detected,	1777	The callback will receive C<EV_STAT> when a change was detected, relative
1326	relative to the attributes at the time the watcher was started (or the	1778	to the attributes at the time the watcher was started (or the last change
1327	last change was detected).	1779	was detected).
1328		1780
1329	=item ev_stat_stat (ev_stat *)	1781	=item ev_stat_stat (loop, ev_stat *)
1330		1782
1331	Updates the stat buffer immediately with new values. If you change the	1783	Updates the stat buffer immediately with new values. If you change the
1332	watched path in your callback, you could call this fucntion to avoid	1784	watched path in your callback, you could call this function to avoid
1333	detecting this change (while introducing a race condition). Can also be	1785	detecting this change (while introducing a race condition if you are not
1334	useful simply to find out the new values.	1786	the only one changing the path). Can also be useful simply to find out the
		1787	new values.
1335		1788
1336	=item ev_statdata attr [read-only]	1789	=item ev_statdata attr [read-only]
1337		1790
1338	The most-recently detected attributes of the file. Although the type is of	1791	The most-recently detected attributes of the file. Although the type is
1339	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	1792	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1340	suitable for your system. If the C<st_nlink> member is C<0>, then there	1793	suitable for your system, but you can only rely on the POSIX-standardised
		1794	members to be present. If the C<st_nlink> member is C<0>, then there was
1341	was some error while C<stat>ing the file.	1795	some error while C<stat>ing the file.
1342		1796
1343	=item ev_statdata prev [read-only]	1797	=item ev_statdata prev [read-only]
1344		1798
1345	The previous attributes of the file. The callback gets invoked whenever	1799	The previous attributes of the file. The callback gets invoked whenever
1346	C<prev> != C<attr>.	1800	C<prev> != C<attr>, or, more precisely, one or more of these members
		1801	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		1802	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1347		1803
1348	=item ev_tstamp interval [read-only]	1804	=item ev_tstamp interval [read-only]
1349		1805
1350	The specified interval.	1806	The specified interval.
1351		1807
1352	=item const char *path [read-only]	1808	=item const char *path [read-only]
1353		1809
1354	The filesystem path that is being watched.	1810	The file system path that is being watched.
1355		1811
1356	=back	1812	=back
1357		1813
		1814	=head3 Examples
		1815
1358	Example: Watch C</etc/passwd> for attribute changes.	1816	Example: Watch C</etc/passwd> for attribute changes.
1359		1817
1360	static void	1818	static void
1361	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	1819	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1362	{	1820	{
1363	/* /etc/passwd changed in some way */	1821	/* /etc/passwd changed in some way */
1364	if (w->attr.st_nlink)	1822	if (w->attr.st_nlink)
1365	{	1823	{
1366	printf ("passwd current size %ld\n", (long)w->attr.st_size);	1824	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1367	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	1825	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1368	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	1826	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1369	}	1827	}
1370	else	1828	else
1371	/* you shalt not abuse printf for puts */	1829	/* you shalt not abuse printf for puts */
1372	puts ("wow, /etc/passwd is not there, expect problems. "	1830	puts ("wow, /etc/passwd is not there, expect problems. "
1373	"if this is windows, they already arrived\n");	1831	"if this is windows, they already arrived\n");
1374	}	1832	}
1375		1833
1376	...	1834	...
1377	ev_stat passwd;	1835	ev_stat passwd;
1378		1836
1379	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");	1837	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1380	ev_stat_start (loop, &passwd);	1838	ev_stat_start (loop, &passwd);
		1839
		1840	Example: Like above, but additionally use a one-second delay so we do not
		1841	miss updates (however, frequent updates will delay processing, too, so
		1842	one might do the work both on C<ev_stat> callback invocation I<and> on
		1843	C<ev_timer> callback invocation).
		1844
		1845	static ev_stat passwd;
		1846	static ev_timer timer;
		1847
		1848	static void
		1849	timer_cb (EV_P_ ev_timer *w, int revents)
		1850	{
		1851	ev_timer_stop (EV_A_ w);
		1852
		1853	/* now it's one second after the most recent passwd change */
		1854	}
		1855
		1856	static void
		1857	stat_cb (EV_P_ ev_stat *w, int revents)
		1858	{
		1859	/* reset the one-second timer */
		1860	ev_timer_again (EV_A_ &timer);
		1861	}
		1862
		1863	...
		1864	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		1865	ev_stat_start (loop, &passwd);
		1866	ev_timer_init (&timer, timer_cb, 0., 1.02);
1381		1867
1382		1868
1383	=head2 C<ev_idle> - when you've got nothing better to do...	1869	=head2 C<ev_idle> - when you've got nothing better to do...
1384		1870
1385	Idle watchers trigger events when no other events of the same or higher	1871	Idle watchers trigger events when no other events of the same or higher
…		…
1399	Apart from keeping your process non-blocking (which is a useful	1885	Apart from keeping your process non-blocking (which is a useful
1400	effect on its own sometimes), idle watchers are a good place to do	1886	effect on its own sometimes), idle watchers are a good place to do
1401	"pseudo-background processing", or delay processing stuff to after the	1887	"pseudo-background processing", or delay processing stuff to after the
1402	event loop has handled all outstanding events.	1888	event loop has handled all outstanding events.
1403		1889
		1890	=head3 Watcher-Specific Functions and Data Members
		1891
1404	=over 4	1892	=over 4
1405		1893
1406	=item ev_idle_init (ev_signal *, callback)	1894	=item ev_idle_init (ev_signal *, callback)
1407		1895
1408	Initialises and configures the idle watcher - it has no parameters of any	1896	Initialises and configures the idle watcher - it has no parameters of any
1409	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1897	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1410	believe me.	1898	believe me.
1411		1899
1412	=back	1900	=back
1413		1901
		1902	=head3 Examples
		1903
1414	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	1904	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1415	callback, free it. Also, use no error checking, as usual.	1905	callback, free it. Also, use no error checking, as usual.
1416		1906
1417	static void	1907	static void
1418	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	1908	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1419	{	1909	{
1420	free (w);	1910	free (w);
1421	// now do something you wanted to do when the program has	1911	// now do something you wanted to do when the program has
1422	// no longer asnything immediate to do.	1912	// no longer anything immediate to do.
1423	}	1913	}
1424		1914
1425	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1915	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1426	ev_idle_init (idle_watcher, idle_cb);	1916	ev_idle_init (idle_watcher, idle_cb);
1427	ev_idle_start (loop, idle_cb);	1917	ev_idle_start (loop, idle_cb);
1428		1918
1429		1919
1430	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	1920	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1431		1921
1432	Prepare and check watchers are usually (but not always) used in tandem:	1922	Prepare and check watchers are usually (but not always) used in tandem:
…		…
1451		1941
1452	This is done by examining in each prepare call which file descriptors need	1942	This is done by examining in each prepare call which file descriptors need
1453	to be watched by the other library, registering C<ev_io> watchers for	1943	to be watched by the other library, registering C<ev_io> watchers for
1454	them and starting an C<ev_timer> watcher for any timeouts (many libraries	1944	them and starting an C<ev_timer> watcher for any timeouts (many libraries
1455	provide just this functionality). Then, in the check watcher you check for	1945	provide just this functionality). Then, in the check watcher you check for
1456	any events that occured (by checking the pending status of all watchers	1946	any events that occurred (by checking the pending status of all watchers
1457	and stopping them) and call back into the library. The I/O and timer	1947	and stopping them) and call back into the library. The I/O and timer
1458	callbacks will never actually be called (but must be valid nevertheless,	1948	callbacks will never actually be called (but must be valid nevertheless,
1459	because you never know, you know?).	1949	because you never know, you know?).
1460		1950
1461	As another example, the Perl Coro module uses these hooks to integrate	1951	As another example, the Perl Coro module uses these hooks to integrate
…		…
1465	with priority higher than or equal to the event loop and one coroutine	1955	with priority higher than or equal to the event loop and one coroutine
1466	of lower priority, but only once, using idle watchers to keep the event	1956	of lower priority, but only once, using idle watchers to keep the event
1467	loop from blocking if lower-priority coroutines are active, thus mapping	1957	loop from blocking if lower-priority coroutines are active, thus mapping
1468	low-priority coroutines to idle/background tasks).	1958	low-priority coroutines to idle/background tasks).
1469		1959
		1960	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		1961	priority, to ensure that they are being run before any other watchers
		1962	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,
		1963	too) should not activate ("feed") events into libev. While libev fully
		1964	supports this, they might get executed before other C<ev_check> watchers
		1965	did their job. As C<ev_check> watchers are often used to embed other
		1966	(non-libev) event loops those other event loops might be in an unusable
		1967	state until their C<ev_check> watcher ran (always remind yourself to
		1968	coexist peacefully with others).
		1969
		1970	=head3 Watcher-Specific Functions and Data Members
		1971
1470	=over 4	1972	=over 4
1471		1973
1472	=item ev_prepare_init (ev_prepare *, callback)	1974	=item ev_prepare_init (ev_prepare *, callback)
1473		1975
1474	=item ev_check_init (ev_check *, callback)	1976	=item ev_check_init (ev_check *, callback)
…		…
1477	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1979	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1478	macros, but using them is utterly, utterly and completely pointless.	1980	macros, but using them is utterly, utterly and completely pointless.
1479		1981
1480	=back	1982	=back
1481		1983
1482	Example: To include a library such as adns, you would add IO watchers	1984	=head3 Examples
1483	and a timeout watcher in a prepare handler, as required by libadns, and	1985
		1986	There are a number of principal ways to embed other event loops or modules
		1987	into libev. Here are some ideas on how to include libadns into libev
		1988	(there is a Perl module named C<EV::ADNS> that does this, which you could
		1989	use as a working example. Another Perl module named C<EV::Glib> embeds a
		1990	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		1991	Glib event loop).
		1992
		1993	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1484	in a check watcher, destroy them and call into libadns. What follows is	1994	and in a check watcher, destroy them and call into libadns. What follows
1485	pseudo-code only of course:	1995	is pseudo-code only of course. This requires you to either use a low
		1996	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		1997	the callbacks for the IO/timeout watchers might not have been called yet.
