ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.34 by root, Fri Nov 23 16:17:12 2007 UTC vs.
Revision 1.112 by root, Wed Dec 26 08:06:09 2007 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
8		8
		9	=head2 EXAMPLE PROGRAM
		10
		11	#include <ev.h>
		12
		13	ev_io stdin_watcher;
		14	ev_timer timeout_watcher;
		15
		16	/* called when data readable on stdin */
		17	static void
		18	stdin_cb (EV_P_ struct ev_io *w, int revents)
		19	{
		20	/* puts ("stdin ready"); */
		21	ev_io_stop (EV_A_ w); /* just a syntax example */
		22	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */
		23	}
		24
		25	static void
		26	timeout_cb (EV_P_ struct ev_timer *w, int revents)
		27	{
		28	/* puts ("timeout"); */
		29	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */
		30	}
		31
		32	int
		33	main (void)
		34	{
		35	struct ev_loop *loop = ev_default_loop (0);
		36
		37	/* initialise an io watcher, then start it */
		38	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
		39	ev_io_start (loop, &stdin_watcher);
		40
		41	/* simple non-repeating 5.5 second timeout */
		42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		43	ev_timer_start (loop, &timeout_watcher);
		44
		45	/* loop till timeout or data ready */
		46	ev_loop (loop, 0);
		47
		48	return 0;
		49	}
		50
9	=head1 DESCRIPTION	51	=head1 DESCRIPTION
10		52
		53	The newest version of this document is also available as a html-formatted
		54	web page you might find easier to navigate when reading it for the first
		55	time: L<http://cvs.schmorp.de/libev/ev.html>.
		56
11	Libev is an event loop: you register interest in certain events (such as a	57	Libev is an event loop: you register interest in certain events (such as a
12	file descriptor being readable or a timeout occuring), and it will manage	58	file descriptor being readable or a timeout occurring), and it will manage
13	these event sources and provide your program with events.	59	these event sources and provide your program with events.
14		60
15	To do this, it must take more or less complete control over your process	61	To do this, it must take more or less complete control over your process
16	(or thread) by executing the I<event loop> handler, and will then	62	(or thread) by executing the I<event loop> handler, and will then
17	communicate events via a callback mechanism.	63	communicate events via a callback mechanism.
…		…
19	You register interest in certain events by registering so-called I<event	65	You register interest in certain events by registering so-called I<event
20	watchers>, which are relatively small C structures you initialise with the	66	watchers>, which are relatively small C structures you initialise with the
21	details of the event, and then hand it over to libev by I<starting> the	67	details of the event, and then hand it over to libev by I<starting> the
22	watcher.	68	watcher.
23		69
24	=head1 FEATURES	70	=head2 FEATURES
25		71
26	Libev supports select, poll, the linux-specific epoll and the bsd-specific	72	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
27	kqueue mechanisms for file descriptor events, relative timers, absolute	73	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
28	timers with customised rescheduling, signal events, process status change	74	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
29	events (related to SIGCHLD), and event watchers dealing with the event	75	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
30	loop mechanism itself (idle, prepare and check watchers). It also is quite	76	with customised rescheduling (C<ev_periodic>), synchronous signals
		77	(C<ev_signal>), process status change events (C<ev_child>), and event
		78	watchers dealing with the event loop mechanism itself (C<ev_idle>,
		79	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
		80	file watchers (C<ev_stat>) and even limited support for fork events
		81	(C<ev_fork>).
		82
		83	It also is quite fast (see this
31	fast (see this L<benchmark|http://libev.schmorp.de/bench.html> comparing	84	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
32	it to libevent for example).	85	for example).
33		86
34	=head1 CONVENTIONS	87	=head2 CONVENTIONS
35		88
36	Libev is very configurable. In this manual the default configuration	89	Libev is very configurable. In this manual the default configuration will
37	will be described, which supports multiple event loops. For more info	90	be described, which supports multiple event loops. For more info about
38	about various configuration options please have a look at the file	91	various configuration options please have a look at B<EMBED> section in
39	F<README.embed> in the libev distribution. If libev was configured without	92	this manual. If libev was configured without support for multiple event
40	support for multiple event loops, then all functions taking an initial	93	loops, then all functions taking an initial argument of name C<loop>
41	argument of name C<loop> (which is always of type C<struct ev_loop *>)	94	(which is always of type C<struct ev_loop *>) will not have this argument.
42	will not have this argument.
43		95
44	=head1 TIME REPRESENTATION	96	=head2 TIME REPRESENTATION
45		97
46	Libev represents time as a single floating point number, representing the	98	Libev represents time as a single floating point number, representing the
47	(fractional) number of seconds since the (POSIX) epoch (somewhere near	99	(fractional) number of seconds since the (POSIX) epoch (somewhere near
48	the beginning of 1970, details are complicated, don't ask). This type is	100	the beginning of 1970, details are complicated, don't ask). This type is
49	called C<ev_tstamp>, which is what you should use too. It usually aliases	101	called C<ev_tstamp>, which is what you should use too. It usually aliases
50	to the C<double> type in C, and when you need to do any calculations on	102	to the C<double> type in C, and when you need to do any calculations on
51	it, you should treat it as such.	103	it, you should treat it as some floatingpoint value. Unlike the name
52		104	component C<stamp> might indicate, it is also used for time differences
		105	throughout libev.
53		106
54	=head1 GLOBAL FUNCTIONS	107	=head1 GLOBAL FUNCTIONS
55		108
56	These functions can be called anytime, even before initialising the	109	These functions can be called anytime, even before initialising the
57	library in any way.	110	library in any way.
…		…
62		115
63	Returns the current time as libev would use it. Please note that the	116	Returns the current time as libev would use it. Please note that the
64	C<ev_now> function is usually faster and also often returns the timestamp	117	C<ev_now> function is usually faster and also often returns the timestamp
65	you actually want to know.	118	you actually want to know.
66		119
		120	=item ev_sleep (ev_tstamp interval)
		121
		122	Sleep for the given interval: The current thread will be blocked until
		123	either it is interrupted or the given time interval has passed. Basically
		124	this is a subsecond-resolution C<sleep ()>.
		125
67	=item int ev_version_major ()	126	=item int ev_version_major ()
68		127
69	=item int ev_version_minor ()	128	=item int ev_version_minor ()
70		129
71	You can find out the major and minor version numbers of the library	130	You can find out the major and minor ABI version numbers of the library
72	you linked against by calling the functions C<ev_version_major> and	131	you linked against by calling the functions C<ev_version_major> and
73	C<ev_version_minor>. If you want, you can compare against the global	132	C<ev_version_minor>. If you want, you can compare against the global
74	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	133	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
75	version of the library your program was compiled against.	134	version of the library your program was compiled against.
76		135
		136	These version numbers refer to the ABI version of the library, not the
		137	release version.
		138
77	Usually, it's a good idea to terminate if the major versions mismatch,	139	Usually, it's a good idea to terminate if the major versions mismatch,
78	as this indicates an incompatible change. Minor versions are usually	140	as this indicates an incompatible change. Minor versions are usually
79	compatible to older versions, so a larger minor version alone is usually	141	compatible to older versions, so a larger minor version alone is usually
80	not a problem.	142	not a problem.
81		143
82	Example: make sure we haven't accidentally been linked against the wrong	144	Example: Make sure we haven't accidentally been linked against the wrong
83	version:	145	version.
