ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.57 by root, Wed Nov 28 11:27:29 2007 UTC vs.
Revision 1.174 by root, Mon Aug 18 23:23:45 2008 UTC

…		…
2		2
3	libev - a high performance full-featured event loop written in C	3	libev - a high performance full-featured event loop written in C
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
8		8
9	=head1 EXAMPLE PROGRAM	9	=head2 EXAMPLE PROGRAM
10		10
		11	// a single header file is required
11	#include <ev.h>	12	#include <ev.h>
12		13
		14	// every watcher type has its own typedef'd struct
		15	// with the name ev_<type>
13	ev_io stdin_watcher;	16	ev_io stdin_watcher;
14	ev_timer timeout_watcher;	17	ev_timer timeout_watcher;
15		18
16	/* called when data readable on stdin */	19	// all watcher callbacks have a similar signature
		20	// this callback is called when data is readable on stdin
17	static void	21	static void
18	stdin_cb (EV_P_ struct ev_io *w, int revents)	22	stdin_cb (EV_P_ struct ev_io *w, int revents)
19	{	23	{
20	/* puts ("stdin ready"); */	24	puts ("stdin ready");
21	ev_io_stop (EV_A_ w); /* just a syntax example */	25	// for one-shot events, one must manually stop the watcher
22	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */	26	// with its corresponding stop function.
		27	ev_io_stop (EV_A_ w);
		28
		29	// this causes all nested ev_loop's to stop iterating
		30	ev_unloop (EV_A_ EVUNLOOP_ALL);
23	}	31	}
24		32
		33	// another callback, this time for a time-out
25	static void	34	static void
26	timeout_cb (EV_P_ struct ev_timer *w, int revents)	35	timeout_cb (EV_P_ struct ev_timer *w, int revents)
27	{	36	{
28	/* puts ("timeout"); */	37	puts ("timeout");
29	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */	38	// this causes the innermost ev_loop to stop iterating
		39	ev_unloop (EV_A_ EVUNLOOP_ONE);
30	}	40	}
31		41
32	int	42	int
33	main (void)	43	main (void)
34	{	44	{
		45	// use the default event loop unless you have special needs
35	struct ev_loop *loop = ev_default_loop (0);	46	struct ev_loop *loop = ev_default_loop (0);
36		47
37	/* initialise an io watcher, then start it */	48	// initialise an io watcher, then start it
		49	// this one will watch for stdin to become readable
38	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);	50	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
39	ev_io_start (loop, &stdin_watcher);	51	ev_io_start (loop, &stdin_watcher);
40		52
		53	// initialise a timer watcher, then start it
41	/* simple non-repeating 5.5 second timeout */	54	// simple non-repeating 5.5 second timeout
42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	55	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
43	ev_timer_start (loop, &timeout_watcher);	56	ev_timer_start (loop, &timeout_watcher);
44		57
45	/* loop till timeout or data ready */	58	// now wait for events to arrive
46	ev_loop (loop, 0);	59	ev_loop (loop, 0);
47		60
		61	// unloop was called, so exit
48	return 0;	62	return 0;
49	}	63	}
50		64
51	=head1 DESCRIPTION	65	=head1 DESCRIPTION
52		66
		67	The newest version of this document is also available as an html-formatted
		68	web page you might find easier to navigate when reading it for the first
		69	time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
		70
53	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
54	file descriptor being readable or a timeout occuring), and it will manage	72	file descriptor being readable or a timeout occurring), and it will manage
55	these event sources and provide your program with events.	73	these event sources and provide your program with events.
56		74
57	To do this, it must take more or less complete control over your process	75	To do this, it must take more or less complete control over your process
58	(or thread) by executing the I<event loop> handler, and will then	76	(or thread) by executing the I<event loop> handler, and will then
59	communicate events via a callback mechanism.	77	communicate events via a callback mechanism.
…		…
61	You register interest in certain events by registering so-called I<event	79	You register interest in certain events by registering so-called I<event
62	watchers>, which are relatively small C structures you initialise with the	80	watchers>, which are relatively small C structures you initialise with the
63	details of the event, and then hand it over to libev by I<starting> the	81	details of the event, and then hand it over to libev by I<starting> the
64	watcher.	82	watcher.
65		83
66	=head1 FEATURES	84	=head2 FEATURES
67		85
68	Libev supports C<select>, C<poll>, the linux-specific C<epoll>, the	86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
69	bsd-specific C<kqueue> and the solaris-specific event port mechanisms	87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
70	for file descriptor events (C<ev_io>), relative timers (C<ev_timer>),	88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
		89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
71	absolute timers with customised rescheduling (C<ev_periodic>), synchronous	90	with customised rescheduling (C<ev_periodic>), synchronous signals
72	signals (C<ev_signal>), process status change events (C<ev_child>), and	91	(C<ev_signal>), process status change events (C<ev_child>), and event
73	event watchers dealing with the event loop mechanism itself (C<ev_idle>,	92	watchers dealing with the event loop mechanism itself (C<ev_idle>,
74	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as	93	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
75	file watchers (C<ev_stat>) and even limited support for fork events	94	file watchers (C<ev_stat>) and even limited support for fork events
76	(C<ev_fork>).	95	(C<ev_fork>).
77		96
78	It also is quite fast (see this	97	It also is quite fast (see this
79	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
80	for example).	99	for example).
81		100
82	=head1 CONVENTIONS	101	=head2 CONVENTIONS
83		102
84	Libev is very configurable. In this manual the default configuration will	103	Libev is very configurable. In this manual the default (and most common)
85	be described, which supports multiple event loops. For more info about	104	configuration will be described, which supports multiple event loops. For
86	various configuration options please have a look at B<EMBED> section in	105	more info about various configuration options please have a look at
87	this manual. If libev was configured without support for multiple event	106	B<EMBED> section in this manual. If libev was configured without support
88	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
89	(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.
90		110
91	=head1 TIME REPRESENTATION	111	=head2 TIME REPRESENTATION
92		112
93	Libev represents time as a single floating point number, representing the	113	Libev represents time as a single floating point number, representing the
94	(fractional) number of seconds since the (POSIX) epoch (somewhere near	114	(fractional) number of seconds since the (POSIX) epoch (somewhere near
95	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
96	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
97	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
98	it, you should treat it as such.	118	it, you should treat it as some floating point value. Unlike the name
		119	component C<stamp> might indicate, it is also used for time differences
		120	throughout libev.
		121
		122	=head1 ERROR HANDLING
		123
		124	Libev knows three classes of errors: operating system errors, usage errors
		125	and internal errors (bugs).
		126
		127	When libev catches an operating system error it cannot handle (for example
		128	a system call indicating a condition libev cannot fix), it calls the callback
		129	set via C<ev_set_syserr_cb>, which is supposed to fix the problem or
		130	abort. The default is to print a diagnostic message and to call C<abort
		131	()>.
		132
		133	When libev detects a usage error such as a negative timer interval, then
		134	it will print a diagnostic message and abort (via the C<assert> mechanism,
		135	so C<NDEBUG> will disable this checking): these are programming errors in
		136	the libev caller and need to be fixed there.
		137
		138	Libev also has a few internal error-checking C<assert>ions, and also has
		139	extensive consistency checking code. These do not trigger under normal
		140	circumstances, as they indicate either a bug in libev or worse.
		141
99		142
100	=head1 GLOBAL FUNCTIONS	143	=head1 GLOBAL FUNCTIONS
101		144
102	These functions can be called anytime, even before initialising the	145	These functions can be called anytime, even before initialising the
103	library in any way.	146	library in any way.
…		…
108		151
109	Returns the current time as libev would use it. Please note that the	152	Returns the current time as libev would use it. Please note that the
110	C<ev_now> function is usually faster and also often returns the timestamp	153	C<ev_now> function is usually faster and also often returns the timestamp
111	you actually want to know.	154	you actually want to know.
112		155
		156	=item ev_sleep (ev_tstamp interval)
		157
		158	Sleep for the given interval: The current thread will be blocked until
		159	either it is interrupted or the given time interval has passed. Basically
		160	this is a sub-second-resolution C<sleep ()>.
		161
113	=item int ev_version_major ()	162	=item int ev_version_major ()
114		163
115	=item int ev_version_minor ()	164	=item int ev_version_minor ()
116		165
117	You can find out the major and minor version numbers of the library	166	You can find out the major and minor ABI version numbers of the library
118	you linked against by calling the functions C<ev_version_major> and	167	you linked against by calling the functions C<ev_version_major> and
119	C<ev_version_minor>. If you want, you can compare against the global	168	C<ev_version_minor>. If you want, you can compare against the global
120	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	169	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
121	version of the library your program was compiled against.	170	version of the library your program was compiled against.
122		171
		172	These version numbers refer to the ABI version of the library, not the
		173	release version.
		174
123	Usually, it's a good idea to terminate if the major versions mismatch,	175	Usually, it's a good idea to terminate if the major versions mismatch,
124	as this indicates an incompatible change. Minor versions are usually	176	as this indicates an incompatible change. Minor versions are usually
125	compatible to older versions, so a larger minor version alone is usually	177	compatible to older versions, so a larger minor version alone is usually
126	not a problem.	178	not a problem.
127		179
128	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
129	version.	181	version.
130		182
131	assert (("libev version mismatch",	183	assert (("libev version mismatch",
132	ev_version_major () == EV_VERSION_MAJOR	184	ev_version_major () == EV_VERSION_MAJOR
133	&& ev_version_minor () >= EV_VERSION_MINOR));	185	&& ev_version_minor () >= EV_VERSION_MINOR));
134		186
135	=item unsigned int ev_supported_backends ()	187	=item unsigned int ev_supported_backends ()
136		188
137	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_*>
138	value) compiled into this binary of libev (independent of their	190	value) compiled into this binary of libev (independent of their
…		…
140	a description of the set values.	192	a description of the set values.
141		193
142	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
143	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
144		196
145	assert (("sorry, no epoll, no sex",	197	assert (("sorry, no epoll, no sex",
146	ev_supported_backends () & EVBACKEND_EPOLL));	198	ev_supported_backends () & EVBACKEND_EPOLL));
147		199
148	=item unsigned int ev_recommended_backends ()	200	=item unsigned int ev_recommended_backends ()
149		201
150	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
151	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
152	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
153	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
154	(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
155	libev will probe for if you specify no backends explicitly.	207	libev will probe for if you specify no backends explicitly.
156		208
157	=item unsigned int ev_embeddable_backends ()	209	=item unsigned int ev_embeddable_backends ()
158		210
…		…
162	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
163	recommended ones.	215	recommended ones.
164		216
165	See the description of C<ev_embed> watchers for more info.	217	See the description of C<ev_embed> watchers for more info.
166		218
167	=item ev_set_allocator (void *(*cb)(void *ptr, size_t size))	219	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
168		220
169	Sets the allocation function to use (the prototype and semantics are	221	Sets the allocation function to use (the prototype is similar - the
170	identical to the realloc C function). It is used to allocate and free	222	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
171	memory (no surprises here). If it returns zero when memory needs to be	223	used to allocate and free memory (no surprises here). If it returns zero
172	allocated, the library might abort or take some potentially destructive	224	when memory needs to be allocated (C<size != 0>), the library might abort
173	action. The default is your system realloc function.	225	or take some potentially destructive action.
		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.
174		230
175	You could override this function in high-availability programs to, say,	231	You could override this function in high-availability programs to, say,
176	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,
177	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.
178		234
179	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
180	retries).	236	retries (example requires a standards-compliant C<realloc>).
181		237
182	static void *	238	static void *
183	persistent_realloc (void *ptr, size_t size)	239	persistent_realloc (void *ptr, size_t size)
184	{	240	{
185	for (;;)	241	for (;;)
…		…
196	...	252	...
197	ev_set_allocator (persistent_realloc);	253	ev_set_allocator (persistent_realloc);
198		254
199	=item ev_set_syserr_cb (void (*cb)(const char *msg));	255	=item ev_set_syserr_cb (void (*cb)(const char *msg));
200		256
201	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
202	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
203	indicating the system call or subsystem causing the problem. If this	259	indicating the system call or subsystem causing the problem. If this
204	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
205	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
206	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
207	(such as abort).	263	(such as abort).
208		264
209	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.
…		…
223	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	279	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
224		280
225	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
226	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
227	events, and dynamically created loops which do not.	283	events, and dynamically created loops which do not.
228
229	If you use threads, a common model is to run the default event loop
230	in your main thread (or in a separate thread) and for each thread you
231	create, you also create another event loop. Libev itself does no locking
232	whatsoever, so if you mix calls to the same event loop in different
233	threads, make sure you lock (this is usually a bad idea, though, even if
234	done correctly, because it's hideous and inefficient).
235		284
236	=over 4	285	=over 4
237		286
238	=item struct ev_loop *ev_default_loop (unsigned int flags)	287	=item struct ev_loop *ev_default_loop (unsigned int flags)
239		288
…		…
243	flags. If that is troubling you, check C<ev_backend ()> afterwards).	292	flags. If that is troubling you, check C<ev_backend ()> afterwards).
244		293
245	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
246	function.	295	function.
247		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
248	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
249	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>).
250		310
251	The following flags are supported:	311	The following flags are supported:
252		312
…		…
257	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
258	thing, believe me).	318	thing, believe me).
259		319
260	=item C<EVFLAG_NOENV>	320	=item C<EVFLAG_NOENV>
261		321
262	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
263	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
264	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	324	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
265	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
266	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
267	around bugs.	327	around bugs.
268		328
		329	=item C<EVFLAG_FORKCHECK>
		330
		331	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after
		332	a fork, you can also make libev check for a fork in each iteration by
		333	enabling this flag.
		334
		335	This works by calling C<getpid ()> on every iteration of the loop,
		336	and thus this might slow down your event loop if you do a lot of loop
		337	iterations and little real work, but is usually not noticeable (on my
		338	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
		339	without a system call and thus I<very> fast, but my GNU/Linux system also has
		340	C<pthread_atfork> which is even faster).
		341
		342	The big advantage of this flag is that you can forget about fork (and
		343	forget about forgetting to tell libev about forking) when you use this
		344	flag.
		345
		346	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
		347	environment variable.
		348
269	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	349	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
270		350
271	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
272	libev tries to roll its own fd_set with no limits on the number of fds,	352	libev tries to roll its own fd_set with no limits on the number of fds,
273	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
274	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	354	using this backend. It doesn't scale too well (O(highest_fd)), but its
275	the fastest backend for a low number of fds.	355	usually the fastest backend for a low number of (low-numbered :) fds.
		356
		357	To get good performance out of this backend you need a high amount of
		358	parallelism (most of the file descriptors should be busy). If you are
		359	writing a server, you should C<accept ()> in a loop to accept as many
		360	connections as possible during one iteration. You might also want to have
		361	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		362	readiness notifications you get per iteration.
276		363
277	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	364	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
278		365
279	And this is your standard poll(2) backend. It's more complicated than	366	And this is your standard poll(2) backend. It's more complicated
280	select, but handles sparse fds better and has no artificial limit on the	367	than select, but handles sparse fds better and has no artificial
281	number of fds you can use (except it will slow down considerably with a	368	limit on the number of fds you can use (except it will slow down
282	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	369	considerably with a lot of inactive fds). It scales similarly to select,
		370	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		371	performance tips.
283		372
284	=item C<EVBACKEND_EPOLL> (value 4, Linux)	373	=item C<EVBACKEND_EPOLL> (value 4, Linux)
285		374
286	For few fds, this backend is a bit little slower than poll and select,	375	For few fds, this backend is a bit little slower than poll and select,
287	but it scales phenomenally better. While poll and select usually scale like	376	but it scales phenomenally better. While poll and select usually scale
288	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	377	like O(total_fds) where n is the total number of fds (or the highest fd),
289	either O(1) or O(active_fds).	378	epoll scales either O(1) or O(active_fds). The epoll design has a number
		379	of shortcomings, such as silently dropping events in some hard-to-detect
		380	cases and requiring a system call per fd change, no fork support and bad
		381	support for dup.
290		382
291	While stopping and starting an I/O watcher in the same iteration will	383	While stopping, setting and starting an I/O watcher in the same iteration
292	result in some caching, there is still a syscall per such incident	384	will result in some caching, there is still a system call per such incident
293	(because the fd could point to a different file description now), so its	385	(because the fd could point to a different file description now), so its
294	best to avoid that. Also, dup()ed file descriptors might not work very	386	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
295	well if you register events for both fds.	387	very well if you register events for both fds.
296		388
297	Please note that epoll sometimes generates spurious notifications, so you	389	Please note that epoll sometimes generates spurious notifications, so you
298	need to use non-blocking I/O or other means to avoid blocking when no data	390	need to use non-blocking I/O or other means to avoid blocking when no data
299	(or space) is available.	391	(or space) is available.
300		392
		393	Best performance from this backend is achieved by not unregistering all
		394	watchers for a file descriptor until it has been closed, if possible, i.e.
		395	keep at least one watcher active per fd at all times.
		396
		397	While nominally embeddable in other event loops, this feature is broken in
		398	all kernel versions tested so far.
		399
301	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	400	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
302		401
303	Kqueue deserves special mention, as at the time of this writing, it	402	Kqueue deserves special mention, as at the time of this writing, it
304	was broken on all BSDs except NetBSD (usually it doesn't work with	403	was broken on all BSDs except NetBSD (usually it doesn't work reliably
305	anything but sockets and pipes, except on Darwin, where of course its	404	with anything but sockets and pipes, except on Darwin, where of course
306	completely useless). For this reason its not being "autodetected"	405	it's completely useless). For this reason it's not being "auto-detected"
307	unless you explicitly specify it explicitly in the flags (i.e. using	406	unless you explicitly specify it explicitly in the flags (i.e. using
308	C<EVBACKEND_KQUEUE>).	407	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		408	system like NetBSD.
		409
		410	You still can embed kqueue into a normal poll or select backend and use it
		411	only for sockets (after having made sure that sockets work with kqueue on
		412	the target platform). See C<ev_embed> watchers for more info.
309		413
310	It scales in the same way as the epoll backend, but the interface to the	414	It scales in the same way as the epoll backend, but the interface to the
311	kernel is more efficient (which says nothing about its actual speed, of	415	kernel is more efficient (which says nothing about its actual speed, of
312	course). While starting and stopping an I/O watcher does not cause an	416	course). While stopping, setting and starting an I/O watcher does never
313	extra syscall as with epoll, it still adds up to four event changes per	417	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
314	incident, so its best to avoid that.	418	two event changes per incident, support for C<fork ()> is very bad and it
		419	drops fds silently in similarly hard-to-detect cases.
		420
		421	This backend usually performs well under most conditions.
		422
		423	While nominally embeddable in other event loops, this doesn't work
		424	everywhere, so you might need to test for this. And since it is broken
		425	almost everywhere, you should only use it when you have a lot of sockets
		426	(for which it usually works), by embedding it into another event loop
		427	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for
		428	sockets.
315		429
316	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	430	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
317		431
318	This is not implemented yet (and might never be).	432	This is not implemented yet (and might never be, unless you send me an
		433	implementation). According to reports, C</dev/poll> only supports sockets
		434	and is not embeddable, which would limit the usefulness of this backend
		435	immensely.
319		436
320	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	437	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
321		438
322	This uses the Solaris 10 port mechanism. As with everything on Solaris,	439	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
323	it's really slow, but it still scales very well (O(active_fds)).	440	it's really slow, but it still scales very well (O(active_fds)).
324		441
325	Please note that solaris ports can result in a lot of spurious	442	Please note that Solaris event ports can deliver a lot of spurious
326	notifications, so you need to use non-blocking I/O or other means to avoid	443	notifications, so you need to use non-blocking I/O or other means to avoid
327	blocking when no data (or space) is available.	444	blocking when no data (or space) is available.
		445
		446	While this backend scales well, it requires one system call per active
		447	file descriptor per loop iteration. For small and medium numbers of file
		448	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		449	might perform better.
		450
		451	On the positive side, ignoring the spurious readiness notifications, this
		452	backend actually performed to specification in all tests and is fully
		453	embeddable, which is a rare feat among the OS-specific backends.
328		454
329	=item C<EVBACKEND_ALL>	455	=item C<EVBACKEND_ALL>
330		456
331	Try all backends (even potentially broken ones that wouldn't be tried	457	Try all backends (even potentially broken ones that wouldn't be tried
332	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	458	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
333	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	459	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
334		460
		461	It is definitely not recommended to use this flag.
		462
335	=back	463	=back
336		464
337	If one or more of these are ored into the flags value, then only these	465	If one or more of these are or'ed into the flags value, then only these
338	backends will be tried (in the reverse order as given here). If none are	466	backends will be tried (in the reverse order as listed here). If none are
339	specified, most compiled-in backend will be tried, usually in reverse	467	specified, all backends in C<ev_recommended_backends ()> will be tried.
