ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.115 by root, Mon Dec 31 01:32:59 2007 UTC vs.
Revision 1.196 by root, Tue Oct 21 20:04:14 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	=head2 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 occurring), 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
…		…
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	=head2 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	=head2 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 some floatingpoint value. Unlike the name	118	it, you should treat it as some floating point value. Unlike the name
104	component C<stamp> might indicate, it is also used for time differences	119	component C<stamp> might indicate, it is also used for time differences
105	throughout libev.	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
106		142
107	=head1 GLOBAL FUNCTIONS	143	=head1 GLOBAL FUNCTIONS
108		144
109	These functions can be called anytime, even before initialising the	145	These functions can be called anytime, even before initialising the
110	library in any way.	146	library in any way.
…		…
119		155
120	=item ev_sleep (ev_tstamp interval)	156	=item ev_sleep (ev_tstamp interval)
121		157
122	Sleep for the given interval: The current thread will be blocked until	158	Sleep for the given interval: The current thread will be blocked until
123	either it is interrupted or the given time interval has passed. Basically	159	either it is interrupted or the given time interval has passed. Basically
124	this is a subsecond-resolution C<sleep ()>.	160	this is a sub-second-resolution C<sleep ()>.
125		161
126	=item int ev_version_major ()	162	=item int ev_version_major ()
127		163
128	=item int ev_version_minor ()	164	=item int ev_version_minor ()
129		165
…		…
142	not a problem.	178	not a problem.
143		179
144	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
145	version.	181	version.
146		182
147	assert (("libev version mismatch",	183	assert (("libev version mismatch",
148	ev_version_major () == EV_VERSION_MAJOR	184	ev_version_major () == EV_VERSION_MAJOR
149	&& ev_version_minor () >= EV_VERSION_MINOR));	185	&& ev_version_minor () >= EV_VERSION_MINOR));
150		186
151	=item unsigned int ev_supported_backends ()	187	=item unsigned int ev_supported_backends ()
152		188
153	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_*>
154	value) compiled into this binary of libev (independent of their	190	value) compiled into this binary of libev (independent of their
…		…
156	a description of the set values.	192	a description of the set values.
157		193
158	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
159	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
160		196
161	assert (("sorry, no epoll, no sex",	197	assert (("sorry, no epoll, no sex",
162	ev_supported_backends () & EVBACKEND_EPOLL));	198	ev_supported_backends () & EVBACKEND_EPOLL));
163		199
164	=item unsigned int ev_recommended_backends ()	200	=item unsigned int ev_recommended_backends ()
165		201
166	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
167	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
168	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
169	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
170	(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
171	libev will probe for if you specify no backends explicitly.	207	libev will probe for if you specify no backends explicitly.
172		208
173	=item unsigned int ev_embeddable_backends ()	209	=item unsigned int ev_embeddable_backends ()
174		210
…		…
178	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
179	recommended ones.	215	recommended ones.
180		216
181	See the description of C<ev_embed> watchers for more info.	217	See the description of C<ev_embed> watchers for more info.
182		218
183	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	219	=item ev_set_allocator (void *(*cb)(void *ptr, long size)) [NOT REENTRANT]
184		220
185	Sets the allocation function to use (the prototype is similar - the	221	Sets the allocation function to use (the prototype is similar - the
186	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
187	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
188	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
189	potentially destructive action. The default is your system realloc	225	or take some potentially destructive action.
190	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.
191		230
192	You could override this function in high-availability programs to, say,	231	You could override this function in high-availability programs to, say,
193	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,
194	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.
195		234
196	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
197	retries).	236	retries (example requires a standards-compliant C<realloc>).
198		237
199	static void *	238	static void *
200	persistent_realloc (void *ptr, size_t size)	239	persistent_realloc (void *ptr, size_t size)
201	{	240	{
202	for (;;)	241	for (;;)
…		…
211	}	250	}
212		251
213	...	252	...
214	ev_set_allocator (persistent_realloc);	253	ev_set_allocator (persistent_realloc);
215		254
216	=item ev_set_syserr_cb (void (*cb)(const char *msg));	255	=item ev_set_syserr_cb (void (*cb)(const char *msg)); [NOT REENTRANT]
217		256
218	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
219	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
220	indicating the system call or subsystem causing the problem. If this	259	indicating the system call or subsystem causing the problem. If this
221	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
222	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
223	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
224	(such as abort).	263	(such as abort).
225		264
226	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.
…		…
241		280
242	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
243	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
244	events, and dynamically created loops which do not.	283	events, and dynamically created loops which do not.
245		284
246	If you use threads, a common model is to run the default event loop
247	in your main thread (or in a separate thread) and for each thread you
248	create, you also create another event loop. Libev itself does no locking
249	whatsoever, so if you mix calls to the same event loop in different
250	threads, make sure you lock (this is usually a bad idea, though, even if
251	done correctly, because it's hideous and inefficient).
252
253	=over 4	285	=over 4
254		286
255	=item struct ev_loop *ev_default_loop (unsigned int flags)	287	=item struct ev_loop *ev_default_loop (unsigned int flags)
256		288
257	This will initialise the default event loop if it hasn't been initialised	289	This will initialise the default event loop if it hasn't been initialised
…		…
260	flags. If that is troubling you, check C<ev_backend ()> afterwards).	292	flags. If that is troubling you, check C<ev_backend ()> afterwards).
261		293
262	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
263	function.	295	function.
264		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
265	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
266	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>).
267		310
268	The following flags are supported:	311	The following flags are supported:
269		312
…		…
274	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
275	thing, believe me).	318	thing, believe me).
276		319
277	=item C<EVFLAG_NOENV>	320	=item C<EVFLAG_NOENV>
278		321
279	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
280	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
281	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	324	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
282	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
283	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
284	around bugs.	327	around bugs.
…		…
290	enabling this flag.	333	enabling this flag.
291		334
292	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,
293	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
294	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
295	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
296	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
297	C<pthread_atfork> which is even faster).	340	C<pthread_atfork> which is even faster).
298		341
299	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
300	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
301	flag.	344	flag.
302		345
303	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>
304	environment variable.	347	environment variable.
305		348
306	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	349	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
307		350
308	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
…		…
310	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
311	using this backend. It doesn't scale too well (O(highest_fd)), but its	354	using this backend. It doesn't scale too well (O(highest_fd)), but its
312	usually the fastest backend for a low number of (low-numbered :) fds.	355	usually the fastest backend for a low number of (low-numbered :) fds.
313		356
314	To get good performance out of this backend you need a high amount of	357	To get good performance out of this backend you need a high amount of
315	parallelity (most of the file descriptors should be busy). If you are	358	parallelism (most of the file descriptors should be busy). If you are
316	writing a server, you should C<accept ()> in a loop to accept as many	359	writing a server, you should C<accept ()> in a loop to accept as many
317	connections as possible during one iteration. You might also want to have	360	connections as possible during one iteration. You might also want to have
318	a look at C<ev_set_io_collect_interval ()> to increase the amount of	361	a look at C<ev_set_io_collect_interval ()> to increase the amount of
319	readyness notifications you get per iteration.	362	readiness notifications you get per iteration.
		363
		364	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		365	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		366	C<exceptfds> set on that platform).
320		367
321	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	368	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
322		369
323	And this is your standard poll(2) backend. It's more complicated	370	And this is your standard poll(2) backend. It's more complicated
324	than select, but handles sparse fds better and has no artificial	371	than select, but handles sparse fds better and has no artificial
325	limit on the number of fds you can use (except it will slow down	372	limit on the number of fds you can use (except it will slow down
326	considerably with a lot of inactive fds). It scales similarly to select,	373	considerably with a lot of inactive fds). It scales similarly to select,
327	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for	374	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
328	performance tips.	375	performance tips.
329		376
		377	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		378	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
		379
330	=item C<EVBACKEND_EPOLL> (value 4, Linux)	380	=item C<EVBACKEND_EPOLL> (value 4, Linux)
331		381
332	For few fds, this backend is a bit little slower than poll and select,	382	For few fds, this backend is a bit little slower than poll and select,
333	but it scales phenomenally better. While poll and select usually scale	383	but it scales phenomenally better. While poll and select usually scale
334	like O(total_fds) where n is the total number of fds (or the highest fd),	384	like O(total_fds) where n is the total number of fds (or the highest fd),
335	epoll scales either O(1) or O(active_fds). The epoll design has a number	385	epoll scales either O(1) or O(active_fds). The epoll design has a number
336	of shortcomings, such as silently dropping events in some hard-to-detect	386	of shortcomings, such as silently dropping events in some hard-to-detect
337	cases and rewiring a syscall per fd change, no fork support and bad	387	cases and requiring a system call per fd change, no fork support and bad
338	support for dup.	388	support for dup.
339		389
340	While stopping, setting and starting an I/O watcher in the same iteration	390	While stopping, setting and starting an I/O watcher in the same iteration
341	will result in some caching, there is still a syscall per such incident	391	will result in some caching, there is still a system call per such incident
342	(because the fd could point to a different file description now), so its	392	(because the fd could point to a different file description now), so its
343	best to avoid that. Also, C<dup ()>'ed file descriptors might not work	393	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
344	very well if you register events for both fds.	394	very well if you register events for both fds.
345		395
346	Please note that epoll sometimes generates spurious notifications, so you	396	Please note that epoll sometimes generates spurious notifications, so you
347	need to use non-blocking I/O or other means to avoid blocking when no data	397	need to use non-blocking I/O or other means to avoid blocking when no data
348	(or space) is available.	398	(or space) is available.
349		399
350	Best performance from this backend is achieved by not unregistering all	400	Best performance from this backend is achieved by not unregistering all
351	watchers for a file descriptor until it has been closed, if possible, i.e.	401	watchers for a file descriptor until it has been closed, if possible,
352	keep at least one watcher active per fd at all times.	402	i.e. keep at least one watcher active per fd at all times. Stopping and
		403	starting a watcher (without re-setting it) also usually doesn't cause
		404	extra overhead.
353		405
354	While nominally embeddeble in other event loops, this feature is broken in	406	While nominally embeddable in other event loops, this feature is broken in
355	all kernel versions tested so far.	407	all kernel versions tested so far.
356		408
		409	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		410	C<EVBACKEND_POLL>.
		411
357	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	412	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
358		413
359	Kqueue deserves special mention, as at the time of this writing, it	414	Kqueue deserves special mention, as at the time of this writing, it was
360	was broken on all BSDs except NetBSD (usually it doesn't work reliably	415	broken on all BSDs except NetBSD (usually it doesn't work reliably with
361	with anything but sockets and pipes, except on Darwin, where of course	416	anything but sockets and pipes, except on Darwin, where of course it's
362	it's completely useless). For this reason it's not being "autodetected"	417	completely useless). For this reason it's not being "auto-detected" unless
363	unless you explicitly specify it explicitly in the flags (i.e. using	418	you explicitly specify it in the flags (i.e. using C<EVBACKEND_KQUEUE>) or
364	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)	419	libev was compiled on a known-to-be-good (-enough) system like NetBSD.
365	system like NetBSD.
366		420
367	You still can embed kqueue into a normal poll or select backend and use it	421	You still can embed kqueue into a normal poll or select backend and use it
368	only for sockets (after having made sure that sockets work with kqueue on	422	only for sockets (after having made sure that sockets work with kqueue on
369	the target platform). See C<ev_embed> watchers for more info.	423	the target platform). See C<ev_embed> watchers for more info.
370		424
371	It scales in the same way as the epoll backend, but the interface to the	425	It scales in the same way as the epoll backend, but the interface to the
372	kernel is more efficient (which says nothing about its actual speed, of	426	kernel is more efficient (which says nothing about its actual speed, of
373	course). While stopping, setting and starting an I/O watcher does never	427	course). While stopping, setting and starting an I/O watcher does never
374	cause an extra syscall as with C<EVBACKEND_EPOLL>, it still adds up to	428	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
375	two event changes per incident, support for C<fork ()> is very bad and it	429	two event changes per incident. Support for C<fork ()> is very bad and it
376	drops fds silently in similarly hard-to-detect cases.	430	drops fds silently in similarly hard-to-detect cases.
377		431
378	This backend usually performs well under most conditions.	432	This backend usually performs well under most conditions.
379		433
380	While nominally embeddable in other event loops, this doesn't work	434	While nominally embeddable in other event loops, this doesn't work
381	everywhere, so you might need to test for this. And since it is broken	435	everywhere, so you might need to test for this. And since it is broken
382	almost everywhere, you should only use it when you have a lot of sockets	436	almost everywhere, you should only use it when you have a lot of sockets
383	(for which it usually works), by embedding it into another event loop	437	(for which it usually works), by embedding it into another event loop
384	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for	438	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and, did I mention it,
385	sockets.	439	using it only for sockets.
		440
		441	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		442	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		443	C<NOTE_EOF>.
386		444
387	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	445	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
388		446
389	This is not implemented yet (and might never be, unless you send me an	447	This is not implemented yet (and might never be, unless you send me an
390	implementation). According to reports, C</dev/poll> only supports sockets	448	implementation). According to reports, C</dev/poll> only supports sockets
…		…
394	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	452	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
395		453
396	This uses the Solaris 10 event port mechanism. As with everything on Solaris,	454	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
397	it's really slow, but it still scales very well (O(active_fds)).	455	it's really slow, but it still scales very well (O(active_fds)).
398		456
399	Please note that solaris event ports can deliver a lot of spurious	457	Please note that Solaris event ports can deliver a lot of spurious
400	notifications, so you need to use non-blocking I/O or other means to avoid	458	notifications, so you need to use non-blocking I/O or other means to avoid
401	blocking when no data (or space) is available.	459	blocking when no data (or space) is available.
402		460
403	While this backend scales well, it requires one system call per active	461	While this backend scales well, it requires one system call per active
404	file descriptor per loop iteration. For small and medium numbers of file	462	file descriptor per loop iteration. For small and medium numbers of file
405	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	463	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
406	might perform better.	464	might perform better.
407		465
		466	On the positive side, with the exception of the spurious readiness
		467	notifications, this backend actually performed fully to specification
		468	in all tests and is fully embeddable, which is a rare feat among the
		469	OS-specific backends.
		470
		471	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		472	C<EVBACKEND_POLL>.
		473
408	=item C<EVBACKEND_ALL>	474	=item C<EVBACKEND_ALL>
409		475
410	Try all backends (even potentially broken ones that wouldn't be tried	476	Try all backends (even potentially broken ones that wouldn't be tried
411	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	477	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
412	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	478	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
413		479
414	It is definitely not recommended to use this flag.	480	It is definitely not recommended to use this flag.
415		481
416	=back	482	=back
417		483
418	If one or more of these are ored into the flags value, then only these	484	If one or more of these are or'ed into the flags value, then only these
419	backends will be tried (in the reverse order as given here). If none are	485	backends will be tried (in the reverse order as listed here). If none are
420	specified, most compiled-in backend will be tried, usually in reverse	486	specified, all backends in C<ev_recommended_backends ()> will be tried.
421	order of their flag values :)
422		487
423	The most typical usage is like this:	488	Example: This is the most typical usage.
424		489
425	if (!ev_default_loop (0))	490	if (!ev_default_loop (0))
426	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	491	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
427		492
428	Restrict libev to the select and poll backends, and do not allow	493	Example: Restrict libev to the select and poll backends, and do not allow
429	environment settings to be taken into account:	494	environment settings to be taken into account:
430		495
431	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);	496	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
432		497
433	Use whatever libev has to offer, but make sure that kqueue is used if	498	Example: Use whatever libev has to offer, but make sure that kqueue is
434	available (warning, breaks stuff, best use only with your own private	499	used if available (warning, breaks stuff, best use only with your own
435	event loop and only if you know the OS supports your types of fds):	500	private event loop and only if you know the OS supports your types of
		501	fds):
436		502
437	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);	503	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
438		504
439	=item struct ev_loop *ev_loop_new (unsigned int flags)	505	=item struct ev_loop *ev_loop_new (unsigned int flags)
440		506
441	Similar to C<ev_default_loop>, but always creates a new event loop that is	507	Similar to C<ev_default_loop>, but always creates a new event loop that is
442	always distinct from the default loop. Unlike the default loop, it cannot	508	always distinct from the default loop. Unlike the default loop, it cannot
443	handle signal and child watchers, and attempts to do so will be greeted by	509	handle signal and child watchers, and attempts to do so will be greeted by
444	undefined behaviour (or a failed assertion if assertions are enabled).	510	undefined behaviour (or a failed assertion if assertions are enabled).
445		511
		512	Note that this function I<is> thread-safe, and the recommended way to use
		513	libev with threads is indeed to create one loop per thread, and using the
		514	default loop in the "main" or "initial" thread.
		515
446	Example: Try to create a event loop that uses epoll and nothing else.	516	Example: Try to create a event loop that uses epoll and nothing else.
447		517
448	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	518	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
449	if (!epoller)	519	if (!epoller)
450	fatal ("no epoll found here, maybe it hides under your chair");	520	fatal ("no epoll found here, maybe it hides under your chair");
451		521
452	=item ev_default_destroy ()	522	=item ev_default_destroy ()
453		523
454	Destroys the default loop again (frees all memory and kernel state	524	Destroys the default loop again (frees all memory and kernel state
455	etc.). None of the active event watchers will be stopped in the normal	525	etc.). None of the active event watchers will be stopped in the normal
456	sense, so e.g. C<ev_is_active> might still return true. It is your	526	sense, so e.g. C<ev_is_active> might still return true. It is your
457	responsibility to either stop all watchers cleanly yoursef I<before>	527	responsibility to either stop all watchers cleanly yourself I<before>
458	calling this function, or cope with the fact afterwards (which is usually	528	calling this function, or cope with the fact afterwards (which is usually
459	the easiest thing, you can just ignore the watchers and/or C<free ()> them	529	the easiest thing, you can just ignore the watchers and/or C<free ()> them
460	for example).	530	for example).
461		531
462	Note that certain global state, such as signal state, will not be freed by	532	Note that certain global state, such as signal state, will not be freed by
…		…
473	Like C<ev_default_destroy>, but destroys an event loop created by an	543	Like C<ev_default_destroy>, but destroys an event loop created by an
474	earlier call to C<ev_loop_new>.	544	earlier call to C<ev_loop_new>.
475		545
476	=item ev_default_fork ()	546	=item ev_default_fork ()
477		547
		548	This function sets a flag that causes subsequent C<ev_loop> iterations
478	This function reinitialises the kernel state for backends that have	549	to reinitialise the kernel state for backends that have one. Despite the
479	one. Despite the name, you can call it anytime, but it makes most sense	550	name, you can call it anytime, but it makes most sense after forking, in
480	after forking, in either the parent or child process (or both, but that	551	the child process (or both child and parent, but that again makes little
481	again makes little sense).	552	sense). You I<must> call it in the child before using any of the libev
		553	functions, and it will only take effect at the next C<ev_loop> iteration.
482		554
483	You I<must> call this function in the child process after forking if and	555	On the other hand, you only need to call this function in the child
484	only if you want to use the event library in both processes. If you just	556	process if and only if you want to use the event library in the child. If
485	fork+exec, you don't have to call it.	557	you just fork+exec, you don't have to call it at all.
486		558
487	The function itself is quite fast and it's usually not a problem to call	559	The function itself is quite fast and it's usually not a problem to call
488	it just in case after a fork. To make this easy, the function will fit in	560	it just in case after a fork. To make this easy, the function will fit in
489	quite nicely into a call to C<pthread_atfork>:	561	quite nicely into a call to C<pthread_atfork>:
490		562
491	pthread_atfork (0, 0, ev_default_fork);	563	pthread_atfork (0, 0, ev_default_fork);
492		564
493	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
494	without calling this function, so if you force one of those backends you
495	do not need to care.
496
497	=item ev_loop_fork (loop)	565	=item ev_loop_fork (loop)
498		566
499	Like C<ev_default_fork>, but acts on an event loop created by	567	Like C<ev_default_fork>, but acts on an event loop created by
500	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	568	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
501	after fork, and how you do this is entirely your own problem.	569	after fork that you want to re-use in the child, and how you do this is
		570	entirely your own problem.
		571
		572	=item int ev_is_default_loop (loop)
		573
		574	Returns true when the given loop is, in fact, the default loop, and false
		575	otherwise.
502		576
503	=item unsigned int ev_loop_count (loop)	577	=item unsigned int ev_loop_count (loop)
504		578
505	Returns the count of loop iterations for the loop, which is identical to	579	Returns the count of loop iterations for the loop, which is identical to
506	the number of times libev did poll for new events. It starts at C<0> and	580	the number of times libev did poll for new events. It starts at C<0> and
…		…
521	received events and started processing them. This timestamp does not	595	received events and started processing them. This timestamp does not
522	change as long as callbacks are being processed, and this is also the base	596	change as long as callbacks are being processed, and this is also the base
523	time used for relative timers. You can treat it as the timestamp of the	597	time used for relative timers. You can treat it as the timestamp of the
524	event occurring (or more correctly, libev finding out about it).	598	event occurring (or more correctly, libev finding out about it).
525		599
		600	=item ev_now_update (loop)
		601
		602	Establishes the current time by querying the kernel, updating the time
		603	returned by C<ev_now ()> in the progress. This is a costly operation and
		604	is usually done automatically within C<ev_loop ()>.
		605
		606	This function is rarely useful, but when some event callback runs for a
		607	very long time without entering the event loop, updating libev's idea of
		608	the current time is a good idea.
		609
		610	See also "The special problem of time updates" in the C<ev_timer> section.
		611
526	=item ev_loop (loop, int flags)	612	=item ev_loop (loop, int flags)
527		613
528	Finally, this is it, the event handler. This function usually is called	614	Finally, this is it, the event handler. This function usually is called
529	after you initialised all your watchers and you want to start handling	615	after you initialised all your watchers and you want to start handling
530	events.	616	events.