84		146
85	assert (("libev version mismatch",	147	assert (("libev version mismatch",
86	ev_version_major () == EV_VERSION_MAJOR	148	ev_version_major () == EV_VERSION_MAJOR
87	&& ev_version_minor () >= EV_VERSION_MINOR));	149	&& ev_version_minor () >= EV_VERSION_MINOR));
88		150
…		…
106	returned by C<ev_supported_backends>, as for example kqueue is broken on	168	returned by C<ev_supported_backends>, as for example kqueue is broken on
107	most BSDs and will not be autodetected unless you explicitly request it	169	most BSDs and will not be autodetected unless you explicitly request it
108	(assuming you know what you are doing). This is the set of backends that	170	(assuming you know what you are doing). This is the set of backends that
109	libev will probe for if you specify no backends explicitly.	171	libev will probe for if you specify no backends explicitly.
110		172
		173	=item unsigned int ev_embeddable_backends ()
		174
		175	Returns the set of backends that are embeddable in other event loops. This
		176	is the theoretical, all-platform, value. To find which backends
		177	might be supported on the current system, you would need to look at
		178	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
		179	recommended ones.
		180
		181	See the description of C<ev_embed> watchers for more info.
		182
111	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	183	=item ev_set_allocator (void *(*cb)(void *ptr, long size))
112		184
113	Sets the allocation function to use (the prototype is similar to the	185	Sets the allocation function to use (the prototype is similar - the
114	realloc C function, the semantics are identical). It is used to allocate	186	semantics is identical - to the realloc C function). It is used to
115	and free memory (no surprises here). If it returns zero when memory	187	allocate and free memory (no surprises here). If it returns zero when
116	needs to be allocated, the library might abort or take some potentially	188	memory needs to be allocated, the library might abort or take some
117	destructive action. The default is your system realloc function.	189	potentially destructive action. The default is your system realloc
		190	function.
118		191
119	You could override this function in high-availability programs to, say,	192	You could override this function in high-availability programs to, say,
120	free some memory if it cannot allocate memory, to use a special allocator,	193	free some memory if it cannot allocate memory, to use a special allocator,
121	or even to sleep a while and retry until some memory is available.	194	or even to sleep a while and retry until some memory is available.
122		195
123	Example: replace the libev allocator with one that waits a bit and then	196	Example: Replace the libev allocator with one that waits a bit and then
124	retries: better than mine).	197	retries).
125		198
126	static void *	199	static void *
127	persistent_realloc (void *ptr, long size)	200	persistent_realloc (void *ptr, size_t size)
128	{	201	{
129	for (;;)	202	for (;;)
130	{	203	{
131	void *newptr = realloc (ptr, size);	204	void *newptr = realloc (ptr, size);
132		205
…		…
148	callback is set, then libev will expect it to remedy the sitution, no	221	callback is set, then libev will expect it to remedy the sitution, no
149	matter what, when it returns. That is, libev will generally retry the	222	matter what, when it returns. That is, libev will generally retry the
150	requested operation, or, if the condition doesn't go away, do bad stuff	223	requested operation, or, if the condition doesn't go away, do bad stuff
151	(such as abort).	224	(such as abort).
152		225
153	Example: do the same thing as libev does internally:	226	Example: This is basically the same thing that libev does internally, too.
154		227
155	static void	228	static void
156	fatal_error (const char *msg)	229	fatal_error (const char *msg)
157	{	230	{
158	perror (msg);	231	perror (msg);
…		…
208	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	281	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
209	override the flags completely if it is found in the environment. This is	282	override the flags completely if it is found in the environment. This is
210	useful to try out specific backends to test their performance, or to work	283	useful to try out specific backends to test their performance, or to work
211	around bugs.	284	around bugs.
212		285
		286	=item C<EVFLAG_FORKCHECK>
		287
		288	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after
		289	a fork, you can also make libev check for a fork in each iteration by
		290	enabling this flag.
		291
		292	This works by calling C<getpid ()> on every iteration of the loop,
		293	and thus this might slow down your event loop if you do a lot of loop
		294	iterations and little real work, but is usually not noticeable (on my
		295	Linux system for example, C<getpid> is actually a simple 5-insn sequence
		296	without a syscall and thus I<very> fast, but my Linux system also has
		297	C<pthread_atfork> which is even faster).
		298
		299	The big advantage of this flag is that you can forget about fork (and
		300	forget about forgetting to tell libev about forking) when you use this
		301	flag.
		302
		303	This flag setting cannot be overriden or specified in the C<LIBEV_FLAGS>
		304	environment variable.
		305
213	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	306	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
214		307
215	This is your standard select(2) backend. Not I<completely> standard, as	308	This is your standard select(2) backend. Not I<completely> standard, as
216	libev tries to roll its own fd_set with no limits on the number of fds,	309	libev tries to roll its own fd_set with no limits on the number of fds,
217	but if that fails, expect a fairly low limit on the number of fds when	310	but if that fails, expect a fairly low limit on the number of fds when
218	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	311	using this backend. It doesn't scale too well (O(highest_fd)), but its
219	the fastest backend for a low number of fds.	312	usually the fastest backend for a low number of (low-numbered :) fds.
		313
		314	To get good performance out of this backend you need a high amount of
		315	parallelity (most of the file descriptors should be busy). If you are
		316	writing a server, you should C<accept ()> in a loop to accept as many
		317	connections as possible during one iteration. You might also want to have
		318	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		319	readyness notifications you get per iteration.
220		320
221	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	321	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
222		322
223	And this is your standard poll(2) backend. It's more complicated than	323	And this is your standard poll(2) backend. It's more complicated
224	select, but handles sparse fds better and has no artificial limit on the	324	than select, but handles sparse fds better and has no artificial
225	number of fds you can use (except it will slow down considerably with a	325	limit on the number of fds you can use (except it will slow down
226	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	326	considerably with a lot of inactive fds). It scales similarly to select,
		327	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		328	performance tips.
227		329
228	=item C<EVBACKEND_EPOLL> (value 4, Linux)	330	=item C<EVBACKEND_EPOLL> (value 4, Linux)
229		331
230	For few fds, this backend is a bit little slower than poll and select,	332	For few fds, this backend is a bit little slower than poll and select,
231	but it scales phenomenally better. While poll and select usually scale like	333	but it scales phenomenally better. While poll and select usually scale
232	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	334	like O(total_fds) where n is the total number of fds (or the highest fd),
233	either O(1) or O(active_fds).	335	epoll scales either O(1) or O(active_fds). The epoll design has a number
		336	of shortcomings, such as silently dropping events in some hard-to-detect
		337	cases and rewiring a syscall per fd change, no fork support and bad
		338	support for dup.
234		339
235	While stopping and starting an I/O watcher in the same iteration will	340	While stopping, setting and starting an I/O watcher in the same iteration
236	result in some caching, there is still a syscall per such incident	341	will result in some caching, there is still a syscall per such incident
237	(because the fd could point to a different file description now), so its	342	(because the fd could point to a different file description now), so its
238	best to avoid that. Also, dup()ed file descriptors might not work very	343	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
239	well if you register events for both fds.	344	very well if you register events for both fds.
240		345
241	Please note that epoll sometimes generates spurious notifications, so you	346	Please note that epoll sometimes generates spurious notifications, so you
242	need to use non-blocking I/O or other means to avoid blocking when no data	347	need to use non-blocking I/O or other means to avoid blocking when no data
243	(or space) is available.	348	(or space) is available.
244		349
		350	Best performance from this backend is achieved by not unregistering all
		351	watchers for a file descriptor until it has been closed, if possible, i.e.
		352	keep at least one watcher active per fd at all times.
		353
		354	While nominally embeddeble in other event loops, this feature is broken in
		355	all kernel versions tested so far.