340	order of their flag values :)
341		468
342	The most typical usage is like this:	469	The most typical usage is like this:
343		470
344	if (!ev_default_loop (0))	471	if (!ev_default_loop (0))
345	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	472	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
346		473
347	Restrict libev to the select and poll backends, and do not allow	474	Restrict libev to the select and poll backends, and do not allow
348	environment settings to be taken into account:	475	environment settings to be taken into account:
349		476
350	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);	477	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
351		478
352	Use whatever libev has to offer, but make sure that kqueue is used if	479	Use whatever libev has to offer, but make sure that kqueue is used if
353	available (warning, breaks stuff, best use only with your own private	480	available (warning, breaks stuff, best use only with your own private
354	event loop and only if you know the OS supports your types of fds):	481	event loop and only if you know the OS supports your types of fds):
355		482
356	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);	483	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
357		484
358	=item struct ev_loop *ev_loop_new (unsigned int flags)	485	=item struct ev_loop *ev_loop_new (unsigned int flags)
359		486
360	Similar to C<ev_default_loop>, but always creates a new event loop that is	487	Similar to C<ev_default_loop>, but always creates a new event loop that is
361	always distinct from the default loop. Unlike the default loop, it cannot	488	always distinct from the default loop. Unlike the default loop, it cannot
362	handle signal and child watchers, and attempts to do so will be greeted by	489	handle signal and child watchers, and attempts to do so will be greeted by
363	undefined behaviour (or a failed assertion if assertions are enabled).	490	undefined behaviour (or a failed assertion if assertions are enabled).
364		491
		492	Note that this function I<is> thread-safe, and the recommended way to use
		493	libev with threads is indeed to create one loop per thread, and using the
		494	default loop in the "main" or "initial" thread.
		495
365	Example: Try to create a event loop that uses epoll and nothing else.	496	Example: Try to create a event loop that uses epoll and nothing else.
366		497
367	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	498	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
368	if (!epoller)	499	if (!epoller)
369	fatal ("no epoll found here, maybe it hides under your chair");	500	fatal ("no epoll found here, maybe it hides under your chair");
370		501
371	=item ev_default_destroy ()	502	=item ev_default_destroy ()
372		503
373	Destroys the default loop again (frees all memory and kernel state	504	Destroys the default loop again (frees all memory and kernel state
374	etc.). None of the active event watchers will be stopped in the normal	505	etc.). None of the active event watchers will be stopped in the normal
375	sense, so e.g. C<ev_is_active> might still return true. It is your	506	sense, so e.g. C<ev_is_active> might still return true. It is your
376	responsibility to either stop all watchers cleanly yoursef I<before>	507	responsibility to either stop all watchers cleanly yourself I<before>
377	calling this function, or cope with the fact afterwards (which is usually	508	calling this function, or cope with the fact afterwards (which is usually
378	the easiest thing, youc na just ignore the watchers and/or C<free ()> them	509	the easiest thing, you can just ignore the watchers and/or C<free ()> them
379	for example).	510	for example).
		511
		512	Note that certain global state, such as signal state, will not be freed by
		513	this function, and related watchers (such as signal and child watchers)
		514	would need to be stopped manually.
		515
		516	In general it is not advisable to call this function except in the
		517	rare occasion where you really need to free e.g. the signal handling
		518	pipe fds. If you need dynamically allocated loops it is better to use
		519	C<ev_loop_new> and C<ev_loop_destroy>).
380		520
381	=item ev_loop_destroy (loop)	521	=item ev_loop_destroy (loop)
382		522
383	Like C<ev_default_destroy>, but destroys an event loop created by an	523	Like C<ev_default_destroy>, but destroys an event loop created by an
384	earlier call to C<ev_loop_new>.	524	earlier call to C<ev_loop_new>.
385		525
386	=item ev_default_fork ()	526	=item ev_default_fork ()
387		527
		528	This function sets a flag that causes subsequent C<ev_loop> iterations
388	This function reinitialises the kernel state for backends that have	529	to reinitialise the kernel state for backends that have one. Despite the
389	one. Despite the name, you can call it anytime, but it makes most sense	530	name, you can call it anytime, but it makes most sense after forking, in
390	after forking, in either the parent or child process (or both, but that	531	the child process (or both child and parent, but that again makes little
391	again makes little sense).	532	sense). You I<must> call it in the child before using any of the libev
		533	functions, and it will only take effect at the next C<ev_loop> iteration.
392		534
393	You I<must> call this function in the child process after forking if and	535	On the other hand, you only need to call this function in the child
394	only if you want to use the event library in both processes. If you just	536	process if and only if you want to use the event library in the child. If
395	fork+exec, you don't have to call it.	537	you just fork+exec, you don't have to call it at all.
396		538
397	The function itself is quite fast and it's usually not a problem to call	539	The function itself is quite fast and it's usually not a problem to call
398	it just in case after a fork. To make this easy, the function will fit in	540	it just in case after a fork. To make this easy, the function will fit in
399	quite nicely into a call to C<pthread_atfork>:	541	quite nicely into a call to C<pthread_atfork>:
400		542
401	pthread_atfork (0, 0, ev_default_fork);	543	pthread_atfork (0, 0, ev_default_fork);
402		544
403	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
404	without calling this function, so if you force one of those backends you
405	do not need to care.
406
407	=item ev_loop_fork (loop)	545	=item ev_loop_fork (loop)
408		546
409	Like C<ev_default_fork>, but acts on an event loop created by	547	Like C<ev_default_fork>, but acts on an event loop created by
410	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	548	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
411	after fork, and how you do this is entirely your own problem.	549	after fork, and how you do this is entirely your own problem.
		550
		551	=item int ev_is_default_loop (loop)
		552
		553	Returns true when the given loop actually is the default loop, false otherwise.
		554
		555	=item unsigned int ev_loop_count (loop)
		556
		557	Returns the count of loop iterations for the loop, which is identical to
		558	the number of times libev did poll for new events. It starts at C<0> and
		559	happily wraps around with enough iterations.
		560
		561	This value can sometimes be useful as a generation counter of sorts (it
		562	"ticks" the number of loop iterations), as it roughly corresponds with
		563	C<ev_prepare> and C<ev_check> calls.
412		564
413	=item unsigned int ev_backend (loop)	565	=item unsigned int ev_backend (loop)
414		566
415	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	567	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
416	use.	568	use.
…		…
419		571
420	Returns the current "event loop time", which is the time the event loop	572	Returns the current "event loop time", which is the time the event loop
421	received events and started processing them. This timestamp does not	573	received events and started processing them. This timestamp does not
422	change as long as callbacks are being processed, and this is also the base	574	change as long as callbacks are being processed, and this is also the base
423	time used for relative timers. You can treat it as the timestamp of the	575	time used for relative timers. You can treat it as the timestamp of the
424	event occuring (or more correctly, libev finding out about it).	576	event occurring (or more correctly, libev finding out about it).
425		577
426	=item ev_loop (loop, int flags)	578	=item ev_loop (loop, int flags)
427		579
428	Finally, this is it, the event handler. This function usually is called	580	Finally, this is it, the event handler. This function usually is called
429	after you initialised all your watchers and you want to start handling	581	after you initialised all your watchers and you want to start handling
…		…
441	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	593	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
442	those events and any outstanding ones, but will not block your process in	594	those events and any outstanding ones, but will not block your process in
443	case there are no events and will return after one iteration of the loop.	595	case there are no events and will return after one iteration of the loop.
444		596
445	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	597	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
446	neccessary) and will handle those and any outstanding ones. It will block	598	necessary) and will handle those and any outstanding ones. It will block
447	your process until at least one new event arrives, and will return after	599	your process until at least one new event arrives, and will return after
448	one iteration of the loop. This is useful if you are waiting for some	600	one iteration of the loop. This is useful if you are waiting for some
449	external event in conjunction with something not expressible using other	601	external event in conjunction with something not expressible using other
450	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	602	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is
451	usually a better approach for this kind of thing.	603	usually a better approach for this kind of thing.
452		604
453	Here are the gory details of what C<ev_loop> does:	605	Here are the gory details of what C<ev_loop> does:
454		606
455	* If there are no active watchers (reference count is zero), return.	607	- Before the first iteration, call any pending watchers.
456	- Queue prepare watchers and then call all outstanding watchers.	608	* If EVFLAG_FORKCHECK was used, check for a fork.
		609	- If a fork was detected (by any means), queue and call all fork watchers.
		610	- Queue and call all prepare watchers.
457	- If we have been forked, recreate the kernel state.	611	- If we have been forked, detach and recreate the kernel state
		612	as to not disturb the other process.
458	- Update the kernel state with all outstanding changes.	613	- Update the kernel state with all outstanding changes.
459	- Update the "event loop time".	614	- Update the "event loop time" (ev_now ()).
460	- Calculate for how long to block.	615	- Calculate for how long to sleep or block, if at all
		616	(active idle watchers, EVLOOP_NONBLOCK or not having
		617	any active watchers at all will result in not sleeping).
		618	- Sleep if the I/O and timer collect interval say so.
461	- Block the process, waiting for any events.	619	- Block the process, waiting for any events.
462	- Queue all outstanding I/O (fd) events.	620	- Queue all outstanding I/O (fd) events.
463	- Update the "event loop time" and do time jump handling.	621	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
464	- Queue all outstanding timers.	622	- Queue all outstanding timers.
465	- Queue all outstanding periodics.	623	- Queue all outstanding periodics.
466	- If no events are pending now, queue all idle watchers.	624	- Unless any events are pending now, queue all idle watchers.
467	- Queue all check watchers.	625	- Queue all check watchers.
468	- Call all queued watchers in reverse order (i.e. check watchers first).	626	- Call all queued watchers in reverse order (i.e. check watchers first).
469	Signals and child watchers are implemented as I/O watchers, and will	627	Signals and child watchers are implemented as I/O watchers, and will
470	be handled here by queueing them when their watcher gets executed.	628	be handled here by queueing them when their watcher gets executed.
471	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	629	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
472	were used, return, otherwise continue with step *.	630	were used, or there are no active watchers, return, otherwise
		631	continue with step *.
473		632
474	Example: Queue some jobs and then loop until no events are outsanding	633	Example: Queue some jobs and then loop until no events are outstanding
475	anymore.	634	anymore.
476		635
477	... queue jobs here, make sure they register event watchers as long	636	... queue jobs here, make sure they register event watchers as long
478	... as they still have work to do (even an idle watcher will do..)	637	... as they still have work to do (even an idle watcher will do..)
479	ev_loop (my_loop, 0);	638	ev_loop (my_loop, 0);
480	... jobs done. yeah!	639	... jobs done or somebody called unloop. yeah!
481		640
482	=item ev_unloop (loop, how)	641	=item ev_unloop (loop, how)
483		642
484	Can be used to make a call to C<ev_loop> return early (but only after it	643	Can be used to make a call to C<ev_loop> return early (but only after it
485	has processed all outstanding events). The C<how> argument must be either	644	has processed all outstanding events). The C<how> argument must be either
486	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	645	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
487	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	646	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
		647
		648	This "unloop state" will be cleared when entering C<ev_loop> again.
488		649
489	=item ev_ref (loop)	650	=item ev_ref (loop)
490		651
491	=item ev_unref (loop)	652	=item ev_unref (loop)
492		653
…		…
497	returning, ev_unref() after starting, and ev_ref() before stopping it. For	658	returning, ev_unref() after starting, and ev_ref() before stopping it. For
498	example, libev itself uses this for its internal signal pipe: It is not	659	example, libev itself uses this for its internal signal pipe: It is not
499	visible to the libev user and should not keep C<ev_loop> from exiting if	660	visible to the libev user and should not keep C<ev_loop> from exiting if
500	no event watchers registered by it are active. It is also an excellent	661	no event watchers registered by it are active. It is also an excellent
501	way to do this for generic recurring timers or from within third-party	662	way to do this for generic recurring timers or from within third-party
502	libraries. Just remember to I<unref after start> and I<ref before stop>.	663	libraries. Just remember to I<unref after start> and I<ref before stop>
		664	(but only if the watcher wasn't active before, or was active before,
		665	respectively).
503		666
504	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	667	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
505	running when nothing else is active.	668	running when nothing else is active.
506		669
507	struct ev_signal exitsig;	670	struct ev_signal exitsig;
508	ev_signal_init (&exitsig, sig_cb, SIGINT);	671	ev_signal_init (&exitsig, sig_cb, SIGINT);
509	ev_signal_start (loop, &exitsig);	672	ev_signal_start (loop, &exitsig);
510	evf_unref (loop);	673	evf_unref (loop);
511		674
512	Example: For some weird reason, unregister the above signal handler again.	675	Example: For some weird reason, unregister the above signal handler again.
513		676
514	ev_ref (loop);	677	ev_ref (loop);
515	ev_signal_stop (loop, &exitsig);	678	ev_signal_stop (loop, &exitsig);
		679
		680	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		681
		682	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		683
		684	These advanced functions influence the time that libev will spend waiting
		685	for events. Both time intervals are by default C<0>, meaning that libev
		686	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		687	latency.
		688
		689	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		690	allows libev to delay invocation of I/O and timer/periodic callbacks
		691	to increase efficiency of loop iterations (or to increase power-saving
		692	opportunities).
		693
		694	The background is that sometimes your program runs just fast enough to
		695	handle one (or very few) event(s) per loop iteration. While this makes
		696	the program responsive, it also wastes a lot of CPU time to poll for new
		697	events, especially with backends like C<select ()> which have a high
		698	overhead for the actual polling but can deliver many events at once.
		699
		700	By setting a higher I<io collect interval> you allow libev to spend more
		701	time collecting I/O events, so you can handle more events per iteration,
		702	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		703	C<ev_timer>) will be not affected. Setting this to a non-null value will
		704	introduce an additional C<ev_sleep ()> call into most loop iterations.
		705
		706	Likewise, by setting a higher I<timeout collect interval> you allow libev
		707	to spend more time collecting timeouts, at the expense of increased
		708	latency (the watcher callback will be called later). C<ev_io> watchers
		709	will not be affected. Setting this to a non-null value will not introduce
		710	any overhead in libev.
		711
		712	Many (busy) programs can usually benefit by setting the I/O collect
		713	interval to a value near C<0.1> or so, which is often enough for
		714	interactive servers (of course not for games), likewise for timeouts. It
		715	usually doesn't make much sense to set it to a lower value than C<0.01>,
		716	as this approaches the timing granularity of most systems.
		717
		718	Setting the I<timeout collect interval> can improve the opportunity for
		719	saving power, as the program will "bundle" timer callback invocations that
		720	are "near" in time together, by delaying some, thus reducing the number of
		721	times the process sleeps and wakes up again. Another useful technique to
		722	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		723	they fire on, say, one-second boundaries only.
		724
		725	=item ev_loop_verify (loop)
		726
		727	This function only does something when C<EV_VERIFY> support has been
		728	compiled in. It tries to go through all internal structures and checks
		729	them for validity. If anything is found to be inconsistent, it will print
		730	an error message to standard error and call C<abort ()>.
		731
		732	This can be used to catch bugs inside libev itself: under normal
		733	circumstances, this function will never abort as of course libev keeps its
		734	data structures consistent.
516		735
517	=back	736	=back
518		737
519		738
520	=head1 ANATOMY OF A WATCHER	739	=head1 ANATOMY OF A WATCHER
521		740
522	A watcher is a structure that you create and register to record your	741	A watcher is a structure that you create and register to record your
523	interest in some event. For instance, if you want to wait for STDIN to	742	interest in some event. For instance, if you want to wait for STDIN to
524	become readable, you would create an C<ev_io> watcher for that:	743	become readable, you would create an C<ev_io> watcher for that:
525		744
526	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	745	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)
527	{	746	{
528	ev_io_stop (w);	747	ev_io_stop (w);
529	ev_unloop (loop, EVUNLOOP_ALL);	748	ev_unloop (loop, EVUNLOOP_ALL);
530	}	749	}
531		750
532	struct ev_loop *loop = ev_default_loop (0);	751	struct ev_loop *loop = ev_default_loop (0);
533	struct ev_io stdin_watcher;	752	struct ev_io stdin_watcher;
534	ev_init (&stdin_watcher, my_cb);	753	ev_init (&stdin_watcher, my_cb);
535	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	754	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
536	ev_io_start (loop, &stdin_watcher);	755	ev_io_start (loop, &stdin_watcher);
537	ev_loop (loop, 0);	756	ev_loop (loop, 0);
538		757
539	As you can see, you are responsible for allocating the memory for your	758	As you can see, you are responsible for allocating the memory for your
540	watcher structures (and it is usually a bad idea to do this on the stack,	759	watcher structures (and it is usually a bad idea to do this on the stack,
541	although this can sometimes be quite valid).	760	although this can sometimes be quite valid).
542		761
543	Each watcher structure must be initialised by a call to C<ev_init	762	Each watcher structure must be initialised by a call to C<ev_init
544	(watcher *, callback)>, which expects a callback to be provided. This	763	(watcher *, callback)>, which expects a callback to be provided. This
545	callback gets invoked each time the event occurs (or, in the case of io	764	callback gets invoked each time the event occurs (or, in the case of I/O
546	watchers, each time the event loop detects that the file descriptor given	765	watchers, each time the event loop detects that the file descriptor given
547	is readable and/or writable).	766	is readable and/or writable).
548		767
549	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	768	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro
550	with arguments specific to this watcher type. There is also a macro	769	with arguments specific to this watcher type. There is also a macro
…		…
620	=item C<EV_FORK>	839	=item C<EV_FORK>
621		840
622	The event loop has been resumed in the child process after fork (see	841	The event loop has been resumed in the child process after fork (see
623	C<ev_fork>).	842	C<ev_fork>).
624		843
		844	=item C<EV_ASYNC>
		845
		846	The given async watcher has been asynchronously notified (see C<ev_async>).
		847
625	=item C<EV_ERROR>	848	=item C<EV_ERROR>
626		849
627	An unspecified error has occured, the watcher has been stopped. This might	850	An unspecified error has occurred, the watcher has been stopped. This might
628	happen because the watcher could not be properly started because libev	851	happen because the watcher could not be properly started because libev
629	ran out of memory, a file descriptor was found to be closed or any other	852	ran out of memory, a file descriptor was found to be closed or any other
630	problem. You best act on it by reporting the problem and somehow coping	853	problem. You best act on it by reporting the problem and somehow coping
631	with the watcher being stopped.	854	with the watcher being stopped.
632		855
633	Libev will usually signal a few "dummy" events together with an error,	856	Libev will usually signal a few "dummy" events together with an error,
634	for example it might indicate that a fd is readable or writable, and if	857	for example it might indicate that a fd is readable or writable, and if
635	your callbacks is well-written it can just attempt the operation and cope	858	your callbacks is well-written it can just attempt the operation and cope
636	with the error from read() or write(). This will not work in multithreaded	859	with the error from read() or write(). This will not work in multi-threaded
637	programs, though, so beware.	860	programs, though, so beware.
638		861
639	=back	862	=back
640		863
641	=head2 GENERIC WATCHER FUNCTIONS	864	=head2 GENERIC WATCHER FUNCTIONS
…		…
671	Although some watcher types do not have type-specific arguments	894	Although some watcher types do not have type-specific arguments
672	(e.g. C<ev_prepare>) you still need to call its C<set> macro.	895	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
673		896
674	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])	897	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
675		898
676	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro	899	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
677	calls into a single call. This is the most convinient method to initialise	900	calls into a single call. This is the most convenient method to initialise
678	a watcher. The same limitations apply, of course.	901	a watcher. The same limitations apply, of course.
679		902
680	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)	903	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
681		904
682	Starts (activates) the given watcher. Only active watchers will receive	905	Starts (activates) the given watcher. Only active watchers will receive
…		…
700	=item bool ev_is_pending (ev_TYPE *watcher)	923	=item bool ev_is_pending (ev_TYPE *watcher)
701		924
702	Returns a true value iff the watcher is pending, (i.e. it has outstanding	925	Returns a true value iff the watcher is pending, (i.e. it has outstanding
703	events but its callback has not yet been invoked). As long as a watcher	926	events but its callback has not yet been invoked). As long as a watcher
704	is pending (but not active) you must not call an init function on it (but	927	is pending (but not active) you must not call an init function on it (but
705	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to	928	C<ev_TYPE_set> is safe), you must not change its priority, and you must
706	libev (e.g. you cnanot C<free ()> it).	929	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		930	it).
707		931
708	=item callback ev_cb (ev_TYPE *watcher)	932	=item callback ev_cb (ev_TYPE *watcher)
709		933
710	Returns the callback currently set on the watcher.	934	Returns the callback currently set on the watcher.
711		935
712	=item ev_cb_set (ev_TYPE *watcher, callback)	936	=item ev_cb_set (ev_TYPE *watcher, callback)
713		937
714	Change the callback. You can change the callback at virtually any time	938	Change the callback. You can change the callback at virtually any time
715	(modulo threads).	939	(modulo threads).