1486		1998
1487	static ev_io iow [nfd];	1999	static ev_io iow [nfd];
1488	static ev_timer tw;	2000	static ev_timer tw;
1489		2001
1490	static void	2002	static void
1491	io_cb (ev_loop *loop, ev_io *w, int revents)	2003	io_cb (ev_loop *loop, ev_io *w, int revents)
1492	{	2004	{
1493	// set the relevant poll flags
1494	// could also call adns_processreadable etc. here
1495	struct pollfd *fd = (struct pollfd *)w->data;
1496	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1497	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1498	}	2005	}
1499		2006
1500	// create io watchers for each fd and a timer before blocking	2007	// create io watchers for each fd and a timer before blocking
1501	static void	2008	static void
1502	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2009	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1503	{	2010	{
1504	int timeout = 3600000;	2011	int timeout = 3600000;
1505	struct pollfd fds [nfd];	2012	struct pollfd fds [nfd];
1506	// actual code will need to loop here and realloc etc.	2013	// actual code will need to loop here and realloc etc.
1507	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2014	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1508		2015
1509	/* the callback is illegal, but won't be called as we stop during check */	2016	/* the callback is illegal, but won't be called as we stop during check */
1510	ev_timer_init (&tw, 0, timeout * 1e-3);	2017	ev_timer_init (&tw, 0, timeout * 1e-3);
1511	ev_timer_start (loop, &tw);	2018	ev_timer_start (loop, &tw);
1512		2019
1513	// create on ev_io per pollfd	2020	// create one ev_io per pollfd
1514	for (int i = 0; i < nfd; ++i)	2021	for (int i = 0; i < nfd; ++i)
1515	{	2022	{
1516	ev_io_init (iow + i, io_cb, fds [i].fd,	2023	ev_io_init (iow + i, io_cb, fds [i].fd,
1517	((fds [i].events & POLLIN ? EV_READ : 0)	2024	((fds [i].events & POLLIN ? EV_READ : 0)
1518	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	2025	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1519		2026
1520	fds [i].revents = 0;	2027	fds [i].revents = 0;
1521	iow [i].data = fds + i;
1522	ev_io_start (loop, iow + i);	2028	ev_io_start (loop, iow + i);
1523	}	2029	}
1524	}	2030	}
1525		2031
1526	// stop all watchers after blocking	2032	// stop all watchers after blocking
1527	static void	2033	static void
1528	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2034	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1529	{	2035	{
1530	ev_timer_stop (loop, &tw);	2036	ev_timer_stop (loop, &tw);
1531		2037
1532	for (int i = 0; i < nfd; ++i)	2038	for (int i = 0; i < nfd; ++i)
		2039	{
		2040	// set the relevant poll flags
		2041	// could also call adns_processreadable etc. here
		2042	struct pollfd *fd = fds + i;
		2043	int revents = ev_clear_pending (iow + i);
		2044	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		2045	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		2046
		2047	// now stop the watcher
1533	ev_io_stop (loop, iow + i);	2048	ev_io_stop (loop, iow + i);
		2049	}
1534		2050
1535	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	2051	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1536	}	2052	}
		2053
		2054	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		2055	in the prepare watcher and would dispose of the check watcher.
		2056
		2057	Method 3: If the module to be embedded supports explicit event
		2058	notification (libadns does), you can also make use of the actual watcher
		2059	callbacks, and only destroy/create the watchers in the prepare watcher.
		2060
		2061	static void
		2062	timer_cb (EV_P_ ev_timer *w, int revents)
		2063	{
		2064	adns_state ads = (adns_state)w->data;
		2065	update_now (EV_A);
		2066
		2067	adns_processtimeouts (ads, &tv_now);
		2068	}
		2069
		2070	static void
		2071	io_cb (EV_P_ ev_io *w, int revents)
		2072	{
		2073	adns_state ads = (adns_state)w->data;
		2074	update_now (EV_A);
		2075
		2076	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		2077	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		2078	}
		2079
		2080	// do not ever call adns_afterpoll
		2081
		2082	Method 4: Do not use a prepare or check watcher because the module you
		2083	want to embed is too inflexible to support it. Instead, you can override
		2084	their poll function. The drawback with this solution is that the main
		2085	loop is now no longer controllable by EV. The C<Glib::EV> module does
		2086	this.
		2087
		2088	static gint
		2089	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		2090	{
		2091	int got_events = 0;
		2092
		2093	for (n = 0; n < nfds; ++n)
		2094	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		2095
		2096	if (timeout >= 0)
		2097	// create/start timer
		2098
		2099	// poll
		2100	ev_loop (EV_A_ 0);
		2101
		2102	// stop timer again
		2103	if (timeout >= 0)
		2104	ev_timer_stop (EV_A_ &to);
		2105
		2106	// stop io watchers again - their callbacks should have set
		2107	for (n = 0; n < nfds; ++n)
		2108	ev_io_stop (EV_A_ iow [n]);
		2109
		2110	return got_events;
		2111	}
1537		2112
1538		2113
1539	=head2 C<ev_embed> - when one backend isn't enough...	2114	=head2 C<ev_embed> - when one backend isn't enough...
1540		2115
1541	This is a rather advanced watcher type that lets you embed one event loop	2116	This is a rather advanced watcher type that lets you embed one event loop
…		…
1583	portable one.	2158	portable one.
1584		2159
1585	So when you want to use this feature you will always have to be prepared	2160	So when you want to use this feature you will always have to be prepared
1586	that you cannot get an embeddable loop. The recommended way to get around	2161	that you cannot get an embeddable loop. The recommended way to get around
1587	this is to have a separate variables for your embeddable loop, try to	2162	this is to have a separate variables for your embeddable loop, try to
1588	create it, and if that fails, use the normal loop for everything:	2163	create it, and if that fails, use the normal loop for everything.
1589		2164
1590	struct ev_loop *loop_hi = ev_default_init (0);	2165	=head3 Watcher-Specific Functions and Data Members
1591	struct ev_loop *loop_lo = 0;
1592	struct ev_embed embed;
1593
1594	// see if there is a chance of getting one that works
1595	// (remember that a flags value of 0 means autodetection)
1596	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1597	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1598	: 0;
1599
1600	// if we got one, then embed it, otherwise default to loop_hi
1601	if (loop_lo)
1602	{
1603	ev_embed_init (&embed, 0, loop_lo);
1604	ev_embed_start (loop_hi, &embed);
1605	}
1606	else
1607	loop_lo = loop_hi;
1608		2166
1609	=over 4	2167	=over 4
1610		2168
1611	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	2169	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
1612		2170
…		…
1614		2172
1615	Configures the watcher to embed the given loop, which must be	2173	Configures the watcher to embed the given loop, which must be
1616	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be	2174	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1617	invoked automatically, otherwise it is the responsibility of the callback	2175	invoked automatically, otherwise it is the responsibility of the callback
1618	to invoke it (it will continue to be called until the sweep has been done,	2176	to invoke it (it will continue to be called until the sweep has been done,
1619	if you do not want thta, you need to temporarily stop the embed watcher).	2177	if you do not want that, you need to temporarily stop the embed watcher).
1620		2178
1621	=item ev_embed_sweep (loop, ev_embed *)	2179	=item ev_embed_sweep (loop, ev_embed *)
1622		2180
1623	Make a single, non-blocking sweep over the embedded loop. This works	2181	Make a single, non-blocking sweep over the embedded loop. This works
1624	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	2182	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
1625	apropriate way for embedded loops.	2183	appropriate way for embedded loops.
1626		2184
1627	=item struct ev_loop *loop [read-only]	2185	=item struct ev_loop *other [read-only]
1628		2186
1629	The embedded event loop.	2187	The embedded event loop.
1630		2188
1631	=back	2189	=back
		2190
		2191	=head3 Examples
		2192
		2193	Example: Try to get an embeddable event loop and embed it into the default
		2194	event loop. If that is not possible, use the default loop. The default
		2195	loop is stored in C<loop_hi>, while the embeddable loop is stored in
		2196	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
		2197	used).
		2198
		2199	struct ev_loop *loop_hi = ev_default_init (0);
		2200	struct ev_loop *loop_lo = 0;
		2201	struct ev_embed embed;
		2202
		2203	// see if there is a chance of getting one that works
		2204	// (remember that a flags value of 0 means autodetection)
		2205	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		2206	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		2207	: 0;
		2208
		2209	// if we got one, then embed it, otherwise default to loop_hi
		2210	if (loop_lo)
		2211	{
		2212	ev_embed_init (&embed, 0, loop_lo);
		2213	ev_embed_start (loop_hi, &embed);
		2214	}
		2215	else
		2216	loop_lo = loop_hi;
		2217
		2218	Example: Check if kqueue is available but not recommended and create
		2219	a kqueue backend for use with sockets (which usually work with any
		2220	kqueue implementation). Store the kqueue/socket-only event loop in
		2221	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		2222
		2223	struct ev_loop *loop = ev_default_init (0);
		2224	struct ev_loop *loop_socket = 0;
		2225	struct ev_embed embed;
		2226
		2227	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2228	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2229	{
		2230	ev_embed_init (&embed, 0, loop_socket);
		2231	ev_embed_start (loop, &embed);
		2232	}
		2233
		2234	if (!loop_socket)
		2235	loop_socket = loop;
		2236
		2237	// now use loop_socket for all sockets, and loop for everything else
1632		2238
1633		2239
1634	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	2240	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
1635		2241
1636	Fork watchers are called when a C<fork ()> was detected (usually because	2242	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
1639	event loop blocks next and before C<ev_check> watchers are being called,	2245	event loop blocks next and before C<ev_check> watchers are being called,
1640	and only in the child after the fork. If whoever good citizen calling	2246	and only in the child after the fork. If whoever good citizen calling
1641	C<ev_default_fork> cheats and calls it in the wrong process, the fork	2247	C<ev_default_fork> cheats and calls it in the wrong process, the fork
1642	handlers will be invoked, too, of course.	2248	handlers will be invoked, too, of course.
1643		2249
		2250	=head3 Watcher-Specific Functions and Data Members
		2251
1644	=over 4	2252	=over 4
1645		2253
1646	=item ev_fork_init (ev_signal *, callback)	2254	=item ev_fork_init (ev_signal *, callback)
1647		2255
1648	Initialises and configures the fork watcher - it has no parameters of any	2256	Initialises and configures the fork watcher - it has no parameters of any
1649	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,	2257	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
1650	believe me.	2258	believe me.