		356
245	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	357	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
246		358
247	Kqueue deserves special mention, as at the time of this writing, it	359	Kqueue deserves special mention, as at the time of this writing, it
248	was broken on all BSDs except NetBSD (usually it doesn't work with	360	was broken on all BSDs except NetBSD (usually it doesn't work reliably
249	anything but sockets and pipes, except on Darwin, where of course its	361	with anything but sockets and pipes, except on Darwin, where of course
250	completely useless). For this reason its not being "autodetected"	362	it's completely useless). For this reason it's not being "autodetected"
251	unless you explicitly specify it explicitly in the flags (i.e. using	363	unless you explicitly specify it explicitly in the flags (i.e. using
252	C<EVBACKEND_KQUEUE>).	364	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		365	system like NetBSD.
		366
		367	You still can embed kqueue into a normal poll or select backend and use it
		368	only for sockets (after having made sure that sockets work with kqueue on
		369	the target platform). See C<ev_embed> watchers for more info.
253		370
254	It scales in the same way as the epoll backend, but the interface to the	371	It scales in the same way as the epoll backend, but the interface to the
255	kernel is more efficient (which says nothing about its actual speed, of	372	kernel is more efficient (which says nothing about its actual speed, of
256	course). While starting and stopping an I/O watcher does not cause an	373	course). While stopping, setting and starting an I/O watcher does never
257	extra syscall as with epoll, it still adds up to four event changes per	374	cause an extra syscall as with C<EVBACKEND_EPOLL>, it still adds up to
258	incident, so its best to avoid that.	375	two event changes per incident, support for C<fork ()> is very bad and it
		376	drops fds silently in similarly hard-to-detect cases.
		377
		378	This backend usually performs well under most conditions.
		379
		380	While nominally embeddable in other event loops, this doesn't work
		381	everywhere, so you might need to test for this. And since it is broken
		382	almost everywhere, you should only use it when you have a lot of sockets
		383	(for which it usually works), by embedding it into another event loop
		384	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for
		385	sockets.
259		386
260	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	387	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
261		388
262	This is not implemented yet (and might never be).	389	This is not implemented yet (and might never be, unless you send me an
		390	implementation). According to reports, C</dev/poll> only supports sockets
		391	and is not embeddable, which would limit the usefulness of this backend
		392	immensely.
263		393
264	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	394	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
265		395
266	This uses the Solaris 10 port mechanism. As with everything on Solaris,	396	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
267	it's really slow, but it still scales very well (O(active_fds)).	397	it's really slow, but it still scales very well (O(active_fds)).
268		398
269	Please note that solaris ports can result in a lot of spurious	399	Please note that solaris event ports can deliver a lot of spurious
270	notifications, so you need to use non-blocking I/O or other means to avoid	400	notifications, so you need to use non-blocking I/O or other means to avoid
271	blocking when no data (or space) is available.	401	blocking when no data (or space) is available.
		402
		403	While this backend scales well, it requires one system call per active
		404	file descriptor per loop iteration. For small and medium numbers of file
		405	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		406	might perform better.
272		407
273	=item C<EVBACKEND_ALL>	408	=item C<EVBACKEND_ALL>
274		409
275	Try all backends (even potentially broken ones that wouldn't be tried	410	Try all backends (even potentially broken ones that wouldn't be tried
276	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	411	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
277	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	412	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
		413
		414	It is definitely not recommended to use this flag.
278		415
279	=back	416	=back
280		417
281	If one or more of these are ored into the flags value, then only these	418	If one or more of these are ored into the flags value, then only these
282	backends will be tried (in the reverse order as given here). If none are	419	backends will be tried (in the reverse order as given here). If none are
…		…
304	Similar to C<ev_default_loop>, but always creates a new event loop that is	441	Similar to C<ev_default_loop>, but always creates a new event loop that is
305	always distinct from the default loop. Unlike the default loop, it cannot	442	always distinct from the default loop. Unlike the default loop, it cannot
306	handle signal and child watchers, and attempts to do so will be greeted by	443	handle signal and child watchers, and attempts to do so will be greeted by
307	undefined behaviour (or a failed assertion if assertions are enabled).	444	undefined behaviour (or a failed assertion if assertions are enabled).
308		445
309	Example: try to create a event loop that uses epoll and nothing else.	446	Example: Try to create a event loop that uses epoll and nothing else.
310		447
311	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);	448	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
312	if (!epoller)	449	if (!epoller)
313	fatal ("no epoll found here, maybe it hides under your chair");	450	fatal ("no epoll found here, maybe it hides under your chair");
314		451
315	=item ev_default_destroy ()	452	=item ev_default_destroy ()
316		453
317	Destroys the default loop again (frees all memory and kernel state	454	Destroys the default loop again (frees all memory and kernel state
318	etc.). This stops all registered event watchers (by not touching them in	455	etc.). None of the active event watchers will be stopped in the normal
319	any way whatsoever, although you cannot rely on this :).	456	sense, so e.g. C<ev_is_active> might still return true. It is your
		457	responsibility to either stop all watchers cleanly yoursef I<before>
		458	calling this function, or cope with the fact afterwards (which is usually
		459	the easiest thing, you can just ignore the watchers and/or C<free ()> them
		460	for example).
		461
		462	Note that certain global state, such as signal state, will not be freed by
		463	this function, and related watchers (such as signal and child watchers)
		464	would need to be stopped manually.
		465
		466	In general it is not advisable to call this function except in the
		467	rare occasion where you really need to free e.g. the signal handling
		468	pipe fds. If you need dynamically allocated loops it is better to use
		469	C<ev_loop_new> and C<ev_loop_destroy>).
320		470
321	=item ev_loop_destroy (loop)	471	=item ev_loop_destroy (loop)
322		472
323	Like C<ev_default_destroy>, but destroys an event loop created by an	473	Like C<ev_default_destroy>, but destroys an event loop created by an
324	earlier call to C<ev_loop_new>.	474	earlier call to C<ev_loop_new>.
…		…
348		498
349	Like C<ev_default_fork>, but acts on an event loop created by	499	Like C<ev_default_fork>, but acts on an event loop created by
350	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	500	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
351	after fork, and how you do this is entirely your own problem.	501	after fork, and how you do this is entirely your own problem.
352		502
		503	=item unsigned int ev_loop_count (loop)
		504
		505	Returns the count of loop iterations for the loop, which is identical to
		506	the number of times libev did poll for new events. It starts at C<0> and
		507	happily wraps around with enough iterations.
		508
		509	This value can sometimes be useful as a generation counter of sorts (it
		510	"ticks" the number of loop iterations), as it roughly corresponds with
		511	C<ev_prepare> and C<ev_check> calls.
		512
353	=item unsigned int ev_backend (loop)	513	=item unsigned int ev_backend (loop)
354		514
355	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	515	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
356	use.	516	use.
357		517
…		…
359		519
360	Returns the current "event loop time", which is the time the event loop	520	Returns the current "event loop time", which is the time the event loop
361	received events and started processing them. This timestamp does not	521	received events and started processing them. This timestamp does not
362	change as long as callbacks are being processed, and this is also the base	522	change as long as callbacks are being processed, and this is also the base
363	time used for relative timers. You can treat it as the timestamp of the	523	time used for relative timers. You can treat it as the timestamp of the
364	event occuring (or more correctly, libev finding out about it).	524	event occurring (or more correctly, libev finding out about it).
365		525
366	=item ev_loop (loop, int flags)	526	=item ev_loop (loop, int flags)
367		527
368	Finally, this is it, the event handler. This function usually is called	528	Finally, this is it, the event handler. This function usually is called
369	after you initialised all your watchers and you want to start handling	529	after you initialised all your watchers and you want to start handling
…		…
390	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is	550	libev watchers. However, a pair of C<ev_prepare>/C<ev_check> watchers is
391	usually a better approach for this kind of thing.	551	usually a better approach for this kind of thing.
392		552
393	Here are the gory details of what C<ev_loop> does:	553	Here are the gory details of what C<ev_loop> does:
394		554
		555	- Before the first iteration, call any pending watchers.