…		…
532	If the flags argument is specified as C<0>, it will not return until	618	If the flags argument is specified as C<0>, it will not return until
533	either no event watchers are active anymore or C<ev_unloop> was called.	619	either no event watchers are active anymore or C<ev_unloop> was called.
534		620
535	Please note that an explicit C<ev_unloop> is usually better than	621	Please note that an explicit C<ev_unloop> is usually better than
536	relying on all watchers to be stopped when deciding when a program has	622	relying on all watchers to be stopped when deciding when a program has
537	finished (especially in interactive programs), but having a program that	623	finished (especially in interactive programs), but having a program
538	automatically loops as long as it has to and no longer by virtue of	624	that automatically loops as long as it has to and no longer by virtue
539	relying on its watchers stopping correctly is a thing of beauty.	625	of relying on its watchers stopping correctly, that is truly a thing of
		626	beauty.
540		627
541	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	628	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
542	those events and any outstanding ones, but will not block your process in	629	those events and any already outstanding ones, but will not block your
543	case there are no events and will return after one iteration of the loop.	630	process in case there are no events and will return after one iteration of
		631	the loop.
544		632
545	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	633	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
546	neccessary) and will handle those and any outstanding ones. It will block	634	necessary) and will handle those and any already outstanding ones. It
547	your process until at least one new event arrives, and will return after	635	will block your process until at least one new event arrives (which could
548	one iteration of the loop. This is useful if you are waiting for some	636	be an event internal to libev itself, so there is no guarentee that a
549	external event in conjunction with something not expressible using other	637	user-registered callback will be called), and will return after one
		638	iteration of the loop.
		639
		640	This is useful if you are waiting for some external event in conjunction
		641	with something not expressible using other libev watchers (i.e. "roll your
550	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	642	own C<ev_loop>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
551	usually a better approach for this kind of thing.	643	usually a better approach for this kind of thing.
552		644
553	Here are the gory details of what C<ev_loop> does:	645	Here are the gory details of what C<ev_loop> does:
554		646
555	- Before the first iteration, call any pending watchers.	647	- Before the first iteration, call any pending watchers.
556	* If EVFLAG_FORKCHECK was used, check for a fork.	648	* If EVFLAG_FORKCHECK was used, check for a fork.
557	- If a fork was detected, queue and call all fork watchers.	649	- If a fork was detected (by any means), queue and call all fork watchers.
558	- Queue and call all prepare watchers.	650	- Queue and call all prepare watchers.
559	- If we have been forked, recreate the kernel state.	651	- If we have been forked, detach and recreate the kernel state
		652	as to not disturb the other process.
560	- Update the kernel state with all outstanding changes.	653	- Update the kernel state with all outstanding changes.
561	- Update the "event loop time".	654	- Update the "event loop time" (ev_now ()).
562	- Calculate for how long to sleep or block, if at all	655	- Calculate for how long to sleep or block, if at all
563	(active idle watchers, EVLOOP_NONBLOCK or not having	656	(active idle watchers, EVLOOP_NONBLOCK or not having
564	any active watchers at all will result in not sleeping).	657	any active watchers at all will result in not sleeping).
565	- Sleep if the I/O and timer collect interval say so.	658	- Sleep if the I/O and timer collect interval say so.
566	- Block the process, waiting for any events.	659	- Block the process, waiting for any events.
567	- Queue all outstanding I/O (fd) events.	660	- Queue all outstanding I/O (fd) events.
568	- Update the "event loop time" and do time jump handling.	661	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
569	- Queue all outstanding timers.	662	- Queue all expired timers.
570	- Queue all outstanding periodics.	663	- Queue all expired periodics.
571	- If no events are pending now, queue all idle watchers.	664	- Unless any events are pending now, queue all idle watchers.
572	- Queue all check watchers.	665	- Queue all check watchers.
573	- Call all queued watchers in reverse order (i.e. check watchers first).	666	- Call all queued watchers in reverse order (i.e. check watchers first).
574	Signals and child watchers are implemented as I/O watchers, and will	667	Signals and child watchers are implemented as I/O watchers, and will
575	be handled here by queueing them when their watcher gets executed.	668	be handled here by queueing them when their watcher gets executed.
576	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	669	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
…		…
581	anymore.	674	anymore.
582		675
583	... queue jobs here, make sure they register event watchers as long	676	... queue jobs here, make sure they register event watchers as long
584	... as they still have work to do (even an idle watcher will do..)	677	... as they still have work to do (even an idle watcher will do..)
585	ev_loop (my_loop, 0);	678	ev_loop (my_loop, 0);
586	... jobs done. yeah!	679	... jobs done or somebody called unloop. yeah!
587		680
588	=item ev_unloop (loop, how)	681	=item ev_unloop (loop, how)
589		682
590	Can be used to make a call to C<ev_loop> return early (but only after it	683	Can be used to make a call to C<ev_loop> return early (but only after it
591	has processed all outstanding events). The C<how> argument must be either	684	has processed all outstanding events). The C<how> argument must be either
592	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	685	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
593	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	686	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
594		687
595	This "unloop state" will be cleared when entering C<ev_loop> again.	688	This "unloop state" will be cleared when entering C<ev_loop> again.
596		689
		690	It is safe to call C<ev_unloop> from otuside any C<ev_loop> calls.
		691
597	=item ev_ref (loop)	692	=item ev_ref (loop)
598		693
599	=item ev_unref (loop)	694	=item ev_unref (loop)
600		695
601	Ref/unref can be used to add or remove a reference count on the event	696	Ref/unref can be used to add or remove a reference count on the event
602	loop: Every watcher keeps one reference, and as long as the reference	697	loop: Every watcher keeps one reference, and as long as the reference
603	count is nonzero, C<ev_loop> will not return on its own. If you have	698	count is nonzero, C<ev_loop> will not return on its own.
		699
604	a watcher you never unregister that should not keep C<ev_loop> from	700	If you have a watcher you never unregister that should not keep C<ev_loop>
605	returning, ev_unref() after starting, and ev_ref() before stopping it. For	701	from returning, call ev_unref() after starting, and ev_ref() before
		702	stopping it.
		703
606	example, libev itself uses this for its internal signal pipe: It is not	704	As an example, libev itself uses this for its internal signal pipe: It is
607	visible to the libev user and should not keep C<ev_loop> from exiting if	705	not visible to the libev user and should not keep C<ev_loop> from exiting
608	no event watchers registered by it are active. It is also an excellent	706	if no event watchers registered by it are active. It is also an excellent
609	way to do this for generic recurring timers or from within third-party	707	way to do this for generic recurring timers or from within third-party
610	libraries. Just remember to I<unref after start> and I<ref before stop>.	708	libraries. Just remember to I<unref after start> and I<ref before stop>
		709	(but only if the watcher wasn't active before, or was active before,
		710	respectively).
611		711
612	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	712	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
613	running when nothing else is active.	713	running when nothing else is active.
614		714
615	struct ev_signal exitsig;	715	struct ev_signal exitsig;
616	ev_signal_init (&exitsig, sig_cb, SIGINT);	716	ev_signal_init (&exitsig, sig_cb, SIGINT);
617	ev_signal_start (loop, &exitsig);	717	ev_signal_start (loop, &exitsig);
618	evf_unref (loop);	718	evf_unref (loop);
619		719
620	Example: For some weird reason, unregister the above signal handler again.	720	Example: For some weird reason, unregister the above signal handler again.
621		721
622	ev_ref (loop);	722	ev_ref (loop);
623	ev_signal_stop (loop, &exitsig);	723	ev_signal_stop (loop, &exitsig);
624		724
625	=item ev_set_io_collect_interval (loop, ev_tstamp interval)	725	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
626		726
627	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)	727	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
628		728
629	These advanced functions influence the time that libev will spend waiting	729	These advanced functions influence the time that libev will spend waiting
630	for events. Both are by default C<0>, meaning that libev will try to	730	for events. Both time intervals are by default C<0>, meaning that libev
631	invoke timer/periodic callbacks and I/O callbacks with minimum latency.	731	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		732	latency.
632		733
633	Setting these to a higher value (the C<interval> I<must> be >= C<0>)	734	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
634	allows libev to delay invocation of I/O and timer/periodic callbacks to	735	allows libev to delay invocation of I/O and timer/periodic callbacks
635	increase efficiency of loop iterations.	736	to increase efficiency of loop iterations (or to increase power-saving
		737	opportunities).
636		738
637	The background is that sometimes your program runs just fast enough to	739	The idea is that sometimes your program runs just fast enough to handle
638	handle one (or very few) event(s) per loop iteration. While this makes	740	one (or very few) event(s) per loop iteration. While this makes the
639	the program responsive, it also wastes a lot of CPU time to poll for new	741	program responsive, it also wastes a lot of CPU time to poll for new
640	events, especially with backends like C<select ()> which have a high	742	events, especially with backends like C<select ()> which have a high
641	overhead for the actual polling but can deliver many events at once.	743	overhead for the actual polling but can deliver many events at once.
642		744
643	By setting a higher I<io collect interval> you allow libev to spend more	745	By setting a higher I<io collect interval> you allow libev to spend more
644	time collecting I/O events, so you can handle more events per iteration,	746	time collecting I/O events, so you can handle more events per iteration,
…		…
646	C<ev_timer>) will be not affected. Setting this to a non-null value will	748	C<ev_timer>) will be not affected. Setting this to a non-null value will
647	introduce an additional C<ev_sleep ()> call into most loop iterations.	749	introduce an additional C<ev_sleep ()> call into most loop iterations.
648		750
649	Likewise, by setting a higher I<timeout collect interval> you allow libev	751	Likewise, by setting a higher I<timeout collect interval> you allow libev
650	to spend more time collecting timeouts, at the expense of increased	752	to spend more time collecting timeouts, at the expense of increased
651	latency (the watcher callback will be called later). C<ev_io> watchers	753	latency/jitter/inexactness (the watcher callback will be called
652	will not be affected. Setting this to a non-null value will not introduce	754	later). C<ev_io> watchers will not be affected. Setting this to a non-null
653	any overhead in libev.	755	value will not introduce any overhead in libev.
654		756
655	Many (busy) programs can usually benefit by setting the io collect	757	Many (busy) programs can usually benefit by setting the I/O collect
656	interval to a value near C<0.1> or so, which is often enough for	758	interval to a value near C<0.1> or so, which is often enough for
657	interactive servers (of course not for games), likewise for timeouts. It	759	interactive servers (of course not for games), likewise for timeouts. It
658	usually doesn't make much sense to set it to a lower value than C<0.01>,	760	usually doesn't make much sense to set it to a lower value than C<0.01>,
659	as this approsaches the timing granularity of most systems.	761	as this approaches the timing granularity of most systems.
		762
		763	Setting the I<timeout collect interval> can improve the opportunity for
		764	saving power, as the program will "bundle" timer callback invocations that
		765	are "near" in time together, by delaying some, thus reducing the number of
		766	times the process sleeps and wakes up again. Another useful technique to
		767	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		768	they fire on, say, one-second boundaries only.
		769
		770	=item ev_loop_verify (loop)
		771
		772	This function only does something when C<EV_VERIFY> support has been
		773	compiled in. which is the default for non-minimal builds. It tries to go
		774	through all internal structures and checks them for validity. If anything
		775	is found to be inconsistent, it will print an error message to standard
		776	error and call C<abort ()>.
		777
		778	This can be used to catch bugs inside libev itself: under normal
		779	circumstances, this function will never abort as of course libev keeps its
		780	data structures consistent.
660		781
661	=back	782	=back
662		783
663		784
664	=head1 ANATOMY OF A WATCHER	785	=head1 ANATOMY OF A WATCHER
665		786
666	A watcher is a structure that you create and register to record your	787	A watcher is a structure that you create and register to record your
667	interest in some event. For instance, if you want to wait for STDIN to	788	interest in some event. For instance, if you want to wait for STDIN to
668	become readable, you would create an C<ev_io> watcher for that:	789	become readable, you would create an C<ev_io> watcher for that:
669		790
670	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	791	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)
671	{	792	{
672	ev_io_stop (w);	793	ev_io_stop (w);
673	ev_unloop (loop, EVUNLOOP_ALL);	794	ev_unloop (loop, EVUNLOOP_ALL);
674	}	795	}
675		796
676	struct ev_loop *loop = ev_default_loop (0);	797	struct ev_loop *loop = ev_default_loop (0);
677	struct ev_io stdin_watcher;	798	struct ev_io stdin_watcher;
678	ev_init (&stdin_watcher, my_cb);	799	ev_init (&stdin_watcher, my_cb);
679	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	800	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
680	ev_io_start (loop, &stdin_watcher);	801	ev_io_start (loop, &stdin_watcher);
681	ev_loop (loop, 0);	802	ev_loop (loop, 0);
682		803
683	As you can see, you are responsible for allocating the memory for your	804	As you can see, you are responsible for allocating the memory for your
684	watcher structures (and it is usually a bad idea to do this on the stack,	805	watcher structures (and it is usually a bad idea to do this on the stack,
685	although this can sometimes be quite valid).	806	although this can sometimes be quite valid).
686		807
687	Each watcher structure must be initialised by a call to C<ev_init	808	Each watcher structure must be initialised by a call to C<ev_init
688	(watcher *, callback)>, which expects a callback to be provided. This	809	(watcher *, callback)>, which expects a callback to be provided. This
689	callback gets invoked each time the event occurs (or, in the case of io	810	callback gets invoked each time the event occurs (or, in the case of I/O
690	watchers, each time the event loop detects that the file descriptor given	811	watchers, each time the event loop detects that the file descriptor given
691	is readable and/or writable).	812	is readable and/or writable).
692		813
693	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	814	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro
694	with arguments specific to this watcher type. There is also a macro	815	with arguments specific to this watcher type. There is also a macro
…		…
764	=item C<EV_FORK>	885	=item C<EV_FORK>
765		886
766	The event loop has been resumed in the child process after fork (see	887	The event loop has been resumed in the child process after fork (see
767	C<ev_fork>).	888	C<ev_fork>).
768		889
		890	=item C<EV_ASYNC>
		891
		892	The given async watcher has been asynchronously notified (see C<ev_async>).
		893
769	=item C<EV_ERROR>	894	=item C<EV_ERROR>
770		895
771	An unspecified error has occured, the watcher has been stopped. This might	896	An unspecified error has occurred, the watcher has been stopped. This might
772	happen because the watcher could not be properly started because libev	897	happen because the watcher could not be properly started because libev
773	ran out of memory, a file descriptor was found to be closed or any other	898	ran out of memory, a file descriptor was found to be closed or any other
774	problem. You best act on it by reporting the problem and somehow coping	899	problem. You best act on it by reporting the problem and somehow coping
775	with the watcher being stopped.	900	with the watcher being stopped.
776		901
777	Libev will usually signal a few "dummy" events together with an error,	902	Libev will usually signal a few "dummy" events together with an error, for
778	for example it might indicate that a fd is readable or writable, and if	903	example it might indicate that a fd is readable or writable, and if your
779	your callbacks is well-written it can just attempt the operation and cope	904	callbacks is well-written it can just attempt the operation and cope with
780	with the error from read() or write(). This will not work in multithreaded	905	the error from read() or write(). This will not work in multi-threaded
781	programs, though, so beware.	906	programs, though, as the fd could already be closed and reused for another
		907	thing, so beware.
782		908
783	=back	909	=back
784		910
785	=head2 GENERIC WATCHER FUNCTIONS	911	=head2 GENERIC WATCHER FUNCTIONS
786		912
…		…
802	(or never started) and there are no pending events outstanding.	928	(or never started) and there are no pending events outstanding.
803		929
804	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,	930	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,
805	int revents)>.	931	int revents)>.
806		932
		933	Example: Initialise an C<ev_io> watcher in two steps.
		934
		935	ev_io w;
		936	ev_init (&w, my_cb);
		937	ev_io_set (&w, STDIN_FILENO, EV_READ);
		938
807	=item C<ev_TYPE_set> (ev_TYPE *, [args])	939	=item C<ev_TYPE_set> (ev_TYPE *, [args])
808		940
809	This macro initialises the type-specific parts of a watcher. You need to	941	This macro initialises the type-specific parts of a watcher. You need to
810	call C<ev_init> at least once before you call this macro, but you can	942	call C<ev_init> at least once before you call this macro, but you can
811	call C<ev_TYPE_set> any number of times. You must not, however, call this	943	call C<ev_TYPE_set> any number of times. You must not, however, call this
…		…
813	difference to the C<ev_init> macro).	945	difference to the C<ev_init> macro).
814		946
815	Although some watcher types do not have type-specific arguments	947	Although some watcher types do not have type-specific arguments
816	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	948	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
817		949
		950	See C<ev_init>, above, for an example.
		951
818	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	952	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
819		953
820	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	954	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
821	calls into a single call. This is the most convinient method to initialise	955	calls into a single call. This is the most convenient method to initialise
822	a watcher. The same limitations apply, of course.	956	a watcher. The same limitations apply, of course.
		957
		958	Example: Initialise and set an C<ev_io> watcher in one step.
		959
		960	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
823		961
824	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	962	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
825		963
826	Starts (activates) the given watcher. Only active watchers will receive	964	Starts (activates) the given watcher. Only active watchers will receive
827	events. If the watcher is already active nothing will happen.	965	events. If the watcher is already active nothing will happen.
828		966
		967	Example: Start the C<ev_io> watcher that is being abused as example in this
		968	whole section.
		969
		970	ev_io_start (EV_DEFAULT_UC, &w);
		971
829	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)	972	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
830		973
831	Stops the given watcher again (if active) and clears the pending	974	Stops the given watcher if active, and clears the pending status (whether
		975	the watcher was active or not).
		976
832	status. It is possible that stopped watchers are pending (for example,	977	It is possible that stopped watchers are pending - for example,
833	non-repeating timers are being stopped when they become pending), but	978	non-repeating timers are being stopped when they become pending - but
834	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	979	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
835	you want to free or reuse the memory used by the watcher it is therefore a	980	pending. If you want to free or reuse the memory used by the watcher it is
836	good idea to always call its C<ev_TYPE_stop> function.	981	therefore a good idea to always call its C<ev_TYPE_stop> function.
837		982
838	=item bool ev_is_active (ev_TYPE *watcher)	983	=item bool ev_is_active (ev_TYPE *watcher)
839		984
840	Returns a true value iff the watcher is active (i.e. it has been started	985	Returns a true value iff the watcher is active (i.e. it has been started
841	and not yet been stopped). As long as a watcher is active you must not modify	986	and not yet been stopped). As long as a watcher is active you must not modify
…		…
889		1034
890	=item ev_invoke (loop, ev_TYPE *watcher, int revents)	1035	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
891		1036
892	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither	1037	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
893	C<loop> nor C<revents> need to be valid as long as the watcher callback	1038	C<loop> nor C<revents> need to be valid as long as the watcher callback
894	can deal with that fact.	1039	can deal with that fact, as both are simply passed through to the
		1040	callback.
895		1041
896	=item int ev_clear_pending (loop, ev_TYPE *watcher)	1042	=item int ev_clear_pending (loop, ev_TYPE *watcher)
897		1043
898	If the watcher is pending, this function returns clears its pending status	1044	If the watcher is pending, this function clears its pending status and
899	and returns its C<revents> bitset (as if its callback was invoked). If the	1045	returns its C<revents> bitset (as if its callback was invoked). If the
900	watcher isn't pending it does nothing and returns C<0>.	1046	watcher isn't pending it does nothing and returns C<0>.
901		1047
		1048	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1049	callback to be invoked, which can be accomplished with this function.
		1050
902	=back	1051	=back
903		1052
904		1053
905	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1054	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
906		1055
907	Each watcher has, by default, a member C<void *data> that you can change	1056	Each watcher has, by default, a member C<void *data> that you can change
908	and read at any time, libev will completely ignore it. This can be used	1057	and read at any time: libev will completely ignore it. This can be used
909	to associate arbitrary data with your watcher. If you need more data and	1058	to associate arbitrary data with your watcher. If you need more data and
910	don't want to allocate memory and store a pointer to it in that data	1059	don't want to allocate memory and store a pointer to it in that data
911	member, you can also "subclass" the watcher type and provide your own	1060	member, you can also "subclass" the watcher type and provide your own
912	data:	1061	data:
913		1062
914	struct my_io	1063	struct my_io
915	{	1064	{
916	struct ev_io io;	1065	struct ev_io io;
917	int otherfd;	1066	int otherfd;
918	void *somedata;	1067	void *somedata;
919	struct whatever *mostinteresting;	1068	struct whatever *mostinteresting;
920	}	1069	};
		1070
		1071	...
		1072	struct my_io w;
		1073	ev_io_init (&w.io, my_cb, fd, EV_READ);
921		1074
922	And since your callback will be called with a pointer to the watcher, you	1075	And since your callback will be called with a pointer to the watcher, you
923	can cast it back to your own type:	1076	can cast it back to your own type:
924		1077
925	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1078	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)
926	{	1079	{
927	struct my_io *w = (struct my_io *)w_;	1080	struct my_io *w = (struct my_io *)w_;
928	...	1081	...
929	}	1082	}
930		1083
931	More interesting and less C-conformant ways of casting your callback type	1084	More interesting and less C-conformant ways of casting your callback type
932	instead have been omitted.	1085	instead have been omitted.