		2259
		2260	=back
		2261
		2262
		2263	=head2 C<ev_async> - how to wake up another event loop
		2264
		2265	In general, you cannot use an C<ev_loop> from multiple threads or other
		2266	asynchronous sources such as signal handlers (as opposed to multiple event
		2267	loops - those are of course safe to use in different threads).
		2268
		2269	Sometimes, however, you need to wake up another event loop you do not
		2270	control, for example because it belongs to another thread. This is what
		2271	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2272	can signal it by calling C<ev_async_send>, which is thread- and signal
		2273	safe.
		2274
		2275	This functionality is very similar to C<ev_signal> watchers, as signals,
		2276	too, are asynchronous in nature, and signals, too, will be compressed
		2277	(i.e. the number of callback invocations may be less than the number of
		2278	C<ev_async_sent> calls).
		2279
		2280	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2281	just the default loop.
		2282
		2283	=head3 Queueing
		2284
		2285	C<ev_async> does not support queueing of data in any way. The reason
		2286	is that the author does not know of a simple (or any) algorithm for a
		2287	multiple-writer-single-reader queue that works in all cases and doesn't
		2288	need elaborate support such as pthreads.
		2289
		2290	That means that if you want to queue data, you have to provide your own
		2291	queue. But at least I can tell you would implement locking around your
		2292	queue:
		2293
		2294	=over 4
		2295
		2296	=item queueing from a signal handler context
		2297
		2298	To implement race-free queueing, you simply add to the queue in the signal
		2299	handler but you block the signal handler in the watcher callback. Here is an example that does that for
		2300	some fictitious SIGUSR1 handler:
		2301
		2302	static ev_async mysig;
		2303
		2304	static void
		2305	sigusr1_handler (void)
		2306	{
		2307	sometype data;
		2308
		2309	// no locking etc.
		2310	queue_put (data);
		2311	ev_async_send (EV_DEFAULT_ &mysig);
		2312	}
		2313
		2314	static void
		2315	mysig_cb (EV_P_ ev_async *w, int revents)
		2316	{
		2317	sometype data;
		2318	sigset_t block, prev;
		2319
		2320	sigemptyset (&block);
		2321	sigaddset (&block, SIGUSR1);
		2322	sigprocmask (SIG_BLOCK, &block, &prev);
		2323
		2324	while (queue_get (&data))
		2325	process (data);
		2326
		2327	if (sigismember (&prev, SIGUSR1)
		2328	sigprocmask (SIG_UNBLOCK, &block, 0);
		2329	}
		2330
		2331	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2332	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2333	either...).
		2334
		2335	=item queueing from a thread context
		2336
		2337	The strategy for threads is different, as you cannot (easily) block
		2338	threads but you can easily preempt them, so to queue safely you need to
		2339	employ a traditional mutex lock, such as in this pthread example:
		2340
		2341	static ev_async mysig;
		2342	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2343
		2344	static void
		2345	otherthread (void)
		2346	{
		2347	// only need to lock the actual queueing operation
		2348	pthread_mutex_lock (&mymutex);
		2349	queue_put (data);
		2350	pthread_mutex_unlock (&mymutex);
		2351
		2352	ev_async_send (EV_DEFAULT_ &mysig);
		2353	}
		2354
		2355	static void
		2356	mysig_cb (EV_P_ ev_async *w, int revents)
		2357	{
		2358	pthread_mutex_lock (&mymutex);
		2359
		2360	while (queue_get (&data))
		2361	process (data);
		2362
		2363	pthread_mutex_unlock (&mymutex);
		2364	}
		2365
		2366	=back
		2367
		2368
		2369	=head3 Watcher-Specific Functions and Data Members
		2370
		2371	=over 4
		2372
		2373	=item ev_async_init (ev_async *, callback)
		2374
		2375	Initialises and configures the async watcher - it has no parameters of any
		2376	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,
		2377	believe me.
		2378
		2379	=item ev_async_send (loop, ev_async *)
		2380
		2381	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2382	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2383	C<ev_feed_event>, this call is safe to do in other threads, signal or
		2384	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
		2385	section below on what exactly this means).
		2386
		2387	This call incurs the overhead of a system call only once per loop iteration,
		2388	so while the overhead might be noticeable, it doesn't apply to repeated
		2389	calls to C<ev_async_send>.
		2390
		2391	=item bool = ev_async_pending (ev_async *)
		2392
		2393	Returns a non-zero value when C<ev_async_send> has been called on the
		2394	watcher but the event has not yet been processed (or even noted) by the
		2395	event loop.
		2396
		2397	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2398	the loop iterates next and checks for the watcher to have become active,
		2399	it will reset the flag again. C<ev_async_pending> can be used to very
		2400	quickly check whether invoking the loop might be a good idea.
		2401
		2402	Not that this does I<not> check whether the watcher itself is pending, only
		2403	whether it has been requested to make this watcher pending.
1651		2404
1652	=back	2405	=back
1653		2406
1654		2407
1655	=head1 OTHER FUNCTIONS	2408	=head1 OTHER FUNCTIONS
…		…
1666	or timeout without having to allocate/configure/start/stop/free one or	2419	or timeout without having to allocate/configure/start/stop/free one or
1667	more watchers yourself.	2420	more watchers yourself.
1668		2421
1669	If C<fd> is less than 0, then no I/O watcher will be started and events	2422	If C<fd> is less than 0, then no I/O watcher will be started and events
1670	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2423	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and
1671	C<events> set will be craeted and started.	2424	C<events> set will be created and started.
1672		2425
1673	If C<timeout> is less than 0, then no timeout watcher will be	2426	If C<timeout> is less than 0, then no timeout watcher will be
1674	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2427	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
1675	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2428	repeat = 0) will be started. While C<0> is a valid timeout, it is of
1676	dubious value.	2429	dubious value.
…		…
1678	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	2431	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
1679	passed an C<revents> set like normal event callbacks (a combination of	2432	passed an C<revents> set like normal event callbacks (a combination of
1680	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2433	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
1681	value passed to C<ev_once>:	2434	value passed to C<ev_once>:
1682		2435
1683	static void stdin_ready (int revents, void *arg)	2436	static void stdin_ready (int revents, void *arg)
1684	{	2437	{
1685	if (revents & EV_TIMEOUT)	2438	if (revents & EV_TIMEOUT)
1686	/* doh, nothing entered */;	2439	/* doh, nothing entered */;
1687	else if (revents & EV_READ)	2440	else if (revents & EV_READ)
1688	/* stdin might have data for us, joy! */;	2441	/* stdin might have data for us, joy! */;
1689	}	2442	}
1690		2443
1691	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2444	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
1692		2445
1693	=item ev_feed_event (ev_loop *, watcher *, int revents)	2446	=item ev_feed_event (ev_loop *, watcher *, int revents)
1694		2447
1695	Feeds the given event set into the event loop, as if the specified event	2448	Feeds the given event set into the event loop, as if the specified event
1696	had happened for the specified watcher (which must be a pointer to an	2449	had happened for the specified watcher (which must be a pointer to an
…		…
1701	Feed an event on the given fd, as if a file descriptor backend detected	2454	Feed an event on the given fd, as if a file descriptor backend detected
1702	the given events it.	2455	the given events it.
1703		2456
1704	=item ev_feed_signal_event (ev_loop *loop, int signum)	2457	=item ev_feed_signal_event (ev_loop *loop, int signum)
1705		2458
1706	Feed an event as if the given signal occured (C<loop> must be the default	2459	Feed an event as if the given signal occurred (C<loop> must be the default
1707	loop!).	2460	loop!).
1708		2461
1709	=back	2462	=back
1710		2463
1711		2464
…		…
1727		2480
1728	=item * Priorities are not currently supported. Initialising priorities	2481	=item * Priorities are not currently supported. Initialising priorities
1729	will fail and all watchers will have the same priority, even though there	2482	will fail and all watchers will have the same priority, even though there
1730	is an ev_pri field.	2483	is an ev_pri field.
1731		2484
		2485	=item * In libevent, the last base created gets the signals, in libev, the
		2486	first base created (== the default loop) gets the signals.
		2487
1732	=item * Other members are not supported.	2488	=item * Other members are not supported.
1733		2489
1734	=item * The libev emulation is I<not> ABI compatible to libevent, you need	2490	=item * The libev emulation is I<not> ABI compatible to libevent, you need
1735	to use the libev header file and library.	2491	to use the libev header file and library.
1736		2492
1737	=back	2493	=back
1738		2494
1739	=head1 C++ SUPPORT	2495	=head1 C++ SUPPORT
1740		2496
1741	Libev comes with some simplistic wrapper classes for C++ that mainly allow	2497	Libev comes with some simplistic wrapper classes for C++ that mainly allow
1742	you to use some convinience methods to start/stop watchers and also change	2498	you to use some convenience methods to start/stop watchers and also change
1743	the callback model to a model using method callbacks on objects.	2499	the callback model to a model using method callbacks on objects.
1744		2500
1745	To use it,	2501	To use it,
1746		2502
1747	#include <ev++.h>	2503	#include <ev++.h>
1748		2504
1749	(it is not installed by default). This automatically includes F<ev.h>	2505	This automatically includes F<ev.h> and puts all of its definitions (many
1750	and puts all of its definitions (many of them macros) into the global	2506	of them macros) into the global namespace. All C++ specific things are
1751	namespace. All C++ specific things are put into the C<ev> namespace.	2507	put into the C<ev> namespace. It should support all the same embedding
		2508	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
1752		2509
1753	It should support all the same embedding options as F<ev.h>, most notably	2510	Care has been taken to keep the overhead low. The only data member the C++
1754	C<EV_MULTIPLICITY>.	2511	classes add (compared to plain C-style watchers) is the event loop pointer
		2512	that the watcher is associated with (or no additional members at all if
		2513	you disable C<EV_MULTIPLICITY> when embedding libev).
		2514
		2515	Currently, functions, and static and non-static member functions can be
		2516	used as callbacks. Other types should be easy to add as long as they only
		2517	need one additional pointer for context. If you need support for other
		2518	types of functors please contact the author (preferably after implementing
		2519	it).