395	* If there are no active watchers (reference count is zero), return.	556	* If there are no active watchers (reference count is zero), return.
396	- Queue prepare watchers and then call all outstanding watchers.	557	- Queue all prepare watchers and then call all outstanding watchers.
397	- If we have been forked, recreate the kernel state.	558	- If we have been forked, recreate the kernel state.
398	- Update the kernel state with all outstanding changes.	559	- Update the kernel state with all outstanding changes.
399	- Update the "event loop time".	560	- Update the "event loop time".
400	- Calculate for how long to block.	561	- Calculate for how long to block.
401	- Block the process, waiting for any events.	562	- Block the process, waiting for any events.
…		…
409	Signals and child watchers are implemented as I/O watchers, and will	570	Signals and child watchers are implemented as I/O watchers, and will
410	be handled here by queueing them when their watcher gets executed.	571	be handled here by queueing them when their watcher gets executed.
411	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	572	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
412	were used, return, otherwise continue with step *.	573	were used, return, otherwise continue with step *.
413		574
414	Example: queue some jobs and then loop until no events are outsanding	575	Example: Queue some jobs and then loop until no events are outsanding
415	anymore.	576	anymore.
416		577
417	... queue jobs here, make sure they register event watchers as long	578	... queue jobs here, make sure they register event watchers as long
418	... as they still have work to do (even an idle watcher will do..)	579	... as they still have work to do (even an idle watcher will do..)
419	ev_loop (my_loop, 0);	580	ev_loop (my_loop, 0);
…		…
439	visible to the libev user and should not keep C<ev_loop> from exiting if	600	visible to the libev user and should not keep C<ev_loop> from exiting if
440	no event watchers registered by it are active. It is also an excellent	601	no event watchers registered by it are active. It is also an excellent
441	way to do this for generic recurring timers or from within third-party	602	way to do this for generic recurring timers or from within third-party
442	libraries. Just remember to I<unref after start> and I<ref before stop>.	603	libraries. Just remember to I<unref after start> and I<ref before stop>.
443		604
444	Example: create a signal watcher, but keep it from keeping C<ev_loop>	605	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
445	running when nothing else is active.	606	running when nothing else is active.
446		607
447	struct dv_signal exitsig;	608	struct ev_signal exitsig;
448	ev_signal_init (&exitsig, sig_cb, SIGINT);	609	ev_signal_init (&exitsig, sig_cb, SIGINT);
449	ev_signal_start (myloop, &exitsig);	610	ev_signal_start (loop, &exitsig);
450	evf_unref (myloop);	611	evf_unref (loop);
451		612
452	Example: for some weird reason, unregister the above signal handler again.	613	Example: For some weird reason, unregister the above signal handler again.
453		614
454	ev_ref (myloop);	615	ev_ref (loop);
455	ev_signal_stop (myloop, &exitsig);	616	ev_signal_stop (loop, &exitsig);
		617
		618	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		619
		620	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		621
		622	These advanced functions influence the time that libev will spend waiting
		623	for events. Both are by default C<0>, meaning that libev will try to
		624	invoke timer/periodic callbacks and I/O callbacks with minimum latency.
		625
		626	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		627	allows libev to delay invocation of I/O and timer/periodic callbacks to
		628	increase efficiency of loop iterations.
		629
		630	The background is that sometimes your program runs just fast enough to
		631	handle one (or very few) event(s) per loop iteration. While this makes
		632	the program responsive, it also wastes a lot of CPU time to poll for new
		633	events, especially with backends like C<select ()> which have a high
		634	overhead for the actual polling but can deliver many events at once.
		635
		636	By setting a higher I<io collect interval> you allow libev to spend more
		637	time collecting I/O events, so you can handle more events per iteration,
		638	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		639	C<ev_timer>) will be not affected. Setting this to a non-null value will
		640	introduce an additional C<ev_sleep ()> call into most loop iterations.
		641
		642	Likewise, by setting a higher I<timeout collect interval> you allow libev
		643	to spend more time collecting timeouts, at the expense of increased
		644	latency (the watcher callback will be called later). C<ev_io> watchers
		645	will not be affected. Setting this to a non-null value will not introduce
		646	any overhead in libev.
		647
		648	Many (busy) programs can usually benefit by setting the io collect
		649	interval to a value near C<0.1> or so, which is often enough for
		650	interactive servers (of course not for games), likewise for timeouts. It
		651	usually doesn't make much sense to set it to a lower value than C<0.01>,
		652	as this approsaches the timing granularity of most systems.
456		653
457	=back	654	=back
		655
458		656
459	=head1 ANATOMY OF A WATCHER	657	=head1 ANATOMY OF A WATCHER
460		658
461	A watcher is a structure that you create and register to record your	659	A watcher is a structure that you create and register to record your
462	interest in some event. For instance, if you want to wait for STDIN to	660	interest in some event. For instance, if you want to wait for STDIN to
…		…
495	*) >>), and you can stop watching for events at any time by calling the	693	*) >>), and you can stop watching for events at any time by calling the
496	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	694	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.
497		695
498	As long as your watcher is active (has been started but not stopped) you	696	As long as your watcher is active (has been started but not stopped) you
499	must not touch the values stored in it. Most specifically you must never	697	must not touch the values stored in it. Most specifically you must never
500	reinitialise it or call its set macro.	698	reinitialise it or call its C<set> macro.
501
502	You can check whether an event is active by calling the C<ev_is_active
503	(watcher *)> macro. To see whether an event is outstanding (but the
504	callback for it has not been called yet) you can use the C<ev_is_pending
505	(watcher *)> macro.
506		699
507	Each and every callback receives the event loop pointer as first, the	700	Each and every callback receives the event loop pointer as first, the
508	registered watcher structure as second, and a bitset of received events as	701	registered watcher structure as second, and a bitset of received events as
509	third argument.	702	third argument.
510		703
…		…
534	The signal specified in the C<ev_signal> watcher has been received by a thread.	727	The signal specified in the C<ev_signal> watcher has been received by a thread.
535		728
536	=item C<EV_CHILD>	729	=item C<EV_CHILD>
537		730
538	The pid specified in the C<ev_child> watcher has received a status change.	731	The pid specified in the C<ev_child> watcher has received a status change.
		732
		733	=item C<EV_STAT>
		734
		735	The path specified in the C<ev_stat> watcher changed its attributes somehow.
539		736
540	=item C<EV_IDLE>	737	=item C<EV_IDLE>
541		738
542	The C<ev_idle> watcher has determined that you have nothing better to do.	739	The C<ev_idle> watcher has determined that you have nothing better to do.
543		740
…		…
551	received events. Callbacks of both watcher types can start and stop as	748	received events. Callbacks of both watcher types can start and stop as
552	many watchers as they want, and all of them will be taken into account	749	many watchers as they want, and all of them will be taken into account
553	(for example, a C<ev_prepare> watcher might start an idle watcher to keep	750	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
554	C<ev_loop> from blocking).	751	C<ev_loop> from blocking).
555		752
		753	=item C<EV_EMBED>
		754
		755	The embedded event loop specified in the C<ev_embed> watcher needs attention.
		756
		757	=item C<EV_FORK>
		758
		759	The event loop has been resumed in the child process after fork (see
		760	C<ev_fork>).
		761
556	=item C<EV_ERROR>	762	=item C<EV_ERROR>
557		763
558	An unspecified error has occured, the watcher has been stopped. This might	764	An unspecified error has occured, the watcher has been stopped. This might
559	happen because the watcher could not be properly started because libev	765	happen because the watcher could not be properly started because libev
560	ran out of memory, a file descriptor was found to be closed or any other	766	ran out of memory, a file descriptor was found to be closed or any other
…		…
566	your callbacks is well-written it can just attempt the operation and cope	772	your callbacks is well-written it can just attempt the operation and cope
567	with the error from read() or write(). This will not work in multithreaded	773	with the error from read() or write(). This will not work in multithreaded
568	programs, though, so beware.	774	programs, though, so beware.