		940
		941	=item ev_set_priority (ev_TYPE *watcher, priority)
		942
		943	=item int ev_priority (ev_TYPE *watcher)
		944
		945	Set and query the priority of the watcher. The priority is a small
		946	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		947	(default: C<-2>). Pending watchers with higher priority will be invoked
		948	before watchers with lower priority, but priority will not keep watchers
		949	from being executed (except for C<ev_idle> watchers).
		950
		951	This means that priorities are I<only> used for ordering callback
		952	invocation after new events have been received. This is useful, for
		953	example, to reduce latency after idling, or more often, to bind two
		954	watchers on the same event and make sure one is called first.
		955
		956	If you need to suppress invocation when higher priority events are pending
		957	you need to look at C<ev_idle> watchers, which provide this functionality.
		958
		959	You I<must not> change the priority of a watcher as long as it is active or
		960	pending.
		961
		962	The default priority used by watchers when no priority has been set is
		963	always C<0>, which is supposed to not be too high and not be too low :).
		964
		965	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		966	fine, as long as you do not mind that the priority value you query might
		967	or might not have been adjusted to be within valid range.
		968
		969	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		970
		971	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		972	C<loop> nor C<revents> need to be valid as long as the watcher callback
		973	can deal with that fact.
		974
		975	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		976
		977	If the watcher is pending, this function returns clears its pending status
		978	and returns its C<revents> bitset (as if its callback was invoked). If the
		979	watcher isn't pending it does nothing and returns C<0>.
716		980
717	=back	981	=back
718		982
719		983
720	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	984	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
…		…
724	to associate arbitrary data with your watcher. If you need more data and	988	to associate arbitrary data with your watcher. If you need more data and
725	don't want to allocate memory and store a pointer to it in that data	989	don't want to allocate memory and store a pointer to it in that data
726	member, you can also "subclass" the watcher type and provide your own	990	member, you can also "subclass" the watcher type and provide your own
727	data:	991	data:
728		992
729	struct my_io	993	struct my_io
730	{	994	{
731	struct ev_io io;	995	struct ev_io io;
732	int otherfd;	996	int otherfd;
733	void *somedata;	997	void *somedata;
734	struct whatever *mostinteresting;	998	struct whatever *mostinteresting;
735	}	999	}
736		1000
737	And since your callback will be called with a pointer to the watcher, you	1001	And since your callback will be called with a pointer to the watcher, you
738	can cast it back to your own type:	1002	can cast it back to your own type:
739		1003
740	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1004	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)
741	{	1005	{
742	struct my_io *w = (struct my_io *)w_;	1006	struct my_io *w = (struct my_io *)w_;
743	...	1007	...
744	}	1008	}
745		1009
746	More interesting and less C-conformant ways of casting your callback type	1010	More interesting and less C-conformant ways of casting your callback type
747	instead have been omitted.	1011	instead have been omitted.
748		1012
749	Another common scenario is having some data structure with multiple	1013	Another common scenario is having some data structure with multiple
750	watchers:	1014	watchers:
751		1015
752	struct my_biggy	1016	struct my_biggy
753	{	1017	{
754	int some_data;	1018	int some_data;
755	ev_timer t1;	1019	ev_timer t1;
756	ev_timer t2;	1020	ev_timer t2;
757	}	1021	}
758		1022
759	In this case getting the pointer to C<my_biggy> is a bit more complicated,	1023	In this case getting the pointer to C<my_biggy> is a bit more complicated,
760	you need to use C<offsetof>:	1024	you need to use C<offsetof>:
761		1025
762	#include <stddef.h>	1026	#include <stddef.h>
763		1027
764	static void	1028	static void
765	t1_cb (EV_P_ struct ev_timer *w, int revents)	1029	t1_cb (EV_P_ struct ev_timer *w, int revents)
766	{	1030	{
767	struct my_biggy big = (struct my_biggy *	1031	struct my_biggy big = (struct my_biggy *
768	(((char *)w) - offsetof (struct my_biggy, t1));	1032	(((char *)w) - offsetof (struct my_biggy, t1));
769	}	1033	}
770		1034
771	static void	1035	static void
772	t2_cb (EV_P_ struct ev_timer *w, int revents)	1036	t2_cb (EV_P_ struct ev_timer *w, int revents)
773	{	1037	{
774	struct my_biggy big = (struct my_biggy *	1038	struct my_biggy big = (struct my_biggy *
775	(((char *)w) - offsetof (struct my_biggy, t2));	1039	(((char *)w) - offsetof (struct my_biggy, t2));
776	}	1040	}
777		1041
778		1042
779	=head1 WATCHER TYPES	1043	=head1 WATCHER TYPES
780		1044
781	This section describes each watcher in detail, but will not repeat	1045	This section describes each watcher in detail, but will not repeat
…		…
805	In general you can register as many read and/or write event watchers per	1069	In general you can register as many read and/or write event watchers per
806	fd as you want (as long as you don't confuse yourself). Setting all file	1070	fd as you want (as long as you don't confuse yourself). Setting all file
807	descriptors to non-blocking mode is also usually a good idea (but not	1071	descriptors to non-blocking mode is also usually a good idea (but not
808	required if you know what you are doing).	1072	required if you know what you are doing).
809		1073
810	You have to be careful with dup'ed file descriptors, though. Some backends
811	(the linux epoll backend is a notable example) cannot handle dup'ed file
812	descriptors correctly if you register interest in two or more fds pointing
813	to the same underlying file/socket/etc. description (that is, they share
814	the same underlying "file open").
815
816	If you must do this, then force the use of a known-to-be-good backend	1074	If you must do this, then force the use of a known-to-be-good backend
817	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1075	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
818	C<EVBACKEND_POLL>).	1076	C<EVBACKEND_POLL>).
819		1077
820	Another thing you have to watch out for is that it is quite easy to	1078	Another thing you have to watch out for is that it is quite easy to
821	receive "spurious" readyness notifications, that is your callback might	1079	receive "spurious" readiness notifications, that is your callback might
822	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1080	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
823	because there is no data. Not only are some backends known to create a	1081	because there is no data. Not only are some backends known to create a
824	lot of those (for example solaris ports), it is very easy to get into	1082	lot of those (for example Solaris ports), it is very easy to get into
825	this situation even with a relatively standard program structure. Thus	1083	this situation even with a relatively standard program structure. Thus
826	it is best to always use non-blocking I/O: An extra C<read>(2) returning	1084	it is best to always use non-blocking I/O: An extra C<read>(2) returning
827	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1085	C<EAGAIN> is far preferable to a program hanging until some data arrives.
828		1086
829	If you cannot run the fd in non-blocking mode (for example you should not	1087	If you cannot run the fd in non-blocking mode (for example you should not
830	play around with an Xlib connection), then you have to seperately re-test	1088	play around with an Xlib connection), then you have to separately re-test
831	wether a file descriptor is really ready with a known-to-be good interface	1089	whether a file descriptor is really ready with a known-to-be good interface
832	such as poll (fortunately in our Xlib example, Xlib already does this on	1090	such as poll (fortunately in our Xlib example, Xlib already does this on
833	its own, so its quite safe to use).	1091	its own, so its quite safe to use).
834		1092
		1093	=head3 The special problem of disappearing file descriptors
		1094
		1095	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1096	descriptor (either by calling C<close> explicitly or by any other means,
		1097	such as C<dup>). The reason is that you register interest in some file
		1098	descriptor, but when it goes away, the operating system will silently drop
		1099	this interest. If another file descriptor with the same number then is
		1100	registered with libev, there is no efficient way to see that this is, in
		1101	fact, a different file descriptor.
		1102
		1103	To avoid having to explicitly tell libev about such cases, libev follows
		1104	the following policy: Each time C<ev_io_set> is being called, libev
		1105	will assume that this is potentially a new file descriptor, otherwise
		1106	it is assumed that the file descriptor stays the same. That means that
		1107	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1108	descriptor even if the file descriptor number itself did not change.
		1109
		1110	This is how one would do it normally anyway, the important point is that
		1111	the libev application should not optimise around libev but should leave
		1112	optimisations to libev.
		1113
		1114	=head3 The special problem of dup'ed file descriptors
		1115
		1116	Some backends (e.g. epoll), cannot register events for file descriptors,
		1117	but only events for the underlying file descriptions. That means when you
		1118	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1119	events for them, only one file descriptor might actually receive events.
		1120
		1121	There is no workaround possible except not registering events
		1122	for potentially C<dup ()>'ed file descriptors, or to resort to
		1123	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1124
		1125	=head3 The special problem of fork
		1126
		1127	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1128	useless behaviour. Libev fully supports fork, but needs to be told about
		1129	it in the child.
		1130
		1131	To support fork in your programs, you either have to call
		1132	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1133	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1134	C<EVBACKEND_POLL>.
		1135
		1136	=head3 The special problem of SIGPIPE
		1137
		1138	While not really specific to libev, it is easy to forget about SIGPIPE:
		1139	when writing to a pipe whose other end has been closed, your program gets
		1140	send a SIGPIPE, which, by default, aborts your program. For most programs
		1141	this is sensible behaviour, for daemons, this is usually undesirable.
		1142
		1143	So when you encounter spurious, unexplained daemon exits, make sure you
		1144	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1145	somewhere, as that would have given you a big clue).
		1146
		1147
		1148	=head3 Watcher-Specific Functions
		1149
835	=over 4	1150	=over 4
836		1151
837	=item ev_io_init (ev_io *, callback, int fd, int events)	1152	=item ev_io_init (ev_io *, callback, int fd, int events)
838		1153
839	=item ev_io_set (ev_io *, int fd, int events)	1154	=item ev_io_set (ev_io *, int fd, int events)
840		1155
841	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to	1156	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
842	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or	1157	receive events for and events is either C<EV_READ>, C<EV_WRITE> or
843	C<EV_READ | EV_WRITE> to receive the given events.	1158	C<EV_READ | EV_WRITE> to receive the given events.
844		1159
845	=item int fd [read-only]	1160	=item int fd [read-only]
846		1161
847	The file descriptor being watched.	1162	The file descriptor being watched.
…		…
849	=item int events [read-only]	1164	=item int events [read-only]
850		1165
851	The events being watched.	1166	The events being watched.
852		1167
853	=back	1168	=back
		1169
		1170	=head3 Examples
854		1171
855	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1172	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
856	readable, but only once. Since it is likely line-buffered, you could	1173	readable, but only once. Since it is likely line-buffered, you could
857	attempt to read a whole line in the callback.	1174	attempt to read a whole line in the callback.
858		1175
859	static void	1176	static void
860	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1177	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
861	{	1178	{
862	ev_io_stop (loop, w);	1179	ev_io_stop (loop, w);
863	.. read from stdin here (or from w->fd) and haqndle any I/O errors	1180	.. read from stdin here (or from w->fd) and haqndle any I/O errors
864	}	1181	}
865		1182
866	...	1183	...
867	struct ev_loop *loop = ev_default_init (0);	1184	struct ev_loop *loop = ev_default_init (0);
868	struct ev_io stdin_readable;	1185	struct ev_io stdin_readable;
869	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1186	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
870	ev_io_start (loop, &stdin_readable);	1187	ev_io_start (loop, &stdin_readable);
871	ev_loop (loop, 0);	1188	ev_loop (loop, 0);
872		1189
873		1190
874	=head2 C<ev_timer> - relative and optionally repeating timeouts	1191	=head2 C<ev_timer> - relative and optionally repeating timeouts
875		1192
876	Timer watchers are simple relative timers that generate an event after a	1193	Timer watchers are simple relative timers that generate an event after a
877	given time, and optionally repeating in regular intervals after that.	1194	given time, and optionally repeating in regular intervals after that.
878		1195
879	The timers are based on real time, that is, if you register an event that	1196	The timers are based on real time, that is, if you register an event that
880	times out after an hour and you reset your system clock to last years	1197	times out after an hour and you reset your system clock to January last
881	time, it will still time out after (roughly) and hour. "Roughly" because	1198	year, it will still time out after (roughly) and hour. "Roughly" because
882	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1199	detecting time jumps is hard, and some inaccuracies are unavoidable (the
883	monotonic clock option helps a lot here).	1200	monotonic clock option helps a lot here).
884		1201
885	The relative timeouts are calculated relative to the C<ev_now ()>	1202	The relative timeouts are calculated relative to the C<ev_now ()>
886	time. This is usually the right thing as this timestamp refers to the time	1203	time. This is usually the right thing as this timestamp refers to the time
…		…
888	you suspect event processing to be delayed and you I<need> to base the timeout	1205	you suspect event processing to be delayed and you I<need> to base the timeout
889	on the current time, use something like this to adjust for this:	1206	on the current time, use something like this to adjust for this:
890		1207
891	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1208	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
892		1209
893	The callback is guarenteed to be invoked only when its timeout has passed,	1210	The callback is guaranteed to be invoked only after its timeout has passed,
894	but if multiple timers become ready during the same loop iteration then	1211	but if multiple timers become ready during the same loop iteration then
895	order of execution is undefined.	1212	order of execution is undefined.
896		1213
		1214	=head3 Watcher-Specific Functions and Data Members
		1215
897	=over 4	1216	=over 4
898		1217
899	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1218	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
900		1219
901	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1220	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
902		1221
903	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1222	Configure the timer to trigger after C<after> seconds. If C<repeat>
904	C<0.>, then it will automatically be stopped. If it is positive, then the	1223	is C<0.>, then it will automatically be stopped once the timeout is
905	timer will automatically be configured to trigger again C<repeat> seconds	1224	reached. If it is positive, then the timer will automatically be
906	later, again, and again, until stopped manually.	1225	configured to trigger again C<repeat> seconds later, again, and again,
		1226	until stopped manually.
907		1227
908	The timer itself will do a best-effort at avoiding drift, that is, if you	1228	The timer itself will do a best-effort at avoiding drift, that is, if
909	configure a timer to trigger every 10 seconds, then it will trigger at	1229	you configure a timer to trigger every 10 seconds, then it will normally
910	exactly 10 second intervals. If, however, your program cannot keep up with	1230	trigger at exactly 10 second intervals. If, however, your program cannot
911	the timer (because it takes longer than those 10 seconds to do stuff) the	1231	keep up with the timer (because it takes longer than those 10 seconds to
912	timer will not fire more than once per event loop iteration.	1232	do stuff) the timer will not fire more than once per event loop iteration.
913		1233
914	=item ev_timer_again (loop)	1234	=item ev_timer_again (loop, ev_timer *)
915		1235
916	This will act as if the timer timed out and restart it again if it is	1236	This will act as if the timer timed out and restart it again if it is
917	repeating. The exact semantics are:	1237	repeating. The exact semantics are:
918		1238
		1239	If the timer is pending, its pending status is cleared.
		1240
919	If the timer is started but nonrepeating, stop it.	1241	If the timer is started but non-repeating, stop it (as if it timed out).
920		1242
921	If the timer is repeating, either start it if necessary (with the repeat	1243	If the timer is repeating, either start it if necessary (with the
922	value), or reset the running timer to the repeat value.	1244	C<repeat> value), or reset the running timer to the C<repeat> value.
923		1245
924	This sounds a bit complicated, but here is a useful and typical	1246	This sounds a bit complicated, but here is a useful and typical
925	example: Imagine you have a tcp connection and you want a so-called	1247	example: Imagine you have a TCP connection and you want a so-called idle
926	idle timeout, that is, you want to be called when there have been,	1248	timeout, that is, you want to be called when there have been, say, 60
927	say, 60 seconds of inactivity on the socket. The easiest way to do	1249	seconds of inactivity on the socket. The easiest way to do this is to
928	this is to configure an C<ev_timer> with C<after>=C<repeat>=C<60> and calling	1250	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
929	C<ev_timer_again> each time you successfully read or write some data. If	1251	C<ev_timer_again> each time you successfully read or write some data. If
930	you go into an idle state where you do not expect data to travel on the	1252	you go into an idle state where you do not expect data to travel on the
931	socket, you can stop the timer, and again will automatically restart it if	1253	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
932	need be.	1254	automatically restart it if need be.
933		1255
934	You can also ignore the C<after> value and C<ev_timer_start> altogether	1256	That means you can ignore the C<after> value and C<ev_timer_start>
935	and only ever use the C<repeat> value:	1257	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
936		1258
937	ev_timer_init (timer, callback, 0., 5.);	1259	ev_timer_init (timer, callback, 0., 5.);
938	ev_timer_again (loop, timer);	1260	ev_timer_again (loop, timer);
939	...	1261	...
940	timer->again = 17.;	1262	timer->again = 17.;
941	ev_timer_again (loop, timer);	1263	ev_timer_again (loop, timer);
942	...	1264	...
943	timer->again = 10.;	1265	timer->again = 10.;
944	ev_timer_again (loop, timer);	1266	ev_timer_again (loop, timer);
945		1267
946	This is more efficient then stopping/starting the timer eahc time you want	1268	This is more slightly efficient then stopping/starting the timer each time
947	to modify its timeout value.	1269	you want to modify its timeout value.
948		1270
949	=item ev_tstamp repeat [read-write]	1271	=item ev_tstamp repeat [read-write]
950		1272
951	The current C<repeat> value. Will be used each time the watcher times out	1273	The current C<repeat> value. Will be used each time the watcher times out
952	or C<ev_timer_again> is called and determines the next timeout (if any),	1274	or C<ev_timer_again> is called and determines the next timeout (if any),
953	which is also when any modifications are taken into account.	1275	which is also when any modifications are taken into account.
954		1276
955	=back	1277	=back
956		1278
		1279	=head3 Examples
		1280
957	Example: Create a timer that fires after 60 seconds.	1281	Example: Create a timer that fires after 60 seconds.
958		1282
959	static void	1283	static void
960	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1284	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
961	{	1285	{
962	.. one minute over, w is actually stopped right here	1286	.. one minute over, w is actually stopped right here
963	}	1287	}
964		1288
965	struct ev_timer mytimer;	1289	struct ev_timer mytimer;
966	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1290	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
967	ev_timer_start (loop, &mytimer);	1291	ev_timer_start (loop, &mytimer);
968		1292
969	Example: Create a timeout timer that times out after 10 seconds of	1293	Example: Create a timeout timer that times out after 10 seconds of
970	inactivity.	1294	inactivity.
971		1295
972	static void	1296	static void
973	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1297	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
974	{	1298	{
975	.. ten seconds without any activity	1299	.. ten seconds without any activity
976	}	1300	}
977		1301
978	struct ev_timer mytimer;	1302	struct ev_timer mytimer;
979	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */	1303	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
980	ev_timer_again (&mytimer); /* start timer */	1304	ev_timer_again (&mytimer); /* start timer */
981	ev_loop (loop, 0);	1305	ev_loop (loop, 0);
982		1306
983	// and in some piece of code that gets executed on any "activity":	1307	// and in some piece of code that gets executed on any "activity":
984	// reset the timeout to start ticking again at 10 seconds	1308	// reset the timeout to start ticking again at 10 seconds
985	ev_timer_again (&mytimer);	1309	ev_timer_again (&mytimer);
986		1310
987		1311
988	=head2 C<ev_periodic> - to cron or not to cron?	1312	=head2 C<ev_periodic> - to cron or not to cron?
989		1313
990	Periodic watchers are also timers of a kind, but they are very versatile	1314	Periodic watchers are also timers of a kind, but they are very versatile
991	(and unfortunately a bit complex).	1315	(and unfortunately a bit complex).
992		1316
993	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1317	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
994	but on wallclock time (absolute time). You can tell a periodic watcher	1318	but on wall clock time (absolute time). You can tell a periodic watcher
995	to trigger "at" some specific point in time. For example, if you tell a	1319	to trigger after some specific point in time. For example, if you tell a
996	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()	1320	periodic watcher to trigger in 10 seconds (by specifying e.g. C<ev_now ()
997	+ 10.>) and then reset your system clock to the last year, then it will	1321	+ 10.>, that is, an absolute time not a delay) and then reset your system
		1322	clock to January of the previous year, then it will take more than year
998	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1323	to trigger the event (unlike an C<ev_timer>, which would still trigger
999	roughly 10 seconds later and of course not if you reset your system time	1324	roughly 10 seconds later as it uses a relative timeout).
1000	again).
1001		1325
1002	They can also be used to implement vastly more complex timers, such as	1326	C<ev_periodic>s can also be used to implement vastly more complex timers,
1003	triggering an event on eahc midnight, local time.	1327	such as triggering an event on each "midnight, local time", or other
		1328	complicated, rules.
1004		1329
1005	As with timers, the callback is guarenteed to be invoked only when the	1330	As with timers, the callback is guaranteed to be invoked only when the
1006	time (C<at>) has been passed, but if multiple periodic timers become ready	1331	time (C<at>) has passed, but if multiple periodic timers become ready
1007	during the same loop iteration then order of execution is undefined.	1332	during the same loop iteration then order of execution is undefined.