933		1086
934	Another common scenario is having some data structure with multiple	1087	Another common scenario is to use some data structure with multiple
935	watchers:	1088	embedded watchers:
936		1089
937	struct my_biggy	1090	struct my_biggy
938	{	1091	{
939	int some_data;	1092	int some_data;
940	ev_timer t1;	1093	ev_timer t1;
941	ev_timer t2;	1094	ev_timer t2;
942	}	1095	}
943		1096
944	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1097	In this case getting the pointer to C<my_biggy> is a bit more
945	you need to use C<offsetof>:	1098	complicated: Either you store the address of your C<my_biggy> struct
		1099	in the C<data> member of the watcher (for woozies), or you need to use
		1100	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1101	programmers):
946		1102
947	#include <stddef.h>	1103	#include <stddef.h>
948		1104
949	static void	1105	static void
950	t1_cb (EV_P_ struct ev_timer *w, int revents)	1106	t1_cb (EV_P_ struct ev_timer *w, int revents)
951	{	1107	{
952	struct my_biggy big = (struct my_biggy *	1108	struct my_biggy big = (struct my_biggy *
953	(((char *)w) - offsetof (struct my_biggy, t1));	1109	(((char *)w) - offsetof (struct my_biggy, t1));
954	}	1110	}
955		1111
956	static void	1112	static void
957	t2_cb (EV_P_ struct ev_timer *w, int revents)	1113	t2_cb (EV_P_ struct ev_timer *w, int revents)
958	{	1114	{
959	struct my_biggy big = (struct my_biggy *	1115	struct my_biggy big = (struct my_biggy *
960	(((char *)w) - offsetof (struct my_biggy, t2));	1116	(((char *)w) - offsetof (struct my_biggy, t2));
961	}	1117	}
962		1118
963		1119
964	=head1 WATCHER TYPES	1120	=head1 WATCHER TYPES
965		1121
966	This section describes each watcher in detail, but will not repeat	1122	This section describes each watcher in detail, but will not repeat
…		…
990	In general you can register as many read and/or write event watchers per	1146	In general you can register as many read and/or write event watchers per
991	fd as you want (as long as you don't confuse yourself). Setting all file	1147	fd as you want (as long as you don't confuse yourself). Setting all file
992	descriptors to non-blocking mode is also usually a good idea (but not	1148	descriptors to non-blocking mode is also usually a good idea (but not
993	required if you know what you are doing).	1149	required if you know what you are doing).
994		1150
995	If you must do this, then force the use of a known-to-be-good backend	1151	If you cannot use non-blocking mode, then force the use of a
996	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1152	known-to-be-good backend (at the time of this writing, this includes only
997	C<EVBACKEND_POLL>).	1153	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>).
998		1154
999	Another thing you have to watch out for is that it is quite easy to	1155	Another thing you have to watch out for is that it is quite easy to
1000	receive "spurious" readyness notifications, that is your callback might	1156	receive "spurious" readiness notifications, that is your callback might
1001	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1157	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
1002	because there is no data. Not only are some backends known to create a	1158	because there is no data. Not only are some backends known to create a
1003	lot of those (for example solaris ports), it is very easy to get into	1159	lot of those (for example Solaris ports), it is very easy to get into
1004	this situation even with a relatively standard program structure. Thus	1160	this situation even with a relatively standard program structure. Thus
1005	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1161	it is best to always use non-blocking I/O: An extra C<read>(2) returning
1006	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1162	C<EAGAIN> is far preferable to a program hanging until some data arrives.
1007		1163
1008	If you cannot run the fd in non-blocking mode (for example you should not	1164	If you cannot run the fd in non-blocking mode (for example you should
1009	play around with an Xlib connection), then you have to seperately re-test	1165	not play around with an Xlib connection), then you have to separately
1010	whether a file descriptor is really ready with a known-to-be good interface	1166	re-test whether a file descriptor is really ready with a known-to-be good
1011	such as poll (fortunately in our Xlib example, Xlib already does this on	1167	interface such as poll (fortunately in our Xlib example, Xlib already
1012	its own, so its quite safe to use).	1168	does this on its own, so its quite safe to use). Some people additionally
		1169	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1170	indefinitely.
		1171
		1172	But really, best use non-blocking mode.
1013		1173
1014	=head3 The special problem of disappearing file descriptors	1174	=head3 The special problem of disappearing file descriptors
1015		1175
1016	Some backends (e.g. kqueue, epoll) need to be told about closing a file	1176	Some backends (e.g. kqueue, epoll) need to be told about closing a file
1017	descriptor (either by calling C<close> explicitly or by any other means,	1177	descriptor (either due to calling C<close> explicitly or any other means,
1018	such as C<dup>). The reason is that you register interest in some file	1178	such as C<dup2>). The reason is that you register interest in some file
1019	descriptor, but when it goes away, the operating system will silently drop	1179	descriptor, but when it goes away, the operating system will silently drop
1020	this interest. If another file descriptor with the same number then is	1180	this interest. If another file descriptor with the same number then is
1021	registered with libev, there is no efficient way to see that this is, in	1181	registered with libev, there is no efficient way to see that this is, in
1022	fact, a different file descriptor.	1182	fact, a different file descriptor.
1023		1183
…		…
1052	To support fork in your programs, you either have to call	1212	To support fork in your programs, you either have to call
1053	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,	1213	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
1054	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1214	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
1055	C<EVBACKEND_POLL>.	1215	C<EVBACKEND_POLL>.
1056		1216
		1217	=head3 The special problem of SIGPIPE
		1218
		1219	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1220	when writing to a pipe whose other end has been closed, your program gets
		1221	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1222	this is sensible behaviour, for daemons, this is usually undesirable.
		1223
		1224	So when you encounter spurious, unexplained daemon exits, make sure you
		1225	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1226	somewhere, as that would have given you a big clue).
		1227
1057		1228
1058	=head3 Watcher-Specific Functions	1229	=head3 Watcher-Specific Functions
1059		1230
1060	=over 4	1231	=over 4
1061		1232
1062	=item ev_io_init (ev_io *, callback, int fd, int events)	1233	=item ev_io_init (ev_io *, callback, int fd, int events)
1063		1234
1064	=item ev_io_set (ev_io *, int fd, int events)	1235	=item ev_io_set (ev_io *, int fd, int events)
1065		1236
1066	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1237	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
1067	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or	1238	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
1068	C<EV_READ | EV_WRITE> to receive the given events.	1239	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
1069		1240
1070	=item int fd [read-only]	1241	=item int fd [read-only]
1071		1242
1072	The file descriptor being watched.	1243	The file descriptor being watched.
1073		1244
…		…
1081		1252
1082	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1253	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
1083	readable, but only once. Since it is likely line-buffered, you could	1254	readable, but only once. Since it is likely line-buffered, you could
1084	attempt to read a whole line in the callback.	1255	attempt to read a whole line in the callback.
1085		1256
1086	static void	1257	static void
1087	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1258	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1088	{	1259	{
1089	ev_io_stop (loop, w);	1260	ev_io_stop (loop, w);
1090	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1261	.. read from stdin here (or from w->fd) and handle any I/O errors
1091	}	1262	}
1092		1263
1093	...	1264	...
1094	struct ev_loop *loop = ev_default_init (0);	1265	struct ev_loop *loop = ev_default_init (0);
1095	struct ev_io stdin_readable;	1266	struct ev_io stdin_readable;
1096	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1267	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
1097	ev_io_start (loop, &stdin_readable);	1268	ev_io_start (loop, &stdin_readable);
1098	ev_loop (loop, 0);	1269	ev_loop (loop, 0);
1099		1270
1100		1271
1101	=head2 C<ev_timer> - relative and optionally repeating timeouts	1272	=head2 C<ev_timer> - relative and optionally repeating timeouts
1102		1273
1103	Timer watchers are simple relative timers that generate an event after a	1274	Timer watchers are simple relative timers that generate an event after a
1104	given time, and optionally repeating in regular intervals after that.	1275	given time, and optionally repeating in regular intervals after that.
1105		1276
1106	The timers are based on real time, that is, if you register an event that	1277	The timers are based on real time, that is, if you register an event that
1107	times out after an hour and you reset your system clock to last years	1278	times out after an hour and you reset your system clock to January last
1108	time, it will still time out after (roughly) and hour. "Roughly" because	1279	year, it will still time out after (roughly) one hour. "Roughly" because
1109	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1280	detecting time jumps is hard, and some inaccuracies are unavoidable (the
1110	monotonic clock option helps a lot here).	1281	monotonic clock option helps a lot here).
		1282
		1283	The callback is guaranteed to be invoked only I<after> its timeout has
		1284	passed, but if multiple timers become ready during the same loop iteration
		1285	then order of execution is undefined.
		1286
		1287	=head3 The special problem of time updates
		1288
		1289	Establishing the current time is a costly operation (it usually takes at
		1290	least two system calls): EV therefore updates its idea of the current
		1291	time only before and after C<ev_loop> collects new events, which causes a
		1292	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1293	lots of events in one iteration.
1111		1294
1112	The relative timeouts are calculated relative to the C<ev_now ()>	1295	The relative timeouts are calculated relative to the C<ev_now ()>
1113	time. This is usually the right thing as this timestamp refers to the time	1296	time. This is usually the right thing as this timestamp refers to the time
1114	of the event triggering whatever timeout you are modifying/starting. If	1297	of the event triggering whatever timeout you are modifying/starting. If
1115	you suspect event processing to be delayed and you I<need> to base the timeout	1298	you suspect event processing to be delayed and you I<need> to base the
1116	on the current time, use something like this to adjust for this:	1299	timeout on the current time, use something like this to adjust for this:
1117		1300
1118	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1301	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
1119		1302
1120	The callback is guarenteed to be invoked only when its timeout has passed,	1303	If the event loop is suspended for a long time, you can also force an
1121	but if multiple timers become ready during the same loop iteration then	1304	update of the time returned by C<ev_now ()> by calling C<ev_now_update
1122	order of execution is undefined.	1305	()>.
1123		1306
1124	=head3 Watcher-Specific Functions and Data Members	1307	=head3 Watcher-Specific Functions and Data Members
1125		1308
1126	=over 4	1309	=over 4
1127		1310
1128	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1311	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
1129		1312
1130	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1313	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
1131		1314
1132	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1315	Configure the timer to trigger after C<after> seconds. If C<repeat>
1133	C<0.>, then it will automatically be stopped. If it is positive, then the	1316	is C<0.>, then it will automatically be stopped once the timeout is
1134	timer will automatically be configured to trigger again C<repeat> seconds	1317	reached. If it is positive, then the timer will automatically be
1135	later, again, and again, until stopped manually.	1318	configured to trigger again C<repeat> seconds later, again, and again,
		1319	until stopped manually.
1136		1320
1137	The timer itself will do a best-effort at avoiding drift, that is, if you	1321	The timer itself will do a best-effort at avoiding drift, that is, if
1138	configure a timer to trigger every 10 seconds, then it will trigger at	1322	you configure a timer to trigger every 10 seconds, then it will normally
1139	exactly 10 second intervals. If, however, your program cannot keep up with	1323	trigger at exactly 10 second intervals. If, however, your program cannot
1140	the timer (because it takes longer than those 10 seconds to do stuff) the	1324	keep up with the timer (because it takes longer than those 10 seconds to
1141	timer will not fire more than once per event loop iteration.	1325	do stuff) the timer will not fire more than once per event loop iteration.
1142		1326
1143	=item ev_timer_again (loop)	1327	=item ev_timer_again (loop, ev_timer *)
1144		1328
1145	This will act as if the timer timed out and restart it again if it is	1329	This will act as if the timer timed out and restart it again if it is
1146	repeating. The exact semantics are:	1330	repeating. The exact semantics are:
1147		1331
1148	If the timer is pending, its pending status is cleared.	1332	If the timer is pending, its pending status is cleared.
1149		1333
1150	If the timer is started but nonrepeating, stop it (as if it timed out).	1334	If the timer is started but non-repeating, stop it (as if it timed out).
1151		1335
1152	If the timer is repeating, either start it if necessary (with the	1336	If the timer is repeating, either start it if necessary (with the
1153	C<repeat> value), or reset the running timer to the C<repeat> value.	1337	C<repeat> value), or reset the running timer to the C<repeat> value.
1154		1338
1155	This sounds a bit complicated, but here is a useful and typical	1339	This sounds a bit complicated, but here is a useful and typical
1156	example: Imagine you have a tcp connection and you want a so-called idle	1340	example: Imagine you have a TCP connection and you want a so-called idle
1157	timeout, that is, you want to be called when there have been, say, 60	1341	timeout, that is, you want to be called when there have been, say, 60
1158	seconds of inactivity on the socket. The easiest way to do this is to	1342	seconds of inactivity on the socket. The easiest way to do this is to
1159	configure an C<ev_timer> with a C<repeat> value of C<60> and then call	1343	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
1160	C<ev_timer_again> each time you successfully read or write some data. If	1344	C<ev_timer_again> each time you successfully read or write some data. If
1161	you go into an idle state where you do not expect data to travel on the	1345	you go into an idle state where you do not expect data to travel on the
…		…
1175	ev_timer_again (loop, timer);	1359	ev_timer_again (loop, timer);
1176		1360
1177	This is more slightly efficient then stopping/starting the timer each time	1361	This is more slightly efficient then stopping/starting the timer each time
1178	you want to modify its timeout value.	1362	you want to modify its timeout value.
1179		1363
		1364	Note, however, that it is often even more efficient to remember the
		1365	time of the last activity and let the timer time-out naturally. In the
		1366	callback, you then check whether the time-out is real, or, if there was
		1367	some activity, you reschedule the watcher to time-out in "last_activity +
		1368	timeout - ev_now ()" seconds.
		1369
1180	=item ev_tstamp repeat [read-write]	1370	=item ev_tstamp repeat [read-write]
1181		1371
1182	The current C<repeat> value. Will be used each time the watcher times out	1372	The current C<repeat> value. Will be used each time the watcher times out
1183	or C<ev_timer_again> is called and determines the next timeout (if any),	1373	or C<ev_timer_again> is called, and determines the next timeout (if any),
1184	which is also when any modifications are taken into account.	1374	which is also when any modifications are taken into account.
1185		1375
1186	=back	1376	=back
1187		1377
1188	=head3 Examples	1378	=head3 Examples
1189		1379
1190	Example: Create a timer that fires after 60 seconds.	1380	Example: Create a timer that fires after 60 seconds.
1191		1381
1192	static void	1382	static void
1193	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1383	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1194	{	1384	{
1195	.. one minute over, w is actually stopped right here	1385	.. one minute over, w is actually stopped right here
1196	}	1386	}
1197		1387
1198	struct ev_timer mytimer;	1388	struct ev_timer mytimer;
1199	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1389	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
1200	ev_timer_start (loop, &mytimer);	1390	ev_timer_start (loop, &mytimer);
1201		1391
1202	Example: Create a timeout timer that times out after 10 seconds of	1392	Example: Create a timeout timer that times out after 10 seconds of
1203	inactivity.	1393	inactivity.
1204		1394
1205	static void	1395	static void
1206	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1396	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
1207	{	1397	{
1208	.. ten seconds without any activity	1398	.. ten seconds without any activity
1209	}	1399	}
1210		1400
1211	struct ev_timer mytimer;	1401	struct ev_timer mytimer;
1212	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	1402	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
1213	ev_timer_again (&mytimer); /* start timer */	1403	ev_timer_again (&mytimer); /* start timer */
1214	ev_loop (loop, 0);	1404	ev_loop (loop, 0);
1215		1405
1216	// and in some piece of code that gets executed on any "activity":	1406	// and in some piece of code that gets executed on any "activity":
1217	// reset the timeout to start ticking again at 10 seconds	1407	// reset the timeout to start ticking again at 10 seconds
1218	ev_timer_again (&mytimer);	1408	ev_timer_again (&mytimer);
1219		1409
1220		1410
1221	=head2 C<ev_periodic> - to cron or not to cron?	1411	=head2 C<ev_periodic> - to cron or not to cron?
1222		1412
1223	Periodic watchers are also timers of a kind, but they are very versatile	1413	Periodic watchers are also timers of a kind, but they are very versatile
1224	(and unfortunately a bit complex).	1414	(and unfortunately a bit complex).
1225		1415
1226	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1416	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
1227	but on wallclock time (absolute time). You can tell a periodic watcher	1417	but on wall clock time (absolute time). You can tell a periodic watcher
1228	to trigger "at" some specific point in time. For example, if you tell a	1418	to trigger after some specific point in time. For example, if you tell a
1229	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	1419	periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()
1230	+ 10.>) and then reset your system clock to the last year, then it will	1420	+ 10.>, that is, an absolute time not a delay) and then reset your system
		1421	clock to January of the previous year, then it will take more than year
1231	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1422	to trigger the event (unlike an C<ev_timer>, which would still trigger
1232	roughly 10 seconds later).	1423	roughly 10 seconds later as it uses a relative timeout).
1233		1424
1234	They can also be used to implement vastly more complex timers, such as	1425	C<ev_periodic>s can also be used to implement vastly more complex timers,
1235	triggering an event on each midnight, local time or other, complicated,	1426	such as triggering an event on each "midnight, local time", or other
1236	rules.	1427	complicated rules.
1237		1428
1238	As with timers, the callback is guarenteed to be invoked only when the	1429	As with timers, the callback is guaranteed to be invoked only when the
1239	time (C<at>) has been passed, but if multiple periodic timers become ready	1430	time (C<at>) has passed, but if multiple periodic timers become ready
1240	during the same loop iteration then order of execution is undefined.	1431	during the same loop iteration, then order of execution is undefined.
1241		1432
1242	=head3 Watcher-Specific Functions and Data Members	1433	=head3 Watcher-Specific Functions and Data Members
1243		1434
1244	=over 4	1435	=over 4
1245		1436
1246	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1437	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
1247		1438
1248	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1439	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)
1249		1440
1250	Lots of arguments, lets sort it out... There are basically three modes of	1441	Lots of arguments, lets sort it out... There are basically three modes of
1251	operation, and we will explain them from simplest to complex:	1442	operation, and we will explain them from simplest to most complex:
1252		1443
1253	=over 4	1444	=over 4
1254		1445
1255	=item * absolute timer (at = time, interval = reschedule_cb = 0)	1446	=item * absolute timer (at = time, interval = reschedule_cb = 0)
1256		1447
1257	In this configuration the watcher triggers an event at the wallclock time	1448	In this configuration the watcher triggers an event after the wall clock
1258	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1449	time C<at> has passed. It will not repeat and will not adjust when a time
1259	that is, if it is to be run at January 1st 2011 then it will run when the	1450	jump occurs, that is, if it is to be run at January 1st 2011 then it will
1260	system time reaches or surpasses this time.	1451	only run when the system clock reaches or surpasses this time.
1261		1452
1262	=item * non-repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	1453	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
1263		1454
1264	In this mode the watcher will always be scheduled to time out at the next	1455	In this mode the watcher will always be scheduled to time out at the next
1265	C<at + N * interval> time (for some integer N, which can also be negative)	1456	C<at + N * interval> time (for some integer N, which can also be negative)
1266	and then repeat, regardless of any time jumps.	1457	and then repeat, regardless of any time jumps.
1267		1458
1268	This can be used to create timers that do not drift with respect to system	1459	This can be used to create timers that do not drift with respect to the
1269	time:	1460	system clock, for example, here is a C<ev_periodic> that triggers each
		1461	hour, on the hour:
1270		1462
1271	ev_periodic_set (&periodic, 0., 3600., 0);	1463	ev_periodic_set (&periodic, 0., 3600., 0);
1272		1464
1273	This doesn't mean there will always be 3600 seconds in between triggers,	1465	This doesn't mean there will always be 3600 seconds in between triggers,
1274	but only that the the callback will be called when the system time shows a	1466	but only that the callback will be called when the system time shows a
1275	full hour (UTC), or more correctly, when the system time is evenly divisible	1467	full hour (UTC), or more correctly, when the system time is evenly divisible
1276	by 3600.	1468	by 3600.
1277		1469
1278	Another way to think about it (for the mathematically inclined) is that	1470	Another way to think about it (for the mathematically inclined) is that
1279	C<ev_periodic> will try to run the callback in this mode at the next possible	1471	C<ev_periodic> will try to run the callback in this mode at the next possible
1280	time where C<time = at (mod interval)>, regardless of any time jumps.	1472	time where C<time = at (mod interval)>, regardless of any time jumps.
1281		1473
1282	For numerical stability it is preferable that the C<at> value is near	1474	For numerical stability it is preferable that the C<at> value is near
1283	C<ev_now ()> (the current time), but there is no range requirement for	1475	C<ev_now ()> (the current time), but there is no range requirement for
1284	this value.	1476	this value, and in fact is often specified as zero.
		1477
		1478	Note also that there is an upper limit to how often a timer can fire (CPU
		1479	speed for example), so if C<interval> is very small then timing stability
		1480	will of course deteriorate. Libev itself tries to be exact to be about one
		1481	millisecond (if the OS supports it and the machine is fast enough).
1285		1482
1286	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)	1483	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
1287		1484
1288	In this mode the values for C<interval> and C<at> are both being	1485	In this mode the values for C<interval> and C<at> are both being
1289	ignored. Instead, each time the periodic watcher gets scheduled, the	1486	ignored. Instead, each time the periodic watcher gets scheduled, the
1290	reschedule callback will be called with the watcher as first, and the	1487	reschedule callback will be called with the watcher as first, and the
1291	current time as second argument.	1488	current time as second argument.
1292		1489
1293	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1490	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
1294	ever, or make any event loop modifications>. If you need to stop it,	1491	ever, or make ANY event loop modifications whatsoever>.
1295	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
1296	starting an C<ev_prepare> watcher, which is legal).
1297		1492
		1493	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1494	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1495	only event loop modification you are allowed to do).
		1496
1298	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1497	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic
1299	ev_tstamp now)>, e.g.:	1498	*w, ev_tstamp now)>, e.g.:
1300		1499
1301	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1500	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
1302	{	1501	{
1303	return now + 60.;	1502	return now + 60.;
1304	}	1503	}
…		…
1306	It must return the next time to trigger, based on the passed time value	1505	It must return the next time to trigger, based on the passed time value
1307	(that is, the lowest time value larger than to the second argument). It	1506	(that is, the lowest time value larger than to the second argument). It
1308	will usually be called just before the callback will be triggered, but	1507	will usually be called just before the callback will be triggered, but
1309	might be called at other times, too.	1508	might be called at other times, too.