569		775
570	=back	776	=back
		777
		778	=head2 GENERIC WATCHER FUNCTIONS
		779
		780	In the following description, C<TYPE> stands for the watcher type,
		781	e.g. C<timer> for C<ev_timer> watchers and C<io> for C<ev_io> watchers.
		782
		783	=over 4
		784
		785	=item C<ev_init> (ev_TYPE *watcher, callback)
		786
		787	This macro initialises the generic portion of a watcher. The contents
		788	of the watcher object can be arbitrary (so C<malloc> will do). Only
		789	the generic parts of the watcher are initialised, you I<need> to call
		790	the type-specific C<ev_TYPE_set> macro afterwards to initialise the
		791	type-specific parts. For each type there is also a C<ev_TYPE_init> macro
		792	which rolls both calls into one.
		793
		794	You can reinitialise a watcher at any time as long as it has been stopped
		795	(or never started) and there are no pending events outstanding.
		796
		797	The callback is always of type C<void (*)(ev_loop *loop, ev_TYPE *watcher,
		798	int revents)>.
		799
		800	=item C<ev_TYPE_set> (ev_TYPE *, [args])
		801
		802	This macro initialises the type-specific parts of a watcher. You need to
		803	call C<ev_init> at least once before you call this macro, but you can
		804	call C<ev_TYPE_set> any number of times. You must not, however, call this
		805	macro on a watcher that is active (it can be pending, however, which is a
		806	difference to the C<ev_init> macro).
		807
		808	Although some watcher types do not have type-specific arguments
		809	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
		810
		811	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
		812
		813	This convinience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
		814	calls into a single call. This is the most convinient method to initialise
		815	a watcher. The same limitations apply, of course.
		816
		817	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
		818
		819	Starts (activates) the given watcher. Only active watchers will receive
		820	events. If the watcher is already active nothing will happen.
		821
		822	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
		823
		824	Stops the given watcher again (if active) and clears the pending
		825	status. It is possible that stopped watchers are pending (for example,
		826	non-repeating timers are being stopped when they become pending), but
		827	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If
		828	you want to free or reuse the memory used by the watcher it is therefore a
		829	good idea to always call its C<ev_TYPE_stop> function.
		830
		831	=item bool ev_is_active (ev_TYPE *watcher)
		832
		833	Returns a true value iff the watcher is active (i.e. it has been started
		834	and not yet been stopped). As long as a watcher is active you must not modify
		835	it.
		836
		837	=item bool ev_is_pending (ev_TYPE *watcher)
		838
		839	Returns a true value iff the watcher is pending, (i.e. it has outstanding
		840	events but its callback has not yet been invoked). As long as a watcher
		841	is pending (but not active) you must not call an init function on it (but
		842	C<ev_TYPE_set> is safe), you must not change its priority, and you must
		843	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		844	it).
		845
		846	=item callback ev_cb (ev_TYPE *watcher)
		847
		848	Returns the callback currently set on the watcher.
		849
		850	=item ev_cb_set (ev_TYPE *watcher, callback)
		851
		852	Change the callback. You can change the callback at virtually any time
		853	(modulo threads).
		854
		855	=item ev_set_priority (ev_TYPE *watcher, priority)
		856
		857	=item int ev_priority (ev_TYPE *watcher)
		858
		859	Set and query the priority of the watcher. The priority is a small
		860	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		861	(default: C<-2>). Pending watchers with higher priority will be invoked
		862	before watchers with lower priority, but priority will not keep watchers
		863	from being executed (except for C<ev_idle> watchers).
		864
		865	This means that priorities are I<only> used for ordering callback
		866	invocation after new events have been received. This is useful, for
		867	example, to reduce latency after idling, or more often, to bind two
		868	watchers on the same event and make sure one is called first.
		869
		870	If you need to suppress invocation when higher priority events are pending
		871	you need to look at C<ev_idle> watchers, which provide this functionality.
		872
		873	You I<must not> change the priority of a watcher as long as it is active or
		874	pending.
		875
		876	The default priority used by watchers when no priority has been set is
		877	always C<0>, which is supposed to not be too high and not be too low :).
		878
		879	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		880	fine, as long as you do not mind that the priority value you query might
		881	or might not have been adjusted to be within valid range.
		882
		883	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		884
		885	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		886	C<loop> nor C<revents> need to be valid as long as the watcher callback
		887	can deal with that fact.
		888
		889	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		890
		891	If the watcher is pending, this function returns clears its pending status
		892	and returns its C<revents> bitset (as if its callback was invoked). If the
		893	watcher isn't pending it does nothing and returns C<0>.
		894
		895	=back
		896
571		897
572	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	898	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
573		899
574	Each watcher has, by default, a member C<void *data> that you can change	900	Each watcher has, by default, a member C<void *data> that you can change
575	and read at any time, libev will completely ignore it. This can be used	901	and read at any time, libev will completely ignore it. This can be used
…		…
593	{	919	{
594	struct my_io *w = (struct my_io *)w_;	920	struct my_io *w = (struct my_io *)w_;
595	...	921	...
596	}	922	}
597		923
598	More interesting and less C-conformant ways of catsing your callback type	924	More interesting and less C-conformant ways of casting your callback type
599	have been omitted....	925	instead have been omitted.
		926
		927	Another common scenario is having some data structure with multiple
		928	watchers:
		929
		930	struct my_biggy
		931	{
		932	int some_data;
		933	ev_timer t1;
		934	ev_timer t2;
		935	}
		936
		937	In this case getting the pointer to C<my_biggy> is a bit more complicated,
		938	you need to use C<offsetof>:
		939
		940	#include <stddef.h>
		941
		942	static void
		943	t1_cb (EV_P_ struct ev_timer *w, int revents)
		944	{
		945	struct my_biggy big = (struct my_biggy *
		946	(((char *)w) - offsetof (struct my_biggy, t1));
		947	}
		948
		949	static void
		950	t2_cb (EV_P_ struct ev_timer *w, int revents)
		951	{
		952	struct my_biggy big = (struct my_biggy *
		953	(((char *)w) - offsetof (struct my_biggy, t2));
		954	}
600		955
601		956
602	=head1 WATCHER TYPES	957	=head1 WATCHER TYPES
603		958
604	This section describes each watcher in detail, but will not repeat	959	This section describes each watcher in detail, but will not repeat
605	information given in the last section.	960	information given in the last section. Any initialisation/set macros,
		961	functions and members specific to the watcher type are explained.
606		962
		963	Members are additionally marked with either I<[read-only]>, meaning that,
		964	while the watcher is active, you can look at the member and expect some
		965	sensible content, but you must not modify it (you can modify it while the
		966	watcher is stopped to your hearts content), or I<[read-write]>, which
		967	means you can expect it to have some sensible content while the watcher
		968	is active, but you can also modify it. Modifying it may not do something
		969	sensible or take immediate effect (or do anything at all), but libev will
		970	not crash or malfunction in any way.
607		971
		972
608	=head2 C<ev_io> - is this file descriptor readable or writable	973	=head2 C<ev_io> - is this file descriptor readable or writable?
609		974
610	I/O watchers check whether a file descriptor is readable or writable	975	I/O watchers check whether a file descriptor is readable or writable
611	in each iteration of the event loop (This behaviour is called	976	in each iteration of the event loop, or, more precisely, when reading
612	level-triggering because you keep receiving events as long as the	977	would not block the process and writing would at least be able to write
613	condition persists. Remember you can stop the watcher if you don't want to	978	some data. This behaviour is called level-triggering because you keep
614	act on the event and neither want to receive future events).	979	receiving events as long as the condition persists. Remember you can stop
		980	the watcher if you don't want to act on the event and neither want to
		981	receive future events.