		1333
		1334	=head3 Watcher-Specific Functions and Data Members
1008		1335
1009	=over 4	1336	=over 4
1010		1337
1011	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1338	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
1012		1339
…		…
1015	Lots of arguments, lets sort it out... There are basically three modes of	1342	Lots of arguments, lets sort it out... There are basically three modes of
1016	operation, and we will explain them from simplest to complex:	1343	operation, and we will explain them from simplest to complex:
1017		1344
1018	=over 4	1345	=over 4
1019		1346
1020	=item * absolute timer (interval = reschedule_cb = 0)	1347	=item * absolute timer (at = time, interval = reschedule_cb = 0)
1021		1348
1022	In this configuration the watcher triggers an event at the wallclock time	1349	In this configuration the watcher triggers an event after the wall clock
1023	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1350	time C<at> has passed and doesn't repeat. It will not adjust when a time
1024	that is, if it is to be run at January 1st 2011 then it will run when the	1351	jump occurs, that is, if it is to be run at January 1st 2011 then it will
1025	system time reaches or surpasses this time.	1352	run when the system time reaches or surpasses this time.
1026		1353
1027	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1354	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
1028		1355
1029	In this mode the watcher will always be scheduled to time out at the next	1356	In this mode the watcher will always be scheduled to time out at the next
1030	C<at + N * interval> time (for some integer N) and then repeat, regardless	1357	C<at + N * interval> time (for some integer N, which can also be negative)
1031	of any time jumps.	1358	and then repeat, regardless of any time jumps.
1032		1359
1033	This can be used to create timers that do not drift with respect to system	1360	This can be used to create timers that do not drift with respect to system
1034	time:	1361	time, for example, here is a C<ev_periodic> that triggers each hour, on
		1362	the hour:
1035		1363
1036	ev_periodic_set (&periodic, 0., 3600., 0);	1364	ev_periodic_set (&periodic, 0., 3600., 0);
1037		1365
1038	This doesn't mean there will always be 3600 seconds in between triggers,	1366	This doesn't mean there will always be 3600 seconds in between triggers,
1039	but only that the the callback will be called when the system time shows a	1367	but only that the callback will be called when the system time shows a
1040	full hour (UTC), or more correctly, when the system time is evenly divisible	1368	full hour (UTC), or more correctly, when the system time is evenly divisible
1041	by 3600.	1369	by 3600.
1042		1370
1043	Another way to think about it (for the mathematically inclined) is that	1371	Another way to think about it (for the mathematically inclined) is that
1044	C<ev_periodic> will try to run the callback in this mode at the next possible	1372	C<ev_periodic> will try to run the callback in this mode at the next possible
1045	time where C<time = at (mod interval)>, regardless of any time jumps.	1373	time where C<time = at (mod interval)>, regardless of any time jumps.
1046		1374
		1375	For numerical stability it is preferable that the C<at> value is near
		1376	C<ev_now ()> (the current time), but there is no range requirement for
		1377	this value, and in fact is often specified as zero.
		1378
		1379	Note also that there is an upper limit to how often a timer can fire (CPU
		1380	speed for example), so if C<interval> is very small then timing stability
		1381	will of course deteriorate. Libev itself tries to be exact to be about one
		1382	millisecond (if the OS supports it and the machine is fast enough).
		1383
1047	=item * manual reschedule mode (reschedule_cb = callback)	1384	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
1048		1385
1049	In this mode the values for C<interval> and C<at> are both being	1386	In this mode the values for C<interval> and C<at> are both being
1050	ignored. Instead, each time the periodic watcher gets scheduled, the	1387	ignored. Instead, each time the periodic watcher gets scheduled, the
1051	reschedule callback will be called with the watcher as first, and the	1388	reschedule callback will be called with the watcher as first, and the
1052	current time as second argument.	1389	current time as second argument.
1053		1390
1054	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1391	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
1055	ever, or make any event loop modifications>. If you need to stop it,	1392	ever, or make ANY event loop modifications whatsoever>.
1056	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
1057	starting a prepare watcher).
1058		1393
		1394	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1395	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1396	only event loop modification you are allowed to do).
		1397
1059	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1398	The callback prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic
1060	ev_tstamp now)>, e.g.:	1399	*w, ev_tstamp now)>, e.g.:
1061		1400
1062	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1401	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
1063	{	1402	{
1064	return now + 60.;	1403	return now + 60.;
1065	}	1404	}
…		…
1067	It must return the next time to trigger, based on the passed time value	1406	It must return the next time to trigger, based on the passed time value
1068	(that is, the lowest time value larger than to the second argument). It	1407	(that is, the lowest time value larger than to the second argument). It
1069	will usually be called just before the callback will be triggered, but	1408	will usually be called just before the callback will be triggered, but
1070	might be called at other times, too.	1409	might be called at other times, too.
1071		1410
1072	NOTE: I<< This callback must always return a time that is later than the	1411	NOTE: I<< This callback must always return a time that is higher than or
1073	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	1412	equal to the passed C<now> value >>.
1074		1413
1075	This can be used to create very complex timers, such as a timer that	1414	This can be used to create very complex timers, such as a timer that
1076	triggers on each midnight, local time. To do this, you would calculate the	1415	triggers on "next midnight, local time". To do this, you would calculate the
1077	next midnight after C<now> and return the timestamp value for this. How	1416	next midnight after C<now> and return the timestamp value for this. How
1078	you do this is, again, up to you (but it is not trivial, which is the main	1417	you do this is, again, up to you (but it is not trivial, which is the main
1079	reason I omitted it as an example).	1418	reason I omitted it as an example).
1080		1419
1081	=back	1420	=back
…		…
1085	Simply stops and restarts the periodic watcher again. This is only useful	1424	Simply stops and restarts the periodic watcher again. This is only useful
1086	when you changed some parameters or the reschedule callback would return	1425	when you changed some parameters or the reschedule callback would return
1087	a different time than the last time it was called (e.g. in a crond like	1426	a different time than the last time it was called (e.g. in a crond like
1088	program when the crontabs have changed).	1427	program when the crontabs have changed).
1089		1428
		1429	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1430
		1431	When active, returns the absolute time that the watcher is supposed to
		1432	trigger next.
		1433
		1434	=item ev_tstamp offset [read-write]
		1435
		1436	When repeating, this contains the offset value, otherwise this is the
		1437	absolute point in time (the C<at> value passed to C<ev_periodic_set>).
		1438
		1439	Can be modified any time, but changes only take effect when the periodic
		1440	timer fires or C<ev_periodic_again> is being called.
		1441
1090	=item ev_tstamp interval [read-write]	1442	=item ev_tstamp interval [read-write]
1091		1443
1092	The current interval value. Can be modified any time, but changes only	1444	The current interval value. Can be modified any time, but changes only
1093	take effect when the periodic timer fires or C<ev_periodic_again> is being	1445	take effect when the periodic timer fires or C<ev_periodic_again> is being
1094	called.	1446	called.
…		…
1099	switched off. Can be changed any time, but changes only take effect when	1451	switched off. Can be changed any time, but changes only take effect when
1100	the periodic timer fires or C<ev_periodic_again> is being called.	1452	the periodic timer fires or C<ev_periodic_again> is being called.
1101		1453
1102	=back	1454	=back
1103		1455
		1456	=head3 Examples
		1457
1104	Example: Call a callback every hour, or, more precisely, whenever the	1458	Example: Call a callback every hour, or, more precisely, whenever the
1105	system clock is divisible by 3600. The callback invocation times have	1459	system clock is divisible by 3600. The callback invocation times have
1106	potentially a lot of jittering, but good long-term stability.	1460	potentially a lot of jitter, but good long-term stability.
1107		1461
1108	static void	1462	static void
1109	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1463	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
1110	{	1464	{
1111	... its now a full hour (UTC, or TAI or whatever your clock follows)	1465	... its now a full hour (UTC, or TAI or whatever your clock follows)
1112	}	1466	}
1113		1467
1114	struct ev_periodic hourly_tick;	1468	struct ev_periodic hourly_tick;
1115	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1469	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
1116	ev_periodic_start (loop, &hourly_tick);	1470	ev_periodic_start (loop, &hourly_tick);
1117		1471
1118	Example: The same as above, but use a reschedule callback to do it:	1472	Example: The same as above, but use a reschedule callback to do it:
1119		1473
1120	#include <math.h>	1474	#include <math.h>
1121		1475
1122	static ev_tstamp	1476	static ev_tstamp
1123	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1477	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
1124	{	1478	{
1125	return fmod (now, 3600.) + 3600.;	1479	return fmod (now, 3600.) + 3600.;
1126	}	1480	}
1127		1481
1128	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1482	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
1129		1483
1130	Example: Call a callback every hour, starting now:	1484	Example: Call a callback every hour, starting now:
1131		1485
1132	struct ev_periodic hourly_tick;	1486	struct ev_periodic hourly_tick;
1133	ev_periodic_init (&hourly_tick, clock_cb,	1487	ev_periodic_init (&hourly_tick, clock_cb,
1134	fmod (ev_now (loop), 3600.), 3600., 0);	1488	fmod (ev_now (loop), 3600.), 3600., 0);
1135	ev_periodic_start (loop, &hourly_tick);	1489	ev_periodic_start (loop, &hourly_tick);
1136		1490
1137		1491
1138	=head2 C<ev_signal> - signal me when a signal gets signalled!	1492	=head2 C<ev_signal> - signal me when a signal gets signalled!
1139		1493
1140	Signal watchers will trigger an event when the process receives a specific	1494	Signal watchers will trigger an event when the process receives a specific
…		…
1147	with the kernel (thus it coexists with your own signal handlers as long	1501	with the kernel (thus it coexists with your own signal handlers as long
1148	as you don't register any with libev). Similarly, when the last signal	1502	as you don't register any with libev). Similarly, when the last signal
1149	watcher for a signal is stopped libev will reset the signal handler to	1503	watcher for a signal is stopped libev will reset the signal handler to
1150	SIG_DFL (regardless of what it was set to before).	1504	SIG_DFL (regardless of what it was set to before).
1151		1505
		1506	If possible and supported, libev will install its handlers with
		1507	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
		1508	interrupted. If you have a problem with system calls getting interrupted by
		1509	signals you can block all signals in an C<ev_check> watcher and unblock
		1510	them in an C<ev_prepare> watcher.
		1511
		1512	=head3 Watcher-Specific Functions and Data Members
		1513
1152	=over 4	1514	=over 4
1153		1515
1154	=item ev_signal_init (ev_signal *, callback, int signum)	1516	=item ev_signal_init (ev_signal *, callback, int signum)
1155		1517
1156	=item ev_signal_set (ev_signal *, int signum)	1518	=item ev_signal_set (ev_signal *, int signum)
…		…
1162		1524
1163	The signal the watcher watches out for.	1525	The signal the watcher watches out for.
1164		1526
1165	=back	1527	=back
1166		1528
		1529	=head3 Examples
		1530
		1531	Example: Try to exit cleanly on SIGINT and SIGTERM.
		1532
		1533	static void
		1534	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
		1535	{
		1536	ev_unloop (loop, EVUNLOOP_ALL);
		1537	}
		1538
		1539	struct ev_signal signal_watcher;
		1540	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1541	ev_signal_start (loop, &sigint_cb);
		1542
1167		1543
1168	=head2 C<ev_child> - watch out for process status changes	1544	=head2 C<ev_child> - watch out for process status changes
1169		1545
1170	Child watchers trigger when your process receives a SIGCHLD in response to	1546	Child watchers trigger when your process receives a SIGCHLD in response to
1171	some child status changes (most typically when a child of yours dies).	1547	some child status changes (most typically when a child of yours dies). It
		1548	is permissible to install a child watcher I<after> the child has been
		1549	forked (which implies it might have already exited), as long as the event
		1550	loop isn't entered (or is continued from a watcher).
		1551
		1552	Only the default event loop is capable of handling signals, and therefore
		1553	you can only register child watchers in the default event loop.
		1554
		1555	=head3 Process Interaction
		1556
		1557	Libev grabs C<SIGCHLD> as soon as the default event loop is
		1558	initialised. This is necessary to guarantee proper behaviour even if
		1559	the first child watcher is started after the child exits. The occurrence
		1560	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		1561	synchronously as part of the event loop processing. Libev always reaps all
		1562	children, even ones not watched.
		1563
		1564	=head3 Overriding the Built-In Processing
		1565
		1566	Libev offers no special support for overriding the built-in child
		1567	processing, but if your application collides with libev's default child
		1568	handler, you can override it easily by installing your own handler for
		1569	C<SIGCHLD> after initialising the default loop, and making sure the
		1570	default loop never gets destroyed. You are encouraged, however, to use an
		1571	event-based approach to child reaping and thus use libev's support for
		1572	that, so other libev users can use C<ev_child> watchers freely.
		1573
		1574	=head3 Stopping the Child Watcher
		1575
		1576	Currently, the child watcher never gets stopped, even when the
		1577	child terminates, so normally one needs to stop the watcher in the
		1578	callback. Future versions of libev might stop the watcher automatically
		1579	when a child exit is detected.
		1580
		1581	=head3 Watcher-Specific Functions and Data Members
1172		1582
1173	=over 4	1583	=over 4
1174		1584
1175	=item ev_child_init (ev_child *, callback, int pid)	1585	=item ev_child_init (ev_child *, callback, int pid, int trace)
1176		1586
1177	=item ev_child_set (ev_child *, int pid)	1587	=item ev_child_set (ev_child *, int pid, int trace)
1178		1588
1179	Configures the watcher to wait for status changes of process C<pid> (or	1589	Configures the watcher to wait for status changes of process C<pid> (or
1180	I<any> process if C<pid> is specified as C<0>). The callback can look	1590	I<any> process if C<pid> is specified as C<0>). The callback can look
1181	at the C<rstatus> member of the C<ev_child> watcher structure to see	1591	at the C<rstatus> member of the C<ev_child> watcher structure to see
1182	the status word (use the macros from C<sys/wait.h> and see your systems	1592	the status word (use the macros from C<sys/wait.h> and see your systems
1183	C<waitpid> documentation). The C<rpid> member contains the pid of the	1593	C<waitpid> documentation). The C<rpid> member contains the pid of the
1184	process causing the status change.	1594	process causing the status change. C<trace> must be either C<0> (only
		1595	activate the watcher when the process terminates) or C<1> (additionally
		1596	activate the watcher when the process is stopped or continued).
1185		1597
1186	=item int pid [read-only]	1598	=item int pid [read-only]
1187		1599
1188	The process id this watcher watches out for, or C<0>, meaning any process id.	1600	The process id this watcher watches out for, or C<0>, meaning any process id.
1189		1601
…		…
1196	The process exit/trace status caused by C<rpid> (see your systems	1608	The process exit/trace status caused by C<rpid> (see your systems
1197	C<waitpid> and C<sys/wait.h> documentation for details).	1609	C<waitpid> and C<sys/wait.h> documentation for details).
1198		1610
1199	=back	1611	=back
1200		1612
1201	Example: Try to exit cleanly on SIGINT and SIGTERM.	1613	=head3 Examples
1202		1614
		1615	Example: C<fork()> a new process and install a child handler to wait for
		1616	its completion.
		1617
		1618	ev_child cw;
		1619
1203	static void	1620	static void
1204	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1621	child_cb (EV_P_ struct ev_child *w, int revents)
1205	{	1622	{
1206	ev_unloop (loop, EVUNLOOP_ALL);	1623	ev_child_stop (EV_A_ w);
		1624	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1207	}	1625	}
1208		1626
1209	struct ev_signal signal_watcher;	1627	pid_t pid = fork ();
1210	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1628
1211	ev_signal_start (loop, &sigint_cb);	1629	if (pid < 0)
		1630	// error
		1631	else if (pid == 0)
		1632	{
		1633	// the forked child executes here
		1634	exit (1);
		1635	}
		1636	else
		1637	{
		1638	ev_child_init (&cw, child_cb, pid, 0);
		1639	ev_child_start (EV_DEFAULT_ &cw);
		1640	}
1212		1641
1213		1642
1214	=head2 C<ev_stat> - did the file attributes just change?	1643	=head2 C<ev_stat> - did the file attributes just change?
1215		1644
1216	This watches a filesystem path for attribute changes. That is, it calls	1645	This watches a file system path for attribute changes. That is, it calls
1217	C<stat> regularly (or when the OS says it changed) and sees if it changed	1646	C<stat> regularly (or when the OS says it changed) and sees if it changed
1218	compared to the last time, invoking the callback if it did.	1647	compared to the last time, invoking the callback if it did.
1219		1648
1220	The path does not need to exist: changing from "path exists" to "path does	1649	The path does not need to exist: changing from "path exists" to "path does
1221	not exist" is a status change like any other. The condition "path does	1650	not exist" is a status change like any other. The condition "path does
1222	not exist" is signified by the C<st_nlink> field being zero (which is	1651	not exist" is signified by the C<st_nlink> field being zero (which is
1223	otherwise always forced to be at least one) and all the other fields of	1652	otherwise always forced to be at least one) and all the other fields of
1224	the stat buffer having unspecified contents.	1653	the stat buffer having unspecified contents.
		1654
		1655	The path I<should> be absolute and I<must not> end in a slash. If it is
		1656	relative and your working directory changes, the behaviour is undefined.
1225		1657
1226	Since there is no standard to do this, the portable implementation simply	1658	Since there is no standard to do this, the portable implementation simply
1227	calls C<stat (2)> regularly on the path to see if it changed somehow. You	1659	calls C<stat (2)> regularly on the path to see if it changed somehow. You
1228	can specify a recommended polling interval for this case. If you specify	1660	can specify a recommended polling interval for this case. If you specify
1229	a polling interval of C<0> (highly recommended!) then a I<suitable,	1661	a polling interval of C<0> (highly recommended!) then a I<suitable,
…		…
1236	as even with OS-supported change notifications, this can be	1668	as even with OS-supported change notifications, this can be
1237	resource-intensive.	1669	resource-intensive.
1238		1670
1239	At the time of this writing, only the Linux inotify interface is	1671	At the time of this writing, only the Linux inotify interface is
1240	implemented (implementing kqueue support is left as an exercise for the	1672	implemented (implementing kqueue support is left as an exercise for the
		1673	reader, note, however, that the author sees no way of implementing ev_stat
1241	reader). Inotify will be used to give hints only and should not change the	1674	semantics with kqueue). Inotify will be used to give hints only and should
1242	semantics of C<ev_stat> watchers, which means that libev sometimes needs	1675	not change the semantics of C<ev_stat> watchers, which means that libev
1243	to fall back to regular polling again even with inotify, but changes are	1676	sometimes needs to fall back to regular polling again even with inotify,
1244	usually detected immediately, and if the file exists there will be no	1677	but changes are usually detected immediately, and if the file exists there
1245	polling.	1678	will be no polling.
		1679
		1680	=head3 ABI Issues (Largefile Support)
		1681
		1682	Libev by default (unless the user overrides this) uses the default
		1683	compilation environment, which means that on systems with large file
		1684	support disabled by default, you get the 32 bit version of the stat
		1685	structure. When using the library from programs that change the ABI to
		1686	use 64 bit file offsets the programs will fail. In that case you have to
		1687	compile libev with the same flags to get binary compatibility. This is
		1688	obviously the case with any flags that change the ABI, but the problem is
		1689	most noticeably disabled with ev_stat and large file support.
		1690
		1691	The solution for this is to lobby your distribution maker to make large
		1692	file interfaces available by default (as e.g. FreeBSD does) and not
		1693	optional. Libev cannot simply switch on large file support because it has
		1694	to exchange stat structures with application programs compiled using the
		1695	default compilation environment.
		1696
		1697	=head3 Inotify
		1698
		1699	When C<inotify (7)> support has been compiled into libev (generally only
		1700	available on Linux) and present at runtime, it will be used to speed up
		1701	change detection where possible. The inotify descriptor will be created lazily
		1702	when the first C<ev_stat> watcher is being started.
		1703
		1704	Inotify presence does not change the semantics of C<ev_stat> watchers
		1705	except that changes might be detected earlier, and in some cases, to avoid
		1706	making regular C<stat> calls. Even in the presence of inotify support
		1707	there are many cases where libev has to resort to regular C<stat> polling.
		1708
		1709	(There is no support for kqueue, as apparently it cannot be used to
		1710	implement this functionality, due to the requirement of having a file
		1711	descriptor open on the object at all times).
		1712
		1713	=head3 The special problem of stat time resolution
		1714
		1715	The C<stat ()> system call only supports full-second resolution portably, and
		1716	even on systems where the resolution is higher, many file systems still
		1717	only support whole seconds.
		1718
		1719	That means that, if the time is the only thing that changes, you can
		1720	easily miss updates: on the first update, C<ev_stat> detects a change and
		1721	calls your callback, which does something. When there is another update
		1722	within the same second, C<ev_stat> will be unable to detect it as the stat
		1723	data does not change.