1755		2520
1756	Here is a list of things available in the C<ev> namespace:	2521	Here is a list of things available in the C<ev> namespace:
1757		2522
1758	=over 4	2523	=over 4
1759		2524
…		…
1775		2540
1776	All of those classes have these methods:	2541	All of those classes have these methods:
1777		2542
1778	=over 4	2543	=over 4
1779		2544
1780	=item ev::TYPE::TYPE (object *, object::method *)	2545	=item ev::TYPE::TYPE ()
1781		2546
1782	=item ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)	2547	=item ev::TYPE::TYPE (struct ev_loop *)
1783		2548
1784	=item ev::TYPE::~TYPE	2549	=item ev::TYPE::~TYPE
1785		2550
1786	The constructor takes a pointer to an object and a method pointer to	2551	The constructor (optionally) takes an event loop to associate the watcher
1787	the event handler callback to call in this class. The constructor calls	2552	with. If it is omitted, it will use C<EV_DEFAULT>.
1788	C<ev_init> for you, which means you have to call the C<set> method	2553
1789	before starting it. If you do not specify a loop then the constructor	2554	The constructor calls C<ev_init> for you, which means you have to call the
1790	automatically associates the default loop with this watcher.	2555	C<set> method before starting it.
		2556
		2557	It will not set a callback, however: You have to call the templated C<set>
		2558	method to set a callback before you can start the watcher.
		2559
		2560	(The reason why you have to use a method is a limitation in C++ which does
		2561	not allow explicit template arguments for constructors).
1791		2562
1792	The destructor automatically stops the watcher if it is active.	2563	The destructor automatically stops the watcher if it is active.
		2564
		2565	=item w->set<class, &class::method> (object *)
		2566
		2567	This method sets the callback method to call. The method has to have a
		2568	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		2569	first argument and the C<revents> as second. The object must be given as
		2570	parameter and is stored in the C<data> member of the watcher.
		2571
		2572	This method synthesizes efficient thunking code to call your method from
		2573	the C callback that libev requires. If your compiler can inline your
		2574	callback (i.e. it is visible to it at the place of the C<set> call and
		2575	your compiler is good :), then the method will be fully inlined into the
		2576	thunking function, making it as fast as a direct C callback.
		2577
		2578	Example: simple class declaration and watcher initialisation
		2579
		2580	struct myclass
		2581	{
		2582	void io_cb (ev::io &w, int revents) { }
		2583	}
		2584
		2585	myclass obj;
		2586	ev::io iow;
		2587	iow.set <myclass, &myclass::io_cb> (&obj);
		2588
		2589	=item w->set<function> (void *data = 0)
		2590
		2591	Also sets a callback, but uses a static method or plain function as
		2592	callback. The optional C<data> argument will be stored in the watcher's
		2593	C<data> member and is free for you to use.
		2594
		2595	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		2596
		2597	See the method-C<set> above for more details.
		2598
		2599	Example:
		2600
		2601	static void io_cb (ev::io &w, int revents) { }
		2602	iow.set <io_cb> ();
1793		2603
1794	=item w->set (struct ev_loop *)	2604	=item w->set (struct ev_loop *)
1795		2605
1796	Associates a different C<struct ev_loop> with this watcher. You can only	2606	Associates a different C<struct ev_loop> with this watcher. You can only
1797	do this when the watcher is inactive (and not pending either).	2607	do this when the watcher is inactive (and not pending either).
1798		2608
1799	=item w->set ([args])	2609	=item w->set ([arguments])
1800		2610
1801	Basically the same as C<ev_TYPE_set>, with the same args. Must be	2611	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be
1802	called at least once. Unlike the C counterpart, an active watcher gets	2612	called at least once. Unlike the C counterpart, an active watcher gets
1803	automatically stopped and restarted.	2613	automatically stopped and restarted when reconfiguring it with this
		2614	method.
1804		2615
1805	=item w->start ()	2616	=item w->start ()
1806		2617
1807	Starts the watcher. Note that there is no C<loop> argument as the	2618	Starts the watcher. Note that there is no C<loop> argument, as the
1808	constructor already takes the loop.	2619	constructor already stores the event loop.
1809		2620
1810	=item w->stop ()	2621	=item w->stop ()
1811		2622
1812	Stops the watcher if it is active. Again, no C<loop> argument.	2623	Stops the watcher if it is active. Again, no C<loop> argument.
1813		2624
1814	=item w->again () C<ev::timer>, C<ev::periodic> only	2625	=item w->again () (C<ev::timer>, C<ev::periodic> only)
1815		2626
1816	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding	2627	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
1817	C<ev_TYPE_again> function.	2628	C<ev_TYPE_again> function.
1818		2629
1819	=item w->sweep () C<ev::embed> only	2630	=item w->sweep () (C<ev::embed> only)
1820		2631
1821	Invokes C<ev_embed_sweep>.	2632	Invokes C<ev_embed_sweep>.
1822		2633
1823	=item w->update () C<ev::stat> only	2634	=item w->update () (C<ev::stat> only)
1824		2635
1825	Invokes C<ev_stat_stat>.	2636	Invokes C<ev_stat_stat>.
1826		2637
1827	=back	2638	=back
1828		2639
1829	=back	2640	=back
1830		2641
1831	Example: Define a class with an IO and idle watcher, start one of them in	2642	Example: Define a class with an IO and idle watcher, start one of them in
1832	the constructor.	2643	the constructor.
1833		2644
1834	class myclass	2645	class myclass
1835	{	2646	{
1836	ev_io io; void io_cb (ev::io &w, int revents);	2647	ev::io io; void io_cb (ev::io &w, int revents);
1837	ev_idle idle void idle_cb (ev::idle &w, int revents);	2648	ev:idle idle void idle_cb (ev::idle &w, int revents);
1838		2649
1839	myclass ();	2650	myclass (int fd)
1840	}	2651	{
		2652	io .set <myclass, &myclass::io_cb > (this);
		2653	idle.set <myclass, &myclass::idle_cb> (this);
1841		2654
1842	myclass::myclass (int fd)
1843	: io (this, &myclass::io_cb),
1844	idle (this, &myclass::idle_cb)
1845	{
1846	io.start (fd, ev::READ);	2655	io.start (fd, ev::READ);
		2656	}
1847	}	2657	};
		2658
		2659
		2660	=head1 OTHER LANGUAGE BINDINGS
		2661
		2662	Libev does not offer other language bindings itself, but bindings for a
		2663	number of languages exist in the form of third-party packages. If you know
		2664	any interesting language binding in addition to the ones listed here, drop
		2665	me a note.
		2666
		2667	=over 4
		2668
		2669	=item Perl
		2670
		2671	The EV module implements the full libev API and is actually used to test
		2672	libev. EV is developed together with libev. Apart from the EV core module,
		2673	there are additional modules that implement libev-compatible interfaces
		2674	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the
		2675	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).
		2676
		2677	It can be found and installed via CPAN, its homepage is at
		2678	L<http://software.schmorp.de/pkg/EV>.
		2679
		2680	=item Python
		2681
		2682	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		2683	seems to be quite complete and well-documented. Note, however, that the
		2684	patch they require for libev is outright dangerous as it breaks the ABI
		2685	for everybody else, and therefore, should never be applied in an installed
		2686	libev (if python requires an incompatible ABI then it needs to embed
		2687	libev).
		2688
		2689	=item Ruby
		2690
		2691	Tony Arcieri has written a ruby extension that offers access to a subset
		2692	of the libev API and adds file handle abstractions, asynchronous DNS and
		2693	more on top of it. It can be found via gem servers. Its homepage is at
		2694	L<http://rev.rubyforge.org/>.
		2695
		2696	=item D
		2697
		2698	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		2699	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		2700
		2701	=back
1848		2702
1849		2703
1850	=head1 MACRO MAGIC	2704	=head1 MACRO MAGIC
1851		2705
1852	Libev can be compiled with a variety of options, the most fundemantal is	2706	Libev can be compiled with a variety of options, the most fundamental
1853	C<EV_MULTIPLICITY>. This option determines whether (most) functions and	2707	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
1854	callbacks have an initial C<struct ev_loop *> argument.	2708	functions and callbacks have an initial C<struct ev_loop *> argument.
1855		2709
1856	To make it easier to write programs that cope with either variant, the	2710	To make it easier to write programs that cope with either variant, the
1857	following macros are defined:	2711	following macros are defined:
1858		2712
1859	=over 4	2713	=over 4
…		…
1862		2716
1863	This provides the loop I<argument> for functions, if one is required ("ev	2717	This provides the loop I<argument> for functions, if one is required ("ev
1864	loop argument"). The C<EV_A> form is used when this is the sole argument,	2718	loop argument"). The C<EV_A> form is used when this is the sole argument,
1865	C<EV_A_> is used when other arguments are following. Example:	2719	C<EV_A_> is used when other arguments are following. Example:
1866		2720
1867	ev_unref (EV_A);	2721	ev_unref (EV_A);
1868	ev_timer_add (EV_A_ watcher);	2722	ev_timer_add (EV_A_ watcher);
1869	ev_loop (EV_A_ 0);	2723	ev_loop (EV_A_ 0);
1870		2724
1871	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	2725	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
1872	which is often provided by the following macro.	2726	which is often provided by the following macro.
1873		2727
1874	=item C<EV_P>, C<EV_P_>	2728	=item C<EV_P>, C<EV_P_>
1875		2729
1876	This provides the loop I<parameter> for functions, if one is required ("ev	2730	This provides the loop I<parameter> for functions, if one is required ("ev
1877	loop parameter"). The C<EV_P> form is used when this is the sole parameter,	2731	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
1878	C<EV_P_> is used when other parameters are following. Example:	2732	C<EV_P_> is used when other parameters are following. Example:
1879		2733
1880	// this is how ev_unref is being declared	2734	// this is how ev_unref is being declared
1881	static void ev_unref (EV_P);	2735	static void ev_unref (EV_P);
1882		2736
1883	// this is how you can declare your typical callback	2737	// this is how you can declare your typical callback
1884	static void cb (EV_P_ ev_timer *w, int revents)	2738	static void cb (EV_P_ ev_timer *w, int revents)
1885		2739
1886	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	2740	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
1887	suitable for use with C<EV_A>.	2741	suitable for use with C<EV_A>.