615		982
616	In general you can register as many read and/or write event watchers per	983	In general you can register as many read and/or write event watchers per
617	fd as you want (as long as you don't confuse yourself). Setting all file	984	fd as you want (as long as you don't confuse yourself). Setting all file
618	descriptors to non-blocking mode is also usually a good idea (but not	985	descriptors to non-blocking mode is also usually a good idea (but not
619	required if you know what you are doing).	986	required if you know what you are doing).
620		987
621	You have to be careful with dup'ed file descriptors, though. Some backends
622	(the linux epoll backend is a notable example) cannot handle dup'ed file
623	descriptors correctly if you register interest in two or more fds pointing
624	to the same underlying file/socket etc. description (that is, they share
625	the same underlying "file open").
626
627	If you must do this, then force the use of a known-to-be-good backend	988	If you must do this, then force the use of a known-to-be-good backend
628	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	989	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
629	C<EVBACKEND_POLL>).	990	C<EVBACKEND_POLL>).
630		991
		992	Another thing you have to watch out for is that it is quite easy to
		993	receive "spurious" readyness notifications, that is your callback might
		994	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
		995	because there is no data. Not only are some backends known to create a
		996	lot of those (for example solaris ports), it is very easy to get into
		997	this situation even with a relatively standard program structure. Thus
		998	it is best to always use non-blocking I/O: An extra C<read>(2) returning
		999	C<EAGAIN> is far preferable to a program hanging until some data arrives.
		1000
		1001	If you cannot run the fd in non-blocking mode (for example you should not
		1002	play around with an Xlib connection), then you have to seperately re-test
		1003	whether a file descriptor is really ready with a known-to-be good interface
		1004	such as poll (fortunately in our Xlib example, Xlib already does this on
		1005	its own, so its quite safe to use).
		1006
		1007	=head3 The special problem of disappearing file descriptors
		1008
		1009	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1010	descriptor (either by calling C<close> explicitly or by any other means,
		1011	such as C<dup>). The reason is that you register interest in some file
		1012	descriptor, but when it goes away, the operating system will silently drop
		1013	this interest. If another file descriptor with the same number then is
		1014	registered with libev, there is no efficient way to see that this is, in
		1015	fact, a different file descriptor.
		1016
		1017	To avoid having to explicitly tell libev about such cases, libev follows
		1018	the following policy: Each time C<ev_io_set> is being called, libev
		1019	will assume that this is potentially a new file descriptor, otherwise
		1020	it is assumed that the file descriptor stays the same. That means that
		1021	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1022	descriptor even if the file descriptor number itself did not change.
		1023
		1024	This is how one would do it normally anyway, the important point is that
		1025	the libev application should not optimise around libev but should leave
		1026	optimisations to libev.
		1027
		1028	=head3 The special problem of dup'ed file descriptors
		1029
		1030	Some backends (e.g. epoll), cannot register events for file descriptors,
		1031	but only events for the underlying file descriptions. That means when you
		1032	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1033	events for them, only one file descriptor might actually receive events.
		1034
		1035	There is no workaround possible except not registering events
		1036	for potentially C<dup ()>'ed file descriptors, or to resort to
		1037	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1038
		1039	=head3 The special problem of fork
		1040
		1041	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1042	useless behaviour. Libev fully supports fork, but needs to be told about
		1043	it in the child.
		1044
		1045	To support fork in your programs, you either have to call
		1046	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1047	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1048	C<EVBACKEND_POLL>.
		1049
		1050
		1051	=head3 Watcher-Specific Functions
		1052
631	=over 4	1053	=over 4
632		1054
633	=item ev_io_init (ev_io *, callback, int fd, int events)	1055	=item ev_io_init (ev_io *, callback, int fd, int events)
634		1056
635	=item ev_io_set (ev_io *, int fd, int events)	1057	=item ev_io_set (ev_io *, int fd, int events)
636		1058
637	Configures an C<ev_io> watcher. The fd is the file descriptor to rceeive	1059	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
638	events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ |	1060	rceeive events for and events is either C<EV_READ>, C<EV_WRITE> or
639	EV_WRITE> to receive the given events.	1061	C<EV_READ | EV_WRITE> to receive the given events.
640		1062
641	Please note that most of the more scalable backend mechanisms (for example	1063	=item int fd [read-only]
642	epoll and solaris ports) can result in spurious readyness notifications	1064
643	for file descriptors, so you practically need to use non-blocking I/O (and	1065	The file descriptor being watched.
644	treat callback invocation as hint only), or retest separately with a safe	1066
645	interface before doing I/O (XLib can do this), or force the use of either	1067	=item int events [read-only]
646	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>, which don't suffer from this	1068
647	problem. Also note that it is quite easy to have your callback invoked	1069	The events being watched.
648	when the readyness condition is no longer valid even when employing
649	typical ways of handling events, so its a good idea to use non-blocking
650	I/O unconditionally.
651		1070
652	=back	1071	=back
653		1072
		1073	=head3 Examples
		1074
654	Example: call C<stdin_readable_cb> when STDIN_FILENO has become, well	1075	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
655	readable, but only once. Since it is likely line-buffered, you could	1076	readable, but only once. Since it is likely line-buffered, you could
656	attempt to read a whole line in the callback:	1077	attempt to read a whole line in the callback.
657		1078
658	static void	1079	static void
659	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1080	stdin_readable_cb (struct ev_loop *loop, struct ev_io *w, int revents)
660	{	1081	{
661	ev_io_stop (loop, w);	1082	ev_io_stop (loop, w);
…		…
668	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);	1089	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
669	ev_io_start (loop, &stdin_readable);	1090	ev_io_start (loop, &stdin_readable);
670	ev_loop (loop, 0);	1091	ev_loop (loop, 0);
671		1092
672		1093
673	=head2 C<ev_timer> - relative and optionally recurring timeouts	1094	=head2 C<ev_timer> - relative and optionally repeating timeouts
674		1095
675	Timer watchers are simple relative timers that generate an event after a	1096	Timer watchers are simple relative timers that generate an event after a
676	given time, and optionally repeating in regular intervals after that.	1097	given time, and optionally repeating in regular intervals after that.
677		1098
678	The timers are based on real time, that is, if you register an event that	1099	The timers are based on real time, that is, if you register an event that
…		…
691		1112
692	The callback is guarenteed to be invoked only when its timeout has passed,	1113	The callback is guarenteed to be invoked only when its timeout has passed,
693	but if multiple timers become ready during the same loop iteration then	1114	but if multiple timers become ready during the same loop iteration then
694	order of execution is undefined.	1115	order of execution is undefined.
695		1116
		1117	=head3 Watcher-Specific Functions and Data Members
		1118
696	=over 4	1119	=over 4
697		1120
698	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1121	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
699		1122
700	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1123	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
…		…
713	=item ev_timer_again (loop)	1136	=item ev_timer_again (loop)
714		1137
715	This will act as if the timer timed out and restart it again if it is	1138	This will act as if the timer timed out and restart it again if it is
716	repeating. The exact semantics are:	1139	repeating. The exact semantics are:
717		1140
		1141	If the timer is pending, its pending status is cleared.
		1142
718	If the timer is started but nonrepeating, stop it.	1143	If the timer is started but nonrepeating, stop it (as if it timed out).
719		1144
720	If the timer is repeating, either start it if necessary (with the repeat	1145	If the timer is repeating, either start it if necessary (with the
721	value), or reset the running timer to the repeat value.	1146	C<repeat> value), or reset the running timer to the C<repeat> value.