		1724
		1725	The solution to this is to delay acting on a change for slightly more
		1726	than a second (or till slightly after the next full second boundary), using
		1727	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		1728	ev_timer_again (loop, w)>).
		1729
		1730	The C<.02> offset is added to work around small timing inconsistencies
		1731	of some operating systems (where the second counter of the current time
		1732	might be be delayed. One such system is the Linux kernel, where a call to
		1733	C<gettimeofday> might return a timestamp with a full second later than
		1734	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		1735	update file times then there will be a small window where the kernel uses
		1736	the previous second to update file times but libev might already execute
		1737	the timer callback).
		1738
		1739	=head3 Watcher-Specific Functions and Data Members
1246		1740
1247	=over 4	1741	=over 4
1248		1742
1249	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)	1743	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
1250		1744
…		…
1254	C<path>. The C<interval> is a hint on how quickly a change is expected to	1748	C<path>. The C<interval> is a hint on how quickly a change is expected to
1255	be detected and should normally be specified as C<0> to let libev choose	1749	be detected and should normally be specified as C<0> to let libev choose
1256	a suitable value. The memory pointed to by C<path> must point to the same	1750	a suitable value. The memory pointed to by C<path> must point to the same
1257	path for as long as the watcher is active.	1751	path for as long as the watcher is active.
1258		1752
1259	The callback will be receive C<EV_STAT> when a change was detected,	1753	The callback will receive C<EV_STAT> when a change was detected, relative
1260	relative to the attributes at the time the watcher was started (or the	1754	to the attributes at the time the watcher was started (or the last change
1261	last change was detected).	1755	was detected).
1262		1756
1263	=item ev_stat_stat (ev_stat *)	1757	=item ev_stat_stat (loop, ev_stat *)
1264		1758
1265	Updates the stat buffer immediately with new values. If you change the	1759	Updates the stat buffer immediately with new values. If you change the
1266	watched path in your callback, you could call this fucntion to avoid	1760	watched path in your callback, you could call this function to avoid
1267	detecting this change (while introducing a race condition). Can also be	1761	detecting this change (while introducing a race condition if you are not
1268	useful simply to find out the new values.	1762	the only one changing the path). Can also be useful simply to find out the
		1763	new values.
1269		1764
1270	=item ev_statdata attr [read-only]	1765	=item ev_statdata attr [read-only]
1271		1766
1272	The most-recently detected attributes of the file. Although the type is of	1767	The most-recently detected attributes of the file. Although the type is
1273	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types	1768	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
1274	suitable for your system. If the C<st_nlink> member is C<0>, then there	1769	suitable for your system, but you can only rely on the POSIX-standardised
		1770	members to be present. If the C<st_nlink> member is C<0>, then there was
1275	was some error while C<stat>ing the file.	1771	some error while C<stat>ing the file.
1276		1772
1277	=item ev_statdata prev [read-only]	1773	=item ev_statdata prev [read-only]
1278		1774
1279	The previous attributes of the file. The callback gets invoked whenever	1775	The previous attributes of the file. The callback gets invoked whenever
1280	C<prev> != C<attr>.	1776	C<prev> != C<attr>, or, more precisely, one or more of these members
		1777	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		1778	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
1281		1779
1282	=item ev_tstamp interval [read-only]	1780	=item ev_tstamp interval [read-only]
1283		1781
1284	The specified interval.	1782	The specified interval.
1285		1783
1286	=item const char *path [read-only]	1784	=item const char *path [read-only]
1287		1785
1288	The filesystem path that is being watched.	1786	The file system path that is being watched.
1289		1787
1290	=back	1788	=back
1291		1789
		1790	=head3 Examples
		1791
1292	Example: Watch C</etc/passwd> for attribute changes.	1792	Example: Watch C</etc/passwd> for attribute changes.
1293		1793
1294	static void	1794	static void
1295	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)	1795	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
1296	{	1796	{
1297	/* /etc/passwd changed in some way */	1797	/* /etc/passwd changed in some way */
1298	if (w->attr.st_nlink)	1798	if (w->attr.st_nlink)
1299	{	1799	{
1300	printf ("passwd current size %ld\n", (long)w->attr.st_size);	1800	printf ("passwd current size %ld\n", (long)w->attr.st_size);
1301	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);	1801	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
1302	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);	1802	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
1303	}	1803	}
1304	else	1804	else
1305	/* you shalt not abuse printf for puts */	1805	/* you shalt not abuse printf for puts */
1306	puts ("wow, /etc/passwd is not there, expect problems. "	1806	puts ("wow, /etc/passwd is not there, expect problems. "
1307	"if this is windows, they already arrived\n");	1807	"if this is windows, they already arrived\n");
1308	}	1808	}
1309		1809
1310	...	1810	...
1311	ev_stat passwd;	1811	ev_stat passwd;
1312		1812
1313	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");	1813	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1314	ev_stat_start (loop, &passwd);	1814	ev_stat_start (loop, &passwd);
		1815
		1816	Example: Like above, but additionally use a one-second delay so we do not
		1817	miss updates (however, frequent updates will delay processing, too, so
		1818	one might do the work both on C<ev_stat> callback invocation I<and> on
		1819	C<ev_timer> callback invocation).
		1820
		1821	static ev_stat passwd;
		1822	static ev_timer timer;
		1823
		1824	static void
		1825	timer_cb (EV_P_ ev_timer *w, int revents)
		1826	{
		1827	ev_timer_stop (EV_A_ w);
		1828
		1829	/* now it's one second after the most recent passwd change */
		1830	}
		1831
		1832	static void
		1833	stat_cb (EV_P_ ev_stat *w, int revents)
		1834	{
		1835	/* reset the one-second timer */
		1836	ev_timer_again (EV_A_ &timer);
		1837	}
		1838
		1839	...
		1840	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		1841	ev_stat_start (loop, &passwd);
		1842	ev_timer_init (&timer, timer_cb, 0., 1.02);
1315		1843
1316		1844
1317	=head2 C<ev_idle> - when you've got nothing better to do...	1845	=head2 C<ev_idle> - when you've got nothing better to do...
1318		1846
1319	Idle watchers trigger events when there are no other events are pending	1847	Idle watchers trigger events when no other events of the same or higher
1320	(prepare, check and other idle watchers do not count). That is, as long	1848	priority are pending (prepare, check and other idle watchers do not
1321	as your process is busy handling sockets or timeouts (or even signals,	1849	count).
1322	imagine) it will not be triggered. But when your process is idle all idle	1850
1323	watchers are being called again and again, once per event loop iteration -	1851	That is, as long as your process is busy handling sockets or timeouts
		1852	(or even signals, imagine) of the same or higher priority it will not be
		1853	triggered. But when your process is idle (or only lower-priority watchers
		1854	are pending), the idle watchers are being called once per event loop
1324	until stopped, that is, or your process receives more events and becomes	1855	iteration - until stopped, that is, or your process receives more events
1325	busy.	1856	and becomes busy again with higher priority stuff.
1326		1857
1327	The most noteworthy effect is that as long as any idle watchers are	1858	The most noteworthy effect is that as long as any idle watchers are
1328	active, the process will not block when waiting for new events.	1859	active, the process will not block when waiting for new events.
1329		1860
1330	Apart from keeping your process non-blocking (which is a useful	1861	Apart from keeping your process non-blocking (which is a useful
1331	effect on its own sometimes), idle watchers are a good place to do	1862	effect on its own sometimes), idle watchers are a good place to do
1332	"pseudo-background processing", or delay processing stuff to after the	1863	"pseudo-background processing", or delay processing stuff to after the
1333	event loop has handled all outstanding events.	1864	event loop has handled all outstanding events.
1334		1865
		1866	=head3 Watcher-Specific Functions and Data Members
		1867
1335	=over 4	1868	=over 4
1336		1869
1337	=item ev_idle_init (ev_signal *, callback)	1870	=item ev_idle_init (ev_signal *, callback)
1338		1871
1339	Initialises and configures the idle watcher - it has no parameters of any	1872	Initialises and configures the idle watcher - it has no parameters of any
1340	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1873	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1341	believe me.	1874	believe me.
1342		1875
1343	=back	1876	=back
1344		1877
		1878	=head3 Examples
		1879
1345	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	1880	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1346	callback, free it. Also, use no error checking, as usual.	1881	callback, free it. Also, use no error checking, as usual.
1347		1882
1348	static void	1883	static void
1349	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	1884	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
1350	{	1885	{
1351	free (w);	1886	free (w);
1352	// now do something you wanted to do when the program has	1887	// now do something you wanted to do when the program has
1353	// no longer asnything immediate to do.	1888	// no longer anything immediate to do.
1354	}	1889	}
1355		1890
1356	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1891	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1357	ev_idle_init (idle_watcher, idle_cb);	1892	ev_idle_init (idle_watcher, idle_cb);
1358	ev_idle_start (loop, idle_cb);	1893	ev_idle_start (loop, idle_cb);
1359		1894
1360		1895
1361	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!	1896	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1362		1897
1363	Prepare and check watchers are usually (but not always) used in tandem:	1898	Prepare and check watchers are usually (but not always) used in tandem:
…		…
1382		1917
1383	This is done by examining in each prepare call which file descriptors need	1918	This is done by examining in each prepare call which file descriptors need
1384	to be watched by the other library, registering C<ev_io> watchers for	1919	to be watched by the other library, registering C<ev_io> watchers for
1385	them and starting an C<ev_timer> watcher for any timeouts (many libraries	1920	them and starting an C<ev_timer> watcher for any timeouts (many libraries
1386	provide just this functionality). Then, in the check watcher you check for	1921	provide just this functionality). Then, in the check watcher you check for
1387	any events that occured (by checking the pending status of all watchers	1922	any events that occurred (by checking the pending status of all watchers
1388	and stopping them) and call back into the library. The I/O and timer	1923	and stopping them) and call back into the library. The I/O and timer
1389	callbacks will never actually be called (but must be valid nevertheless,	1924	callbacks will never actually be called (but must be valid nevertheless,
1390	because you never know, you know?).	1925	because you never know, you know?).
1391		1926
1392	As another example, the Perl Coro module uses these hooks to integrate	1927	As another example, the Perl Coro module uses these hooks to integrate
…		…
1396	with priority higher than or equal to the event loop and one coroutine	1931	with priority higher than or equal to the event loop and one coroutine
1397	of lower priority, but only once, using idle watchers to keep the event	1932	of lower priority, but only once, using idle watchers to keep the event
1398	loop from blocking if lower-priority coroutines are active, thus mapping	1933	loop from blocking if lower-priority coroutines are active, thus mapping
1399	low-priority coroutines to idle/background tasks).	1934	low-priority coroutines to idle/background tasks).
1400		1935
		1936	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		1937	priority, to ensure that they are being run before any other watchers
		1938	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,
		1939	too) should not activate ("feed") events into libev. While libev fully
		1940	supports this, they might get executed before other C<ev_check> watchers
		1941	did their job. As C<ev_check> watchers are often used to embed other
		1942	(non-libev) event loops those other event loops might be in an unusable
		1943	state until their C<ev_check> watcher ran (always remind yourself to
		1944	coexist peacefully with others).
		1945
		1946	=head3 Watcher-Specific Functions and Data Members
		1947
1401	=over 4	1948	=over 4
1402		1949
1403	=item ev_prepare_init (ev_prepare *, callback)	1950	=item ev_prepare_init (ev_prepare *, callback)
1404		1951
1405	=item ev_check_init (ev_check *, callback)	1952	=item ev_check_init (ev_check *, callback)
…		…
1408	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1955	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1409	macros, but using them is utterly, utterly and completely pointless.	1956	macros, but using them is utterly, utterly and completely pointless.
1410		1957
1411	=back	1958	=back
1412		1959
1413	Example: To include a library such as adns, you would add IO watchers	1960	=head3 Examples
1414	and a timeout watcher in a prepare handler, as required by libadns, and	1961
		1962	There are a number of principal ways to embed other event loops or modules
		1963	into libev. Here are some ideas on how to include libadns into libev
		1964	(there is a Perl module named C<EV::ADNS> that does this, which you could
		1965	use as a working example. Another Perl module named C<EV::Glib> embeds a
		1966	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		1967	Glib event loop).
		1968
		1969	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
1415	in a check watcher, destroy them and call into libadns. What follows is	1970	and in a check watcher, destroy them and call into libadns. What follows
1416	pseudo-code only of course:	1971	is pseudo-code only of course. This requires you to either use a low
		1972	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		1973	the callbacks for the IO/timeout watchers might not have been called yet.
1417		1974
1418	static ev_io iow [nfd];	1975	static ev_io iow [nfd];
1419	static ev_timer tw;	1976	static ev_timer tw;
1420		1977
1421	static void	1978	static void
1422	io_cb (ev_loop *loop, ev_io *w, int revents)	1979	io_cb (ev_loop *loop, ev_io *w, int revents)
1423	{	1980	{
1424	// set the relevant poll flags
1425	// could also call adns_processreadable etc. here
1426	struct pollfd *fd = (struct pollfd *)w->data;
1427	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
1428	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
1429	}	1981	}
1430		1982
1431	// create io watchers for each fd and a timer before blocking	1983	// create io watchers for each fd and a timer before blocking
1432	static void	1984	static void
1433	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)	1985	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
1434	{	1986	{
1435	int timeout = 3600000;truct pollfd fds [nfd];	1987	int timeout = 3600000;
		1988	struct pollfd fds [nfd];
1436	// actual code will need to loop here and realloc etc.	1989	// actual code will need to loop here and realloc etc.
1437	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	1990	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1438		1991
1439	/* the callback is illegal, but won't be called as we stop during check */	1992	/* the callback is illegal, but won't be called as we stop during check */
1440	ev_timer_init (&tw, 0, timeout * 1e-3);	1993	ev_timer_init (&tw, 0, timeout * 1e-3);
1441	ev_timer_start (loop, &tw);	1994	ev_timer_start (loop, &tw);
1442		1995
1443	// create on ev_io per pollfd	1996	// create one ev_io per pollfd
1444	for (int i = 0; i < nfd; ++i)	1997	for (int i = 0; i < nfd; ++i)
1445	{	1998	{
1446	ev_io_init (iow + i, io_cb, fds [i].fd,	1999	ev_io_init (iow + i, io_cb, fds [i].fd,
1447	((fds [i].events & POLLIN ? EV_READ : 0)	2000	((fds [i].events & POLLIN ? EV_READ : 0)
1448	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));	2001	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
1449		2002
1450	fds [i].revents = 0;	2003	fds [i].revents = 0;
1451	iow [i].data = fds + i;
1452	ev_io_start (loop, iow + i);	2004	ev_io_start (loop, iow + i);
1453	}	2005	}
1454	}	2006	}
1455		2007
1456	// stop all watchers after blocking	2008	// stop all watchers after blocking
1457	static void	2009	static void
1458	adns_check_cb (ev_loop *loop, ev_check *w, int revents)	2010	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
1459	{	2011	{
1460	ev_timer_stop (loop, &tw);	2012	ev_timer_stop (loop, &tw);
1461		2013
1462	for (int i = 0; i < nfd; ++i)	2014	for (int i = 0; i < nfd; ++i)
		2015	{
		2016	// set the relevant poll flags
		2017	// could also call adns_processreadable etc. here
		2018	struct pollfd *fd = fds + i;
		2019	int revents = ev_clear_pending (iow + i);
		2020	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		2021	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		2022
		2023	// now stop the watcher
1463	ev_io_stop (loop, iow + i);	2024	ev_io_stop (loop, iow + i);
		2025	}
1464		2026
1465	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));	2027	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
1466	}	2028	}
		2029
		2030	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		2031	in the prepare watcher and would dispose of the check watcher.
		2032
		2033	Method 3: If the module to be embedded supports explicit event
		2034	notification (libadns does), you can also make use of the actual watcher
		2035	callbacks, and only destroy/create the watchers in the prepare watcher.
		2036
		2037	static void
		2038	timer_cb (EV_P_ ev_timer *w, int revents)
		2039	{
		2040	adns_state ads = (adns_state)w->data;
		2041	update_now (EV_A);
		2042
		2043	adns_processtimeouts (ads, &tv_now);
		2044	}
		2045
		2046	static void
		2047	io_cb (EV_P_ ev_io *w, int revents)
		2048	{
		2049	adns_state ads = (adns_state)w->data;
		2050	update_now (EV_A);
		2051
		2052	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		2053	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		2054	}
		2055
		2056	// do not ever call adns_afterpoll
		2057
		2058	Method 4: Do not use a prepare or check watcher because the module you
		2059	want to embed is too inflexible to support it. Instead, you can override
		2060	their poll function. The drawback with this solution is that the main
		2061	loop is now no longer controllable by EV. The C<Glib::EV> module does
		2062	this.
		2063
		2064	static gint
		2065	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		2066	{
		2067	int got_events = 0;
		2068
		2069	for (n = 0; n < nfds; ++n)
		2070	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		2071
		2072	if (timeout >= 0)
		2073	// create/start timer
		2074
		2075	// poll
		2076	ev_loop (EV_A_ 0);
		2077
		2078	// stop timer again
		2079	if (timeout >= 0)
		2080	ev_timer_stop (EV_A_ &to);
		2081
		2082	// stop io watchers again - their callbacks should have set
		2083	for (n = 0; n < nfds; ++n)
		2084	ev_io_stop (EV_A_ iow [n]);
		2085
		2086	return got_events;
		2087	}
1467		2088
1468		2089
1469	=head2 C<ev_embed> - when one backend isn't enough...	2090	=head2 C<ev_embed> - when one backend isn't enough...
1470		2091
1471	This is a rather advanced watcher type that lets you embed one event loop	2092	This is a rather advanced watcher type that lets you embed one event loop
…		…
1513	portable one.	2134	portable one.
1514		2135
1515	So when you want to use this feature you will always have to be prepared	2136	So when you want to use this feature you will always have to be prepared
1516	that you cannot get an embeddable loop. The recommended way to get around	2137	that you cannot get an embeddable loop. The recommended way to get around
1517	this is to have a separate variables for your embeddable loop, try to	2138	this is to have a separate variables for your embeddable loop, try to
1518	create it, and if that fails, use the normal loop for everything:	2139	create it, and if that fails, use the normal loop for everything.
1519		2140
1520	struct ev_loop *loop_hi = ev_default_init (0);	2141	=head3 Watcher-Specific Functions and Data Members
1521	struct ev_loop *loop_lo = 0;
1522	struct ev_embed embed;
1523
1524	// see if there is a chance of getting one that works
1525	// (remember that a flags value of 0 means autodetection)
1526	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
1527	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
1528	: 0;
1529
1530	// if we got one, then embed it, otherwise default to loop_hi
1531	if (loop_lo)
1532	{
1533	ev_embed_init (&embed, 0, loop_lo);
1534	ev_embed_start (loop_hi, &embed);
1535	}
1536	else
1537	loop_lo = loop_hi;
1538		2142
1539	=over 4	2143	=over 4
1540		2144
1541	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)	2145	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
1542		2146
…		…
1544		2148
1545	Configures the watcher to embed the given loop, which must be	2149	Configures the watcher to embed the given loop, which must be
1546	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be	2150	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1547	invoked automatically, otherwise it is the responsibility of the callback	2151	invoked automatically, otherwise it is the responsibility of the callback
1548	to invoke it (it will continue to be called until the sweep has been done,	2152	to invoke it (it will continue to be called until the sweep has been done,
1549	if you do not want thta, you need to temporarily stop the embed watcher).	2153	if you do not want that, you need to temporarily stop the embed watcher).
1550		2154
1551	=item ev_embed_sweep (loop, ev_embed *)	2155	=item ev_embed_sweep (loop, ev_embed *)
1552		2156
1553	Make a single, non-blocking sweep over the embedded loop. This works	2157	Make a single, non-blocking sweep over the embedded loop. This works
1554	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most	2158	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
1555	apropriate way for embedded loops.	2159	appropriate way for embedded loops.
1556		2160
1557	=item struct ev_loop *loop [read-only]	2161	=item struct ev_loop *other [read-only]
1558		2162
1559	The embedded event loop.	2163	The embedded event loop.
1560		2164
1561	=back	2165	=back
		2166
		2167	=head3 Examples
		2168
		2169	Example: Try to get an embeddable event loop and embed it into the default
		2170	event loop. If that is not possible, use the default loop. The default
		2171	loop is stored in C<loop_hi>, while the embeddable loop is stored in
		2172	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
		2173	used).