1310		1509
1311	NOTE: I<< This callback must always return a time that is later than the	1510	NOTE: I<< This callback must always return a time that is higher than or
1312	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	1511	equal to the passed C<now> value >>.
1313		1512
1314	This can be used to create very complex timers, such as a timer that	1513	This can be used to create very complex timers, such as a timer that
1315	triggers on each midnight, local time. To do this, you would calculate the	1514	triggers on "next midnight, local time". To do this, you would calculate the
1316	next midnight after C<now> and return the timestamp value for this. How	1515	next midnight after C<now> and return the timestamp value for this. How
1317	you do this is, again, up to you (but it is not trivial, which is the main	1516	you do this is, again, up to you (but it is not trivial, which is the main
1318	reason I omitted it as an example).	1517	reason I omitted it as an example).
1319		1518
1320	=back	1519	=back
…		…
1324	Simply stops and restarts the periodic watcher again. This is only useful	1523	Simply stops and restarts the periodic watcher again. This is only useful
1325	when you changed some parameters or the reschedule callback would return	1524	when you changed some parameters or the reschedule callback would return
1326	a different time than the last time it was called (e.g. in a crond like	1525	a different time than the last time it was called (e.g. in a crond like
1327	program when the crontabs have changed).	1526	program when the crontabs have changed).
1328		1527
		1528	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1529
		1530	When active, returns the absolute time that the watcher is supposed to
		1531	trigger next.
		1532
1329	=item ev_tstamp offset [read-write]	1533	=item ev_tstamp offset [read-write]
1330		1534
1331	When repeating, this contains the offset value, otherwise this is the	1535	When repeating, this contains the offset value, otherwise this is the
1332	absolute point in time (the C<at> value passed to C<ev_periodic_set>).	1536	absolute point in time (the C<at> value passed to C<ev_periodic_set>).
1333		1537
…		…
1344		1548
1345	The current reschedule callback, or C<0>, if this functionality is	1549	The current reschedule callback, or C<0>, if this functionality is
1346	switched off. Can be changed any time, but changes only take effect when	1550	switched off. Can be changed any time, but changes only take effect when
1347	the periodic timer fires or C<ev_periodic_again> is being called.	1551	the periodic timer fires or C<ev_periodic_again> is being called.
1348		1552
1349	=item ev_tstamp at [read-only]
1350
1351	When active, contains the absolute time that the watcher is supposed to
1352	trigger next.
1353
1354	=back	1553	=back
1355		1554
1356	=head3 Examples	1555	=head3 Examples
1357		1556
1358	Example: Call a callback every hour, or, more precisely, whenever the	1557	Example: Call a callback every hour, or, more precisely, whenever the
1359	system clock is divisible by 3600. The callback invocation times have	1558	system time is divisible by 3600. The callback invocation times have
1360	potentially a lot of jittering, but good long-term stability.	1559	potentially a lot of jitter, but good long-term stability.
1361		1560
1362	static void	1561	static void
1363	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1562	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1364	{	1563	{
1365	... its now a full hour (UTC, or TAI or whatever your clock follows)	1564	... its now a full hour (UTC, or TAI or whatever your clock follows)
1366	}	1565	}
1367		1566
1368	struct ev_periodic hourly_tick;	1567	struct ev_periodic hourly_tick;
1369	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1568	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1370	ev_periodic_start (loop, &hourly_tick);	1569	ev_periodic_start (loop, &hourly_tick);
1371		1570
1372	Example: The same as above, but use a reschedule callback to do it:	1571	Example: The same as above, but use a reschedule callback to do it:
1373		1572
1374	#include <math.h>	1573	#include <math.h>
1375		1574
1376	static ev_tstamp	1575	static ev_tstamp
1377	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1576	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1378	{	1577	{
1379	return fmod (now, 3600.) + 3600.;	1578	return now + (3600. - fmod (now, 3600.));
1380	}	1579	}
1381		1580
1382	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1581	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1383		1582
1384	Example: Call a callback every hour, starting now:	1583	Example: Call a callback every hour, starting now:
1385		1584
1386	struct ev_periodic hourly_tick;	1585	struct ev_periodic hourly_tick;
1387	ev_periodic_init (&hourly_tick, clock_cb,	1586	ev_periodic_init (&hourly_tick, clock_cb,
1388	fmod (ev_now (loop), 3600.), 3600., 0);	1587	fmod (ev_now (loop), 3600.), 3600., 0);
1389	ev_periodic_start (loop, &hourly_tick);	1588	ev_periodic_start (loop, &hourly_tick);
1390		1589
1391		1590
1392	=head2 C<ev_signal> - signal me when a signal gets signalled!	1591	=head2 C<ev_signal> - signal me when a signal gets signalled!
1393		1592
1394	Signal watchers will trigger an event when the process receives a specific	1593	Signal watchers will trigger an event when the process receives a specific
1395	signal one or more times. Even though signals are very asynchronous, libev	1594	signal one or more times. Even though signals are very asynchronous, libev
1396	will try it's best to deliver signals synchronously, i.e. as part of the	1595	will try it's best to deliver signals synchronously, i.e. as part of the
1397	normal event processing, like any other event.	1596	normal event processing, like any other event.
1398		1597
		1598	If you want signals asynchronously, just use C<sigaction> as you would
		1599	do without libev and forget about sharing the signal. You can even use
		1600	C<ev_async> from a signal handler to synchronously wake up an event loop.
		1601
1399	You can configure as many watchers as you like per signal. Only when the	1602	You can configure as many watchers as you like per signal. Only when the
1400	first watcher gets started will libev actually register a signal watcher	1603	first watcher gets started will libev actually register a signal handler
1401	with the kernel (thus it coexists with your own signal handlers as long	1604	with the kernel (thus it coexists with your own signal handlers as long as
1402	as you don't register any with libev). Similarly, when the last signal	1605	you don't register any with libev for the same signal). Similarly, when
1403	watcher for a signal is stopped libev will reset the signal handler to	1606	the last signal watcher for a signal is stopped, libev will reset the
1404	SIG_DFL (regardless of what it was set to before).	1607	signal handler to SIG_DFL (regardless of what it was set to before).
		1608
		1609	If possible and supported, libev will install its handlers with
		1610	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
		1611	interrupted. If you have a problem with system calls getting interrupted by
		1612	signals you can block all signals in an C<ev_check> watcher and unblock
		1613	them in an C<ev_prepare> watcher.
1405		1614
1406	=head3 Watcher-Specific Functions and Data Members	1615	=head3 Watcher-Specific Functions and Data Members
1407		1616
1408	=over 4	1617	=over 4
1409		1618
…		…
1418		1627
1419	The signal the watcher watches out for.	1628	The signal the watcher watches out for.
1420		1629
1421	=back	1630	=back
1422		1631
		1632	=head3 Examples
		1633
		1634	Example: Try to exit cleanly on SIGINT.
		1635
		1636	static void
		1637	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
		1638	{
		1639	ev_unloop (loop, EVUNLOOP_ALL);
		1640	}
		1641
		1642	struct ev_signal signal_watcher;
		1643	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1644	ev_signal_start (loop, &signal_watcher);
		1645
1423		1646
1424	=head2 C<ev_child> - watch out for process status changes	1647	=head2 C<ev_child> - watch out for process status changes
1425		1648
1426	Child watchers trigger when your process receives a SIGCHLD in response to	1649	Child watchers trigger when your process receives a SIGCHLD in response to
1427	some child status changes (most typically when a child of yours dies).	1650	some child status changes (most typically when a child of yours dies or
		1651	exits). It is permissible to install a child watcher I<after> the child
		1652	has been forked (which implies it might have already exited), as long
		1653	as the event loop isn't entered (or is continued from a watcher), i.e.,
		1654	forking and then immediately registering a watcher for the child is fine,
		1655	but forking and registering a watcher a few event loop iterations later is
		1656	not.
		1657
		1658	Only the default event loop is capable of handling signals, and therefore
		1659	you can only register child watchers in the default event loop.
		1660
		1661	=head3 Process Interaction
		1662
		1663	Libev grabs C<SIGCHLD> as soon as the default event loop is
		1664	initialised. This is necessary to guarantee proper behaviour even if
		1665	the first child watcher is started after the child exits. The occurrence
		1666	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		1667	synchronously as part of the event loop processing. Libev always reaps all
		1668	children, even ones not watched.
		1669
		1670	=head3 Overriding the Built-In Processing
		1671
		1672	Libev offers no special support for overriding the built-in child
		1673	processing, but if your application collides with libev's default child
		1674	handler, you can override it easily by installing your own handler for
		1675	C<SIGCHLD> after initialising the default loop, and making sure the
		1676	default loop never gets destroyed. You are encouraged, however, to use an
		1677	event-based approach to child reaping and thus use libev's support for
		1678	that, so other libev users can use C<ev_child> watchers freely.
		1679
		1680	=head3 Stopping the Child Watcher
		1681
		1682	Currently, the child watcher never gets stopped, even when the
		1683	child terminates, so normally one needs to stop the watcher in the
		1684	callback. Future versions of libev might stop the watcher automatically
		1685	when a child exit is detected.
1428		1686
1429	=head3 Watcher-Specific Functions and Data Members	1687	=head3 Watcher-Specific Functions and Data Members
1430		1688
1431	=over 4	1689	=over 4
1432		1690
1433	=item ev_child_init (ev_child *, callback, int pid)	1691	=item ev_child_init (ev_child *, callback, int pid, int trace)
1434		1692
1435	=item ev_child_set (ev_child *, int pid)	1693	=item ev_child_set (ev_child *, int pid, int trace)
1436		1694
1437	Configures the watcher to wait for status changes of process C<pid> (or	1695	Configures the watcher to wait for status changes of process C<pid> (or
1438	I<any> process if C<pid> is specified as C<0>). The callback can look	1696	I<any> process if C<pid> is specified as C<0>). The callback can look
1439	at the C<rstatus> member of the C<ev_child> watcher structure to see	1697	at the C<rstatus> member of the C<ev_child> watcher structure to see
1440	the status word (use the macros from C<sys/wait.h> and see your systems	1698	the status word (use the macros from C<sys/wait.h> and see your systems
1441	C<waitpid> documentation). The C<rpid> member contains the pid of the	1699	C<waitpid> documentation). The C<rpid> member contains the pid of the
1442	process causing the status change.	1700	process causing the status change. C<trace> must be either C<0> (only
		1701	activate the watcher when the process terminates) or C<1> (additionally
		1702	activate the watcher when the process is stopped or continued).
1443		1703
1444	=item int pid [read-only]	1704	=item int pid [read-only]
1445		1705
1446	The process id this watcher watches out for, or C<0>, meaning any process id.	1706	The process id this watcher watches out for, or C<0>, meaning any process id.
1447		1707
…		…
1456		1716
1457	=back	1717	=back
1458		1718
1459	=head3 Examples	1719	=head3 Examples
1460		1720
1461	Example: Try to exit cleanly on SIGINT and SIGTERM.	1721	Example: C<fork()> a new process and install a child handler to wait for
		1722	its completion.
1462		1723
		1724	ev_child cw;
		1725
1463	static void	1726	static void
1464	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1727	child_cb (EV_P_ struct ev_child *w, int revents)
1465	{	1728	{
1466	ev_unloop (loop, EVUNLOOP_ALL);	1729	ev_child_stop (EV_A_ w);
		1730	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1467	}	1731	}
1468		1732
1469	struct ev_signal signal_watcher;	1733	pid_t pid = fork ();
1470	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1734
1471	ev_signal_start (loop, &sigint_cb);	1735	if (pid < 0)
		1736	// error
		1737	else if (pid == 0)
		1738	{
		1739	// the forked child executes here
		1740	exit (1);
		1741	}
		1742	else
		1743	{
		1744	ev_child_init (&cw, child_cb, pid, 0);
		1745	ev_child_start (EV_DEFAULT_ &cw);
		1746	}
1472		1747
1473		1748
1474	=head2 C<ev_stat> - did the file attributes just change?	1749	=head2 C<ev_stat> - did the file attributes just change?
1475		1750
1476	This watches a filesystem path for attribute changes. That is, it calls	1751	This watches a file system path for attribute changes. That is, it calls
1477	C<stat> regularly (or when the OS says it changed) and sees if it changed	1752	C<stat> regularly (or when the OS says it changed) and sees if it changed
1478	compared to the last time, invoking the callback if it did.	1753	compared to the last time, invoking the callback if it did.
1479		1754
1480	The path does not need to exist: changing from "path exists" to "path does	1755	The path does not need to exist: changing from "path exists" to "path does
1481	not exist" is a status change like any other. The condition "path does	1756	not exist" is a status change like any other. The condition "path does
…		…
1484	the stat buffer having unspecified contents.	1759	the stat buffer having unspecified contents.
1485		1760
1486	The path I<should> be absolute and I<must not> end in a slash. If it is	1761	The path I<should> be absolute and I<must not> end in a slash. If it is
1487	relative and your working directory changes, the behaviour is undefined.	1762	relative and your working directory changes, the behaviour is undefined.
1488		1763
1489	Since there is no standard to do this, the portable implementation simply	1764	Since there is no standard kernel interface to do this, the portable
1490	calls C<stat (2)> regularly on the path to see if it changed somehow. You	1765	implementation simply calls C<stat (2)> regularly on the path to see if
1491	can specify a recommended polling interval for this case. If you specify	1766	it changed somehow. You can specify a recommended polling interval for
1492	a polling interval of C<0> (highly recommended!) then a I<suitable,	1767	this case. If you specify a polling interval of C<0> (highly recommended!)
1493	unspecified default> value will be used (which you can expect to be around	1768	then a I<suitable, unspecified default> value will be used (which
1494	five seconds, although this might change dynamically). Libev will also	1769	you can expect to be around five seconds, although this might change
1495	impose a minimum interval which is currently around C<0.1>, but thats	1770	dynamically). Libev will also impose a minimum interval which is currently
1496	usually overkill.	1771	around C<0.1>, but thats usually overkill.
1497		1772
1498	This watcher type is not meant for massive numbers of stat watchers,	1773	This watcher type is not meant for massive numbers of stat watchers,
1499	as even with OS-supported change notifications, this can be	1774	as even with OS-supported change notifications, this can be
1500	resource-intensive.	1775	resource-intensive.
1501		1776
1502	At the time of this writing, only the Linux inotify interface is	1777	At the time of this writing, the only OS-specific interface implemented
1503	implemented (implementing kqueue support is left as an exercise for the	1778	is the Linux inotify interface (implementing kqueue support is left as
1504	reader). Inotify will be used to give hints only and should not change the	1779	an exercise for the reader. Note, however, that the author sees no way
1505	semantics of C<ev_stat> watchers, which means that libev sometimes needs	1780	of implementing C<ev_stat> semantics with kqueue).
1506	to fall back to regular polling again even with inotify, but changes are
1507	usually detected immediately, and if the file exists there will be no
1508	polling.
1509		1781
1510	=head3 Inotify	1782	=head3 ABI Issues (Largefile Support)
1511		1783
		1784	Libev by default (unless the user overrides this) uses the default
		1785	compilation environment, which means that on systems with large file
		1786	support disabled by default, you get the 32 bit version of the stat
		1787	structure. When using the library from programs that change the ABI to
		1788	use 64 bit file offsets the programs will fail. In that case you have to
		1789	compile libev with the same flags to get binary compatibility. This is
		1790	obviously the case with any flags that change the ABI, but the problem is
		1791	most noticeably disabled with ev_stat and large file support.
		1792
		1793	The solution for this is to lobby your distribution maker to make large
		1794	file interfaces available by default (as e.g. FreeBSD does) and not
		1795	optional. Libev cannot simply switch on large file support because it has
		1796	to exchange stat structures with application programs compiled using the
		1797	default compilation environment.
		1798
		1799	=head3 Inotify and Kqueue
		1800
1512	When C<inotify (7)> support has been compiled into libev (generally only	1801	When C<inotify (7)> support has been compiled into libev (generally
		1802	only available with Linux 2.6.25 or above due to bugs in earlier
1513	available on Linux) and present at runtime, it will be used to speed up	1803	implementations) and present at runtime, it will be used to speed up
1514	change detection where possible. The inotify descriptor will be created lazily	1804	change detection where possible. The inotify descriptor will be created
1515	when the first C<ev_stat> watcher is being started.	1805	lazily when the first C<ev_stat> watcher is being started.
1516		1806
1517	Inotify presense does not change the semantics of C<ev_stat> watchers	1807	Inotify presence does not change the semantics of C<ev_stat> watchers
1518	except that changes might be detected earlier, and in some cases, to avoid	1808	except that changes might be detected earlier, and in some cases, to avoid
1519	making regular C<stat> calls. Even in the presense of inotify support	1809	making regular C<stat> calls. Even in the presence of inotify support
1520	there are many cases where libev has to resort to regular C<stat> polling.	1810	there are many cases where libev has to resort to regular C<stat> polling,
		1811	but as long as the path exists, libev usually gets away without polling.
1521		1812
1522	(There is no support for kqueue, as apparently it cannot be used to	1813	There is no support for kqueue, as apparently it cannot be used to
1523	implement this functionality, due to the requirement of having a file	1814	implement this functionality, due to the requirement of having a file
1524	descriptor open on the object at all times).	1815	descriptor open on the object at all times, and detecting renames, unlinks
		1816	etc. is difficult.
1525		1817
1526	=head3 The special problem of stat time resolution	1818	=head3 The special problem of stat time resolution
1527		1819
1528	The C<stat ()> syscall only supports full-second resolution portably, and	1820	The C<stat ()> system call only supports full-second resolution portably, and
1529	even on systems where the resolution is higher, many filesystems still	1821	even on systems where the resolution is higher, most file systems still
1530	only support whole seconds.	1822	only support whole seconds.
1531		1823
1532	That means that, if the time is the only thing that changes, you might	1824	That means that, if the time is the only thing that changes, you can
1533	miss updates: on the first update, C<ev_stat> detects a change and calls	1825	easily miss updates: on the first update, C<ev_stat> detects a change and
1534	your callback, which does something. When there is another update within	1826	calls your callback, which does something. When there is another update
1535	the same second, C<ev_stat> will be unable to detect it.	1827	within the same second, C<ev_stat> will be unable to detect unless the
		1828	stat data does change in other ways (e.g. file size).
1536		1829
1537	The solution to this is to delay acting on a change for a second (or till	1830	The solution to this is to delay acting on a change for slightly more
1538	the next second boundary), using a roughly one-second delay C<ev_timer>	1831	than a second (or till slightly after the next full second boundary), using
1539	(C<ev_timer_set (w, 0., 1.01); ev_timer_again (loop, w)>). The C<.01>	1832	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
1540	is added to work around small timing inconsistencies of some operating	1833	ev_timer_again (loop, w)>).
1541	systems.	1834
		1835	The C<.02> offset is added to work around small timing inconsistencies
		1836	of some operating systems (where the second counter of the current time
		1837	might be be delayed. One such system is the Linux kernel, where a call to
		1838	C<gettimeofday> might return a timestamp with a full second later than
		1839	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		1840	update file times then there will be a small window where the kernel uses
		1841	the previous second to update file times but libev might already execute
		1842	the timer callback).
1542		1843
1543	=head3 Watcher-Specific Functions and Data Members	1844	=head3 Watcher-Specific Functions and Data Members
1544		1845
1545	=over 4	1846	=over 4
1546		1847
…		…
1552	C<path>. The C<interval> is a hint on how quickly a change is expected to	1853	C<path>. The C<interval> is a hint on how quickly a change is expected to
1553	be detected and should normally be specified as C<0> to let libev choose	1854	be detected and should normally be specified as C<0> to let libev choose
1554	a suitable value. The memory pointed to by C<path> must point to the same	1855	a suitable value. The memory pointed to by C<path> must point to the same
1555	path for as long as the watcher is active.	1856	path for as long as the watcher is active.
1556		1857
1557	The callback will be receive C<EV_STAT> when a change was detected,	1858	The callback will receive an C<EV_STAT> event when a change was detected,
1558	relative to the attributes at the time the watcher was started (or the	1859	relative to the attributes at the time the watcher was started (or the
1559	last change was detected).	1860	last change was detected).
1560		1861
1561	=item ev_stat_stat (ev_stat *)	1862	=item ev_stat_stat (loop, ev_stat *)
1562		1863
1563	Updates the stat buffer immediately with new values. If you change the	1864	Updates the stat buffer immediately with new values. If you change the
1564	watched path in your callback, you could call this fucntion to avoid	1865	watched path in your callback, you could call this function to avoid
1565	detecting this change (while introducing a race condition). Can also be	1866	detecting this change (while introducing a race condition if you are not
1566	useful simply to find out the new values.	1867	the only one changing the path). Can also be useful simply to find out the
		1868	new values.
1567		1869
1568	=item ev_statdata attr [read-only]	1870	=item ev_statdata attr [read-only]
1569		1871
1570	The most-recently detected attributes of the file. Although the type is of	1872	The most-recently detected attributes of the file. Although the type is
1571	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	1873	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1572	suitable for your system. If the C<st_nlink> member is C<0>, then there	1874	suitable for your system, but you can only rely on the POSIX-standardised
		1875	members to be present. If the C<st_nlink> member is C<0>, then there was
1573	was some error while C<stat>ing the file.	1876	some error while C<stat>ing the file.