1888		2742
1889	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	2743	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
1890		2744
1891	Similar to the other two macros, this gives you the value of the default	2745	Similar to the other two macros, this gives you the value of the default
1892	loop, if multiple loops are supported ("ev loop default").	2746	loop, if multiple loops are supported ("ev loop default").
		2747
		2748	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		2749
		2750	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		2751	default loop has been initialised (C<UC> == unchecked). Their behaviour
		2752	is undefined when the default loop has not been initialised by a previous
		2753	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		2754
		2755	It is often prudent to use C<EV_DEFAULT> when initialising the first
		2756	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
1893		2757
1894	=back	2758	=back
1895		2759
1896	Example: Declare and initialise a check watcher, utilising the above	2760	Example: Declare and initialise a check watcher, utilising the above
1897	macros so it will work regardless of whether multiple loops are supported	2761	macros so it will work regardless of whether multiple loops are supported
1898	or not.	2762	or not.
1899		2763
1900	static void	2764	static void
1901	check_cb (EV_P_ ev_timer *w, int revents)	2765	check_cb (EV_P_ ev_timer *w, int revents)
1902	{	2766	{
1903	ev_check_stop (EV_A_ w);	2767	ev_check_stop (EV_A_ w);
1904	}	2768	}
1905		2769
1906	ev_check check;	2770	ev_check check;
1907	ev_check_init (&check, check_cb);	2771	ev_check_init (&check, check_cb);
1908	ev_check_start (EV_DEFAULT_ &check);	2772	ev_check_start (EV_DEFAULT_ &check);
1909	ev_loop (EV_DEFAULT_ 0);	2773	ev_loop (EV_DEFAULT_ 0);
1910		2774
1911	=head1 EMBEDDING	2775	=head1 EMBEDDING
1912		2776
1913	Libev can (and often is) directly embedded into host	2777	Libev can (and often is) directly embedded into host
1914	applications. Examples of applications that embed it include the Deliantra	2778	applications. Examples of applications that embed it include the Deliantra
1915	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)	2779	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
1916	and rxvt-unicode.	2780	and rxvt-unicode.
1917		2781
1918	The goal is to enable you to just copy the neecssary files into your	2782	The goal is to enable you to just copy the necessary files into your
1919	source directory without having to change even a single line in them, so	2783	source directory without having to change even a single line in them, so
1920	you can easily upgrade by simply copying (or having a checked-out copy of	2784	you can easily upgrade by simply copying (or having a checked-out copy of
1921	libev somewhere in your source tree).	2785	libev somewhere in your source tree).
1922		2786
1923	=head2 FILESETS	2787	=head2 FILESETS
1924		2788
1925	Depending on what features you need you need to include one or more sets of files	2789	Depending on what features you need you need to include one or more sets of files
1926	in your app.	2790	in your application.
1927		2791
1928	=head3 CORE EVENT LOOP	2792	=head3 CORE EVENT LOOP
1929		2793
1930	To include only the libev core (all the C<ev_*> functions), with manual	2794	To include only the libev core (all the C<ev_*> functions), with manual
1931	configuration (no autoconf):	2795	configuration (no autoconf):
1932		2796
1933	#define EV_STANDALONE 1	2797	#define EV_STANDALONE 1
1934	#include "ev.c"	2798	#include "ev.c"
1935		2799
1936	This will automatically include F<ev.h>, too, and should be done in a	2800	This will automatically include F<ev.h>, too, and should be done in a
1937	single C source file only to provide the function implementations. To use	2801	single C source file only to provide the function implementations. To use
1938	it, do the same for F<ev.h> in all files wishing to use this API (best	2802	it, do the same for F<ev.h> in all files wishing to use this API (best
1939	done by writing a wrapper around F<ev.h> that you can include instead and	2803	done by writing a wrapper around F<ev.h> that you can include instead and
1940	where you can put other configuration options):	2804	where you can put other configuration options):
1941		2805
1942	#define EV_STANDALONE 1	2806	#define EV_STANDALONE 1
1943	#include "ev.h"	2807	#include "ev.h"
1944		2808
1945	Both header files and implementation files can be compiled with a C++	2809	Both header files and implementation files can be compiled with a C++
1946	compiler (at least, thats a stated goal, and breakage will be treated	2810	compiler (at least, thats a stated goal, and breakage will be treated
1947	as a bug).	2811	as a bug).
1948		2812
1949	You need the following files in your source tree, or in a directory	2813	You need the following files in your source tree, or in a directory
1950	in your include path (e.g. in libev/ when using -Ilibev):	2814	in your include path (e.g. in libev/ when using -Ilibev):
1951		2815
1952	ev.h	2816	ev.h
1953	ev.c	2817	ev.c
1954	ev_vars.h	2818	ev_vars.h
1955	ev_wrap.h	2819	ev_wrap.h
1956		2820
1957	ev_win32.c required on win32 platforms only	2821	ev_win32.c required on win32 platforms only
1958		2822
1959	ev_select.c only when select backend is enabled (which is enabled by default)	2823	ev_select.c only when select backend is enabled (which is enabled by default)
1960	ev_poll.c only when poll backend is enabled (disabled by default)	2824	ev_poll.c only when poll backend is enabled (disabled by default)
1961	ev_epoll.c only when the epoll backend is enabled (disabled by default)	2825	ev_epoll.c only when the epoll backend is enabled (disabled by default)
1962	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	2826	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
1963	ev_port.c only when the solaris port backend is enabled (disabled by default)	2827	ev_port.c only when the solaris port backend is enabled (disabled by default)
1964		2828
1965	F<ev.c> includes the backend files directly when enabled, so you only need	2829	F<ev.c> includes the backend files directly when enabled, so you only need
1966	to compile this single file.	2830	to compile this single file.
1967		2831
1968	=head3 LIBEVENT COMPATIBILITY API	2832	=head3 LIBEVENT COMPATIBILITY API
1969		2833
1970	To include the libevent compatibility API, also include:	2834	To include the libevent compatibility API, also include:
1971		2835
1972	#include "event.c"	2836	#include "event.c"
1973		2837
1974	in the file including F<ev.c>, and:	2838	in the file including F<ev.c>, and:
1975		2839
1976	#include "event.h"	2840	#include "event.h"
1977		2841
1978	in the files that want to use the libevent API. This also includes F<ev.h>.	2842	in the files that want to use the libevent API. This also includes F<ev.h>.
1979		2843
1980	You need the following additional files for this:	2844	You need the following additional files for this:
1981		2845
1982	event.h	2846	event.h
1983	event.c	2847	event.c
1984		2848
1985	=head3 AUTOCONF SUPPORT	2849	=head3 AUTOCONF SUPPORT
1986		2850
1987	Instead of using C<EV_STANDALONE=1> and providing your config in	2851	Instead of using C<EV_STANDALONE=1> and providing your configuration in
1988	whatever way you want, you can also C<m4_include([libev.m4])> in your	2852	whatever way you want, you can also C<m4_include([libev.m4])> in your
1989	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then	2853	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
1990	include F<config.h> and configure itself accordingly.	2854	include F<config.h> and configure itself accordingly.
1991		2855
1992	For this of course you need the m4 file:	2856	For this of course you need the m4 file:
1993		2857
1994	libev.m4	2858	libev.m4
1995		2859
1996	=head2 PREPROCESSOR SYMBOLS/MACROS	2860	=head2 PREPROCESSOR SYMBOLS/MACROS
1997		2861
1998	Libev can be configured via a variety of preprocessor symbols you have to define	2862	Libev can be configured via a variety of preprocessor symbols you have to
1999	before including any of its files. The default is not to build for multiplicity	2863	define before including any of its files. The default in the absence of
2000	and only include the select backend.	2864	autoconf is noted for every option.
2001		2865
2002	=over 4	2866	=over 4
2003		2867
2004	=item EV_STANDALONE	2868	=item EV_STANDALONE
2005		2869
…		…
2010	F<event.h> that are not directly supported by the libev core alone.	2874	F<event.h> that are not directly supported by the libev core alone.
2011		2875
2012	=item EV_USE_MONOTONIC	2876	=item EV_USE_MONOTONIC
2013		2877
2014	If defined to be C<1>, libev will try to detect the availability of the	2878	If defined to be C<1>, libev will try to detect the availability of the
2015	monotonic clock option at both compiletime and runtime. Otherwise no use	2879	monotonic clock option at both compile time and runtime. Otherwise no use
2016	of the monotonic clock option will be attempted. If you enable this, you	2880	of the monotonic clock option will be attempted. If you enable this, you
2017	usually have to link against librt or something similar. Enabling it when	2881	usually have to link against librt or something similar. Enabling it when
2018	the functionality isn't available is safe, though, althoguh you have	2882	the functionality isn't available is safe, though, although you have
2019	to make sure you link against any libraries where the C<clock_gettime>	2883	to make sure you link against any libraries where the C<clock_gettime>
2020	function is hiding in (often F<-lrt>).	2884	function is hiding in (often F<-lrt>).
2021		2885
2022	=item EV_USE_REALTIME	2886	=item EV_USE_REALTIME
2023		2887
2024	If defined to be C<1>, libev will try to detect the availability of the	2888	If defined to be C<1>, libev will try to detect the availability of the
2025	realtime clock option at compiletime (and assume its availability at	2889	real-time clock option at compile time (and assume its availability at
2026	runtime if successful). Otherwise no use of the realtime clock option will	2890	runtime if successful). Otherwise no use of the real-time clock option will
2027	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	2891	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
2028	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See tzhe note about libraries	2892	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
2029	in the description of C<EV_USE_MONOTONIC>, though.	2893	note about libraries in the description of C<EV_USE_MONOTONIC>, though.
		2894
		2895	=item EV_USE_NANOSLEEP
		2896
		2897	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		2898	and will use it for delays. Otherwise it will use C<select ()>.
		2899
		2900	=item EV_USE_EVENTFD
		2901
		2902	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		2903	available and will probe for kernel support at runtime. This will improve
		2904	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		2905	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		2906	2.7 or newer, otherwise disabled.