		2174
		2175	struct ev_loop *loop_hi = ev_default_init (0);
		2176	struct ev_loop *loop_lo = 0;
		2177	struct ev_embed embed;
		2178
		2179	// see if there is a chance of getting one that works
		2180	// (remember that a flags value of 0 means autodetection)
		2181	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		2182	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		2183	: 0;
		2184
		2185	// if we got one, then embed it, otherwise default to loop_hi
		2186	if (loop_lo)
		2187	{
		2188	ev_embed_init (&embed, 0, loop_lo);
		2189	ev_embed_start (loop_hi, &embed);
		2190	}
		2191	else
		2192	loop_lo = loop_hi;
		2193
		2194	Example: Check if kqueue is available but not recommended and create
		2195	a kqueue backend for use with sockets (which usually work with any
		2196	kqueue implementation). Store the kqueue/socket-only event loop in
		2197	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		2198
		2199	struct ev_loop *loop = ev_default_init (0);
		2200	struct ev_loop *loop_socket = 0;
		2201	struct ev_embed embed;
		2202
		2203	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2204	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2205	{
		2206	ev_embed_init (&embed, 0, loop_socket);
		2207	ev_embed_start (loop, &embed);
		2208	}
		2209
		2210	if (!loop_socket)
		2211	loop_socket = loop;
		2212
		2213	// now use loop_socket for all sockets, and loop for everything else
1562		2214
1563		2215
1564	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	2216	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
1565		2217
1566	Fork watchers are called when a C<fork ()> was detected (usually because	2218	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
1569	event loop blocks next and before C<ev_check> watchers are being called,	2221	event loop blocks next and before C<ev_check> watchers are being called,
1570	and only in the child after the fork. If whoever good citizen calling	2222	and only in the child after the fork. If whoever good citizen calling
1571	C<ev_default_fork> cheats and calls it in the wrong process, the fork	2223	C<ev_default_fork> cheats and calls it in the wrong process, the fork
1572	handlers will be invoked, too, of course.	2224	handlers will be invoked, too, of course.
1573		2225
		2226	=head3 Watcher-Specific Functions and Data Members
		2227
1574	=over 4	2228	=over 4
1575		2229
1576	=item ev_fork_init (ev_signal *, callback)	2230	=item ev_fork_init (ev_signal *, callback)
1577		2231
1578	Initialises and configures the fork watcher - it has no parameters of any	2232	Initialises and configures the fork watcher - it has no parameters of any
1579	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,	2233	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
1580	believe me.	2234	believe me.
		2235
		2236	=back
		2237
		2238
		2239	=head2 C<ev_async> - how to wake up another event loop
		2240
		2241	In general, you cannot use an C<ev_loop> from multiple threads or other
		2242	asynchronous sources such as signal handlers (as opposed to multiple event
		2243	loops - those are of course safe to use in different threads).
		2244
		2245	Sometimes, however, you need to wake up another event loop you do not
		2246	control, for example because it belongs to another thread. This is what
		2247	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2248	can signal it by calling C<ev_async_send>, which is thread- and signal
		2249	safe.
		2250
		2251	This functionality is very similar to C<ev_signal> watchers, as signals,
		2252	too, are asynchronous in nature, and signals, too, will be compressed
		2253	(i.e. the number of callback invocations may be less than the number of
		2254	C<ev_async_sent> calls).
		2255
		2256	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2257	just the default loop.
		2258
		2259	=head3 Queueing
		2260
		2261	C<ev_async> does not support queueing of data in any way. The reason
		2262	is that the author does not know of a simple (or any) algorithm for a
		2263	multiple-writer-single-reader queue that works in all cases and doesn't
		2264	need elaborate support such as pthreads.
		2265
		2266	That means that if you want to queue data, you have to provide your own
		2267	queue. But at least I can tell you would implement locking around your
		2268	queue:
		2269
		2270	=over 4
		2271
		2272	=item queueing from a signal handler context
		2273
		2274	To implement race-free queueing, you simply add to the queue in the signal
		2275	handler but you block the signal handler in the watcher callback. Here is an example that does that for
		2276	some fictitious SIGUSR1 handler:
		2277
		2278	static ev_async mysig;
		2279
		2280	static void
		2281	sigusr1_handler (void)
		2282	{
		2283	sometype data;
		2284
		2285	// no locking etc.
		2286	queue_put (data);
		2287	ev_async_send (EV_DEFAULT_ &mysig);
		2288	}
		2289
		2290	static void
		2291	mysig_cb (EV_P_ ev_async *w, int revents)
		2292	{
		2293	sometype data;
		2294	sigset_t block, prev;
		2295
		2296	sigemptyset (&block);
		2297	sigaddset (&block, SIGUSR1);
		2298	sigprocmask (SIG_BLOCK, &block, &prev);
		2299
		2300	while (queue_get (&data))
		2301	process (data);
		2302
		2303	if (sigismember (&prev, SIGUSR1)
		2304	sigprocmask (SIG_UNBLOCK, &block, 0);
		2305	}
		2306
		2307	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2308	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2309	either...).
		2310
		2311	=item queueing from a thread context
		2312
		2313	The strategy for threads is different, as you cannot (easily) block
		2314	threads but you can easily preempt them, so to queue safely you need to
		2315	employ a traditional mutex lock, such as in this pthread example:
		2316
		2317	static ev_async mysig;
		2318	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2319
		2320	static void
		2321	otherthread (void)
		2322	{
		2323	// only need to lock the actual queueing operation
		2324	pthread_mutex_lock (&mymutex);
		2325	queue_put (data);
		2326	pthread_mutex_unlock (&mymutex);
		2327
		2328	ev_async_send (EV_DEFAULT_ &mysig);
		2329	}
		2330
		2331	static void
		2332	mysig_cb (EV_P_ ev_async *w, int revents)
		2333	{
		2334	pthread_mutex_lock (&mymutex);
		2335
		2336	while (queue_get (&data))
		2337	process (data);
		2338
		2339	pthread_mutex_unlock (&mymutex);
		2340	}
		2341
		2342	=back
		2343
		2344
		2345	=head3 Watcher-Specific Functions and Data Members
		2346
		2347	=over 4
		2348
		2349	=item ev_async_init (ev_async *, callback)
		2350
		2351	Initialises and configures the async watcher - it has no parameters of any
		2352	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,
		2353	believe me.
		2354
		2355	=item ev_async_send (loop, ev_async *)
		2356
		2357	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2358	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2359	C<ev_feed_event>, this call is safe to do in other threads, signal or
		2360	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
		2361	section below on what exactly this means).
		2362
		2363	This call incurs the overhead of a system call only once per loop iteration,
		2364	so while the overhead might be noticeable, it doesn't apply to repeated
		2365	calls to C<ev_async_send>.
		2366
		2367	=item bool = ev_async_pending (ev_async *)
		2368
		2369	Returns a non-zero value when C<ev_async_send> has been called on the
		2370	watcher but the event has not yet been processed (or even noted) by the
		2371	event loop.
		2372
		2373	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2374	the loop iterates next and checks for the watcher to have become active,
		2375	it will reset the flag again. C<ev_async_pending> can be used to very
		2376	quickly check whether invoking the loop might be a good idea.
		2377
		2378	Not that this does I<not> check whether the watcher itself is pending, only
		2379	whether it has been requested to make this watcher pending.
1581		2380
1582	=back	2381	=back
1583		2382
1584		2383
1585	=head1 OTHER FUNCTIONS	2384	=head1 OTHER FUNCTIONS
…		…
1596	or timeout without having to allocate/configure/start/stop/free one or	2395	or timeout without having to allocate/configure/start/stop/free one or
1597	more watchers yourself.	2396	more watchers yourself.
1598		2397
1599	If C<fd> is less than 0, then no I/O watcher will be started and events	2398	If C<fd> is less than 0, then no I/O watcher will be started and events
1600	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2399	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and
1601	C<events> set will be craeted and started.	2400	C<events> set will be created and started.
1602		2401
1603	If C<timeout> is less than 0, then no timeout watcher will be	2402	If C<timeout> is less than 0, then no timeout watcher will be
1604	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2403	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
1605	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2404	repeat = 0) will be started. While C<0> is a valid timeout, it is of
1606	dubious value.	2405	dubious value.
…		…
1608	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	2407	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
1609	passed an C<revents> set like normal event callbacks (a combination of	2408	passed an C<revents> set like normal event callbacks (a combination of
1610	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2409	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
1611	value passed to C<ev_once>:	2410	value passed to C<ev_once>:
1612		2411
1613	static void stdin_ready (int revents, void *arg)	2412	static void stdin_ready (int revents, void *arg)
1614	{	2413	{
1615	if (revents & EV_TIMEOUT)	2414	if (revents & EV_TIMEOUT)
1616	/* doh, nothing entered */;	2415	/* doh, nothing entered */;
1617	else if (revents & EV_READ)	2416	else if (revents & EV_READ)
1618	/* stdin might have data for us, joy! */;	2417	/* stdin might have data for us, joy! */;
1619	}	2418	}
1620		2419
1621	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2420	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
1622		2421
1623	=item ev_feed_event (ev_loop *, watcher *, int revents)	2422	=item ev_feed_event (ev_loop *, watcher *, int revents)
1624		2423
1625	Feeds the given event set into the event loop, as if the specified event	2424	Feeds the given event set into the event loop, as if the specified event
1626	had happened for the specified watcher (which must be a pointer to an	2425	had happened for the specified watcher (which must be a pointer to an
…		…
1631	Feed an event on the given fd, as if a file descriptor backend detected	2430	Feed an event on the given fd, as if a file descriptor backend detected
1632	the given events it.	2431	the given events it.
1633		2432
1634	=item ev_feed_signal_event (ev_loop *loop, int signum)	2433	=item ev_feed_signal_event (ev_loop *loop, int signum)
1635		2434
1636	Feed an event as if the given signal occured (C<loop> must be the default	2435	Feed an event as if the given signal occurred (C<loop> must be the default
1637	loop!).	2436	loop!).
1638		2437
1639	=back	2438	=back
1640		2439
1641		2440
…		…
1657		2456
1658	=item * Priorities are not currently supported. Initialising priorities	2457	=item * Priorities are not currently supported. Initialising priorities
1659	will fail and all watchers will have the same priority, even though there	2458	will fail and all watchers will have the same priority, even though there
1660	is an ev_pri field.	2459	is an ev_pri field.
1661		2460
		2461	=item * In libevent, the last base created gets the signals, in libev, the
		2462	first base created (== the default loop) gets the signals.
		2463
1662	=item * Other members are not supported.	2464	=item * Other members are not supported.
1663		2465
1664	=item * The libev emulation is I<not> ABI compatible to libevent, you need	2466	=item * The libev emulation is I<not> ABI compatible to libevent, you need
1665	to use the libev header file and library.	2467	to use the libev header file and library.
1666		2468
1667	=back	2469	=back
1668		2470
1669	=head1 C++ SUPPORT	2471	=head1 C++ SUPPORT
1670		2472
1671	Libev comes with some simplistic wrapper classes for C++ that mainly allow	2473	Libev comes with some simplistic wrapper classes for C++ that mainly allow
1672	you to use some convinience methods to start/stop watchers and also change	2474	you to use some convenience methods to start/stop watchers and also change
1673	the callback model to a model using method callbacks on objects.	2475	the callback model to a model using method callbacks on objects.
1674		2476
1675	To use it,	2477	To use it,
1676		2478
1677	#include <ev++.h>	2479	#include <ev++.h>
1678		2480
1679	(it is not installed by default). This automatically includes F<ev.h>	2481	This automatically includes F<ev.h> and puts all of its definitions (many
1680	and puts all of its definitions (many of them macros) into the global	2482	of them macros) into the global namespace. All C++ specific things are
1681	namespace. All C++ specific things are put into the C<ev> namespace.	2483	put into the C<ev> namespace. It should support all the same embedding
		2484	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
1682		2485
1683	It should support all the same embedding options as F<ev.h>, most notably	2486	Care has been taken to keep the overhead low. The only data member the C++
1684	C<EV_MULTIPLICITY>.	2487	classes add (compared to plain C-style watchers) is the event loop pointer
		2488	that the watcher is associated with (or no additional members at all if
		2489	you disable C<EV_MULTIPLICITY> when embedding libev).
		2490
		2491	Currently, functions, and static and non-static member functions can be
		2492	used as callbacks. Other types should be easy to add as long as they only
		2493	need one additional pointer for context. If you need support for other
		2494	types of functors please contact the author (preferably after implementing
		2495	it).
1685		2496
1686	Here is a list of things available in the C<ev> namespace:	2497	Here is a list of things available in the C<ev> namespace:
1687		2498
1688	=over 4	2499	=over 4
1689		2500
…		…
1705		2516
1706	All of those classes have these methods:	2517	All of those classes have these methods:
1707		2518
1708	=over 4	2519	=over 4
1709		2520
1710	=item ev::TYPE::TYPE (object *, object::method *)	2521	=item ev::TYPE::TYPE ()
1711		2522
1712	=item ev::TYPE::TYPE (object *, object::method *, struct ev_loop *)	2523	=item ev::TYPE::TYPE (struct ev_loop *)
1713		2524
1714	=item ev::TYPE::~TYPE	2525	=item ev::TYPE::~TYPE
1715		2526
1716	The constructor takes a pointer to an object and a method pointer to	2527	The constructor (optionally) takes an event loop to associate the watcher
1717	the event handler callback to call in this class. The constructor calls	2528	with. If it is omitted, it will use C<EV_DEFAULT>.
1718	C<ev_init> for you, which means you have to call the C<set> method	2529
1719	before starting it. If you do not specify a loop then the constructor	2530	The constructor calls C<ev_init> for you, which means you have to call the
1720	automatically associates the default loop with this watcher.	2531	C<set> method before starting it.
		2532
		2533	It will not set a callback, however: You have to call the templated C<set>
		2534	method to set a callback before you can start the watcher.
		2535
		2536	(The reason why you have to use a method is a limitation in C++ which does
		2537	not allow explicit template arguments for constructors).
1721		2538
1722	The destructor automatically stops the watcher if it is active.	2539	The destructor automatically stops the watcher if it is active.
		2540
		2541	=item w->set<class, &class::method> (object *)
		2542
		2543	This method sets the callback method to call. The method has to have a
		2544	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		2545	first argument and the C<revents> as second. The object must be given as
		2546	parameter and is stored in the C<data> member of the watcher.
		2547
		2548	This method synthesizes efficient thunking code to call your method from
		2549	the C callback that libev requires. If your compiler can inline your
		2550	callback (i.e. it is visible to it at the place of the C<set> call and
		2551	your compiler is good :), then the method will be fully inlined into the
		2552	thunking function, making it as fast as a direct C callback.
		2553
		2554	Example: simple class declaration and watcher initialisation
		2555
		2556	struct myclass
		2557	{
		2558	void io_cb (ev::io &w, int revents) { }
		2559	}
		2560
		2561	myclass obj;
		2562	ev::io iow;
		2563	iow.set <myclass, &myclass::io_cb> (&obj);
		2564
		2565	=item w->set<function> (void *data = 0)
		2566
		2567	Also sets a callback, but uses a static method or plain function as
		2568	callback. The optional C<data> argument will be stored in the watcher's
		2569	C<data> member and is free for you to use.
		2570
		2571	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		2572
		2573	See the method-C<set> above for more details.
		2574
		2575	Example:
		2576
		2577	static void io_cb (ev::io &w, int revents) { }
		2578	iow.set <io_cb> ();
1723		2579
1724	=item w->set (struct ev_loop *)	2580	=item w->set (struct ev_loop *)
1725		2581
1726	Associates a different C<struct ev_loop> with this watcher. You can only	2582	Associates a different C<struct ev_loop> with this watcher. You can only
1727	do this when the watcher is inactive (and not pending either).	2583	do this when the watcher is inactive (and not pending either).
1728		2584
1729	=item w->set ([args])	2585	=item w->set ([arguments])
1730		2586
1731	Basically the same as C<ev_TYPE_set>, with the same args. Must be	2587	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be
1732	called at least once. Unlike the C counterpart, an active watcher gets	2588	called at least once. Unlike the C counterpart, an active watcher gets
1733	automatically stopped and restarted.	2589	automatically stopped and restarted when reconfiguring it with this
		2590	method.
1734		2591
1735	=item w->start ()	2592	=item w->start ()
1736		2593
1737	Starts the watcher. Note that there is no C<loop> argument as the	2594	Starts the watcher. Note that there is no C<loop> argument, as the
1738	constructor already takes the loop.	2595	constructor already stores the event loop.
1739		2596
1740	=item w->stop ()	2597	=item w->stop ()
1741		2598
1742	Stops the watcher if it is active. Again, no C<loop> argument.	2599	Stops the watcher if it is active. Again, no C<loop> argument.
1743		2600
1744	=item w->again () C<ev::timer>, C<ev::periodic> only	2601	=item w->again () (C<ev::timer>, C<ev::periodic> only)
1745		2602
1746	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding	2603	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
1747	C<ev_TYPE_again> function.	2604	C<ev_TYPE_again> function.
1748		2605
1749	=item w->sweep () C<ev::embed> only	2606	=item w->sweep () (C<ev::embed> only)
1750		2607
1751	Invokes C<ev_embed_sweep>.	2608	Invokes C<ev_embed_sweep>.
1752		2609
1753	=item w->update () C<ev::stat> only	2610	=item w->update () (C<ev::stat> only)
1754		2611
1755	Invokes C<ev_stat_stat>.	2612	Invokes C<ev_stat_stat>.
1756		2613
1757	=back	2614	=back
1758		2615
1759	=back	2616	=back
1760		2617
1761	Example: Define a class with an IO and idle watcher, start one of them in	2618	Example: Define a class with an IO and idle watcher, start one of them in
1762	the constructor.	2619	the constructor.
1763		2620
1764	class myclass	2621	class myclass
1765	{	2622	{
1766	ev_io io; void io_cb (ev::io &w, int revents);	2623	ev::io io; void io_cb (ev::io &w, int revents);
1767	ev_idle idle void idle_cb (ev::idle &w, int revents);	2624	ev:idle idle void idle_cb (ev::idle &w, int revents);
1768		2625
1769	myclass ();	2626	myclass (int fd)
1770	}	2627	{
		2628	io .set <myclass, &myclass::io_cb > (this);
		2629	idle.set <myclass, &myclass::idle_cb> (this);
1771		2630
1772	myclass::myclass (int fd)
1773	: io (this, &myclass::io_cb),
1774	idle (this, &myclass::idle_cb)
1775	{
1776	io.start (fd, ev::READ);	2631	io.start (fd, ev::READ);
		2632	}
1777	}	2633	};
		2634
		2635
		2636	=head1 OTHER LANGUAGE BINDINGS
		2637
		2638	Libev does not offer other language bindings itself, but bindings for a
		2639	number of languages exist in the form of third-party packages. If you know
		2640	any interesting language binding in addition to the ones listed here, drop
		2641	me a note.
		2642
		2643	=over 4
		2644
		2645	=item Perl
		2646
		2647	The EV module implements the full libev API and is actually used to test
		2648	libev. EV is developed together with libev. Apart from the EV core module,
		2649	there are additional modules that implement libev-compatible interfaces
		2650	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the
		2651	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).
		2652
		2653	It can be found and installed via CPAN, its homepage is at
		2654	L<http://software.schmorp.de/pkg/EV>.
		2655
		2656	=item Python
		2657
		2658	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		2659	seems to be quite complete and well-documented. Note, however, that the
		2660	patch they require for libev is outright dangerous as it breaks the ABI
		2661	for everybody else, and therefore, should never be applied in an installed
		2662	libev (if python requires an incompatible ABI then it needs to embed
		2663	libev).
		2664
		2665	=item Ruby
		2666
		2667	Tony Arcieri has written a ruby extension that offers access to a subset
		2668	of the libev API and adds file handle abstractions, asynchronous DNS and
		2669	more on top of it. It can be found via gem servers. Its homepage is at
		2670	L<http://rev.rubyforge.org/>.
		2671
		2672	=item D
		2673
		2674	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		2675	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		2676
		2677	=back
1778		2678
1779		2679
1780	=head1 MACRO MAGIC	2680	=head1 MACRO MAGIC
1781		2681
1782	Libev can be compiled with a variety of options, the most fundemantal is	2682	Libev can be compiled with a variety of options, the most fundamental
1783	C<EV_MULTIPLICITY>. This option determines wether (most) functions and	2683	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
1784	callbacks have an initial C<struct ev_loop *> argument.	2684	functions and callbacks have an initial C<struct ev_loop *> argument.
1785		2685
1786	To make it easier to write programs that cope with either variant, the	2686	To make it easier to write programs that cope with either variant, the
1787	following macros are defined:	2687	following macros are defined:
1788		2688
1789	=over 4	2689	=over 4
…		…
1792		2692
1793	This provides the loop I<argument> for functions, if one is required ("ev	2693	This provides the loop I<argument> for functions, if one is required ("ev
1794	loop argument"). The C<EV_A> form is used when this is the sole argument,	2694	loop argument"). The C<EV_A> form is used when this is the sole argument,
1795	C<EV_A_> is used when other arguments are following. Example:	2695	C<EV_A_> is used when other arguments are following. Example:
1796		2696
1797	ev_unref (EV_A);	2697	ev_unref (EV_A);
1798	ev_timer_add (EV_A_ watcher);	2698	ev_timer_add (EV_A_ watcher);
1799	ev_loop (EV_A_ 0);	2699	ev_loop (EV_A_ 0);
1800		2700
1801	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,	2701	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
1802	which is often provided by the following macro.	2702	which is often provided by the following macro.