1574		1877
1575	=item ev_statdata prev [read-only]	1878	=item ev_statdata prev [read-only]
1576		1879
1577	The previous attributes of the file. The callback gets invoked whenever	1880	The previous attributes of the file. The callback gets invoked whenever
1578	C<prev> != C<attr>.	1881	C<prev> != C<attr>, or, more precisely, one or more of these members
		1882	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		1883	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1579		1884
1580	=item ev_tstamp interval [read-only]	1885	=item ev_tstamp interval [read-only]
1581		1886
1582	The specified interval.	1887	The specified interval.
1583		1888
1584	=item const char *path [read-only]	1889	=item const char *path [read-only]
1585		1890
1586	The filesystem path that is being watched.	1891	The file system path that is being watched.
1587		1892
1588	=back	1893	=back
1589		1894
1590	=head3 Examples	1895	=head3 Examples
1591		1896
1592	Example: Watch C</etc/passwd> for attribute changes.	1897	Example: Watch C</etc/passwd> for attribute changes.
1593		1898
1594	static void	1899	static void
1595	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	1900	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1596	{	1901	{
1597	/* /etc/passwd changed in some way */	1902	/* /etc/passwd changed in some way */
1598	if (w->attr.st_nlink)	1903	if (w->attr.st_nlink)
1599	{	1904	{
1600	printf ("passwd current size %ld\n", (long)w->attr.st_size);	1905	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1601	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	1906	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1602	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	1907	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1603	}	1908	}
1604	else	1909	else
1605	/* you shalt not abuse printf for puts */	1910	/* you shalt not abuse printf for puts */
1606	puts ("wow, /etc/passwd is not there, expect problems. "	1911	puts ("wow, /etc/passwd is not there, expect problems. "
1607	"if this is windows, they already arrived\n");	1912	"if this is windows, they already arrived\n");
1608	}	1913	}
1609		1914
1610	...	1915	...
1611	ev_stat passwd;	1916	ev_stat passwd;
1612		1917
1613	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);	1918	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1614	ev_stat_start (loop, &passwd);	1919	ev_stat_start (loop, &passwd);
1615		1920
1616	Example: Like above, but additionally use a one-second delay so we do not	1921	Example: Like above, but additionally use a one-second delay so we do not
1617	miss updates (however, frequent updates will delay processing, too, so	1922	miss updates (however, frequent updates will delay processing, too, so
1618	one might do the work both on C<ev_stat> callback invocation I<and> on	1923	one might do the work both on C<ev_stat> callback invocation I<and> on
1619	C<ev_timer> callback invocation).	1924	C<ev_timer> callback invocation).
1620		1925
1621	static ev_stat passwd;	1926	static ev_stat passwd;
1622	static ev_timer timer;	1927	static ev_timer timer;
1623		1928
1624	static void	1929	static void
1625	timer_cb (EV_P_ ev_timer *w, int revents)	1930	timer_cb (EV_P_ ev_timer *w, int revents)
1626	{	1931	{
1627	ev_timer_stop (EV_A_ w);	1932	ev_timer_stop (EV_A_ w);
1628		1933
1629	/* now it's one second after the most recent passwd change */	1934	/* now it's one second after the most recent passwd change */
1630	}	1935	}
1631		1936
1632	static void	1937	static void
1633	stat_cb (EV_P_ ev_stat *w, int revents)	1938	stat_cb (EV_P_ ev_stat *w, int revents)
1634	{	1939	{
1635	/* reset the one-second timer */	1940	/* reset the one-second timer */
1636	ev_timer_again (EV_A_ &timer);	1941	ev_timer_again (EV_A_ &timer);
1637	}	1942	}
1638		1943
1639	...	1944	...
1640	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);	1945	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
1641	ev_stat_start (loop, &passwd);	1946	ev_stat_start (loop, &passwd);
1642	ev_timer_init (&timer, timer_cb, 0., 1.01);	1947	ev_timer_init (&timer, timer_cb, 0., 1.02);
1643		1948
1644		1949
1645	=head2 C<ev_idle> - when you've got nothing better to do...	1950	=head2 C<ev_idle> - when you've got nothing better to do...
1646		1951
1647	Idle watchers trigger events when no other events of the same or higher	1952	Idle watchers trigger events when no other events of the same or higher
1648	priority are pending (prepare, check and other idle watchers do not	1953	priority are pending (prepare, check and other idle watchers do not count
1649	count).	1954	as receiving "events").
1650		1955
1651	That is, as long as your process is busy handling sockets or timeouts	1956	That is, as long as your process is busy handling sockets or timeouts
1652	(or even signals, imagine) of the same or higher priority it will not be	1957	(or even signals, imagine) of the same or higher priority it will not be
1653	triggered. But when your process is idle (or only lower-priority watchers	1958	triggered. But when your process is idle (or only lower-priority watchers
1654	are pending), the idle watchers are being called once per event loop	1959	are pending), the idle watchers are being called once per event loop
…		…
1678	=head3 Examples	1983	=head3 Examples
1679		1984
1680	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	1985	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1681	callback, free it. Also, use no error checking, as usual.	1986	callback, free it. Also, use no error checking, as usual.
1682		1987
1683	static void	1988	static void
1684	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	1989	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1685	{	1990	{
1686	free (w);	1991	free (w);
1687	// now do something you wanted to do when the program has	1992	// now do something you wanted to do when the program has
1688	// no longer asnything immediate to do.	1993	// no longer anything immediate to do.
1689	}	1994	}
1690		1995
1691	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1996	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1692	ev_idle_init (idle_watcher, idle_cb);	1997	ev_idle_init (idle_watcher, idle_cb);
1693	ev_idle_start (loop, idle_cb);	1998	ev_idle_start (loop, idle_cb);
1694		1999
1695		2000
1696	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	2001	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1697		2002
1698	Prepare and check watchers are usually (but not always) used in tandem:	2003	Prepare and check watchers are usually (but not always) used in pairs:
1699	prepare watchers get invoked before the process blocks and check watchers	2004	prepare watchers get invoked before the process blocks and check watchers
1700	afterwards.	2005	afterwards.
1701		2006
1702	You I<must not> call C<ev_loop> or similar functions that enter	2007	You I<must not> call C<ev_loop> or similar functions that enter
1703	the current event loop from either C<ev_prepare> or C<ev_check>	2008	the current event loop from either C<ev_prepare> or C<ev_check>
…		…
1706	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,	2011	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
1707	C<ev_check> so if you have one watcher of each kind they will always be	2012	C<ev_check> so if you have one watcher of each kind they will always be
1708	called in pairs bracketing the blocking call.	2013	called in pairs bracketing the blocking call.
1709		2014
1710	Their main purpose is to integrate other event mechanisms into libev and	2015	Their main purpose is to integrate other event mechanisms into libev and
1711	their use is somewhat advanced. This could be used, for example, to track	2016	their use is somewhat advanced. They could be used, for example, to track
1712	variable changes, implement your own watchers, integrate net-snmp or a	2017	variable changes, implement your own watchers, integrate net-snmp or a
1713	coroutine library and lots more. They are also occasionally useful if	2018	coroutine library and lots more. They are also occasionally useful if
1714	you cache some data and want to flush it before blocking (for example,	2019	you cache some data and want to flush it before blocking (for example,
1715	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>	2020	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
1716	watcher).	2021	watcher).
1717		2022
1718	This is done by examining in each prepare call which file descriptors need	2023	This is done by examining in each prepare call which file descriptors
1719	to be watched by the other library, registering C<ev_io> watchers for	2024	need to be watched by the other library, registering C<ev_io> watchers
1720	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2025	for them and starting an C<ev_timer> watcher for any timeouts (many
1721	provide just this functionality). Then, in the check watcher you check for	2026	libraries provide exactly this functionality). Then, in the check watcher,
1722	any events that occured (by checking the pending status of all watchers	2027	you check for any events that occurred (by checking the pending status
1723	and stopping them) and call back into the library. The I/O and timer	2028	of all watchers and stopping them) and call back into the library. The
1724	callbacks will never actually be called (but must be valid nevertheless,	2029	I/O and timer callbacks will never actually be called (but must be valid
1725	because you never know, you know?).	2030	nevertheless, because you never know, you know?).
1726		2031
1727	As another example, the Perl Coro module uses these hooks to integrate	2032	As another example, the Perl Coro module uses these hooks to integrate
1728	coroutines into libev programs, by yielding to other active coroutines	2033	coroutines into libev programs, by yielding to other active coroutines
1729	during each prepare and only letting the process block if no coroutines	2034	during each prepare and only letting the process block if no coroutines
1730	are ready to run (it's actually more complicated: it only runs coroutines	2035	are ready to run (it's actually more complicated: it only runs coroutines
…		…
1733	loop from blocking if lower-priority coroutines are active, thus mapping	2038	loop from blocking if lower-priority coroutines are active, thus mapping
1734	low-priority coroutines to idle/background tasks).	2039	low-priority coroutines to idle/background tasks).
1735		2040
1736	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	2041	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
1737	priority, to ensure that they are being run before any other watchers	2042	priority, to ensure that they are being run before any other watchers
		2043	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2044
1738	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	2045	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
1739	too) should not activate ("feed") events into libev. While libev fully	2046	activate ("feed") events into libev. While libev fully supports this, they
1740	supports this, they will be called before other C<ev_check> watchers	2047	might get executed before other C<ev_check> watchers did their job. As
1741	did their job. As C<ev_check> watchers are often used to embed other	2048	C<ev_check> watchers are often used to embed other (non-libev) event
1742	(non-libev) event loops those other event loops might be in an unusable	2049	loops those other event loops might be in an unusable state until their
1743	state until their C<ev_check> watcher ran (always remind yourself to	2050	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
1744	coexist peacefully with others).	2051	others).
1745		2052
1746	=head3 Watcher-Specific Functions and Data Members	2053	=head3 Watcher-Specific Functions and Data Members
1747		2054
1748	=over 4	2055	=over 4
1749		2056
…		…
1751		2058
1752	=item ev_check_init (ev_check *, callback)	2059	=item ev_check_init (ev_check *, callback)
1753		2060
1754	Initialises and configures the prepare or check watcher - they have no	2061	Initialises and configures the prepare or check watcher - they have no
1755	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2062	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1756	macros, but using them is utterly, utterly and completely pointless.	2063	macros, but using them is utterly, utterly, utterly and completely
		2064	pointless.
1757		2065
1758	=back	2066	=back
1759		2067
1760	=head3 Examples	2068	=head3 Examples
1761		2069
1762	There are a number of principal ways to embed other event loops or modules	2070	There are a number of principal ways to embed other event loops or modules
1763	into libev. Here are some ideas on how to include libadns into libev	2071	into libev. Here are some ideas on how to include libadns into libev
1764	(there is a Perl module named C<EV::ADNS> that does this, which you could	2072	(there is a Perl module named C<EV::ADNS> that does this, which you could
1765	use for an actually working example. Another Perl module named C<EV::Glib>	2073	use as a working example. Another Perl module named C<EV::Glib> embeds a
1766	embeds a Glib main context into libev, and finally, C<Glib::EV> embeds EV	2074	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
1767	into the Glib event loop).	2075	Glib event loop).
1768		2076
1769	Method 1: Add IO watchers and a timeout watcher in a prepare handler,	2077	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1770	and in a check watcher, destroy them and call into libadns. What follows	2078	and in a check watcher, destroy them and call into libadns. What follows
1771	is pseudo-code only of course. This requires you to either use a low	2079	is pseudo-code only of course. This requires you to either use a low
1772	priority for the check watcher or use C<ev_clear_pending> explicitly, as	2080	priority for the check watcher or use C<ev_clear_pending> explicitly, as
1773	the callbacks for the IO/timeout watchers might not have been called yet.	2081	the callbacks for the IO/timeout watchers might not have been called yet.
1774		2082
1775	static ev_io iow [nfd];	2083	static ev_io iow [nfd];
1776	static ev_timer tw;	2084	static ev_timer tw;
1777		2085
1778	static void	2086	static void
1779	io_cb (ev_loop *loop, ev_io *w, int revents)	2087	io_cb (ev_loop *loop, ev_io *w, int revents)
1780	{	2088	{
1781	}	2089	}
1782		2090
1783	// create io watchers for each fd and a timer before blocking	2091	// create io watchers for each fd and a timer before blocking
1784	static void	2092	static void
1785	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	2093	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1786	{	2094	{
1787	int timeout = 3600000;	2095	int timeout = 3600000;
1788	struct pollfd fds [nfd];	2096	struct pollfd fds [nfd];
1789	// actual code will need to loop here and realloc etc.	2097	// actual code will need to loop here and realloc etc.
1790	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	2098	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1791		2099
1792	/* the callback is illegal, but won't be called as we stop during check */	2100	/* the callback is illegal, but won't be called as we stop during check */
1793	ev_timer_init (&tw, 0, timeout * 1e-3);	2101	ev_timer_init (&tw, 0, timeout * 1e-3);
1794	ev_timer_start (loop, &tw);	2102	ev_timer_start (loop, &tw);
1795		2103
1796	// create one ev_io per pollfd	2104	// create one ev_io per pollfd
1797	for (int i = 0; i < nfd; ++i)	2105	for (int i = 0; i < nfd; ++i)
1798	{	2106	{
1799	ev_io_init (iow + i, io_cb, fds [i].fd,	2107	ev_io_init (iow + i, io_cb, fds [i].fd,
1800	((fds [i].events & POLLIN ? EV_READ : 0)	2108	((fds [i].events & POLLIN ? EV_READ : 0)
1801	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	2109	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1802		2110
1803	fds [i].revents = 0;	2111	fds [i].revents = 0;
1804	ev_io_start (loop, iow + i);	2112	ev_io_start (loop, iow + i);
1805	}	2113	}
1806	}	2114	}
1807		2115
1808	// stop all watchers after blocking	2116	// stop all watchers after blocking
1809	static void	2117	static void
1810	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2118	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1811	{	2119	{
1812	ev_timer_stop (loop, &tw);	2120	ev_timer_stop (loop, &tw);
1813		2121
1814	for (int i = 0; i < nfd; ++i)	2122	for (int i = 0; i < nfd; ++i)
1815	{	2123	{
1816	// set the relevant poll flags	2124	// set the relevant poll flags
1817	// could also call adns_processreadable etc. here	2125	// could also call adns_processreadable etc. here
1818	struct pollfd *fd = fds + i;	2126	struct pollfd *fd = fds + i;
1819	int revents = ev_clear_pending (iow + i);	2127	int revents = ev_clear_pending (iow + i);
1820	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;	2128	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1821	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;	2129	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1822		2130
1823	// now stop the watcher	2131	// now stop the watcher
1824	ev_io_stop (loop, iow + i);	2132	ev_io_stop (loop, iow + i);
1825	}	2133	}
1826		2134
1827	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	2135	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1828	}	2136	}
1829		2137
1830	Method 2: This would be just like method 1, but you run C<adns_afterpoll>	2138	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
1831	in the prepare watcher and would dispose of the check watcher.	2139	in the prepare watcher and would dispose of the check watcher.
1832		2140
1833	Method 3: If the module to be embedded supports explicit event	2141	Method 3: If the module to be embedded supports explicit event
1834	notification (adns does), you can also make use of the actual watcher	2142	notification (libadns does), you can also make use of the actual watcher
1835	callbacks, and only destroy/create the watchers in the prepare watcher.	2143	callbacks, and only destroy/create the watchers in the prepare watcher.
1836		2144
1837	static void	2145	static void
1838	timer_cb (EV_P_ ev_timer *w, int revents)	2146	timer_cb (EV_P_ ev_timer *w, int revents)
1839	{	2147	{
1840	adns_state ads = (adns_state)w->data;	2148	adns_state ads = (adns_state)w->data;
1841	update_now (EV_A);	2149	update_now (EV_A);
1842		2150
1843	adns_processtimeouts (ads, &tv_now);	2151	adns_processtimeouts (ads, &tv_now);
1844	}	2152	}
1845		2153
1846	static void	2154	static void
1847	io_cb (EV_P_ ev_io *w, int revents)	2155	io_cb (EV_P_ ev_io *w, int revents)
1848	{	2156	{
1849	adns_state ads = (adns_state)w->data;	2157	adns_state ads = (adns_state)w->data;
1850	update_now (EV_A);	2158	update_now (EV_A);
1851		2159
1852	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);	2160	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
1853	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);	2161	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
1854	}	2162	}
1855		2163
1856	// do not ever call adns_afterpoll	2164	// do not ever call adns_afterpoll
1857		2165
1858	Method 4: Do not use a prepare or check watcher because the module you	2166	Method 4: Do not use a prepare or check watcher because the module you
1859	want to embed is too inflexible to support it. Instead, youc na override	2167	want to embed is not flexible enough to support it. Instead, you can
1860	their poll function. The drawback with this solution is that the main	2168	override their poll function. The drawback with this solution is that the
1861	loop is now no longer controllable by EV. The C<Glib::EV> module does	2169	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
1862	this.	2170	this approach, effectively embedding EV as a client into the horrible
		2171	libglib event loop.
1863		2172
1864	static gint	2173	static gint
1865	event_poll_func (GPollFD *fds, guint nfds, gint timeout)	2174	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
1866	{	2175	{
1867	int got_events = 0;	2176	int got_events = 0;
1868		2177
1869	for (n = 0; n < nfds; ++n)	2178	for (n = 0; n < nfds; ++n)
1870	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events	2179	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
1871		2180
1872	if (timeout >= 0)	2181	if (timeout >= 0)
1873	// create/start timer	2182	// create/start timer
1874		2183
1875	// poll	2184	// poll
1876	ev_loop (EV_A_ 0);	2185	ev_loop (EV_A_ 0);
1877		2186
1878	// stop timer again	2187	// stop timer again
1879	if (timeout >= 0)	2188	if (timeout >= 0)
1880	ev_timer_stop (EV_A_ &to);	2189	ev_timer_stop (EV_A_ &to);
1881		2190
1882	// stop io watchers again - their callbacks should have set	2191	// stop io watchers again - their callbacks should have set
1883	for (n = 0; n < nfds; ++n)	2192	for (n = 0; n < nfds; ++n)
1884	ev_io_stop (EV_A_ iow [n]);	2193	ev_io_stop (EV_A_ iow [n]);
1885		2194
1886	return got_events;	2195	return got_events;
1887	}	2196	}
1888		2197
1889		2198
1890	=head2 C<ev_embed> - when one backend isn't enough...	2199	=head2 C<ev_embed> - when one backend isn't enough...
1891		2200
1892	This is a rather advanced watcher type that lets you embed one event loop	2201	This is a rather advanced watcher type that lets you embed one event loop
…		…
1898	prioritise I/O.	2207	prioritise I/O.
1899		2208
1900	As an example for a bug workaround, the kqueue backend might only support	2209	As an example for a bug workaround, the kqueue backend might only support
1901	sockets on some platform, so it is unusable as generic backend, but you	2210	sockets on some platform, so it is unusable as generic backend, but you
1902	still want to make use of it because you have many sockets and it scales	2211	still want to make use of it because you have many sockets and it scales
1903	so nicely. In this case, you would create a kqueue-based loop and embed it	2212	so nicely. In this case, you would create a kqueue-based loop and embed
1904	into your default loop (which might use e.g. poll). Overall operation will	2213	it into your default loop (which might use e.g. poll). Overall operation
1905	be a bit slower because first libev has to poll and then call kevent, but	2214	will be a bit slower because first libev has to call C<poll> and then
1906	at least you can use both at what they are best.	2215	C<kevent>, but at least you can use both mechanisms for what they are
		2216	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
1907		2217
1908	As for prioritising I/O: rarely you have the case where some fds have	2218	As for prioritising I/O: under rare circumstances you have the case where
1909	to be watched and handled very quickly (with low latency), and even	2219	some fds have to be watched and handled very quickly (with low latency),
1910	priorities and idle watchers might have too much overhead. In this case	2220	and even priorities and idle watchers might have too much overhead. In
1911	you would put all the high priority stuff in one loop and all the rest in	2221	this case you would put all the high priority stuff in one loop and all
1912	a second one, and embed the second one in the first.	2222	the rest in a second one, and embed the second one in the first.
1913		2223
1914	As long as the watcher is active, the callback will be invoked every time	2224	As long as the watcher is active, the callback will be invoked every time
1915	there might be events pending in the embedded loop. The callback must then	2225	there might be events pending in the embedded loop. The callback must then
1916	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke	2226	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke
1917	their callbacks (you could also start an idle watcher to give the embedded	2227	their callbacks (you could also start an idle watcher to give the embedded
…		…
1925	interested in that.	2235	interested in that.
1926		2236
1927	Also, there have not currently been made special provisions for forking:	2237	Also, there have not currently been made special provisions for forking:
1928	when you fork, you not only have to call C<ev_loop_fork> on both loops,	2238	when you fork, you not only have to call C<ev_loop_fork> on both loops,
1929	but you will also have to stop and restart any C<ev_embed> watchers	2239	but you will also have to stop and restart any C<ev_embed> watchers
1930	yourself.	2240	yourself - but you can use a fork watcher to handle this automatically,
		2241	and future versions of libev might do just that.
1931		2242
1932	Unfortunately, not all backends are embeddable, only the ones returned by	2243	Unfortunately, not all backends are embeddable: only the ones returned by
1933	C<ev_embeddable_backends> are, which, unfortunately, does not include any	2244	C<ev_embeddable_backends> are, which, unfortunately, does not include any
1934	portable one.	2245	portable one.
1935		2246
1936	So when you want to use this feature you will always have to be prepared	2247	So when you want to use this feature you will always have to be prepared
1937	that you cannot get an embeddable loop. The recommended way to get around	2248	that you cannot get an embeddable loop. The recommended way to get around
1938	this is to have a separate variables for your embeddable loop, try to	2249	this is to have a separate variables for your embeddable loop, try to
1939	create it, and if that fails, use the normal loop for everything.	2250	create it, and if that fails, use the normal loop for everything.
1940		2251
		2252	=head3 C<ev_embed> and fork
		2253
		2254	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2255	automatically be applied to the embedded loop as well, so no special
		2256	fork handling is required in that case. When the watcher is not running,
		2257	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2258	as applicable.