2030		2907
2031	=item EV_USE_SELECT	2908	=item EV_USE_SELECT
2032		2909
2033	If undefined or defined to be C<1>, libev will compile in support for the	2910	If undefined or defined to be C<1>, libev will compile in support for the
2034	C<select>(2) backend. No attempt at autodetection will be done: if no	2911	C<select>(2) backend. No attempt at auto-detection will be done: if no
2035	other method takes over, select will be it. Otherwise the select backend	2912	other method takes over, select will be it. Otherwise the select backend
2036	will not be compiled in.	2913	will not be compiled in.
2037		2914
2038	=item EV_SELECT_USE_FD_SET	2915	=item EV_SELECT_USE_FD_SET
2039		2916
2040	If defined to C<1>, then the select backend will use the system C<fd_set>	2917	If defined to C<1>, then the select backend will use the system C<fd_set>
2041	structure. This is useful if libev doesn't compile due to a missing	2918	structure. This is useful if libev doesn't compile due to a missing
2042	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on	2919	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on
2043	exotic systems. This usually limits the range of file descriptors to some	2920	exotic systems. This usually limits the range of file descriptors to some
2044	low limit such as 1024 or might have other limitations (winsocket only	2921	low limit such as 1024 or might have other limitations (winsocket only
2045	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	2922	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
2046	influence the size of the C<fd_set> used.	2923	influence the size of the C<fd_set> used.
2047		2924
…		…
2053	be used is the winsock select). This means that it will call	2930	be used is the winsock select). This means that it will call
2054	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	2931	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
2055	it is assumed that all these functions actually work on fds, even	2932	it is assumed that all these functions actually work on fds, even
2056	on win32. Should not be defined on non-win32 platforms.	2933	on win32. Should not be defined on non-win32 platforms.
2057		2934
		2935	=item EV_FD_TO_WIN32_HANDLE
		2936
		2937	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		2938	file descriptors to socket handles. When not defining this symbol (the
		2939	default), then libev will call C<_get_osfhandle>, which is usually
		2940	correct. In some cases, programs use their own file descriptor management,
		2941	in which case they can provide this function to map fds to socket handles.
		2942
2058	=item EV_USE_POLL	2943	=item EV_USE_POLL
2059		2944
2060	If defined to be C<1>, libev will compile in support for the C<poll>(2)	2945	If defined to be C<1>, libev will compile in support for the C<poll>(2)
2061	backend. Otherwise it will be enabled on non-win32 platforms. It	2946	backend. Otherwise it will be enabled on non-win32 platforms. It
2062	takes precedence over select.	2947	takes precedence over select.
2063		2948
2064	=item EV_USE_EPOLL	2949	=item EV_USE_EPOLL
2065		2950
2066	If defined to be C<1>, libev will compile in support for the Linux	2951	If defined to be C<1>, libev will compile in support for the Linux
2067	C<epoll>(7) backend. Its availability will be detected at runtime,	2952	C<epoll>(7) backend. Its availability will be detected at runtime,
2068	otherwise another method will be used as fallback. This is the	2953	otherwise another method will be used as fallback. This is the preferred
2069	preferred backend for GNU/Linux systems.	2954	backend for GNU/Linux systems. If undefined, it will be enabled if the
		2955	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2070		2956
2071	=item EV_USE_KQUEUE	2957	=item EV_USE_KQUEUE
2072		2958
2073	If defined to be C<1>, libev will compile in support for the BSD style	2959	If defined to be C<1>, libev will compile in support for the BSD style
2074	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	2960	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
2087	otherwise another method will be used as fallback. This is the preferred	2973	otherwise another method will be used as fallback. This is the preferred
2088	backend for Solaris 10 systems.	2974	backend for Solaris 10 systems.
2089		2975
2090	=item EV_USE_DEVPOLL	2976	=item EV_USE_DEVPOLL
2091		2977
2092	reserved for future expansion, works like the USE symbols above.	2978	Reserved for future expansion, works like the USE symbols above.
2093		2979
2094	=item EV_USE_INOTIFY	2980	=item EV_USE_INOTIFY
2095		2981
2096	If defined to be C<1>, libev will compile in support for the Linux inotify	2982	If defined to be C<1>, libev will compile in support for the Linux inotify
2097	interface to speed up C<ev_stat> watchers. Its actual availability will	2983	interface to speed up C<ev_stat> watchers. Its actual availability will
2098	be detected at runtime.	2984	be detected at runtime. If undefined, it will be enabled if the headers
		2985	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		2986
		2987	=item EV_ATOMIC_T
		2988
		2989	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		2990	access is atomic with respect to other threads or signal contexts. No such
		2991	type is easily found in the C language, so you can provide your own type
		2992	that you know is safe for your purposes. It is used both for signal handler "locking"
		2993	as well as for signal and thread safety in C<ev_async> watchers.
		2994
		2995	In the absence of this define, libev will use C<sig_atomic_t volatile>
		2996	(from F<signal.h>), which is usually good enough on most platforms.
2099		2997
2100	=item EV_H	2998	=item EV_H
2101		2999
2102	The name of the F<ev.h> header file used to include it. The default if	3000	The name of the F<ev.h> header file used to include it. The default if
2103	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This	3001	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2104	can be used to virtually rename the F<ev.h> header file in case of conflicts.	3002	used to virtually rename the F<ev.h> header file in case of conflicts.
2105		3003
2106	=item EV_CONFIG_H	3004	=item EV_CONFIG_H
2107		3005
2108	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	3006	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2109	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	3007	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2110	C<EV_H>, above.	3008	C<EV_H>, above.
2111		3009
2112	=item EV_EVENT_H	3010	=item EV_EVENT_H
2113		3011
2114	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	3012	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2115	of how the F<event.h> header can be found.	3013	of how the F<event.h> header can be found, the default is C<"event.h">.
2116		3014
2117	=item EV_PROTOTYPES	3015	=item EV_PROTOTYPES
2118		3016
2119	If defined to be C<0>, then F<ev.h> will not define any function	3017	If defined to be C<0>, then F<ev.h> will not define any function
2120	prototypes, but still define all the structs and other symbols. This is	3018	prototypes, but still define all the structs and other symbols. This is
…		…
2141	When doing priority-based operations, libev usually has to linearly search	3039	When doing priority-based operations, libev usually has to linearly search
2142	all the priorities, so having many of them (hundreds) uses a lot of space	3040	all the priorities, so having many of them (hundreds) uses a lot of space
2143	and time, so using the defaults of five priorities (-2 .. +2) is usually	3041	and time, so using the defaults of five priorities (-2 .. +2) is usually
2144	fine.	3042	fine.
2145		3043
2146	If your embedding app does not need any priorities, defining these both to	3044	If your embedding application does not need any priorities, defining these both to
2147	C<0> will save some memory and cpu.	3045	C<0> will save some memory and CPU.
2148		3046
2149	=item EV_PERIODIC_ENABLE	3047	=item EV_PERIODIC_ENABLE
2150		3048
2151	If undefined or defined to be C<1>, then periodic timers are supported. If	3049	If undefined or defined to be C<1>, then periodic timers are supported. If
2152	defined to be C<0>, then they are not. Disabling them saves a few kB of	3050	defined to be C<0>, then they are not. Disabling them saves a few kB of
…		…
2171	=item EV_FORK_ENABLE	3069	=item EV_FORK_ENABLE
2172		3070
2173	If undefined or defined to be C<1>, then fork watchers are supported. If	3071	If undefined or defined to be C<1>, then fork watchers are supported. If
2174	defined to be C<0>, then they are not.	3072	defined to be C<0>, then they are not.
2175		3073
		3074	=item EV_ASYNC_ENABLE
		3075
		3076	If undefined or defined to be C<1>, then async watchers are supported. If
		3077	defined to be C<0>, then they are not.
		3078
2176	=item EV_MINIMAL	3079	=item EV_MINIMAL
2177		3080
2178	If you need to shave off some kilobytes of code at the expense of some	3081	If you need to shave off some kilobytes of code at the expense of some
2179	speed, define this symbol to C<1>. Currently only used for gcc to override	3082	speed, define this symbol to C<1>. Currently this is used to override some
2180	some inlining decisions, saves roughly 30% codesize of amd64.	3083	inlining decisions, saves roughly 30% code size on amd64. It also selects a
		3084	much smaller 2-heap for timer management over the default 4-heap.
2181		3085
2182	=item EV_PID_HASHSIZE	3086	=item EV_PID_HASHSIZE
2183		3087
2184	C<ev_child> watchers use a small hash table to distribute workload by	3088	C<ev_child> watchers use a small hash table to distribute workload by
2185	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	3089	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
2186	than enough. If you need to manage thousands of children you might want to	3090	than enough. If you need to manage thousands of children you might want to
2187	increase this value (I<must> be a power of two).	3091	increase this value (I<must> be a power of two).
2188		3092
2189	=item EV_INOTIFY_HASHSIZE	3093	=item EV_INOTIFY_HASHSIZE
2190		3094
2191	C<ev_staz> watchers use a small hash table to distribute workload by	3095	C<ev_stat> watchers use a small hash table to distribute workload by
2192	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	3096	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
2193	usually more than enough. If you need to manage thousands of C<ev_stat>	3097	usually more than enough. If you need to manage thousands of C<ev_stat>
2194	watchers you might want to increase this value (I<must> be a power of	3098	watchers you might want to increase this value (I<must> be a power of
2195	two).	3099	two).
2196		3100
		3101	=item EV_USE_4HEAP
		3102
		3103	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3104	timer and periodics heap, libev uses a 4-heap when this symbol is defined
		3105	to C<1>. The 4-heap uses more complicated (longer) code but has
		3106	noticeably faster performance with many (thousands) of watchers.
		3107
		3108	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3109	(disabled).
		3110
		3111	=item EV_HEAP_CACHE_AT
		3112
		3113	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3114	timer and periodics heap, libev can cache the timestamp (I<at>) within
		3115	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3116	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3117	but avoids random read accesses on heap changes. This improves performance
		3118	noticeably with with many (hundreds) of watchers.
		3119
		3120	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3121	(disabled).