1803		2703
1804	=item C<EV_P>, C<EV_P_>	2704	=item C<EV_P>, C<EV_P_>
1805		2705
1806	This provides the loop I<parameter> for functions, if one is required ("ev	2706	This provides the loop I<parameter> for functions, if one is required ("ev
1807	loop parameter"). The C<EV_P> form is used when this is the sole parameter,	2707	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
1808	C<EV_P_> is used when other parameters are following. Example:	2708	C<EV_P_> is used when other parameters are following. Example:
1809		2709
1810	// this is how ev_unref is being declared	2710	// this is how ev_unref is being declared
1811	static void ev_unref (EV_P);	2711	static void ev_unref (EV_P);
1812		2712
1813	// this is how you can declare your typical callback	2713	// this is how you can declare your typical callback
1814	static void cb (EV_P_ ev_timer *w, int revents)	2714	static void cb (EV_P_ ev_timer *w, int revents)
1815		2715
1816	It declares a parameter C<loop> of type C<struct ev_loop *>, quite	2716	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
1817	suitable for use with C<EV_A>.	2717	suitable for use with C<EV_A>.
1818		2718
1819	=item C<EV_DEFAULT>, C<EV_DEFAULT_>	2719	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
1820		2720
1821	Similar to the other two macros, this gives you the value of the default	2721	Similar to the other two macros, this gives you the value of the default
1822	loop, if multiple loops are supported ("ev loop default").	2722	loop, if multiple loops are supported ("ev loop default").
1823		2723
		2724	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		2725
		2726	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		2727	default loop has been initialised (C<UC> == unchecked). Their behaviour
		2728	is undefined when the default loop has not been initialised by a previous
		2729	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		2730
		2731	It is often prudent to use C<EV_DEFAULT> when initialising the first
		2732	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
		2733
1824	=back	2734	=back
1825		2735
1826	Example: Declare and initialise a check watcher, working regardless of	2736	Example: Declare and initialise a check watcher, utilising the above
1827	wether multiple loops are supported or not.	2737	macros so it will work regardless of whether multiple loops are supported
		2738	or not.
1828		2739
1829	static void	2740	static void
1830	check_cb (EV_P_ ev_timer *w, int revents)	2741	check_cb (EV_P_ ev_timer *w, int revents)
1831	{	2742	{
1832	ev_check_stop (EV_A_ w);	2743	ev_check_stop (EV_A_ w);
1833	}	2744	}
1834		2745
1835	ev_check check;	2746	ev_check check;
1836	ev_check_init (&check, check_cb);	2747	ev_check_init (&check, check_cb);
1837	ev_check_start (EV_DEFAULT_ &check);	2748	ev_check_start (EV_DEFAULT_ &check);
1838	ev_loop (EV_DEFAULT_ 0);	2749	ev_loop (EV_DEFAULT_ 0);
1839
1840		2750
1841	=head1 EMBEDDING	2751	=head1 EMBEDDING
1842		2752
1843	Libev can (and often is) directly embedded into host	2753	Libev can (and often is) directly embedded into host
1844	applications. Examples of applications that embed it include the Deliantra	2754	applications. Examples of applications that embed it include the Deliantra
1845	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)	2755	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
1846	and rxvt-unicode.	2756	and rxvt-unicode.
1847		2757
1848	The goal is to enable you to just copy the neecssary files into your	2758	The goal is to enable you to just copy the necessary files into your
1849	source directory without having to change even a single line in them, so	2759	source directory without having to change even a single line in them, so
1850	you can easily upgrade by simply copying (or having a checked-out copy of	2760	you can easily upgrade by simply copying (or having a checked-out copy of
1851	libev somewhere in your source tree).	2761	libev somewhere in your source tree).
1852		2762
1853	=head2 FILESETS	2763	=head2 FILESETS
1854		2764
1855	Depending on what features you need you need to include one or more sets of files	2765	Depending on what features you need you need to include one or more sets of files
1856	in your app.	2766	in your application.
1857		2767
1858	=head3 CORE EVENT LOOP	2768	=head3 CORE EVENT LOOP
1859		2769
1860	To include only the libev core (all the C<ev_*> functions), with manual	2770	To include only the libev core (all the C<ev_*> functions), with manual
1861	configuration (no autoconf):	2771	configuration (no autoconf):
1862		2772
1863	#define EV_STANDALONE 1	2773	#define EV_STANDALONE 1
1864	#include "ev.c"	2774	#include "ev.c"
1865		2775
1866	This will automatically include F<ev.h>, too, and should be done in a	2776	This will automatically include F<ev.h>, too, and should be done in a
1867	single C source file only to provide the function implementations. To use	2777	single C source file only to provide the function implementations. To use
1868	it, do the same for F<ev.h> in all files wishing to use this API (best	2778	it, do the same for F<ev.h> in all files wishing to use this API (best
1869	done by writing a wrapper around F<ev.h> that you can include instead and	2779	done by writing a wrapper around F<ev.h> that you can include instead and
1870	where you can put other configuration options):	2780	where you can put other configuration options):
1871		2781
1872	#define EV_STANDALONE 1	2782	#define EV_STANDALONE 1
1873	#include "ev.h"	2783	#include "ev.h"
1874		2784
1875	Both header files and implementation files can be compiled with a C++	2785	Both header files and implementation files can be compiled with a C++
1876	compiler (at least, thats a stated goal, and breakage will be treated	2786	compiler (at least, thats a stated goal, and breakage will be treated
1877	as a bug).	2787	as a bug).
1878		2788
1879	You need the following files in your source tree, or in a directory	2789	You need the following files in your source tree, or in a directory
1880	in your include path (e.g. in libev/ when using -Ilibev):	2790	in your include path (e.g. in libev/ when using -Ilibev):
1881		2791
1882	ev.h	2792	ev.h
1883	ev.c	2793	ev.c
1884	ev_vars.h	2794	ev_vars.h
1885	ev_wrap.h	2795	ev_wrap.h
1886		2796
1887	ev_win32.c required on win32 platforms only	2797	ev_win32.c required on win32 platforms only
1888		2798
1889	ev_select.c only when select backend is enabled (which is by default)	2799	ev_select.c only when select backend is enabled (which is enabled by default)
1890	ev_poll.c only when poll backend is enabled (disabled by default)	2800	ev_poll.c only when poll backend is enabled (disabled by default)
1891	ev_epoll.c only when the epoll backend is enabled (disabled by default)	2801	ev_epoll.c only when the epoll backend is enabled (disabled by default)
1892	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)	2802	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
1893	ev_port.c only when the solaris port backend is enabled (disabled by default)	2803	ev_port.c only when the solaris port backend is enabled (disabled by default)
1894		2804
1895	F<ev.c> includes the backend files directly when enabled, so you only need	2805	F<ev.c> includes the backend files directly when enabled, so you only need
1896	to compile this single file.	2806	to compile this single file.
1897		2807
1898	=head3 LIBEVENT COMPATIBILITY API	2808	=head3 LIBEVENT COMPATIBILITY API
1899		2809
1900	To include the libevent compatibility API, also include:	2810	To include the libevent compatibility API, also include:
1901		2811
1902	#include "event.c"	2812	#include "event.c"
1903		2813
1904	in the file including F<ev.c>, and:	2814	in the file including F<ev.c>, and:
1905		2815
1906	#include "event.h"	2816	#include "event.h"
1907		2817
1908	in the files that want to use the libevent API. This also includes F<ev.h>.	2818	in the files that want to use the libevent API. This also includes F<ev.h>.
1909		2819
1910	You need the following additional files for this:	2820	You need the following additional files for this:
1911		2821
1912	event.h	2822	event.h
1913	event.c	2823	event.c
1914		2824
1915	=head3 AUTOCONF SUPPORT	2825	=head3 AUTOCONF SUPPORT
1916		2826
1917	Instead of using C<EV_STANDALONE=1> and providing your config in	2827	Instead of using C<EV_STANDALONE=1> and providing your configuration in
1918	whatever way you want, you can also C<m4_include([libev.m4])> in your	2828	whatever way you want, you can also C<m4_include([libev.m4])> in your
1919	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then	2829	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
1920	include F<config.h> and configure itself accordingly.	2830	include F<config.h> and configure itself accordingly.
1921		2831
1922	For this of course you need the m4 file:	2832	For this of course you need the m4 file:
1923		2833
1924	libev.m4	2834	libev.m4
1925		2835
1926	=head2 PREPROCESSOR SYMBOLS/MACROS	2836	=head2 PREPROCESSOR SYMBOLS/MACROS
1927		2837
1928	Libev can be configured via a variety of preprocessor symbols you have to define	2838	Libev can be configured via a variety of preprocessor symbols you have to
1929	before including any of its files. The default is not to build for multiplicity	2839	define before including any of its files. The default in the absence of
1930	and only include the select backend.	2840	autoconf is noted for every option.
1931		2841
1932	=over 4	2842	=over 4
1933		2843
1934	=item EV_STANDALONE	2844	=item EV_STANDALONE
1935		2845
…		…
1940	F<event.h> that are not directly supported by the libev core alone.	2850	F<event.h> that are not directly supported by the libev core alone.
1941		2851
1942	=item EV_USE_MONOTONIC	2852	=item EV_USE_MONOTONIC
1943		2853
1944	If defined to be C<1>, libev will try to detect the availability of the	2854	If defined to be C<1>, libev will try to detect the availability of the
1945	monotonic clock option at both compiletime and runtime. Otherwise no use	2855	monotonic clock option at both compile time and runtime. Otherwise no use
1946	of the monotonic clock option will be attempted. If you enable this, you	2856	of the monotonic clock option will be attempted. If you enable this, you
1947	usually have to link against librt or something similar. Enabling it when	2857	usually have to link against librt or something similar. Enabling it when
1948	the functionality isn't available is safe, though, althoguh you have	2858	the functionality isn't available is safe, though, although you have
1949	to make sure you link against any libraries where the C<clock_gettime>	2859	to make sure you link against any libraries where the C<clock_gettime>
1950	function is hiding in (often F<-lrt>).	2860	function is hiding in (often F<-lrt>).
1951		2861
1952	=item EV_USE_REALTIME	2862	=item EV_USE_REALTIME
1953		2863
1954	If defined to be C<1>, libev will try to detect the availability of the	2864	If defined to be C<1>, libev will try to detect the availability of the
1955	realtime clock option at compiletime (and assume its availability at	2865	real-time clock option at compile time (and assume its availability at
1956	runtime if successful). Otherwise no use of the realtime clock option will	2866	runtime if successful). Otherwise no use of the real-time clock option will
1957	be attempted. This effectively replaces C<gettimeofday> by C<clock_get	2867	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
1958	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See tzhe note about libraries	2868	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
1959	in the description of C<EV_USE_MONOTONIC>, though.	2869	note about libraries in the description of C<EV_USE_MONOTONIC>, though.
		2870
		2871	=item EV_USE_NANOSLEEP
		2872
		2873	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		2874	and will use it for delays. Otherwise it will use C<select ()>.
		2875
		2876	=item EV_USE_EVENTFD
		2877
		2878	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		2879	available and will probe for kernel support at runtime. This will improve
		2880	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		2881	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		2882	2.7 or newer, otherwise disabled.
1960		2883
1961	=item EV_USE_SELECT	2884	=item EV_USE_SELECT
1962		2885
1963	If undefined or defined to be C<1>, libev will compile in support for the	2886	If undefined or defined to be C<1>, libev will compile in support for the
1964	C<select>(2) backend. No attempt at autodetection will be done: if no	2887	C<select>(2) backend. No attempt at auto-detection will be done: if no
1965	other method takes over, select will be it. Otherwise the select backend	2888	other method takes over, select will be it. Otherwise the select backend
1966	will not be compiled in.	2889	will not be compiled in.
1967		2890
1968	=item EV_SELECT_USE_FD_SET	2891	=item EV_SELECT_USE_FD_SET
1969		2892
1970	If defined to C<1>, then the select backend will use the system C<fd_set>	2893	If defined to C<1>, then the select backend will use the system C<fd_set>
1971	structure. This is useful if libev doesn't compile due to a missing	2894	structure. This is useful if libev doesn't compile due to a missing
1972	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on	2895	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout on
1973	exotic systems. This usually limits the range of file descriptors to some	2896	exotic systems. This usually limits the range of file descriptors to some
1974	low limit such as 1024 or might have other limitations (winsocket only	2897	low limit such as 1024 or might have other limitations (winsocket only
1975	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might	2898	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
1976	influence the size of the C<fd_set> used.	2899	influence the size of the C<fd_set> used.
1977		2900
…		…
1983	be used is the winsock select). This means that it will call	2906	be used is the winsock select). This means that it will call
1984	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	2907	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
1985	it is assumed that all these functions actually work on fds, even	2908	it is assumed that all these functions actually work on fds, even
1986	on win32. Should not be defined on non-win32 platforms.	2909	on win32. Should not be defined on non-win32 platforms.
1987		2910
		2911	=item EV_FD_TO_WIN32_HANDLE
		2912
		2913	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		2914	file descriptors to socket handles. When not defining this symbol (the
		2915	default), then libev will call C<_get_osfhandle>, which is usually
		2916	correct. In some cases, programs use their own file descriptor management,
		2917	in which case they can provide this function to map fds to socket handles.
		2918
1988	=item EV_USE_POLL	2919	=item EV_USE_POLL
1989		2920
1990	If defined to be C<1>, libev will compile in support for the C<poll>(2)	2921	If defined to be C<1>, libev will compile in support for the C<poll>(2)
1991	backend. Otherwise it will be enabled on non-win32 platforms. It	2922	backend. Otherwise it will be enabled on non-win32 platforms. It
1992	takes precedence over select.	2923	takes precedence over select.
1993		2924
1994	=item EV_USE_EPOLL	2925	=item EV_USE_EPOLL
1995		2926
1996	If defined to be C<1>, libev will compile in support for the Linux	2927	If defined to be C<1>, libev will compile in support for the Linux
1997	C<epoll>(7) backend. Its availability will be detected at runtime,	2928	C<epoll>(7) backend. Its availability will be detected at runtime,
1998	otherwise another method will be used as fallback. This is the	2929	otherwise another method will be used as fallback. This is the preferred
1999	preferred backend for GNU/Linux systems.	2930	backend for GNU/Linux systems. If undefined, it will be enabled if the
		2931	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
2000		2932
2001	=item EV_USE_KQUEUE	2933	=item EV_USE_KQUEUE
2002		2934
2003	If defined to be C<1>, libev will compile in support for the BSD style	2935	If defined to be C<1>, libev will compile in support for the BSD style
2004	C<kqueue>(2) backend. Its actual availability will be detected at runtime,	2936	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
…		…
2017	otherwise another method will be used as fallback. This is the preferred	2949	otherwise another method will be used as fallback. This is the preferred
2018	backend for Solaris 10 systems.	2950	backend for Solaris 10 systems.
2019		2951
2020	=item EV_USE_DEVPOLL	2952	=item EV_USE_DEVPOLL
2021		2953
2022	reserved for future expansion, works like the USE symbols above.	2954	Reserved for future expansion, works like the USE symbols above.
2023		2955
2024	=item EV_USE_INOTIFY	2956	=item EV_USE_INOTIFY
2025		2957
2026	If defined to be C<1>, libev will compile in support for the Linux inotify	2958	If defined to be C<1>, libev will compile in support for the Linux inotify
2027	interface to speed up C<ev_stat> watchers. Its actual availability will	2959	interface to speed up C<ev_stat> watchers. Its actual availability will
2028	be detected at runtime.	2960	be detected at runtime. If undefined, it will be enabled if the headers
		2961	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		2962
		2963	=item EV_ATOMIC_T
		2964
		2965	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		2966	access is atomic with respect to other threads or signal contexts. No such
		2967	type is easily found in the C language, so you can provide your own type
		2968	that you know is safe for your purposes. It is used both for signal handler "locking"
		2969	as well as for signal and thread safety in C<ev_async> watchers.
		2970
		2971	In the absence of this define, libev will use C<sig_atomic_t volatile>
		2972	(from F<signal.h>), which is usually good enough on most platforms.
2029		2973
2030	=item EV_H	2974	=item EV_H
2031		2975
2032	The name of the F<ev.h> header file used to include it. The default if	2976	The name of the F<ev.h> header file used to include it. The default if
2033	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This	2977	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2034	can be used to virtually rename the F<ev.h> header file in case of conflicts.	2978	used to virtually rename the F<ev.h> header file in case of conflicts.
2035		2979
2036	=item EV_CONFIG_H	2980	=item EV_CONFIG_H
2037		2981
2038	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	2982	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2039	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	2983	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2040	C<EV_H>, above.	2984	C<EV_H>, above.
2041		2985
2042	=item EV_EVENT_H	2986	=item EV_EVENT_H
2043		2987
2044	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	2988	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2045	of how the F<event.h> header can be found.	2989	of how the F<event.h> header can be found, the default is C<"event.h">.
2046		2990
2047	=item EV_PROTOTYPES	2991	=item EV_PROTOTYPES
2048		2992
2049	If defined to be C<0>, then F<ev.h> will not define any function	2993	If defined to be C<0>, then F<ev.h> will not define any function
2050	prototypes, but still define all the structs and other symbols. This is	2994	prototypes, but still define all the structs and other symbols. This is
…		…
2057	will have the C<struct ev_loop *> as first argument, and you can create	3001	will have the C<struct ev_loop *> as first argument, and you can create
2058	additional independent event loops. Otherwise there will be no support	3002	additional independent event loops. Otherwise there will be no support
2059	for multiple event loops and there is no first event loop pointer	3003	for multiple event loops and there is no first event loop pointer
2060	argument. Instead, all functions act on the single default loop.	3004	argument. Instead, all functions act on the single default loop.
2061		3005
		3006	=item EV_MINPRI
		3007
		3008	=item EV_MAXPRI
		3009
		3010	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		3011	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		3012	provide for more priorities by overriding those symbols (usually defined
		3013	to be C<-2> and C<2>, respectively).
		3014
		3015	When doing priority-based operations, libev usually has to linearly search
		3016	all the priorities, so having many of them (hundreds) uses a lot of space
		3017	and time, so using the defaults of five priorities (-2 .. +2) is usually
		3018	fine.
		3019
		3020	If your embedding application does not need any priorities, defining these both to
		3021	C<0> will save some memory and CPU.
		3022
2062	=item EV_PERIODIC_ENABLE	3023	=item EV_PERIODIC_ENABLE
2063		3024
2064	If undefined or defined to be C<1>, then periodic timers are supported. If	3025	If undefined or defined to be C<1>, then periodic timers are supported. If
2065	defined to be C<0>, then they are not. Disabling them saves a few kB of	3026	defined to be C<0>, then they are not. Disabling them saves a few kB of
2066	code.	3027	code.
2067		3028
		3029	=item EV_IDLE_ENABLE
		3030
		3031	If undefined or defined to be C<1>, then idle watchers are supported. If
		3032	defined to be C<0>, then they are not. Disabling them saves a few kB of
		3033	code.
		3034
2068	=item EV_EMBED_ENABLE	3035	=item EV_EMBED_ENABLE
2069		3036
2070	If undefined or defined to be C<1>, then embed watchers are supported. If	3037	If undefined or defined to be C<1>, then embed watchers are supported. If
2071	defined to be C<0>, then they are not.	3038	defined to be C<0>, then they are not.
2072		3039
…		…
2078	=item EV_FORK_ENABLE	3045	=item EV_FORK_ENABLE
2079		3046
2080	If undefined or defined to be C<1>, then fork watchers are supported. If	3047	If undefined or defined to be C<1>, then fork watchers are supported. If
2081	defined to be C<0>, then they are not.	3048	defined to be C<0>, then they are not.
2082		3049
		3050	=item EV_ASYNC_ENABLE
		3051
		3052	If undefined or defined to be C<1>, then async watchers are supported. If
		3053	defined to be C<0>, then they are not.
		3054
2083	=item EV_MINIMAL	3055	=item EV_MINIMAL
2084		3056
2085	If you need to shave off some kilobytes of code at the expense of some	3057	If you need to shave off some kilobytes of code at the expense of some
2086	speed, define this symbol to C<1>. Currently only used for gcc to override	3058	speed, define this symbol to C<1>. Currently this is used to override some
2087	some inlining decisions, saves roughly 30% codesize of amd64.	3059	inlining decisions, saves roughly 30% code size on amd64. It also selects a
		3060	much smaller 2-heap for timer management over the default 4-heap.