		2259
1941	=head3 Watcher-Specific Functions and Data Members	2260	=head3 Watcher-Specific Functions and Data Members
1942		2261
1943	=over 4	2262	=over 4
1944		2263
1945	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	2264	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
…		…
1948		2267
1949	Configures the watcher to embed the given loop, which must be	2268	Configures the watcher to embed the given loop, which must be
1950	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be	2269	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1951	invoked automatically, otherwise it is the responsibility of the callback	2270	invoked automatically, otherwise it is the responsibility of the callback
1952	to invoke it (it will continue to be called until the sweep has been done,	2271	to invoke it (it will continue to be called until the sweep has been done,
1953	if you do not want thta, you need to temporarily stop the embed watcher).	2272	if you do not want that, you need to temporarily stop the embed watcher).
1954		2273
1955	=item ev_embed_sweep (loop, ev_embed *)	2274	=item ev_embed_sweep (loop, ev_embed *)
1956		2275
1957	Make a single, non-blocking sweep over the embedded loop. This works	2276	Make a single, non-blocking sweep over the embedded loop. This works
1958	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	2277	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
1959	apropriate way for embedded loops.	2278	appropriate way for embedded loops.
1960		2279
1961	=item struct ev_loop *other [read-only]	2280	=item struct ev_loop *other [read-only]
1962		2281
1963	The embedded event loop.	2282	The embedded event loop.
1964		2283
…		…
1966		2285
1967	=head3 Examples	2286	=head3 Examples
1968		2287
1969	Example: Try to get an embeddable event loop and embed it into the default	2288	Example: Try to get an embeddable event loop and embed it into the default
1970	event loop. If that is not possible, use the default loop. The default	2289	event loop. If that is not possible, use the default loop. The default
1971	loop is stored in C<loop_hi>, while the mebeddable loop is stored in	2290	loop is stored in C<loop_hi>, while the embeddable loop is stored in
1972	C<loop_lo> (which is C<loop_hi> in the acse no embeddable loop can be	2291	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
1973	used).	2292	used).
1974		2293
1975	struct ev_loop *loop_hi = ev_default_init (0);	2294	struct ev_loop *loop_hi = ev_default_init (0);
1976	struct ev_loop *loop_lo = 0;	2295	struct ev_loop *loop_lo = 0;
1977	struct ev_embed embed;	2296	struct ev_embed embed;
1978		2297
1979	// see if there is a chance of getting one that works	2298	// see if there is a chance of getting one that works
1980	// (remember that a flags value of 0 means autodetection)	2299	// (remember that a flags value of 0 means autodetection)
1981	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()	2300	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1982	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())	2301	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1983	: 0;	2302	: 0;
1984		2303
1985	// if we got one, then embed it, otherwise default to loop_hi	2304	// if we got one, then embed it, otherwise default to loop_hi
1986	if (loop_lo)	2305	if (loop_lo)
1987	{	2306	{
1988	ev_embed_init (&embed, 0, loop_lo);	2307	ev_embed_init (&embed, 0, loop_lo);
1989	ev_embed_start (loop_hi, &embed);	2308	ev_embed_start (loop_hi, &embed);
1990	}	2309	}
1991	else	2310	else
1992	loop_lo = loop_hi;	2311	loop_lo = loop_hi;
1993		2312
1994	Example: Check if kqueue is available but not recommended and create	2313	Example: Check if kqueue is available but not recommended and create
1995	a kqueue backend for use with sockets (which usually work with any	2314	a kqueue backend for use with sockets (which usually work with any
1996	kqueue implementation). Store the kqueue/socket-only event loop in	2315	kqueue implementation). Store the kqueue/socket-only event loop in
1997	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).	2316	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
1998		2317
1999	struct ev_loop *loop = ev_default_init (0);	2318	struct ev_loop *loop = ev_default_init (0);
2000	struct ev_loop *loop_socket = 0;	2319	struct ev_loop *loop_socket = 0;
2001	struct ev_embed embed;	2320	struct ev_embed embed;
2002		2321
2003	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)	2322	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
2004	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))	2323	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
2005	{	2324	{
2006	ev_embed_init (&embed, 0, loop_socket);	2325	ev_embed_init (&embed, 0, loop_socket);
2007	ev_embed_start (loop, &embed);	2326	ev_embed_start (loop, &embed);
2008	}	2327	}
2009		2328
2010	if (!loop_socket)	2329	if (!loop_socket)
2011	loop_socket = loop;	2330	loop_socket = loop;
2012		2331
2013	// now use loop_socket for all sockets, and loop for everything else	2332	// now use loop_socket for all sockets, and loop for everything else
2014		2333
2015		2334
2016	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	2335	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
2017		2336
2018	Fork watchers are called when a C<fork ()> was detected (usually because	2337	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
2034	believe me.	2353	believe me.
2035		2354
2036	=back	2355	=back
2037		2356
2038		2357
		2358	=head2 C<ev_async> - how to wake up another event loop
		2359
		2360	In general, you cannot use an C<ev_loop> from multiple threads or other
		2361	asynchronous sources such as signal handlers (as opposed to multiple event
		2362	loops - those are of course safe to use in different threads).
		2363
		2364	Sometimes, however, you need to wake up another event loop you do not
		2365	control, for example because it belongs to another thread. This is what
		2366	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2367	can signal it by calling C<ev_async_send>, which is thread- and signal
		2368	safe.
		2369
		2370	This functionality is very similar to C<ev_signal> watchers, as signals,
		2371	too, are asynchronous in nature, and signals, too, will be compressed
		2372	(i.e. the number of callback invocations may be less than the number of
		2373	C<ev_async_sent> calls).
		2374
		2375	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2376	just the default loop.
		2377
		2378	=head3 Queueing
		2379
		2380	C<ev_async> does not support queueing of data in any way. The reason
		2381	is that the author does not know of a simple (or any) algorithm for a
		2382	multiple-writer-single-reader queue that works in all cases and doesn't
		2383	need elaborate support such as pthreads.
		2384
		2385	That means that if you want to queue data, you have to provide your own
		2386	queue. But at least I can tell you how to implement locking around your
		2387	queue:
		2388
		2389	=over 4
		2390
		2391	=item queueing from a signal handler context
		2392
		2393	To implement race-free queueing, you simply add to the queue in the signal
		2394	handler but you block the signal handler in the watcher callback. Here is
		2395	an example that does that for some fictitious SIGUSR1 handler:
		2396
		2397	static ev_async mysig;
		2398
		2399	static void
		2400	sigusr1_handler (void)
		2401	{
		2402	sometype data;
		2403
		2404	// no locking etc.
		2405	queue_put (data);
		2406	ev_async_send (EV_DEFAULT_ &mysig);
		2407	}
		2408
		2409	static void
		2410	mysig_cb (EV_P_ ev_async *w, int revents)
		2411	{
		2412	sometype data;
		2413	sigset_t block, prev;
		2414
		2415	sigemptyset (&block);
		2416	sigaddset (&block, SIGUSR1);
		2417	sigprocmask (SIG_BLOCK, &block, &prev);
		2418
		2419	while (queue_get (&data))
		2420	process (data);
		2421
		2422	if (sigismember (&prev, SIGUSR1)
		2423	sigprocmask (SIG_UNBLOCK, &block, 0);
		2424	}
		2425
		2426	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2427	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2428	either...).
		2429
		2430	=item queueing from a thread context
		2431
		2432	The strategy for threads is different, as you cannot (easily) block
		2433	threads but you can easily preempt them, so to queue safely you need to
		2434	employ a traditional mutex lock, such as in this pthread example:
		2435
		2436	static ev_async mysig;
		2437	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2438
		2439	static void
		2440	otherthread (void)
		2441	{
		2442	// only need to lock the actual queueing operation
		2443	pthread_mutex_lock (&mymutex);
		2444	queue_put (data);
		2445	pthread_mutex_unlock (&mymutex);
		2446
		2447	ev_async_send (EV_DEFAULT_ &mysig);
		2448	}
		2449
		2450	static void
		2451	mysig_cb (EV_P_ ev_async *w, int revents)
		2452	{
		2453	pthread_mutex_lock (&mymutex);
		2454
		2455	while (queue_get (&data))
		2456	process (data);
		2457
		2458	pthread_mutex_unlock (&mymutex);
		2459	}
		2460
		2461	=back
		2462
		2463
		2464	=head3 Watcher-Specific Functions and Data Members
		2465
		2466	=over 4
		2467
		2468	=item ev_async_init (ev_async *, callback)
		2469
		2470	Initialises and configures the async watcher - it has no parameters of any
		2471	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,
		2472	trust me.
		2473
		2474	=item ev_async_send (loop, ev_async *)
		2475
		2476	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2477	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2478	C<ev_feed_event>, this call is safe to do from other threads, signal or
		2479	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
		2480	section below on what exactly this means).
		2481
		2482	This call incurs the overhead of a system call only once per loop iteration,
		2483	so while the overhead might be noticeable, it doesn't apply to repeated
		2484	calls to C<ev_async_send>.
		2485
		2486	=item bool = ev_async_pending (ev_async *)
		2487
		2488	Returns a non-zero value when C<ev_async_send> has been called on the
		2489	watcher but the event has not yet been processed (or even noted) by the
		2490	event loop.
		2491
		2492	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2493	the loop iterates next and checks for the watcher to have become active,
		2494	it will reset the flag again. C<ev_async_pending> can be used to very
		2495	quickly check whether invoking the loop might be a good idea.
		2496
		2497	Not that this does I<not> check whether the watcher itself is pending, only
		2498	whether it has been requested to make this watcher pending.
		2499
		2500	=back
		2501
		2502
2039	=head1 OTHER FUNCTIONS	2503	=head1 OTHER FUNCTIONS
2040		2504
2041	There are some other functions of possible interest. Described. Here. Now.	2505	There are some other functions of possible interest. Described. Here. Now.
2042		2506
2043	=over 4	2507	=over 4
2044		2508
2045	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	2509	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2046		2510
2047	This function combines a simple timer and an I/O watcher, calls your	2511	This function combines a simple timer and an I/O watcher, calls your
2048	callback on whichever event happens first and automatically stop both	2512	callback on whichever event happens first and automatically stops both
2049	watchers. This is useful if you want to wait for a single event on an fd	2513	watchers. This is useful if you want to wait for a single event on an fd
2050	or timeout without having to allocate/configure/start/stop/free one or	2514	or timeout without having to allocate/configure/start/stop/free one or
2051	more watchers yourself.	2515	more watchers yourself.
2052		2516
2053	If C<fd> is less than 0, then no I/O watcher will be started and events	2517	If C<fd> is less than 0, then no I/O watcher will be started and the
2054	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2518	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2055	C<events> set will be craeted and started.	2519	the given C<fd> and C<events> set will be created and started.
2056		2520
2057	If C<timeout> is less than 0, then no timeout watcher will be	2521	If C<timeout> is less than 0, then no timeout watcher will be
2058	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2522	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
2059	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2523	repeat = 0) will be started. C<0> is a valid timeout.
2060	dubious value.
2061		2524
2062	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	2525	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
2063	passed an C<revents> set like normal event callbacks (a combination of	2526	passed an C<revents> set like normal event callbacks (a combination of
2064	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2527	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
2065	value passed to C<ev_once>:	2528	value passed to C<ev_once>. Note that it is possible to receive I<both>
		2529	a timeout and an io event at the same time - you probably should give io
		2530	events precedence.
2066		2531
		2532	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		2533
2067	static void stdin_ready (int revents, void *arg)	2534	static void stdin_ready (int revents, void *arg)
2068	{	2535	{
2069	if (revents & EV_TIMEOUT)
2070	/* doh, nothing entered */;
2071	else if (revents & EV_READ)	2536	if (revents & EV_READ)
2072	/* stdin might have data for us, joy! */;	2537	/* stdin might have data for us, joy! */;
		2538	else if (revents & EV_TIMEOUT)
		2539	/* doh, nothing entered */;
2073	}	2540	}
2074		2541
2075	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2542	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2076		2543
2077	=item ev_feed_event (ev_loop *, watcher *, int revents)	2544	=item ev_feed_event (ev_loop *, watcher *, int revents)
2078		2545
2079	Feeds the given event set into the event loop, as if the specified event	2546	Feeds the given event set into the event loop, as if the specified event
2080	had happened for the specified watcher (which must be a pointer to an	2547	had happened for the specified watcher (which must be a pointer to an
…		…
2085	Feed an event on the given fd, as if a file descriptor backend detected	2552	Feed an event on the given fd, as if a file descriptor backend detected
2086	the given events it.	2553	the given events it.
2087		2554
2088	=item ev_feed_signal_event (ev_loop *loop, int signum)	2555	=item ev_feed_signal_event (ev_loop *loop, int signum)
2089		2556
2090	Feed an event as if the given signal occured (C<loop> must be the default	2557	Feed an event as if the given signal occurred (C<loop> must be the default
2091	loop!).	2558	loop!).
2092		2559
2093	=back	2560	=back
2094		2561
2095		2562
…		…
2111		2578
2112	=item * Priorities are not currently supported. Initialising priorities	2579	=item * Priorities are not currently supported. Initialising priorities
2113	will fail and all watchers will have the same priority, even though there	2580	will fail and all watchers will have the same priority, even though there
2114	is an ev_pri field.	2581	is an ev_pri field.
2115		2582
		2583	=item * In libevent, the last base created gets the signals, in libev, the
		2584	first base created (== the default loop) gets the signals.
		2585
2116	=item * Other members are not supported.	2586	=item * Other members are not supported.
2117		2587
2118	=item * The libev emulation is I<not> ABI compatible to libevent, you need	2588	=item * The libev emulation is I<not> ABI compatible to libevent, you need
2119	to use the libev header file and library.	2589	to use the libev header file and library.
2120		2590
2121	=back	2591	=back
2122		2592
2123	=head1 C++ SUPPORT	2593	=head1 C++ SUPPORT
2124		2594
2125	Libev comes with some simplistic wrapper classes for C++ that mainly allow	2595	Libev comes with some simplistic wrapper classes for C++ that mainly allow
2126	you to use some convinience methods to start/stop watchers and also change	2596	you to use some convenience methods to start/stop watchers and also change
2127	the callback model to a model using method callbacks on objects.	2597	the callback model to a model using method callbacks on objects.
2128		2598
2129	To use it,	2599	To use it,
2130		2600
2131	#include <ev++.h>	2601	#include <ev++.h>
2132		2602
2133	This automatically includes F<ev.h> and puts all of its definitions (many	2603	This automatically includes F<ev.h> and puts all of its definitions (many
2134	of them macros) into the global namespace. All C++ specific things are	2604	of them macros) into the global namespace. All C++ specific things are
2135	put into the C<ev> namespace. It should support all the same embedding	2605	put into the C<ev> namespace. It should support all the same embedding
2136	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.	2606	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
…		…
2203	your compiler is good :), then the method will be fully inlined into the	2673	your compiler is good :), then the method will be fully inlined into the
2204	thunking function, making it as fast as a direct C callback.	2674	thunking function, making it as fast as a direct C callback.
2205		2675
2206	Example: simple class declaration and watcher initialisation	2676	Example: simple class declaration and watcher initialisation
2207		2677
2208	struct myclass	2678	struct myclass
2209	{	2679	{
2210	void io_cb (ev::io &w, int revents) { }	2680	void io_cb (ev::io &w, int revents) { }
2211	}	2681	}
2212		2682
2213	myclass obj;	2683	myclass obj;
2214	ev::io iow;	2684	ev::io iow;
2215	iow.set <myclass, &myclass::io_cb> (&obj);	2685	iow.set <myclass, &myclass::io_cb> (&obj);
2216		2686
2217	=item w->set<function> (void *data = 0)	2687	=item w->set<function> (void *data = 0)
2218		2688
2219	Also sets a callback, but uses a static method or plain function as	2689	Also sets a callback, but uses a static method or plain function as
2220	callback. The optional C<data> argument will be stored in the watcher's	2690	callback. The optional C<data> argument will be stored in the watcher's
…		…
2222		2692
2223	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.	2693	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
2224		2694
2225	See the method-C<set> above for more details.	2695	See the method-C<set> above for more details.
2226		2696
2227	Example:	2697	Example: Use a plain function as callback.
2228		2698
2229	static void io_cb (ev::io &w, int revents) { }	2699	static void io_cb (ev::io &w, int revents) { }
2230	iow.set <io_cb> ();	2700	iow.set <io_cb> ();
2231		2701
2232	=item w->set (struct ev_loop *)	2702	=item w->set (struct ev_loop *)
2233		2703
2234	Associates a different C<struct ev_loop> with this watcher. You can only	2704	Associates a different C<struct ev_loop> with this watcher. You can only
2235	do this when the watcher is inactive (and not pending either).	2705	do this when the watcher is inactive (and not pending either).
2236		2706
2237	=item w->set ([args])	2707	=item w->set ([arguments])
2238		2708
2239	Basically the same as C<ev_TYPE_set>, with the same args. Must be	2709	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be
2240	called at least once. Unlike the C counterpart, an active watcher gets	2710	called at least once. Unlike the C counterpart, an active watcher gets
2241	automatically stopped and restarted when reconfiguring it with this	2711	automatically stopped and restarted when reconfiguring it with this
2242	method.	2712	method.
2243		2713
2244	=item w->start ()	2714	=item w->start ()
…		…
2268	=back	2738	=back
2269		2739
2270	Example: Define a class with an IO and idle watcher, start one of them in	2740	Example: Define a class with an IO and idle watcher, start one of them in
2271	the constructor.	2741	the constructor.
2272		2742
2273	class myclass	2743	class myclass
2274	{	2744	{
2275	ev_io io; void io_cb (ev::io &w, int revents);	2745	ev::io io ; void io_cb (ev::io &w, int revents);
2276	ev_idle idle void idle_cb (ev::idle &w, int revents);	2746	ev::idle idle; void idle_cb (ev::idle &w, int revents);
2277		2747
2278	myclass ();	2748	myclass (int fd)
2279	}	2749	{
2280
2281	myclass::myclass (int fd)
2282	{
2283	io .set <myclass, &myclass::io_cb > (this);	2750	io .set <myclass, &myclass::io_cb > (this);
2284	idle.set <myclass, &myclass::idle_cb> (this);	2751	idle.set <myclass, &myclass::idle_cb> (this);
2285		2752
2286	io.start (fd, ev::READ);	2753	io.start (fd, ev::READ);
		2754	}
2287	}	2755	};
		2756
		2757
		2758	=head1 OTHER LANGUAGE BINDINGS
		2759
		2760	Libev does not offer other language bindings itself, but bindings for a
		2761	number of languages exist in the form of third-party packages. If you know
		2762	any interesting language binding in addition to the ones listed here, drop
		2763	me a note.
		2764
		2765	=over 4
		2766
		2767	=item Perl
		2768
		2769	The EV module implements the full libev API and is actually used to test
		2770	libev. EV is developed together with libev. Apart from the EV core module,
		2771	there are additional modules that implement libev-compatible interfaces
		2772	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
		2773	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		2774	and C<EV::Glib>).
		2775
		2776	It can be found and installed via CPAN, its homepage is at
		2777	L<http://software.schmorp.de/pkg/EV>.
		2778
		2779	=item Python
		2780
		2781	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		2782	seems to be quite complete and well-documented. Note, however, that the
		2783	patch they require for libev is outright dangerous as it breaks the ABI
		2784	for everybody else, and therefore, should never be applied in an installed
		2785	libev (if python requires an incompatible ABI then it needs to embed
		2786	libev).
		2787
		2788	=item Ruby
		2789
		2790	Tony Arcieri has written a ruby extension that offers access to a subset
		2791	of the libev API and adds file handle abstractions, asynchronous DNS and
		2792	more on top of it. It can be found via gem servers. Its homepage is at
		2793	L<http://rev.rubyforge.org/>.
		2794
		2795	=item D
		2796
		2797	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		2798	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		2799
		2800	=back
2288		2801
2289		2802
2290	=head1 MACRO MAGIC	2803	=head1 MACRO MAGIC
2291		2804
2292	Libev can be compiled with a variety of options, the most fundamantal	2805	Libev can be compiled with a variety of options, the most fundamental
2293	of which is C<EV_MULTIPLICITY>. This option determines whether (most)	2806	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
2294	functions and callbacks have an initial C<struct ev_loop *> argument.	2807	functions and callbacks have an initial C<struct ev_loop *> argument.
2295		2808
2296	To make it easier to write programs that cope with either variant, the	2809	To make it easier to write programs that cope with either variant, the
2297	following macros are defined:	2810	following macros are defined:
…		…
2302		2815
2303	This provides the loop I<argument> for functions, if one is required ("ev	2816	This provides the loop I<argument> for functions, if one is required ("ev
2304	loop argument"). The C<EV_A> form is used when this is the sole argument,	2817	loop argument"). The C<EV_A> form is used when this is the sole argument,
2305	C<EV_A_> is used when other arguments are following. Example:	2818	C<EV_A_> is used when other arguments are following. Example:
2306		2819
2307	ev_unref (EV_A);	2820	ev_unref (EV_A);
2308	ev_timer_add (EV_A_ watcher);	2821	ev_timer_add (EV_A_ watcher);
2309	ev_loop (EV_A_ 0);	2822	ev_loop (EV_A_ 0);
2310		2823
2311	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	2824	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
2312	which is often provided by the following macro.	2825	which is often provided by the following macro.