		3122
		3123	=item EV_VERIFY
		3124
		3125	Controls how much internal verification (see C<ev_loop_verify ()>) will
		3126	be done: If set to C<0>, no internal verification code will be compiled
		3127	in. If set to C<1>, then verification code will be compiled in, but not
		3128	called. If set to C<2>, then the internal verification code will be
		3129	called once per loop, which can slow down libev. If set to C<3>, then the
		3130	verification code will be called very frequently, which will slow down
		3131	libev considerably.
		3132
		3133	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
		3134	C<0.>
		3135
2197	=item EV_COMMON	3136	=item EV_COMMON
2198		3137
2199	By default, all watchers have a C<void *data> member. By redefining	3138	By default, all watchers have a C<void *data> member. By redefining
2200	this macro to a something else you can include more and other types of	3139	this macro to a something else you can include more and other types of
2201	members. You have to define it each time you include one of the files,	3140	members. You have to define it each time you include one of the files,
2202	though, and it must be identical each time.	3141	though, and it must be identical each time.
2203		3142
2204	For example, the perl EV module uses something like this:	3143	For example, the perl EV module uses something like this:
2205		3144
2206	#define EV_COMMON \	3145	#define EV_COMMON \
2207	SV *self; /* contains this struct */ \	3146	SV *self; /* contains this struct */ \
2208	SV *cb_sv, *fh /* note no trailing ";" */	3147	SV *cb_sv, *fh /* note no trailing ";" */
2209		3148
2210	=item EV_CB_DECLARE (type)	3149	=item EV_CB_DECLARE (type)
2211		3150
2212	=item EV_CB_INVOKE (watcher, revents)	3151	=item EV_CB_INVOKE (watcher, revents)
2213		3152
2214	=item ev_set_cb (ev, cb)	3153	=item ev_set_cb (ev, cb)
2215		3154
2216	Can be used to change the callback member declaration in each watcher,	3155	Can be used to change the callback member declaration in each watcher,
2217	and the way callbacks are invoked and set. Must expand to a struct member	3156	and the way callbacks are invoked and set. Must expand to a struct member
2218	definition and a statement, respectively. See the F<ev.v> header file for	3157	definition and a statement, respectively. See the F<ev.h> header file for
2219	their default definitions. One possible use for overriding these is to	3158	their default definitions. One possible use for overriding these is to
2220	avoid the C<struct ev_loop *> as first argument in all cases, or to use	3159	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2221	method calls instead of plain function calls in C++.	3160	method calls instead of plain function calls in C++.
		3161
		3162	=head2 EXPORTED API SYMBOLS
		3163
		3164	If you need to re-export the API (e.g. via a DLL) and you need a list of
		3165	exported symbols, you can use the provided F<Symbol.*> files which list
		3166	all public symbols, one per line:
		3167
		3168	Symbols.ev for libev proper
		3169	Symbols.event for the libevent emulation
		3170
		3171	This can also be used to rename all public symbols to avoid clashes with
		3172	multiple versions of libev linked together (which is obviously bad in
		3173	itself, but sometimes it is inconvenient to avoid this).
		3174
		3175	A sed command like this will create wrapper C<#define>'s that you need to
		3176	include before including F<ev.h>:
		3177
		3178	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		3179
		3180	This would create a file F<wrap.h> which essentially looks like this:
		3181
		3182	#define ev_backend myprefix_ev_backend
		3183	#define ev_check_start myprefix_ev_check_start
		3184	#define ev_check_stop myprefix_ev_check_stop
		3185	...
2222		3186
2223	=head2 EXAMPLES	3187	=head2 EXAMPLES
2224		3188
2225	For a real-world example of a program the includes libev	3189	For a real-world example of a program the includes libev
2226	verbatim, you can have a look at the EV perl module	3190	verbatim, you can have a look at the EV perl module
…		…
2231	file.	3195	file.
2232		3196
2233	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	3197	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2234	that everybody includes and which overrides some configure choices:	3198	that everybody includes and which overrides some configure choices:
2235		3199
2236	#define EV_MINIMAL 1	3200	#define EV_MINIMAL 1
2237	#define EV_USE_POLL 0	3201	#define EV_USE_POLL 0
2238	#define EV_MULTIPLICITY 0	3202	#define EV_MULTIPLICITY 0
2239	#define EV_PERIODIC_ENABLE 0	3203	#define EV_PERIODIC_ENABLE 0
2240	#define EV_STAT_ENABLE 0	3204	#define EV_STAT_ENABLE 0
2241	#define EV_FORK_ENABLE 0	3205	#define EV_FORK_ENABLE 0
2242	#define EV_CONFIG_H <config.h>	3206	#define EV_CONFIG_H <config.h>
2243	#define EV_MINPRI 0	3207	#define EV_MINPRI 0
2244	#define EV_MAXPRI 0	3208	#define EV_MAXPRI 0
2245		3209
2246	#include "ev++.h"	3210	#include "ev++.h"
2247		3211
2248	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3212	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
2249		3213
2250	#include "ev_cpp.h"	3214	#include "ev_cpp.h"
2251	#include "ev.c"	3215	#include "ev.c"
		3216
		3217
		3218	=head1 THREADS AND COROUTINES
		3219
		3220	=head2 THREADS
		3221
		3222	Libev itself is completely thread-safe, but it uses no locking. This
		3223	means that you can use as many loops as you want in parallel, as long as
		3224	only one thread ever calls into one libev function with the same loop
		3225	parameter.
		3226
		3227	Or put differently: calls with different loop parameters can be done in
		3228	parallel from multiple threads, calls with the same loop parameter must be
		3229	done serially (but can be done from different threads, as long as only one
		3230	thread ever is inside a call at any point in time, e.g. by using a mutex
		3231	per loop).
		3232
		3233	If you want to know which design (one loop, locking, or multiple loops
		3234	without or something else still) is best for your problem, then I cannot
		3235	help you. I can give some generic advice however:
		3236
		3237	=over 4
		3238
		3239	=item * most applications have a main thread: use the default libev loop
		3240	in that thread, or create a separate thread running only the default loop.
		3241
		3242	This helps integrating other libraries or software modules that use libev
		3243	themselves and don't care/know about threading.
		3244
		3245	=item * one loop per thread is usually a good model.
		3246
		3247	Doing this is almost never wrong, sometimes a better-performance model
		3248	exists, but it is always a good start.
		3249
		3250	=item * other models exist, such as the leader/follower pattern, where one
		3251	loop is handed through multiple threads in a kind of round-robin fashion.
		3252
		3253	Choosing a model is hard - look around, learn, know that usually you can do
		3254	better than you currently do :-)
		3255
		3256	=item * often you need to talk to some other thread which blocks in the
		3257	event loop - C<ev_async> watchers can be used to wake them up from other
		3258	threads safely (or from signal contexts...).
		3259
		3260	=back
		3261
		3262	=head2 COROUTINES
		3263
		3264	Libev is much more accommodating to coroutines ("cooperative threads"):
		3265	libev fully supports nesting calls to it's functions from different
		3266	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		3267	different coroutines and switch freely between both coroutines running the
		3268	loop, as long as you don't confuse yourself). The only exception is that
		3269	you must not do this from C<ev_periodic> reschedule callbacks.
		3270
		3271	Care has been invested into making sure that libev does not keep local
		3272	state inside C<ev_loop>, and other calls do not usually allow coroutine
		3273	switches.
2252		3274
2253		3275
2254	=head1 COMPLEXITIES	3276	=head1 COMPLEXITIES
2255		3277
2256	In this section the complexities of (many of) the algorithms used inside	3278	In this section the complexities of (many of) the algorithms used inside
2257	libev will be explained. For complexity discussions about backends see the	3279	libev will be explained. For complexity discussions about backends see the
2258	documentation for C<ev_default_init>.	3280	documentation for C<ev_default_init>.
2259		3281
		3282	All of the following are about amortised time: If an array needs to be
		3283	extended, libev needs to realloc and move the whole array, but this
		3284	happens asymptotically never with higher number of elements, so O(1) might
		3285	mean it might do a lengthy realloc operation in rare cases, but on average
		3286	it is much faster and asymptotically approaches constant time.
		3287
2260	=over 4	3288	=over 4
2261		3289
2262	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3290	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2263		3291
2264	This means that, when you have a watcher that triggers in one hour and	3292	This means that, when you have a watcher that triggers in one hour and
2265	there are 100 watchers that would trigger before that then inserting will	3293	there are 100 watchers that would trigger before that then inserting will
2266	have to skip those 100 watchers.	3294	have to skip roughly seven (C<ld 100>) of these watchers.
2267		3295
2268	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	3296	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
2269		3297
2270	That means that for changing a timer costs less than removing/adding them	3298	That means that changing a timer costs less than removing/adding them
2271	as only the relative motion in the event queue has to be paid for.	3299	as only the relative motion in the event queue has to be paid for.
2272		3300
2273	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	3301	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
2274		3302
2275	These just add the watcher into an array or at the head of a list. If	3303	These just add the watcher into an array or at the head of a list.
2276	the array needs to be extended libev needs to realloc and move the whole
2277	array, but this happen asymptotically less and less with more watchers,
2278	thus amortised O(1).
2279		3304
2280	=item Stopping check/prepare/idle watchers: O(1)	3305	=item Stopping check/prepare/idle/fork/async watchers: O(1)
2281		3306
2282	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3307	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2283		3308
2284	These watchers are stored in lists then need to be walked to find the	3309	These watchers are stored in lists then need to be walked to find the
2285	correct watcher to remove. The lists are usually short (you don't usually	3310	correct watcher to remove. The lists are usually short (you don't usually
2286	have many watchers waiting for the same fd or signal).	3311	have many watchers waiting for the same fd or signal).
2287		3312
2288	=item Finding the next timer per loop iteration: O(1)	3313	=item Finding the next timer in each loop iteration: O(1)
		3314
		3315	By virtue of using a binary or 4-heap, the next timer is always found at a
		3316	fixed position in the storage array.
2289		3317
2290	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	3318	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2291		3319
2292	A change means an I/O watcher gets started or stopped, which requires	3320	A change means an I/O watcher gets started or stopped, which requires
2293	libev to recalculate its status (and possibly tell the kernel).	3321	libev to recalculate its status (and possibly tell the kernel, depending
		3322	on backend and whether C<ev_io_set> was used).