2088		3061
2089	=item EV_PID_HASHSIZE	3062	=item EV_PID_HASHSIZE
2090		3063
2091	C<ev_child> watchers use a small hash table to distribute workload by	3064	C<ev_child> watchers use a small hash table to distribute workload by
2092	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more	3065	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
2093	than enough. If you need to manage thousands of children you might want to	3066	than enough. If you need to manage thousands of children you might want to
2094	increase this value (I<must> be a power of two).	3067	increase this value (I<must> be a power of two).
2095		3068
2096	=item EV_INOTIFY_HASHSIZE	3069	=item EV_INOTIFY_HASHSIZE
2097		3070
2098	C<ev_staz> watchers use a small hash table to distribute workload by	3071	C<ev_stat> watchers use a small hash table to distribute workload by
2099	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	3072	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
2100	usually more than enough. If you need to manage thousands of C<ev_stat>	3073	usually more than enough. If you need to manage thousands of C<ev_stat>
2101	watchers you might want to increase this value (I<must> be a power of	3074	watchers you might want to increase this value (I<must> be a power of
2102	two).	3075	two).
2103		3076
		3077	=item EV_USE_4HEAP
		3078
		3079	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3080	timer and periodics heap, libev uses a 4-heap when this symbol is defined
		3081	to C<1>. The 4-heap uses more complicated (longer) code but has
		3082	noticeably faster performance with many (thousands) of watchers.
		3083
		3084	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3085	(disabled).
		3086
		3087	=item EV_HEAP_CACHE_AT
		3088
		3089	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3090	timer and periodics heap, libev can cache the timestamp (I<at>) within
		3091	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3092	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3093	but avoids random read accesses on heap changes. This improves performance
		3094	noticeably with with many (hundreds) of watchers.
		3095
		3096	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3097	(disabled).
		3098
		3099	=item EV_VERIFY
		3100
		3101	Controls how much internal verification (see C<ev_loop_verify ()>) will
		3102	be done: If set to C<0>, no internal verification code will be compiled
		3103	in. If set to C<1>, then verification code will be compiled in, but not
		3104	called. If set to C<2>, then the internal verification code will be
		3105	called once per loop, which can slow down libev. If set to C<3>, then the
		3106	verification code will be called very frequently, which will slow down
		3107	libev considerably.
		3108
		3109	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
		3110	C<0.>
		3111
2104	=item EV_COMMON	3112	=item EV_COMMON
2105		3113
2106	By default, all watchers have a C<void *data> member. By redefining	3114	By default, all watchers have a C<void *data> member. By redefining
2107	this macro to a something else you can include more and other types of	3115	this macro to a something else you can include more and other types of
2108	members. You have to define it each time you include one of the files,	3116	members. You have to define it each time you include one of the files,
2109	though, and it must be identical each time.	3117	though, and it must be identical each time.
2110		3118
2111	For example, the perl EV module uses something like this:	3119	For example, the perl EV module uses something like this:
2112		3120
2113	#define EV_COMMON \	3121	#define EV_COMMON \
2114	SV *self; /* contains this struct */ \	3122	SV *self; /* contains this struct */ \
2115	SV *cb_sv, *fh /* note no trailing ";" */	3123	SV *cb_sv, *fh /* note no trailing ";" */
2116		3124
2117	=item EV_CB_DECLARE (type)	3125	=item EV_CB_DECLARE (type)
2118		3126
2119	=item EV_CB_INVOKE (watcher, revents)	3127	=item EV_CB_INVOKE (watcher, revents)
2120		3128
2121	=item ev_set_cb (ev, cb)	3129	=item ev_set_cb (ev, cb)
2122		3130
2123	Can be used to change the callback member declaration in each watcher,	3131	Can be used to change the callback member declaration in each watcher,
2124	and the way callbacks are invoked and set. Must expand to a struct member	3132	and the way callbacks are invoked and set. Must expand to a struct member
2125	definition and a statement, respectively. See the F<ev.v> header file for	3133	definition and a statement, respectively. See the F<ev.h> header file for
2126	their default definitions. One possible use for overriding these is to	3134	their default definitions. One possible use for overriding these is to
2127	avoid the C<struct ev_loop *> as first argument in all cases, or to use	3135	avoid the C<struct ev_loop *> as first argument in all cases, or to use
2128	method calls instead of plain function calls in C++.	3136	method calls instead of plain function calls in C++.
		3137
		3138	=head2 EXPORTED API SYMBOLS
		3139
		3140	If you need to re-export the API (e.g. via a DLL) and you need a list of
		3141	exported symbols, you can use the provided F<Symbol.*> files which list
		3142	all public symbols, one per line:
		3143
		3144	Symbols.ev for libev proper
		3145	Symbols.event for the libevent emulation
		3146
		3147	This can also be used to rename all public symbols to avoid clashes with
		3148	multiple versions of libev linked together (which is obviously bad in
		3149	itself, but sometimes it is inconvenient to avoid this).
		3150
		3151	A sed command like this will create wrapper C<#define>'s that you need to
		3152	include before including F<ev.h>:
		3153
		3154	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		3155
		3156	This would create a file F<wrap.h> which essentially looks like this:
		3157
		3158	#define ev_backend myprefix_ev_backend
		3159	#define ev_check_start myprefix_ev_check_start
		3160	#define ev_check_stop myprefix_ev_check_stop
		3161	...
2129		3162
2130	=head2 EXAMPLES	3163	=head2 EXAMPLES
2131		3164
2132	For a real-world example of a program the includes libev	3165	For a real-world example of a program the includes libev
2133	verbatim, you can have a look at the EV perl module	3166	verbatim, you can have a look at the EV perl module
…		…
2136	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file	3169	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
2137	will be compiled. It is pretty complex because it provides its own header	3170	will be compiled. It is pretty complex because it provides its own header
2138	file.	3171	file.
2139		3172
2140	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file	3173	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
2141	that everybody includes and which overrides some autoconf choices:	3174	that everybody includes and which overrides some configure choices:
2142		3175
		3176	#define EV_MINIMAL 1
2143	#define EV_USE_POLL 0	3177	#define EV_USE_POLL 0
2144	#define EV_MULTIPLICITY 0	3178	#define EV_MULTIPLICITY 0
2145	#define EV_PERIODICS 0	3179	#define EV_PERIODIC_ENABLE 0
		3180	#define EV_STAT_ENABLE 0
		3181	#define EV_FORK_ENABLE 0
2146	#define EV_CONFIG_H <config.h>	3182	#define EV_CONFIG_H <config.h>
		3183	#define EV_MINPRI 0
		3184	#define EV_MAXPRI 0
2147		3185
2148	#include "ev++.h"	3186	#include "ev++.h"
2149		3187
2150	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3188	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
2151		3189
2152	#include "ev_cpp.h"	3190	#include "ev_cpp.h"
2153	#include "ev.c"	3191	#include "ev.c"
		3192
		3193
		3194	=head1 THREADS AND COROUTINES
		3195
		3196	=head2 THREADS
		3197
		3198	Libev itself is completely thread-safe, but it uses no locking. This
		3199	means that you can use as many loops as you want in parallel, as long as
		3200	only one thread ever calls into one libev function with the same loop
		3201	parameter.
		3202
		3203	Or put differently: calls with different loop parameters can be done in
		3204	parallel from multiple threads, calls with the same loop parameter must be
		3205	done serially (but can be done from different threads, as long as only one
		3206	thread ever is inside a call at any point in time, e.g. by using a mutex
		3207	per loop).
		3208
		3209	If you want to know which design (one loop, locking, or multiple loops
		3210	without or something else still) is best for your problem, then I cannot
		3211	help you. I can give some generic advice however:
		3212
		3213	=over 4
		3214
		3215	=item * most applications have a main thread: use the default libev loop
		3216	in that thread, or create a separate thread running only the default loop.
		3217
		3218	This helps integrating other libraries or software modules that use libev
		3219	themselves and don't care/know about threading.
		3220
		3221	=item * one loop per thread is usually a good model.
		3222
		3223	Doing this is almost never wrong, sometimes a better-performance model
		3224	exists, but it is always a good start.
		3225
		3226	=item * other models exist, such as the leader/follower pattern, where one
		3227	loop is handed through multiple threads in a kind of round-robin fashion.
		3228
		3229	Choosing a model is hard - look around, learn, know that usually you can do
		3230	better than you currently do :-)
		3231
		3232	=item * often you need to talk to some other thread which blocks in the
		3233	event loop - C<ev_async> watchers can be used to wake them up from other
		3234	threads safely (or from signal contexts...).
		3235
		3236	=back
		3237
		3238	=head2 COROUTINES
		3239
		3240	Libev is much more accommodating to coroutines ("cooperative threads"):
		3241	libev fully supports nesting calls to it's functions from different
		3242	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		3243	different coroutines and switch freely between both coroutines running the
		3244	loop, as long as you don't confuse yourself). The only exception is that
		3245	you must not do this from C<ev_periodic> reschedule callbacks.
		3246
		3247	Care has been invested into making sure that libev does not keep local
		3248	state inside C<ev_loop>, and other calls do not usually allow coroutine
		3249	switches.
2154		3250
2155		3251
2156	=head1 COMPLEXITIES	3252	=head1 COMPLEXITIES
2157		3253
2158	In this section the complexities of (many of) the algorithms used inside	3254	In this section the complexities of (many of) the algorithms used inside
2159	libev will be explained. For complexity discussions about backends see the	3255	libev will be explained. For complexity discussions about backends see the
2160	documentation for C<ev_default_init>.	3256	documentation for C<ev_default_init>.
2161		3257
		3258	All of the following are about amortised time: If an array needs to be
		3259	extended, libev needs to realloc and move the whole array, but this
		3260	happens asymptotically never with higher number of elements, so O(1) might
		3261	mean it might do a lengthy realloc operation in rare cases, but on average
		3262	it is much faster and asymptotically approaches constant time.
		3263
2162	=over 4	3264	=over 4
2163		3265
2164	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3266	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2165		3267
		3268	This means that, when you have a watcher that triggers in one hour and
		3269	there are 100 watchers that would trigger before that then inserting will
		3270	have to skip roughly seven (C<ld 100>) of these watchers.
		3271
2166	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	3272	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
2167		3273
		3274	That means that changing a timer costs less than removing/adding them
		3275	as only the relative motion in the event queue has to be paid for.
		3276
2168	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	3277	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
2169		3278
		3279	These just add the watcher into an array or at the head of a list.
		3280
2170	=item Stopping check/prepare/idle watchers: O(1)	3281	=item Stopping check/prepare/idle/fork/async watchers: O(1)
2171		3282
2172	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3283	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2173		3284
		3285	These watchers are stored in lists then need to be walked to find the
		3286	correct watcher to remove. The lists are usually short (you don't usually
		3287	have many watchers waiting for the same fd or signal).
		3288
2174	=item Finding the next timer per loop iteration: O(1)	3289	=item Finding the next timer in each loop iteration: O(1)
		3290
		3291	By virtue of using a binary or 4-heap, the next timer is always found at a
		3292	fixed position in the storage array.
2175		3293
2176	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	3294	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2177		3295
2178	=item Activating one watcher: O(1)	3296	A change means an I/O watcher gets started or stopped, which requires
		3297	libev to recalculate its status (and possibly tell the kernel, depending
		3298	on backend and whether C<ev_io_set> was used).
		3299
		3300	=item Activating one watcher (putting it into the pending state): O(1)
		3301
		3302	=item Priority handling: O(number_of_priorities)
		3303
		3304	Priorities are implemented by allocating some space for each
		3305	priority. When doing priority-based operations, libev usually has to
		3306	linearly search all the priorities, but starting/stopping and activating
		3307	watchers becomes O(1) w.r.t. priority handling.
		3308
		3309	=item Sending an ev_async: O(1)
		3310
		3311	=item Processing ev_async_send: O(number_of_async_watchers)
		3312
		3313	=item Processing signals: O(max_signal_number)
		3314
		3315	Sending involves a system call I<iff> there were no other C<ev_async_send>
		3316	calls in the current loop iteration. Checking for async and signal events
		3317	involves iterating over all running async watchers or all signal numbers.
2179		3318
2180	=back	3319	=back
2181		3320
2182		3321
		3322	=head1 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		3323
		3324	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		3325	requires, and its I/O model is fundamentally incompatible with the POSIX
		3326	model. Libev still offers limited functionality on this platform in
		3327	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		3328	descriptors. This only applies when using Win32 natively, not when using
		3329	e.g. cygwin.
		3330
		3331	Lifting these limitations would basically require the full
		3332	re-implementation of the I/O system. If you are into these kinds of
		3333	things, then note that glib does exactly that for you in a very portable
		3334	way (note also that glib is the slowest event library known to man).
		3335
		3336	There is no supported compilation method available on windows except
		3337	embedding it into other applications.
		3338
		3339	Not a libev limitation but worth mentioning: windows apparently doesn't
		3340	accept large writes: instead of resulting in a partial write, windows will
		3341	either accept everything or return C<ENOBUFS> if the buffer is too large,
		3342	so make sure you only write small amounts into your sockets (less than a
		3343	megabyte seems safe, but thsi apparently depends on the amount of memory
		3344	available).
		3345
		3346	Due to the many, low, and arbitrary limits on the win32 platform and
		3347	the abysmal performance of winsockets, using a large number of sockets
		3348	is not recommended (and not reasonable). If your program needs to use
		3349	more than a hundred or so sockets, then likely it needs to use a totally
		3350	different implementation for windows, as libev offers the POSIX readiness
		3351	notification model, which cannot be implemented efficiently on windows
		3352	(Microsoft monopoly games).
		3353
		3354	A typical way to use libev under windows is to embed it (see the embedding
		3355	section for details) and use the following F<evwrap.h> header file instead
		3356	of F<ev.h>:
		3357
		3358	#define EV_STANDALONE /* keeps ev from requiring config.h */
		3359	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		3360
		3361	#include "ev.h"
		3362
		3363	And compile the following F<evwrap.c> file into your project (make sure
		3364	you do I<not> compile the F<ev.c> or any other embedded soruce files!):
		3365
		3366	#include "evwrap.h"
		3367	#include "ev.c"
		3368
		3369	=over 4
		3370
		3371	=item The winsocket select function
		3372
		3373	The winsocket C<select> function doesn't follow POSIX in that it
		3374	requires socket I<handles> and not socket I<file descriptors> (it is
		3375	also extremely buggy). This makes select very inefficient, and also
		3376	requires a mapping from file descriptors to socket handles (the Microsoft
		3377	C runtime provides the function C<_open_osfhandle> for this). See the
		3378	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		3379	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
		3380
		3381	The configuration for a "naked" win32 using the Microsoft runtime
		3382	libraries and raw winsocket select is:
		3383
		3384	#define EV_USE_SELECT 1
		3385	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		3386
		3387	Note that winsockets handling of fd sets is O(n), so you can easily get a
		3388	complexity in the O(n²) range when using win32.
		3389
		3390	=item Limited number of file descriptors
		3391
		3392	Windows has numerous arbitrary (and low) limits on things.
		3393
		3394	Early versions of winsocket's select only supported waiting for a maximum
		3395	of C<64> handles (probably owning to the fact that all windows kernels
		3396	can only wait for C<64> things at the same time internally; Microsoft
		3397	recommends spawning a chain of threads and wait for 63 handles and the
		3398	previous thread in each. Great).
		3399
		3400	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		3401	to some high number (e.g. C<2048>) before compiling the winsocket select
		3402	call (which might be in libev or elsewhere, for example, perl does its own
		3403	select emulation on windows).
		3404
		3405	Another limit is the number of file descriptors in the Microsoft runtime
		3406	libraries, which by default is C<64> (there must be a hidden I<64> fetish
		3407	or something like this inside Microsoft). You can increase this by calling
		3408	C<_setmaxstdio>, which can increase this limit to C<2048> (another
		3409	arbitrary limit), but is broken in many versions of the Microsoft runtime
		3410	libraries.
		3411
		3412	This might get you to about C<512> or C<2048> sockets (depending on
		3413	windows version and/or the phase of the moon). To get more, you need to
		3414	wrap all I/O functions and provide your own fd management, but the cost of
		3415	calling select (O(n²)) will likely make this unworkable.
		3416
		3417	=back
		3418
		3419
		3420	=head1 PORTABILITY REQUIREMENTS
		3421
		3422	In addition to a working ISO-C implementation, libev relies on a few
		3423	additional extensions:
		3424
		3425	=over 4
		3426
		3427	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		3428	calling conventions regardless of C<ev_watcher_type *>.
		3429
		3430	Libev assumes not only that all watcher pointers have the same internal
		3431	structure (guaranteed by POSIX but not by ISO C for example), but it also
		3432	assumes that the same (machine) code can be used to call any watcher
		3433	callback: The watcher callbacks have different type signatures, but libev
		3434	calls them using an C<ev_watcher *> internally.
		3435
		3436	=item C<sig_atomic_t volatile> must be thread-atomic as well
		3437
		3438	The type C<sig_atomic_t volatile> (or whatever is defined as
		3439	C<EV_ATOMIC_T>) must be atomic w.r.t. accesses from different
		3440	threads. This is not part of the specification for C<sig_atomic_t>, but is
		3441	believed to be sufficiently portable.
		3442
		3443	=item C<sigprocmask> must work in a threaded environment
		3444
		3445	Libev uses C<sigprocmask> to temporarily block signals. This is not
		3446	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		3447	pthread implementations will either allow C<sigprocmask> in the "main
		3448	thread" or will block signals process-wide, both behaviours would
		3449	be compatible with libev. Interaction between C<sigprocmask> and
		3450	C<pthread_sigmask> could complicate things, however.
		3451
		3452	The most portable way to handle signals is to block signals in all threads
		3453	except the initial one, and run the default loop in the initial thread as
		3454	well.
		3455
		3456	=item C<long> must be large enough for common memory allocation sizes
		3457
		3458	To improve portability and simplify using libev, libev uses C<long>
		3459	internally instead of C<size_t> when allocating its data structures. On
		3460	non-POSIX systems (Microsoft...) this might be unexpectedly low, but
		3461	is still at least 31 bits everywhere, which is enough for hundreds of
		3462	millions of watchers.
		3463
		3464	=item C<double> must hold a time value in seconds with enough accuracy
		3465
		3466	The type C<double> is used to represent timestamps. It is required to
		3467	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		3468	enough for at least into the year 4000. This requirement is fulfilled by
		3469	implementations implementing IEEE 754 (basically all existing ones).
		3470
		3471	=back
		3472
		3473	If you know of other additional requirements drop me a note.
		3474
		3475
		3476	=head1 COMPILER WARNINGS
		3477
		3478	Depending on your compiler and compiler settings, you might get no or a
		3479	lot of warnings when compiling libev code. Some people are apparently
		3480	scared by this.
		3481
		3482	However, these are unavoidable for many reasons. For one, each compiler
		3483	has different warnings, and each user has different tastes regarding
		3484	warning options. "Warn-free" code therefore cannot be a goal except when
		3485	targeting a specific compiler and compiler-version.
		3486
		3487	Another reason is that some compiler warnings require elaborate
		3488	workarounds, or other changes to the code that make it less clear and less
		3489	maintainable.
		3490
		3491	And of course, some compiler warnings are just plain stupid, or simply
		3492	wrong (because they don't actually warn about the condition their message
		3493	seems to warn about).
		3494
		3495	While libev is written to generate as few warnings as possible,
		3496	"warn-free" code is not a goal, and it is recommended not to build libev
		3497	with any compiler warnings enabled unless you are prepared to cope with
		3498	them (e.g. by ignoring them). Remember that warnings are just that:
		3499	warnings, not errors, or proof of bugs.
		3500
		3501
		3502	=head1 VALGRIND
		3503
		3504	Valgrind has a special section here because it is a popular tool that is
		3505	highly useful, but valgrind reports are very hard to interpret.
		3506
		3507	If you think you found a bug (memory leak, uninitialised data access etc.)
		3508	in libev, then check twice: If valgrind reports something like:
		3509
		3510	==2274== definitely lost: 0 bytes in 0 blocks.
		3511	==2274== possibly lost: 0 bytes in 0 blocks.
		3512	==2274== still reachable: 256 bytes in 1 blocks.
		3513
		3514	Then there is no memory leak. Similarly, under some circumstances,
		3515	valgrind might report kernel bugs as if it were a bug in libev, or it
		3516	might be confused (it is a very good tool, but only a tool).
		3517
		3518	If you are unsure about something, feel free to contact the mailing list
		3519	with the full valgrind report and an explanation on why you think this is
		3520	a bug in libev. However, don't be annoyed when you get a brisk "this is
		3521	no bug" answer and take the chance of learning how to interpret valgrind
		3522	properly.
		3523
		3524	If you need, for some reason, empty reports from valgrind for your project
		3525	I suggest using suppression lists.
		3526
		3527
2183	=head1 AUTHOR	3528	=head1 AUTHOR
2184		3529
2185	Marc Lehmann <libev@schmorp.de>.	3530	Marc Lehmann <libev@schmorp.de>.
2186		3531

Diff Legend

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