722		1147
723	This sounds a bit complicated, but here is a useful and typical	1148	This sounds a bit complicated, but here is a useful and typical
724	example: Imagine you have a tcp connection and you want a so-called idle	1149	example: Imagine you have a tcp connection and you want a so-called idle
725	timeout, that is, you want to be called when there have been, say, 60	1150	timeout, that is, you want to be called when there have been, say, 60
726	seconds of inactivity on the socket. The easiest way to do this is to	1151	seconds of inactivity on the socket. The easiest way to do this is to
727	configure an C<ev_timer> with after=repeat=60 and calling ev_timer_again each	1152	configure an C<ev_timer> with a C<repeat> value of C<60> and then call
728	time you successfully read or write some data. If you go into an idle	1153	C<ev_timer_again> each time you successfully read or write some data. If
729	state where you do not expect data to travel on the socket, you can stop	1154	you go into an idle state where you do not expect data to travel on the
730	the timer, and again will automatically restart it if need be.	1155	socket, you can C<ev_timer_stop> the timer, and C<ev_timer_again> will
		1156	automatically restart it if need be.
		1157
		1158	That means you can ignore the C<after> value and C<ev_timer_start>
		1159	altogether and only ever use the C<repeat> value and C<ev_timer_again>:
		1160
		1161	ev_timer_init (timer, callback, 0., 5.);
		1162	ev_timer_again (loop, timer);
		1163	...
		1164	timer->again = 17.;
		1165	ev_timer_again (loop, timer);
		1166	...
		1167	timer->again = 10.;
		1168	ev_timer_again (loop, timer);
		1169
		1170	This is more slightly efficient then stopping/starting the timer each time
		1171	you want to modify its timeout value.
		1172
		1173	=item ev_tstamp repeat [read-write]
		1174
		1175	The current C<repeat> value. Will be used each time the watcher times out
		1176	or C<ev_timer_again> is called and determines the next timeout (if any),
		1177	which is also when any modifications are taken into account.
731		1178
732	=back	1179	=back
733		1180
		1181	=head3 Examples
		1182
734	Example: create a timer that fires after 60 seconds.	1183	Example: Create a timer that fires after 60 seconds.
735		1184
736	static void	1185	static void
737	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1186	one_minute_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
738	{	1187	{
739	.. one minute over, w is actually stopped right here	1188	.. one minute over, w is actually stopped right here
…		…
741		1190
742	struct ev_timer mytimer;	1191	struct ev_timer mytimer;
743	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);	1192	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
744	ev_timer_start (loop, &mytimer);	1193	ev_timer_start (loop, &mytimer);
745		1194
746	Example: create a timeout timer that times out after 10 seconds of	1195	Example: Create a timeout timer that times out after 10 seconds of
747	inactivity.	1196	inactivity.
748		1197
749	static void	1198	static void
750	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)	1199	timeout_cb (struct ev_loop *loop, struct ev_timer *w, int revents)
751	{	1200	{
…		…
760	// and in some piece of code that gets executed on any "activity":	1209	// and in some piece of code that gets executed on any "activity":
761	// reset the timeout to start ticking again at 10 seconds	1210	// reset the timeout to start ticking again at 10 seconds
762	ev_timer_again (&mytimer);	1211	ev_timer_again (&mytimer);
763		1212
764		1213
765	=head2 C<ev_periodic> - to cron or not to cron	1214	=head2 C<ev_periodic> - to cron or not to cron?
766		1215
767	Periodic watchers are also timers of a kind, but they are very versatile	1216	Periodic watchers are also timers of a kind, but they are very versatile
768	(and unfortunately a bit complex).	1217	(and unfortunately a bit complex).
769		1218
770	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1219	Unlike C<ev_timer>'s, they are not based on real time (or relative time)
771	but on wallclock time (absolute time). You can tell a periodic watcher	1220	but on wallclock time (absolute time). You can tell a periodic watcher
772	to trigger "at" some specific point in time. For example, if you tell a	1221	to trigger "at" some specific point in time. For example, if you tell a
773	periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now ()	1222	periodic watcher to trigger in 10 seconds (by specifiying e.g. C<ev_now ()
774	+ 10.>) and then reset your system clock to the last year, then it will	1223	+ 10.>) and then reset your system clock to the last year, then it will
775	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1224	take a year to trigger the event (unlike an C<ev_timer>, which would trigger
776	roughly 10 seconds later and of course not if you reset your system time	1225	roughly 10 seconds later).
777	again).
778		1226
779	They can also be used to implement vastly more complex timers, such as	1227	They can also be used to implement vastly more complex timers, such as
780	triggering an event on eahc midnight, local time.	1228	triggering an event on each midnight, local time or other, complicated,
		1229	rules.
781		1230
782	As with timers, the callback is guarenteed to be invoked only when the	1231	As with timers, the callback is guarenteed to be invoked only when the
783	time (C<at>) has been passed, but if multiple periodic timers become ready	1232	time (C<at>) has been passed, but if multiple periodic timers become ready
784	during the same loop iteration then order of execution is undefined.	1233	during the same loop iteration then order of execution is undefined.
785		1234
		1235	=head3 Watcher-Specific Functions and Data Members
		1236
786	=over 4	1237	=over 4
787		1238
788	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1239	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)
789		1240
790	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1241	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)
…		…
792	Lots of arguments, lets sort it out... There are basically three modes of	1243	Lots of arguments, lets sort it out... There are basically three modes of
793	operation, and we will explain them from simplest to complex:	1244	operation, and we will explain them from simplest to complex:
794		1245
795	=over 4	1246	=over 4
796		1247
797	=item * absolute timer (interval = reschedule_cb = 0)	1248	=item * absolute timer (at = time, interval = reschedule_cb = 0)
798		1249
799	In this configuration the watcher triggers an event at the wallclock time	1250	In this configuration the watcher triggers an event at the wallclock time
800	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1251	C<at> and doesn't repeat. It will not adjust when a time jump occurs,
801	that is, if it is to be run at January 1st 2011 then it will run when the	1252	that is, if it is to be run at January 1st 2011 then it will run when the
802	system time reaches or surpasses this time.	1253	system time reaches or surpasses this time.
803		1254
804	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1255	=item * non-repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
805		1256
806	In this mode the watcher will always be scheduled to time out at the next	1257	In this mode the watcher will always be scheduled to time out at the next
807	C<at + N * interval> time (for some integer N) and then repeat, regardless	1258	C<at + N * interval> time (for some integer N, which can also be negative)
808	of any time jumps.	1259	and then repeat, regardless of any time jumps.
809		1260
810	This can be used to create timers that do not drift with respect to system	1261	This can be used to create timers that do not drift with respect to system
811	time:	1262	time:
812		1263
813	ev_periodic_set (&periodic, 0., 3600., 0);	1264	ev_periodic_set (&periodic, 0., 3600., 0);
…		…
819		1270
820	Another way to think about it (for the mathematically inclined) is that	1271	Another way to think about it (for the mathematically inclined) is that
821	C<ev_periodic> will try to run the callback in this mode at the next possible	1272	C<ev_periodic> will try to run the callback in this mode at the next possible
822	time where C<time = at (mod interval)>, regardless of any time jumps.	1273	time where C<time = at (mod interval)>, regardless of any time jumps.
823		1274
		1275	For numerical stability it is preferable that the C<at> value is near
		1276	C<ev_now ()> (the current time), but there is no range requirement for
		1277	this value.
		1278
824	=item * manual reschedule mode (reschedule_cb = callback)	1279	=item * manual reschedule mode (at and interval ignored, reschedule_cb = callback)
825		1280
826	In this mode the values for C<interval> and C<at> are both being	1281	In this mode the values for C<interval> and C<at> are both being
827	ignored. Instead, each time the periodic watcher gets scheduled, the	1282	ignored. Instead, each time the periodic watcher gets scheduled, the
828	reschedule callback will be called with the watcher as first, and the	1283	reschedule callback will be called with the watcher as first, and the
829	current time as second argument.	1284	current time as second argument.