2294		3323
2295	=item Activating one watcher: O(1)	3324	=item Activating one watcher (putting it into the pending state): O(1)
2296		3325
2297	=item Priority handling: O(number_of_priorities)	3326	=item Priority handling: O(number_of_priorities)
2298		3327
2299	Priorities are implemented by allocating some space for each	3328	Priorities are implemented by allocating some space for each
2300	priority. When doing priority-based operations, libev usually has to	3329	priority. When doing priority-based operations, libev usually has to
2301	linearly search all the priorities.	3330	linearly search all the priorities, but starting/stopping and activating
		3331	watchers becomes O(1) w.r.t. priority handling.
		3332
		3333	=item Sending an ev_async: O(1)
		3334
		3335	=item Processing ev_async_send: O(number_of_async_watchers)
		3336
		3337	=item Processing signals: O(max_signal_number)
		3338
		3339	Sending involves a system call I<iff> there were no other C<ev_async_send>
		3340	calls in the current loop iteration. Checking for async and signal events
		3341	involves iterating over all running async watchers or all signal numbers.
2302		3342
2303	=back	3343	=back
2304		3344
2305		3345
		3346	=head1 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		3347
		3348	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		3349	requires, and its I/O model is fundamentally incompatible with the POSIX
		3350	model. Libev still offers limited functionality on this platform in
		3351	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		3352	descriptors. This only applies when using Win32 natively, not when using
		3353	e.g. cygwin.
		3354
		3355	Lifting these limitations would basically require the full
		3356	re-implementation of the I/O system. If you are into these kinds of
		3357	things, then note that glib does exactly that for you in a very portable
		3358	way (note also that glib is the slowest event library known to man).
		3359
		3360	There is no supported compilation method available on windows except
		3361	embedding it into other applications.
		3362
		3363	Not a libev limitation but worth mentioning: windows apparently doesn't
		3364	accept large writes: instead of resulting in a partial write, windows will
		3365	either accept everything or return C<ENOBUFS> if the buffer is too large,
		3366	so make sure you only write small amounts into your sockets (less than a
		3367	megabyte seems safe, but thsi apparently depends on the amount of memory
		3368	available).
		3369
		3370	Due to the many, low, and arbitrary limits on the win32 platform and
		3371	the abysmal performance of winsockets, using a large number of sockets
		3372	is not recommended (and not reasonable). If your program needs to use
		3373	more than a hundred or so sockets, then likely it needs to use a totally
		3374	different implementation for windows, as libev offers the POSIX readiness
		3375	notification model, which cannot be implemented efficiently on windows
		3376	(Microsoft monopoly games).
		3377
		3378	A typical way to use libev under windows is to embed it (see the embedding
		3379	section for details) and use the following F<evwrap.h> header file instead
		3380	of F<ev.h>:
		3381
		3382	#define EV_STANDALONE /* keeps ev from requiring config.h */
		3383	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		3384
		3385	#include "ev.h"
		3386
		3387	And compile the following F<evwrap.c> file into your project (make sure
		3388	you do I<not> compile the F<ev.c> or any other embedded soruce files!):
		3389
		3390	#include "evwrap.h"
		3391	#include "ev.c"
		3392
		3393	=over 4
		3394
		3395	=item The winsocket select function
		3396
		3397	The winsocket C<select> function doesn't follow POSIX in that it
		3398	requires socket I<handles> and not socket I<file descriptors> (it is
		3399	also extremely buggy). This makes select very inefficient, and also
		3400	requires a mapping from file descriptors to socket handles (the Microsoft
		3401	C runtime provides the function C<_open_osfhandle> for this). See the
		3402	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		3403	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
		3404
		3405	The configuration for a "naked" win32 using the Microsoft runtime
		3406	libraries and raw winsocket select is:
		3407
		3408	#define EV_USE_SELECT 1
		3409	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		3410
		3411	Note that winsockets handling of fd sets is O(n), so you can easily get a
		3412	complexity in the O(n²) range when using win32.
		3413
		3414	=item Limited number of file descriptors
		3415
		3416	Windows has numerous arbitrary (and low) limits on things.
		3417
		3418	Early versions of winsocket's select only supported waiting for a maximum
		3419	of C<64> handles (probably owning to the fact that all windows kernels
		3420	can only wait for C<64> things at the same time internally; Microsoft
		3421	recommends spawning a chain of threads and wait for 63 handles and the
		3422	previous thread in each. Great).
		3423
		3424	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		3425	to some high number (e.g. C<2048>) before compiling the winsocket select
		3426	call (which might be in libev or elsewhere, for example, perl does its own
		3427	select emulation on windows).
		3428
		3429	Another limit is the number of file descriptors in the Microsoft runtime
		3430	libraries, which by default is C<64> (there must be a hidden I<64> fetish
		3431	or something like this inside Microsoft). You can increase this by calling
		3432	C<_setmaxstdio>, which can increase this limit to C<2048> (another
		3433	arbitrary limit), but is broken in many versions of the Microsoft runtime
		3434	libraries.
		3435
		3436	This might get you to about C<512> or C<2048> sockets (depending on
		3437	windows version and/or the phase of the moon). To get more, you need to
		3438	wrap all I/O functions and provide your own fd management, but the cost of
		3439	calling select (O(n²)) will likely make this unworkable.
		3440
		3441	=back
		3442
		3443
		3444	=head1 PORTABILITY REQUIREMENTS
		3445
		3446	In addition to a working ISO-C implementation, libev relies on a few
		3447	additional extensions:
		3448
		3449	=over 4
		3450
		3451	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		3452	calling conventions regardless of C<ev_watcher_type *>.
		3453
		3454	Libev assumes not only that all watcher pointers have the same internal
		3455	structure (guaranteed by POSIX but not by ISO C for example), but it also
		3456	assumes that the same (machine) code can be used to call any watcher
		3457	callback: The watcher callbacks have different type signatures, but libev
		3458	calls them using an C<ev_watcher *> internally.
		3459
		3460	=item C<sig_atomic_t volatile> must be thread-atomic as well
		3461
		3462	The type C<sig_atomic_t volatile> (or whatever is defined as
		3463	C<EV_ATOMIC_T>) must be atomic w.r.t. accesses from different
		3464	threads. This is not part of the specification for C<sig_atomic_t>, but is
		3465	believed to be sufficiently portable.
		3466
		3467	=item C<sigprocmask> must work in a threaded environment
		3468
		3469	Libev uses C<sigprocmask> to temporarily block signals. This is not
		3470	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		3471	pthread implementations will either allow C<sigprocmask> in the "main
		3472	thread" or will block signals process-wide, both behaviours would
		3473	be compatible with libev. Interaction between C<sigprocmask> and
		3474	C<pthread_sigmask> could complicate things, however.
		3475
		3476	The most portable way to handle signals is to block signals in all threads
		3477	except the initial one, and run the default loop in the initial thread as
		3478	well.
		3479
		3480	=item C<long> must be large enough for common memory allocation sizes
		3481
		3482	To improve portability and simplify using libev, libev uses C<long>
		3483	internally instead of C<size_t> when allocating its data structures. On
		3484	non-POSIX systems (Microsoft...) this might be unexpectedly low, but
		3485	is still at least 31 bits everywhere, which is enough for hundreds of
		3486	millions of watchers.
		3487
		3488	=item C<double> must hold a time value in seconds with enough accuracy
		3489
		3490	The type C<double> is used to represent timestamps. It is required to
		3491	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		3492	enough for at least into the year 4000. This requirement is fulfilled by
		3493	implementations implementing IEEE 754 (basically all existing ones).
		3494
		3495	=back
		3496
		3497	If you know of other additional requirements drop me a note.
		3498
		3499
		3500	=head1 COMPILER WARNINGS
		3501
		3502	Depending on your compiler and compiler settings, you might get no or a
		3503	lot of warnings when compiling libev code. Some people are apparently
		3504	scared by this.
		3505
		3506	However, these are unavoidable for many reasons. For one, each compiler
		3507	has different warnings, and each user has different tastes regarding
		3508	warning options. "Warn-free" code therefore cannot be a goal except when
		3509	targeting a specific compiler and compiler-version.
		3510
		3511	Another reason is that some compiler warnings require elaborate
		3512	workarounds, or other changes to the code that make it less clear and less
		3513	maintainable.
		3514
		3515	And of course, some compiler warnings are just plain stupid, or simply
		3516	wrong (because they don't actually warn about the condition their message
		3517	seems to warn about).
		3518
		3519	While libev is written to generate as few warnings as possible,
		3520	"warn-free" code is not a goal, and it is recommended not to build libev
		3521	with any compiler warnings enabled unless you are prepared to cope with
		3522	them (e.g. by ignoring them). Remember that warnings are just that:
		3523	warnings, not errors, or proof of bugs.
		3524
		3525
		3526	=head1 VALGRIND
		3527
		3528	Valgrind has a special section here because it is a popular tool that is
		3529	highly useful, but valgrind reports are very hard to interpret.
		3530
		3531	If you think you found a bug (memory leak, uninitialised data access etc.)
		3532	in libev, then check twice: If valgrind reports something like:
		3533
		3534	==2274== definitely lost: 0 bytes in 0 blocks.
		3535	==2274== possibly lost: 0 bytes in 0 blocks.
		3536	==2274== still reachable: 256 bytes in 1 blocks.
		3537
		3538	Then there is no memory leak. Similarly, under some circumstances,
		3539	valgrind might report kernel bugs as if it were a bug in libev, or it
		3540	might be confused (it is a very good tool, but only a tool).
		3541
		3542	If you are unsure about something, feel free to contact the mailing list
		3543	with the full valgrind report and an explanation on why you think this is
		3544	a bug in libev. However, don't be annoyed when you get a brisk "this is
		3545	no bug" answer and take the chance of learning how to interpret valgrind
		3546	properly.
		3547
		3548	If you need, for some reason, empty reports from valgrind for your project
		3549	I suggest using suppression lists.
		3550
		3551
2306	=head1 AUTHOR	3552	=head1 AUTHOR
2307		3553
2308	Marc Lehmann <libev@schmorp.de>.	3554	Marc Lehmann <libev@schmorp.de>.
2309		3555

Diff Legend

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