2313		2826
2314	=item C<EV_P>, C<EV_P_>	2827	=item C<EV_P>, C<EV_P_>
2315		2828
2316	This provides the loop I<parameter> for functions, if one is required ("ev	2829	This provides the loop I<parameter> for functions, if one is required ("ev
2317	loop parameter"). The C<EV_P> form is used when this is the sole parameter,	2830	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
2318	C<EV_P_> is used when other parameters are following. Example:	2831	C<EV_P_> is used when other parameters are following. Example:
2319		2832
2320	// this is how ev_unref is being declared	2833	// this is how ev_unref is being declared
2321	static void ev_unref (EV_P);	2834	static void ev_unref (EV_P);
2322		2835
2323	// this is how you can declare your typical callback	2836	// this is how you can declare your typical callback
2324	static void cb (EV_P_ ev_timer *w, int revents)	2837	static void cb (EV_P_ ev_timer *w, int revents)
2325		2838
2326	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	2839	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
2327	suitable for use with C<EV_A>.	2840	suitable for use with C<EV_A>.
2328		2841
2329	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	2842	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
2330		2843
2331	Similar to the other two macros, this gives you the value of the default	2844	Similar to the other two macros, this gives you the value of the default
2332	loop, if multiple loops are supported ("ev loop default").	2845	loop, if multiple loops are supported ("ev loop default").
		2846
		2847	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		2848
		2849	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		2850	default loop has been initialised (C<UC> == unchecked). Their behaviour
		2851	is undefined when the default loop has not been initialised by a previous
		2852	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		2853
		2854	It is often prudent to use C<EV_DEFAULT> when initialising the first
		2855	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
2333		2856
2334	=back	2857	=back
2335		2858
2336	Example: Declare and initialise a check watcher, utilising the above	2859	Example: Declare and initialise a check watcher, utilising the above
2337	macros so it will work regardless of whether multiple loops are supported	2860	macros so it will work regardless of whether multiple loops are supported
2338	or not.	2861	or not.
2339		2862
2340	static void	2863	static void
2341	check_cb (EV_P_ ev_timer *w, int revents)	2864	check_cb (EV_P_ ev_timer *w, int revents)
2342	{	2865	{
2343	ev_check_stop (EV_A_ w);	2866	ev_check_stop (EV_A_ w);
2344	}	2867	}
2345		2868
2346	ev_check check;	2869	ev_check check;
2347	ev_check_init (&check, check_cb);	2870	ev_check_init (&check, check_cb);
2348	ev_check_start (EV_DEFAULT_ &check);	2871	ev_check_start (EV_DEFAULT_ &check);
2349	ev_loop (EV_DEFAULT_ 0);	2872	ev_loop (EV_DEFAULT_ 0);
2350		2873
2351	=head1 EMBEDDING	2874	=head1 EMBEDDING
2352		2875
2353	Libev can (and often is) directly embedded into host	2876	Libev can (and often is) directly embedded into host
2354	applications. Examples of applications that embed it include the Deliantra	2877	applications. Examples of applications that embed it include the Deliantra
…		…
2361	libev somewhere in your source tree).	2884	libev somewhere in your source tree).
2362		2885
2363	=head2 FILESETS	2886	=head2 FILESETS
2364		2887
2365	Depending on what features you need you need to include one or more sets of files	2888	Depending on what features you need you need to include one or more sets of files
2366	in your app.	2889	in your application.
2367		2890
2368	=head3 CORE EVENT LOOP	2891	=head3 CORE EVENT LOOP
2369		2892
2370	To include only the libev core (all the C<ev_*> functions), with manual	2893	To include only the libev core (all the C<ev_*> functions), with manual
2371	configuration (no autoconf):	2894	configuration (no autoconf):
2372		2895
2373	#define EV_STANDALONE 1	2896	#define EV_STANDALONE 1
2374	#include "ev.c"	2897	#include "ev.c"
2375		2898
2376	This will automatically include F<ev.h>, too, and should be done in a	2899	This will automatically include F<ev.h>, too, and should be done in a
2377	single C source file only to provide the function implementations. To use	2900	single C source file only to provide the function implementations. To use
2378	it, do the same for F<ev.h> in all files wishing to use this API (best	2901	it, do the same for F<ev.h> in all files wishing to use this API (best
2379	done by writing a wrapper around F<ev.h> that you can include instead and	2902	done by writing a wrapper around F<ev.h> that you can include instead and
2380	where you can put other configuration options):	2903	where you can put other configuration options):
2381		2904
2382	#define EV_STANDALONE 1	2905	#define EV_STANDALONE 1
2383	#include "ev.h"	2906	#include "ev.h"
2384		2907
2385	Both header files and implementation files can be compiled with a C++	2908	Both header files and implementation files can be compiled with a C++
2386	compiler (at least, thats a stated goal, and breakage will be treated	2909	compiler (at least, thats a stated goal, and breakage will be treated
2387	as a bug).	2910	as a bug).
2388		2911
2389	You need the following files in your source tree, or in a directory	2912	You need the following files in your source tree, or in a directory
2390	in your include path (e.g. in libev/ when using -Ilibev):	2913	in your include path (e.g. in libev/ when using -Ilibev):
2391		2914
2392	ev.h	2915	ev.h
2393	ev.c	2916	ev.c
2394	ev_vars.h	2917	ev_vars.h
2395	ev_wrap.h	2918	ev_wrap.h
2396		2919
2397	ev_win32.c required on win32 platforms only	2920	ev_win32.c required on win32 platforms only
2398		2921
2399	ev_select.c only when select backend is enabled (which is enabled by default)	2922	ev_select.c only when select backend is enabled (which is enabled by default)
2400	ev_poll.c only when poll backend is enabled (disabled by default)	2923	ev_poll.c only when poll backend is enabled (disabled by default)
2401	ev_epoll.c only when the epoll backend is enabled (disabled by default)	2924	ev_epoll.c only when the epoll backend is enabled (disabled by default)
2402	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	2925	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
2403	ev_port.c only when the solaris port backend is enabled (disabled by default)	2926	ev_port.c only when the solaris port backend is enabled (disabled by default)
2404		2927
2405	F<ev.c> includes the backend files directly when enabled, so you only need	2928	F<ev.c> includes the backend files directly when enabled, so you only need
2406	to compile this single file.	2929	to compile this single file.
2407		2930
2408	=head3 LIBEVENT COMPATIBILITY API	2931	=head3 LIBEVENT COMPATIBILITY API
2409		2932
2410	To include the libevent compatibility API, also include:	2933	To include the libevent compatibility API, also include:
2411		2934
2412	#include "event.c"	2935	#include "event.c"
2413		2936
2414	in the file including F<ev.c>, and:	2937	in the file including F<ev.c>, and:
2415		2938
2416	#include "event.h"	2939	#include "event.h"
2417		2940
2418	in the files that want to use the libevent API. This also includes F<ev.h>.	2941	in the files that want to use the libevent API. This also includes F<ev.h>.
2419		2942
2420	You need the following additional files for this:	2943	You need the following additional files for this:
2421		2944
2422	event.h	2945	event.h
2423	event.c	2946	event.c
2424		2947
2425	=head3 AUTOCONF SUPPORT	2948	=head3 AUTOCONF SUPPORT
2426		2949
2427	Instead of using C<EV_STANDALONE=1> and providing your config in	2950	Instead of using C<EV_STANDALONE=1> and providing your configuration in
2428	whatever way you want, you can also C<m4_include([libev.m4])> in your	2951	whatever way you want, you can also C<m4_include([libev.m4])> in your
2429	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then	2952	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
2430	include F<config.h> and configure itself accordingly.	2953	include F<config.h> and configure itself accordingly.
2431		2954
2432	For this of course you need the m4 file:	2955	For this of course you need the m4 file:
2433		2956
2434	libev.m4	2957	libev.m4
2435		2958
2436	=head2 PREPROCESSOR SYMBOLS/MACROS	2959	=head2 PREPROCESSOR SYMBOLS/MACROS
2437		2960
2438	Libev can be configured via a variety of preprocessor symbols you have to define	2961	Libev can be configured via a variety of preprocessor symbols you have to
2439	before including any of its files. The default is not to build for multiplicity	2962	define before including any of its files. The default in the absence of
2440	and only include the select backend.	2963	autoconf is documented for every option.
2441		2964
2442	=over 4	2965	=over 4
2443		2966
2444	=item EV_STANDALONE	2967	=item EV_STANDALONE
2445		2968
…		…
2450	F<event.h> that are not directly supported by the libev core alone.	2973	F<event.h> that are not directly supported by the libev core alone.
2451		2974
2452	=item EV_USE_MONOTONIC	2975	=item EV_USE_MONOTONIC
2453		2976
2454	If defined to be C<1>, libev will try to detect the availability of the	2977	If defined to be C<1>, libev will try to detect the availability of the
2455	monotonic clock option at both compiletime and runtime. Otherwise no use	2978	monotonic clock option at both compile time and runtime. Otherwise no use
2456	of the monotonic clock option will be attempted. If you enable this, you	2979	of the monotonic clock option will be attempted. If you enable this, you
2457	usually have to link against librt or something similar. Enabling it when	2980	usually have to link against librt or something similar. Enabling it when
2458	the functionality isn't available is safe, though, although you have	2981	the functionality isn't available is safe, though, although you have
2459	to make sure you link against any libraries where the C<clock_gettime>	2982	to make sure you link against any libraries where the C<clock_gettime>
2460	function is hiding in (often F<-lrt>).	2983	function is hiding in (often F<-lrt>).
2461		2984
2462	=item EV_USE_REALTIME	2985	=item EV_USE_REALTIME
2463		2986
2464	If defined to be C<1>, libev will try to detect the availability of the	2987	If defined to be C<1>, libev will try to detect the availability of the
2465	realtime clock option at compiletime (and assume its availability at	2988	real-time clock option at compile time (and assume its availability at
2466	runtime if successful). Otherwise no use of the realtime clock option will	2989	runtime if successful). Otherwise no use of the real-time clock option will
2467	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	2990	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
2468	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the	2991	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
2469	note about libraries in the description of C<EV_USE_MONOTONIC>, though.	2992	note about libraries in the description of C<EV_USE_MONOTONIC>, though.
2470		2993
2471	=item EV_USE_NANOSLEEP	2994	=item EV_USE_NANOSLEEP
2472		2995
2473	If defined to be C<1>, libev will assume that C<nanosleep ()> is available	2996	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
2474	and will use it for delays. Otherwise it will use C<select ()>.	2997	and will use it for delays. Otherwise it will use C<select ()>.
2475		2998
		2999	=item EV_USE_EVENTFD
		3000
		3001	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		3002	available and will probe for kernel support at runtime. This will improve
		3003	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		3004	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		3005	2.7 or newer, otherwise disabled.
		3006
2476	=item EV_USE_SELECT	3007	=item EV_USE_SELECT
2477		3008
2478	If undefined or defined to be C<1>, libev will compile in support for the	3009	If undefined or defined to be C<1>, libev will compile in support for the
2479	C<select>(2) backend. No attempt at autodetection will be done: if no	3010	C<select>(2) backend. No attempt at auto-detection will be done: if no
2480	other method takes over, select will be it. Otherwise the select backend	3011	other method takes over, select will be it. Otherwise the select backend
2481	will not be compiled in.	3012	will not be compiled in.
2482		3013
2483	=item EV_SELECT_USE_FD_SET	3014	=item EV_SELECT_USE_FD_SET
2484		3015
2485	If defined to C<1>, then the select backend will use the system C<fd_set>	3016	If defined to C<1>, then the select backend will use the system C<fd_set>
2486	structure. This is useful if libev doesn't compile due to a missing	3017	structure. This is useful if libev doesn't compile due to a missing
2487	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on	3018	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on
2488	exotic systems. This usually limits the range of file descriptors to some	3019	exotic systems. This usually limits the range of file descriptors to some
2489	low limit such as 1024 or might have other limitations (winsocket only	3020	low limit such as 1024 or might have other limitations (winsocket only
2490	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	3021	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
2491	influence the size of the C<fd_set> used.	3022	influence the size of the C<fd_set> used.
2492		3023
…		…
2516		3047
2517	=item EV_USE_EPOLL	3048	=item EV_USE_EPOLL
2518		3049
2519	If defined to be C<1>, libev will compile in support for the Linux	3050	If defined to be C<1>, libev will compile in support for the Linux
2520	C<epoll>(7) backend. Its availability will be detected at runtime,	3051	C<epoll>(7) backend. Its availability will be detected at runtime,
2521	otherwise another method will be used as fallback. This is the	3052	otherwise another method will be used as fallback. This is the preferred
2522	preferred backend for GNU/Linux systems.	3053	backend for GNU/Linux systems. If undefined, it will be enabled if the
		3054	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2523		3055
2524	=item EV_USE_KQUEUE	3056	=item EV_USE_KQUEUE
2525		3057
2526	If defined to be C<1>, libev will compile in support for the BSD style	3058	If defined to be C<1>, libev will compile in support for the BSD style
2527	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	3059	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
2540	otherwise another method will be used as fallback. This is the preferred	3072	otherwise another method will be used as fallback. This is the preferred
2541	backend for Solaris 10 systems.	3073	backend for Solaris 10 systems.
2542		3074
2543	=item EV_USE_DEVPOLL	3075	=item EV_USE_DEVPOLL
2544		3076
2545	reserved for future expansion, works like the USE symbols above.	3077	Reserved for future expansion, works like the USE symbols above.
2546		3078
2547	=item EV_USE_INOTIFY	3079	=item EV_USE_INOTIFY
2548		3080
2549	If defined to be C<1>, libev will compile in support for the Linux inotify	3081	If defined to be C<1>, libev will compile in support for the Linux inotify
2550	interface to speed up C<ev_stat> watchers. Its actual availability will	3082	interface to speed up C<ev_stat> watchers. Its actual availability will
2551	be detected at runtime.	3083	be detected at runtime. If undefined, it will be enabled if the headers
		3084	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		3085
		3086	=item EV_ATOMIC_T
		3087
		3088	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		3089	access is atomic with respect to other threads or signal contexts. No such
		3090	type is easily found in the C language, so you can provide your own type
		3091	that you know is safe for your purposes. It is used both for signal handler "locking"
		3092	as well as for signal and thread safety in C<ev_async> watchers.
		3093
		3094	In the absence of this define, libev will use C<sig_atomic_t volatile>
		3095	(from F<signal.h>), which is usually good enough on most platforms.
2552		3096
2553	=item EV_H	3097	=item EV_H
2554		3098
2555	The name of the F<ev.h> header file used to include it. The default if	3099	The name of the F<ev.h> header file used to include it. The default if
2556	undefined is C<"ev.h"> in F<event.h> and F<ev.c>. This can be used to	3100	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2557	virtually rename the F<ev.h> header file in case of conflicts.	3101	used to virtually rename the F<ev.h> header file in case of conflicts.
2558		3102
2559	=item EV_CONFIG_H	3103	=item EV_CONFIG_H
2560		3104
2561	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	3105	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2562	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	3106	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2563	C<EV_H>, above.	3107	C<EV_H>, above.
2564		3108
2565	=item EV_EVENT_H	3109	=item EV_EVENT_H
2566		3110
2567	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	3111	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2568	of how the F<event.h> header can be found, the dfeault is C<"event.h">.	3112	of how the F<event.h> header can be found, the default is C<"event.h">.
2569		3113
2570	=item EV_PROTOTYPES	3114	=item EV_PROTOTYPES
2571		3115
2572	If defined to be C<0>, then F<ev.h> will not define any function	3116	If defined to be C<0>, then F<ev.h> will not define any function
2573	prototypes, but still define all the structs and other symbols. This is	3117	prototypes, but still define all the structs and other symbols. This is
…		…
2594	When doing priority-based operations, libev usually has to linearly search	3138	When doing priority-based operations, libev usually has to linearly search
2595	all the priorities, so having many of them (hundreds) uses a lot of space	3139	all the priorities, so having many of them (hundreds) uses a lot of space
2596	and time, so using the defaults of five priorities (-2 .. +2) is usually	3140	and time, so using the defaults of five priorities (-2 .. +2) is usually
2597	fine.	3141	fine.
2598		3142
2599	If your embedding app does not need any priorities, defining these both to	3143	If your embedding application does not need any priorities, defining these
2600	C<0> will save some memory and cpu.	3144	both to C<0> will save some memory and CPU.
2601		3145
2602	=item EV_PERIODIC_ENABLE	3146	=item EV_PERIODIC_ENABLE
2603		3147
2604	If undefined or defined to be C<1>, then periodic timers are supported. If	3148	If undefined or defined to be C<1>, then periodic timers are supported. If
2605	defined to be C<0>, then they are not. Disabling them saves a few kB of	3149	defined to be C<0>, then they are not. Disabling them saves a few kB of
…		…
2612	code.	3156	code.
2613		3157
2614	=item EV_EMBED_ENABLE	3158	=item EV_EMBED_ENABLE
2615		3159
2616	If undefined or defined to be C<1>, then embed watchers are supported. If	3160	If undefined or defined to be C<1>, then embed watchers are supported. If
2617	defined to be C<0>, then they are not.	3161	defined to be C<0>, then they are not. Embed watchers rely on most other
		3162	watcher types, which therefore must not be disabled.
2618		3163
2619	=item EV_STAT_ENABLE	3164	=item EV_STAT_ENABLE
2620		3165
2621	If undefined or defined to be C<1>, then stat watchers are supported. If	3166	If undefined or defined to be C<1>, then stat watchers are supported. If
2622	defined to be C<0>, then they are not.	3167	defined to be C<0>, then they are not.
…		…
2624	=item EV_FORK_ENABLE	3169	=item EV_FORK_ENABLE
2625		3170
2626	If undefined or defined to be C<1>, then fork watchers are supported. If	3171	If undefined or defined to be C<1>, then fork watchers are supported. If
2627	defined to be C<0>, then they are not.	3172	defined to be C<0>, then they are not.
2628		3173
		3174	=item EV_ASYNC_ENABLE
		3175
		3176	If undefined or defined to be C<1>, then async watchers are supported. If
		3177	defined to be C<0>, then they are not.
		3178
2629	=item EV_MINIMAL	3179	=item EV_MINIMAL
2630		3180
2631	If you need to shave off some kilobytes of code at the expense of some	3181	If you need to shave off some kilobytes of code at the expense of some
2632	speed, define this symbol to C<1>. Currently only used for gcc to override	3182	speed, define this symbol to C<1>. Currently this is used to override some
2633	some inlining decisions, saves roughly 30% codesize of amd64.	3183	inlining decisions, saves roughly 30% code size on amd64. It also selects a
		3184	much smaller 2-heap for timer management over the default 4-heap.
2634		3185
2635	=item EV_PID_HASHSIZE	3186	=item EV_PID_HASHSIZE
2636		3187
2637	C<ev_child> watchers use a small hash table to distribute workload by	3188	C<ev_child> watchers use a small hash table to distribute workload by
2638	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	3189	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
…		…
2645	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	3196	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
2646	usually more than enough. If you need to manage thousands of C<ev_stat>	3197	usually more than enough. If you need to manage thousands of C<ev_stat>
2647	watchers you might want to increase this value (I<must> be a power of	3198	watchers you might want to increase this value (I<must> be a power of
2648	two).	3199	two).
2649		3200
		3201	=item EV_USE_4HEAP
		3202
		3203	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3204	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		3205	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		3206	faster performance with many (thousands) of watchers.
		3207
		3208	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3209	(disabled).
		3210
		3211	=item EV_HEAP_CACHE_AT
		3212
		3213	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3214	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		3215	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3216	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3217	but avoids random read accesses on heap changes. This improves performance
		3218	noticeably with many (hundreds) of watchers.
		3219
		3220	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3221	(disabled).
		3222
		3223	=item EV_VERIFY
		3224
		3225	Controls how much internal verification (see C<ev_loop_verify ()>) will
		3226	be done: If set to C<0>, no internal verification code will be compiled
		3227	in. If set to C<1>, then verification code will be compiled in, but not
		3228	called. If set to C<2>, then the internal verification code will be
		3229	called once per loop, which can slow down libev. If set to C<3>, then the
		3230	verification code will be called very frequently, which will slow down
		3231	libev considerably.
		3232
		3233	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
		3234	C<0>.
		3235
2650	=item EV_COMMON	3236	=item EV_COMMON
2651		3237
2652	By default, all watchers have a C<void *data> member. By redefining	3238	By default, all watchers have a C<void *data> member. By redefining
2653	this macro to a something else you can include more and other types of	3239	this macro to a something else you can include more and other types of
2654	members. You have to define it each time you include one of the files,	3240	members. You have to define it each time you include one of the files,
2655	though, and it must be identical each time.	3241	though, and it must be identical each time.
2656		3242
2657	For example, the perl EV module uses something like this:	3243	For example, the perl EV module uses something like this:
2658		3244
2659	#define EV_COMMON \	3245	#define EV_COMMON \
2660	SV *self; /* contains this struct */ \	3246	SV *self; /* contains this struct */ \
2661	SV *cb_sv, *fh /* note no trailing ";" */	3247	SV *cb_sv, *fh /* note no trailing ";" */
2662		3248
2663	=item EV_CB_DECLARE (type)	3249	=item EV_CB_DECLARE (type)
2664		3250
2665	=item EV_CB_INVOKE (watcher, revents)	3251	=item EV_CB_INVOKE (watcher, revents)
2666		3252
…		…
2671	definition and a statement, respectively. See the F<ev.h> header file for	3257	definition and a statement, respectively. See the F<ev.h> header file for
2672	their default definitions. One possible use for overriding these is to	3258	their default definitions. One possible use for overriding these is to
2673	avoid the C<struct ev_loop *> as first argument in all cases, or to use	3259	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2674	method calls instead of plain function calls in C++.	3260	method calls instead of plain function calls in C++.
2675		3261
		3262	=back
		3263
2676	=head2 EXPORTED API SYMBOLS	3264	=head2 EXPORTED API SYMBOLS
2677		3265
2678	If you need to re-export the API (e.g. via a dll) and you need a list of	3266	If you need to re-export the API (e.g. via a DLL) and you need a list of
2679	exported symbols, you can use the provided F<Symbol.*> files which list	3267	exported symbols, you can use the provided F<Symbol.*> files which list
2680	all public symbols, one per line:	3268	all public symbols, one per line:
2681		3269
2682	Symbols.ev for libev proper	3270	Symbols.ev for libev proper
2683	Symbols.event for the libevent emulation	3271	Symbols.event for the libevent emulation
2684		3272
2685	This can also be used to rename all public symbols to avoid clashes with	3273	This can also be used to rename all public symbols to avoid clashes with
2686	multiple versions of libev linked together (which is obviously bad in	3274	multiple versions of libev linked together (which is obviously bad in
2687	itself, but sometimes it is inconvinient to avoid this).	3275	itself, but sometimes it is inconvenient to avoid this).
2688		3276
2689	A sed command like this will create wrapper C<#define>'s that you need to	3277	A sed command like this will create wrapper C<#define>'s that you need to
2690	include before including F<ev.h>:	3278	include before including F<ev.h>:
2691		3279
2692	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h	3280	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
…		…
2709	file.	3297	file.
2710		3298
2711	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	3299	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2712	that everybody includes and which overrides some configure choices:	3300	that everybody includes and which overrides some configure choices:
2713		3301
2714	#define EV_MINIMAL 1	3302	#define EV_MINIMAL 1
2715	#define EV_USE_POLL 0	3303	#define EV_USE_POLL 0
2716	#define EV_MULTIPLICITY 0	3304	#define EV_MULTIPLICITY 0
2717	#define EV_PERIODIC_ENABLE 0	3305	#define EV_PERIODIC_ENABLE 0
2718	#define EV_STAT_ENABLE 0	3306	#define EV_STAT_ENABLE 0
2719	#define EV_FORK_ENABLE 0	3307	#define EV_FORK_ENABLE 0
2720	#define EV_CONFIG_H <config.h>	3308	#define EV_CONFIG_H <config.h>
2721	#define EV_MINPRI 0	3309	#define EV_MINPRI 0
2722	#define EV_MAXPRI 0	3310	#define EV_MAXPRI 0
2723		3311
2724	#include "ev++.h"	3312	#include "ev++.h"
2725		3313
2726	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3314	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
2727		3315
2728	#include "ev_cpp.h"	3316	#include "ev_cpp.h"
2729	#include "ev.c"	3317	#include "ev.c"
2730		3318
		3319	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
2731		3320
2732	=head1 COMPLEXITIES	3321	=head2 THREADS AND COROUTINES
2733		3322
2734	In this section the complexities of (many of) the algorithms used inside	3323	=head3 THREADS
2735	libev will be explained. For complexity discussions about backends see the
2736	documentation for C<ev_default_init>.
2737		3324
2738	All of the following are about amortised time: If an array needs to be	3325	All libev functions are reentrant and thread-safe unless explicitly
2739	extended, libev needs to realloc and move the whole array, but this	3326	documented otherwise, but libev implements no locking itself. This means
2740	happens asymptotically never with higher number of elements, so O(1) might	3327	that you can use as many loops as you want in parallel, as long as there
2741	mean it might do a lengthy realloc operation in rare cases, but on average	3328	are no concurrent calls into any libev function with the same loop
2742	it is much faster and asymptotically approaches constant time.	3329	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		3330	of course): libev guarantees that different event loops share no data
		3331	structures that need any locking.
		3332
		3333	Or to put it differently: calls with different loop parameters can be done
		3334	concurrently from multiple threads, calls with the same loop parameter
		3335	must be done serially (but can be done from different threads, as long as
		3336	only one thread ever is inside a call at any point in time, e.g. by using
		3337	a mutex per loop).
		3338
		3339	Specifically to support threads (and signal handlers), libev implements
		3340	so-called C<ev_async> watchers, which allow some limited form of
		3341	concurrency on the same event loop, namely waking it up "from the
		3342	outside".
		3343
		3344	If you want to know which design (one loop, locking, or multiple loops
		3345	without or something else still) is best for your problem, then I cannot
		3346	help you, but here is some generic advice:
2743		3347
2744	=over 4	3348	=over 4
2745		3349
2746	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3350	=item * most applications have a main thread: use the default libev loop
		3351	in that thread, or create a separate thread running only the default loop.
2747		3352
2748	This means that, when you have a watcher that triggers in one hour and	3353	This helps integrating other libraries or software modules that use libev
2749	there are 100 watchers that would trigger before that then inserting will	3354	themselves and don't care/know about threading.
2750	have to skip roughly seven (C<ld 100>) of these watchers.
2751		3355
2752	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	3356	=item * one loop per thread is usually a good model.
2753		3357
2754	That means that changing a timer costs less than removing/adding them	3358	Doing this is almost never wrong, sometimes a better-performance model
2755	as only the relative motion in the event queue has to be paid for.	3359	exists, but it is always a good start.
2756		3360
2757	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	3361	=item * other models exist, such as the leader/follower pattern, where one
		3362	loop is handed through multiple threads in a kind of round-robin fashion.
2758		3363
2759	These just add the watcher into an array or at the head of a list.	3364	Choosing a model is hard - look around, learn, know that usually you can do
		3365	better than you currently do :-)
2760		3366
2761	=item Stopping check/prepare/idle watchers: O(1)	3367	=item * often you need to talk to some other thread which blocks in the
		3368	event loop.
2762		3369
2763	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3370	C<ev_async> watchers can be used to wake them up from other threads safely
		3371	(or from signal contexts...).
2764		3372
2765	These watchers are stored in lists then need to be walked to find the	3373	An example use would be to communicate signals or other events that only
2766	correct watcher to remove. The lists are usually short (you don't usually	3374	work in the default loop by registering the signal watcher with the
2767	have many watchers waiting for the same fd or signal).	3375	default loop and triggering an C<ev_async> watcher from the default loop
2768		3376	watcher callback into the event loop interested in the signal.
2769	=item Finding the next timer in each loop iteration: O(1)
2770
2771	By virtue of using a binary heap, the next timer is always found at the
2772	beginning of the storage array.
2773
2774	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2775
2776	A change means an I/O watcher gets started or stopped, which requires
2777	libev to recalculate its status (and possibly tell the kernel, depending
2778	on backend and wether C<ev_io_set> was used).
2779
2780	=item Activating one watcher (putting it into the pending state): O(1)
2781
2782	=item Priority handling: O(number_of_priorities)
2783
2784	Priorities are implemented by allocating some space for each
2785	priority. When doing priority-based operations, libev usually has to
2786	linearly search all the priorities, but starting/stopping and activating
2787	watchers becomes O(1) w.r.t. prioritiy handling.
2788		3377
2789	=back	3378	=back
2790		3379
		3380	=head3 COROUTINES
2791		3381
2792	=head1 Win32 platform limitations and workarounds	3382	Libev is very accommodating to coroutines ("cooperative threads"):
		3383	libev fully supports nesting calls to its functions from different
		3384	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		3385	different coroutines, and switch freely between both coroutines running the
		3386	loop, as long as you don't confuse yourself). The only exception is that
		3387	you must not do this from C<ev_periodic> reschedule callbacks.
		3388
		3389	Care has been taken to ensure that libev does not keep local state inside
		3390	C<ev_loop>, and other calls do not usually allow for coroutine switches as
		3391	they do not clal any callbacks.
		3392
		3393	=head2 COMPILER WARNINGS
		3394
		3395	Depending on your compiler and compiler settings, you might get no or a
		3396	lot of warnings when compiling libev code. Some people are apparently
		3397	scared by this.
		3398
		3399	However, these are unavoidable for many reasons. For one, each compiler
		3400	has different warnings, and each user has different tastes regarding
		3401	warning options. "Warn-free" code therefore cannot be a goal except when
		3402	targeting a specific compiler and compiler-version.
		3403
		3404	Another reason is that some compiler warnings require elaborate
		3405	workarounds, or other changes to the code that make it less clear and less
		3406	maintainable.
		3407
		3408	And of course, some compiler warnings are just plain stupid, or simply
		3409	wrong (because they don't actually warn about the condition their message
		3410	seems to warn about). For example, certain older gcc versions had some
		3411	warnings that resulted an extreme number of false positives. These have
		3412	been fixed, but some people still insist on making code warn-free with
		3413	such buggy versions.
		3414
		3415	While libev is written to generate as few warnings as possible,
		3416	"warn-free" code is not a goal, and it is recommended not to build libev
		3417	with any compiler warnings enabled unless you are prepared to cope with
		3418	them (e.g. by ignoring them). Remember that warnings are just that:
		3419	warnings, not errors, or proof of bugs.
		3420
		3421
		3422	=head2 VALGRIND
		3423
		3424	Valgrind has a special section here because it is a popular tool that is
		3425	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		3426
		3427	If you think you found a bug (memory leak, uninitialised data access etc.)
		3428	in libev, then check twice: If valgrind reports something like:
		3429
		3430	==2274== definitely lost: 0 bytes in 0 blocks.
		3431	==2274== possibly lost: 0 bytes in 0 blocks.
		3432	==2274== still reachable: 256 bytes in 1 blocks.
		3433
		3434	Then there is no memory leak, just as memory accounted to global variables
		3435	is not a memleak - the memory is still being refernced, and didn't leak.
		3436
		3437	Similarly, under some circumstances, valgrind might report kernel bugs
		3438	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		3439	although an acceptable workaround has been found here), or it might be
		3440	confused.
		3441
		3442	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		3443	make it into some kind of religion.
		3444
		3445	If you are unsure about something, feel free to contact the mailing list
		3446	with the full valgrind report and an explanation on why you think this
		3447	is a bug in libev (best check the archives, too :). However, don't be
		3448	annoyed when you get a brisk "this is no bug" answer and take the chance
		3449	of learning how to interpret valgrind properly.
		3450
		3451	If you need, for some reason, empty reports from valgrind for your project
		3452	I suggest using suppression lists.
		3453
		3454
		3455	=head1 PORTABILITY NOTES
		3456
		3457	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
2793		3458
2794	Win32 doesn't support any of the standards (e.g. POSIX) that libev	3459	Win32 doesn't support any of the standards (e.g. POSIX) that libev
2795	requires, and its I/O model is fundamentally incompatible with the POSIX	3460	requires, and its I/O model is fundamentally incompatible with the POSIX
2796	model. Libev still offers limited functionality on this platform in	3461	model. Libev still offers limited functionality on this platform in
2797	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	3462	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
2798	descriptors. This only applies when using Win32 natively, not when using	3463	descriptors. This only applies when using Win32 natively, not when using
2799	e.g. cygwin.	3464	e.g. cygwin.
2800		3465
		3466	Lifting these limitations would basically require the full
		3467	re-implementation of the I/O system. If you are into these kinds of
		3468	things, then note that glib does exactly that for you in a very portable
		3469	way (note also that glib is the slowest event library known to man).
		3470
2801	There is no supported compilation method available on windows except	3471	There is no supported compilation method available on windows except
2802	embedding it into other applications.	3472	embedding it into other applications.
2803		3473
		3474	Not a libev limitation but worth mentioning: windows apparently doesn't
		3475	accept large writes: instead of resulting in a partial write, windows will
		3476	either accept everything or return C<ENOBUFS> if the buffer is too large,
		3477	so make sure you only write small amounts into your sockets (less than a
		3478	megabyte seems safe, but this apparently depends on the amount of memory
		3479	available).
		3480
2804	Due to the many, low, and arbitrary limits on the win32 platform and the	3481	Due to the many, low, and arbitrary limits on the win32 platform and
2805	abysmal performance of winsockets, using a large number of sockets is not	3482	the abysmal performance of winsockets, using a large number of sockets
2806	recommended (and not reasonable). If your program needs to use more than	3483	is not recommended (and not reasonable). If your program needs to use
2807	a hundred or so sockets, then likely it needs to use a totally different	3484	more than a hundred or so sockets, then likely it needs to use a totally
2808	implementation for windows, as libev offers the POSIX model, which cannot	3485	different implementation for windows, as libev offers the POSIX readiness
2809	be implemented efficiently on windows (microsoft monopoly games).	3486	notification model, which cannot be implemented efficiently on windows
		3487	(Microsoft monopoly games).
		3488
		3489	A typical way to use libev under windows is to embed it (see the embedding
		3490	section for details) and use the following F<evwrap.h> header file instead
		3491	of F<ev.h>:
		3492
		3493	#define EV_STANDALONE /* keeps ev from requiring config.h */
		3494	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		3495
		3496	#include "ev.h"
		3497
		3498	And compile the following F<evwrap.c> file into your project (make sure
		3499	you do I<not> compile the F<ev.c> or any other embedded source files!):
		3500
		3501	#include "evwrap.h"
		3502	#include "ev.c"
2810		3503
2811	=over 4	3504	=over 4
2812		3505
2813	=item The winsocket select function	3506	=item The winsocket select function
2814		3507
2815	The winsocket C<select> function doesn't follow POSIX in that it requires	3508	The winsocket C<select> function doesn't follow POSIX in that it
2816	socket I<handles> and not socket I<file descriptors>. This makes select	3509	requires socket I<handles> and not socket I<file descriptors> (it is
2817	very inefficient, and also requires a mapping from file descriptors	3510	also extremely buggy). This makes select very inefficient, and also
2818	to socket handles. See the discussion of the C<EV_SELECT_USE_FD_SET>,	3511	requires a mapping from file descriptors to socket handles (the Microsoft
2819	C<EV_SELECT_IS_WINSOCKET> and C<EV_FD_TO_WIN32_HANDLE> preprocessor	3512	C runtime provides the function C<_open_osfhandle> for this). See the
2820	symbols for more info.	3513	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		3514	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
2821		3515
2822	The configuration for a "naked" win32 using the microsoft runtime	3516	The configuration for a "naked" win32 using the Microsoft runtime
2823	libraries and raw winsocket select is:	3517	libraries and raw winsocket select is:
2824		3518
2825	#define EV_USE_SELECT 1	3519	#define EV_USE_SELECT 1
2826	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */	3520	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
2827		3521
2828	Note that winsockets handling of fd sets is O(n), so you can easily get a	3522	Note that winsockets handling of fd sets is O(n), so you can easily get a
2829	complexity in the O(n²) range when using win32.	3523	complexity in the O(n²) range when using win32.
2830		3524
2831	=item Limited number of file descriptors	3525	=item Limited number of file descriptors
2832		3526
2833	Windows has numerous arbitrary (and low) limits on things. Early versions	3527	Windows has numerous arbitrary (and low) limits on things.
2834	of winsocket's select only supported waiting for a max. of C<64> handles	3528
		3529	Early versions of winsocket's select only supported waiting for a maximum
2835	(probably owning to the fact that all windows kernels can only wait for	3530	of C<64> handles (probably owning to the fact that all windows kernels
2836	C<64> things at the same time internally; microsoft recommends spawning a	3531	can only wait for C<64> things at the same time internally; Microsoft
2837	chain of threads and wait for 63 handles and the previous thread in each).	3532	recommends spawning a chain of threads and wait for 63 handles and the
		3533	previous thread in each. Great).
2838		3534
2839	Newer versions support more handles, but you need to define C<FD_SETSIZE>	3535	Newer versions support more handles, but you need to define C<FD_SETSIZE>
2840	to some high number (e.g. C<2048>) before compiling the winsocket select	3536	to some high number (e.g. C<2048>) before compiling the winsocket select
2841	call (which might be in libev or elsewhere, for example, perl does its own	3537	call (which might be in libev or elsewhere, for example, perl does its own
2842	select emulation on windows).	3538	select emulation on windows).
2843		3539
2844	Another limit is the number of file descriptors in the microsoft runtime	3540	Another limit is the number of file descriptors in the Microsoft runtime
2845	libraries, which by default is C<64> (there must be a hidden I<64> fetish	3541	libraries, which by default is C<64> (there must be a hidden I<64> fetish
2846	or something like this inside microsoft). You can increase this by calling	3542	or something like this inside Microsoft). You can increase this by calling
2847	C<_setmaxstdio>, which can increase this limit to C<2048> (another	3543	C<_setmaxstdio>, which can increase this limit to C<2048> (another
2848	arbitrary limit), but is broken in many versions of the microsoft runtime	3544	arbitrary limit), but is broken in many versions of the Microsoft runtime
2849	libraries.	3545	libraries.
2850		3546
2851	This might get you to about C<512> or C<2048> sockets (depending on	3547	This might get you to about C<512> or C<2048> sockets (depending on
2852	windows version and/or the phase of the moon). To get more, you need to	3548	windows version and/or the phase of the moon). To get more, you need to
2853	wrap all I/O functions and provide your own fd management, but the cost of	3549	wrap all I/O functions and provide your own fd management, but the cost of
2854	calling select (O(n²)) will likely make this unworkable.	3550	calling select (O(n²)) will likely make this unworkable.
2855		3551
2856	=back	3552	=back
2857		3553
		3554	=head2 PORTABILITY REQUIREMENTS
		3555
		3556	In addition to a working ISO-C implementation and of course the
		3557	backend-specific APIs, libev relies on a few additional extensions:
		3558
		3559	=over 4
		3560
		3561	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		3562	calling conventions regardless of C<ev_watcher_type *>.
		3563
		3564	Libev assumes not only that all watcher pointers have the same internal
		3565	structure (guaranteed by POSIX but not by ISO C for example), but it also
		3566	assumes that the same (machine) code can be used to call any watcher
		3567	callback: The watcher callbacks have different type signatures, but libev
		3568	calls them using an C<ev_watcher *> internally.
		3569
		3570	=item C<sig_atomic_t volatile> must be thread-atomic as well
		3571
		3572	The type C<sig_atomic_t volatile> (or whatever is defined as
		3573	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		3574	threads. This is not part of the specification for C<sig_atomic_t>, but is
		3575	believed to be sufficiently portable.
		3576
		3577	=item C<sigprocmask> must work in a threaded environment
		3578
		3579	Libev uses C<sigprocmask> to temporarily block signals. This is not
		3580	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		3581	pthread implementations will either allow C<sigprocmask> in the "main
		3582	thread" or will block signals process-wide, both behaviours would
		3583	be compatible with libev. Interaction between C<sigprocmask> and
		3584	C<pthread_sigmask> could complicate things, however.
		3585
		3586	The most portable way to handle signals is to block signals in all threads
		3587	except the initial one, and run the default loop in the initial thread as
		3588	well.
		3589
		3590	=item C<long> must be large enough for common memory allocation sizes
		3591
		3592	To improve portability and simplify its API, libev uses C<long> internally
		3593	instead of C<size_t> when allocating its data structures. On non-POSIX
		3594	systems (Microsoft...) this might be unexpectedly low, but is still at
		3595	least 31 bits everywhere, which is enough for hundreds of millions of
		3596	watchers.
		3597
		3598	=item C<double> must hold a time value in seconds with enough accuracy
		3599
		3600	The type C<double> is used to represent timestamps. It is required to
		3601	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		3602	enough for at least into the year 4000. This requirement is fulfilled by
		3603	implementations implementing IEEE 754 (basically all existing ones).
		3604
		3605	=back
		3606
		3607	If you know of other additional requirements drop me a note.
		3608
		3609
		3610	=head1 ALGORITHMIC COMPLEXITIES
		3611
		3612	In this section the complexities of (many of) the algorithms used inside
		3613	libev will be documented. For complexity discussions about backends see
		3614	the documentation for C<ev_default_init>.
		3615
		3616	All of the following are about amortised time: If an array needs to be
		3617	extended, libev needs to realloc and move the whole array, but this
		3618	happens asymptotically rarer with higher number of elements, so O(1) might
		3619	mean that libev does a lengthy realloc operation in rare cases, but on
		3620	average it is much faster and asymptotically approaches constant time.
		3621
		3622	=over 4
		3623
		3624	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		3625
		3626	This means that, when you have a watcher that triggers in one hour and
		3627	there are 100 watchers that would trigger before that, then inserting will
		3628	have to skip roughly seven (C<ld 100>) of these watchers.
		3629
		3630	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		3631
		3632	That means that changing a timer costs less than removing/adding them,
		3633	as only the relative motion in the event queue has to be paid for.
		3634
		3635	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
		3636
		3637	These just add the watcher into an array or at the head of a list.
		3638
		3639	=item Stopping check/prepare/idle/fork/async watchers: O(1)
		3640
		3641	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		3642
		3643	These watchers are stored in lists, so they need to be walked to find the
		3644	correct watcher to remove. The lists are usually short (you don't usually
		3645	have many watchers waiting for the same fd or signal: one is typical, two
		3646	is rare).
		3647
		3648	=item Finding the next timer in each loop iteration: O(1)
		3649
		3650	By virtue of using a binary or 4-heap, the next timer is always found at a
		3651	fixed position in the storage array.
		3652
		3653	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		3654
		3655	A change means an I/O watcher gets started or stopped, which requires
		3656	libev to recalculate its status (and possibly tell the kernel, depending
		3657	on backend and whether C<ev_io_set> was used).
		3658
		3659	=item Activating one watcher (putting it into the pending state): O(1)
		3660
		3661	=item Priority handling: O(number_of_priorities)
		3662
		3663	Priorities are implemented by allocating some space for each
		3664	priority. When doing priority-based operations, libev usually has to
		3665	linearly search all the priorities, but starting/stopping and activating
		3666	watchers becomes O(1) with respect to priority handling.
		3667
		3668	=item Sending an ev_async: O(1)
		3669
		3670	=item Processing ev_async_send: O(number_of_async_watchers)
		3671
		3672	=item Processing signals: O(max_signal_number)
		3673
		3674	Sending involves a system call I<iff> there were no other C<ev_async_send>
		3675	calls in the current loop iteration. Checking for async and signal events
		3676	involves iterating over all running async watchers or all signal numbers.
		3677
		3678	=back
		3679
2858		3680
2859	=head1 AUTHOR	3681	=head1 AUTHOR
2860		3682
2861	Marc Lehmann <libev@schmorp.de>.	3683	Marc Lehmann <libev@schmorp.de>.
2862		3684

Diff Legend

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