830		1285
831	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1286	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,
832	ever, or make any event loop modifications>. If you need to stop it,	1287	ever, or make any event loop modifications>. If you need to stop it,
833	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by	1288	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by
834	starting a prepare watcher).	1289	starting an C<ev_prepare> watcher, which is legal).
835		1290
836	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1291	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,
837	ev_tstamp now)>, e.g.:	1292	ev_tstamp now)>, e.g.:
838		1293
839	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1294	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)
…		…
862	Simply stops and restarts the periodic watcher again. This is only useful	1317	Simply stops and restarts the periodic watcher again. This is only useful
863	when you changed some parameters or the reschedule callback would return	1318	when you changed some parameters or the reschedule callback would return
864	a different time than the last time it was called (e.g. in a crond like	1319	a different time than the last time it was called (e.g. in a crond like
865	program when the crontabs have changed).	1320	program when the crontabs have changed).
866		1321
		1322	=item ev_tstamp offset [read-write]
		1323
		1324	When repeating, this contains the offset value, otherwise this is the
		1325	absolute point in time (the C<at> value passed to C<ev_periodic_set>).
		1326
		1327	Can be modified any time, but changes only take effect when the periodic
		1328	timer fires or C<ev_periodic_again> is being called.
		1329
		1330	=item ev_tstamp interval [read-write]
		1331
		1332	The current interval value. Can be modified any time, but changes only
		1333	take effect when the periodic timer fires or C<ev_periodic_again> is being
		1334	called.
		1335
		1336	=item ev_tstamp (*reschedule_cb)(struct ev_periodic *w, ev_tstamp now) [read-write]
		1337
		1338	The current reschedule callback, or C<0>, if this functionality is
		1339	switched off. Can be changed any time, but changes only take effect when
		1340	the periodic timer fires or C<ev_periodic_again> is being called.
		1341
		1342	=item ev_tstamp at [read-only]
		1343
		1344	When active, contains the absolute time that the watcher is supposed to
		1345	trigger next.
		1346
867	=back	1347	=back
868		1348
		1349	=head3 Examples
		1350
869	Example: call a callback every hour, or, more precisely, whenever the	1351	Example: Call a callback every hour, or, more precisely, whenever the
870	system clock is divisible by 3600. The callback invocation times have	1352	system clock is divisible by 3600. The callback invocation times have
871	potentially a lot of jittering, but good long-term stability.	1353	potentially a lot of jittering, but good long-term stability.
872		1354
873	static void	1355	static void
874	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)	1356	clock_cb (struct ev_loop *loop, struct ev_io *w, int revents)
…		…
878		1360
879	struct ev_periodic hourly_tick;	1361	struct ev_periodic hourly_tick;
880	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);	1362	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
881	ev_periodic_start (loop, &hourly_tick);	1363	ev_periodic_start (loop, &hourly_tick);
882		1364
883	Example: the same as above, but use a reschedule callback to do it:	1365	Example: The same as above, but use a reschedule callback to do it:
884		1366
885	#include <math.h>	1367	#include <math.h>
886		1368
887	static ev_tstamp	1369	static ev_tstamp
888	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)	1370	my_scheduler_cb (struct ev_periodic *w, ev_tstamp now)
…		…
890	return fmod (now, 3600.) + 3600.;	1372	return fmod (now, 3600.) + 3600.;
891	}	1373	}
892		1374
893	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);	1375	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
894		1376
895	Example: call a callback every hour, starting now:	1377	Example: Call a callback every hour, starting now:
896		1378
897	struct ev_periodic hourly_tick;	1379	struct ev_periodic hourly_tick;
898	ev_periodic_init (&hourly_tick, clock_cb,	1380	ev_periodic_init (&hourly_tick, clock_cb,
899	fmod (ev_now (loop), 3600.), 3600., 0);	1381	fmod (ev_now (loop), 3600.), 3600., 0);
900	ev_periodic_start (loop, &hourly_tick);	1382	ev_periodic_start (loop, &hourly_tick);
901		1383
902		1384
903	=head2 C<ev_signal> - signal me when a signal gets signalled	1385	=head2 C<ev_signal> - signal me when a signal gets signalled!
904		1386
905	Signal watchers will trigger an event when the process receives a specific	1387	Signal watchers will trigger an event when the process receives a specific
906	signal one or more times. Even though signals are very asynchronous, libev	1388	signal one or more times. Even though signals are very asynchronous, libev
907	will try it's best to deliver signals synchronously, i.e. as part of the	1389	will try it's best to deliver signals synchronously, i.e. as part of the
908	normal event processing, like any other event.	1390	normal event processing, like any other event.
…		…
912	with the kernel (thus it coexists with your own signal handlers as long	1394	with the kernel (thus it coexists with your own signal handlers as long
913	as you don't register any with libev). Similarly, when the last signal	1395	as you don't register any with libev). Similarly, when the last signal
914	watcher for a signal is stopped libev will reset the signal handler to	1396	watcher for a signal is stopped libev will reset the signal handler to
915	SIG_DFL (regardless of what it was set to before).	1397	SIG_DFL (regardless of what it was set to before).
916		1398
		1399	=head3 Watcher-Specific Functions and Data Members
		1400
917	=over 4	1401	=over 4
918		1402
919	=item ev_signal_init (ev_signal *, callback, int signum)	1403	=item ev_signal_init (ev_signal *, callback, int signum)
920		1404
921	=item ev_signal_set (ev_signal *, int signum)	1405	=item ev_signal_set (ev_signal *, int signum)
922		1406
923	Configures the watcher to trigger on the given signal number (usually one	1407	Configures the watcher to trigger on the given signal number (usually one
924	of the C<SIGxxx> constants).	1408	of the C<SIGxxx> constants).
925		1409
		1410	=item int signum [read-only]
		1411
		1412	The signal the watcher watches out for.
		1413
926	=back	1414	=back
927		1415
		1416
928	=head2 C<ev_child> - wait for pid status changes	1417	=head2 C<ev_child> - watch out for process status changes
929		1418
930	Child watchers trigger when your process receives a SIGCHLD in response to	1419	Child watchers trigger when your process receives a SIGCHLD in response to
931	some child status changes (most typically when a child of yours dies).	1420	some child status changes (most typically when a child of yours dies).
		1421
		1422	=head3 Watcher-Specific Functions and Data Members
932		1423
933	=over 4	1424	=over 4
934		1425
935	=item ev_child_init (ev_child *, callback, int pid)	1426	=item ev_child_init (ev_child *, callback, int pid)
936		1427
…		…
941	at the C<rstatus> member of the C<ev_child> watcher structure to see	1432	at the C<rstatus> member of the C<ev_child> watcher structure to see
942	the status word (use the macros from C<sys/wait.h> and see your systems	1433	the status word (use the macros from C<sys/wait.h> and see your systems
943	C<waitpid> documentation). The C<rpid> member contains the pid of the	1434	C<waitpid> documentation). The C<rpid> member contains the pid of the
944	process causing the status change.	1435	process causing the status change.
945		1436
		1437	=item int pid [read-only]
		1438
		1439	The process id this watcher watches out for, or C<0>, meaning any process id.
		1440
		1441	=item int rpid [read-write]
		1442
		1443	The process id that detected a status change.
		1444
		1445	=item int rstatus [read-write]
		1446
		1447	The process exit/trace status caused by C<rpid> (see your systems
		1448	C<waitpid> and C<sys/wait.h> documentation for details).
		1449
946	=back	1450	=back
947		1451
		1452	=head3 Examples
		1453
948	Example: try to exit cleanly on SIGINT and SIGTERM.	1454	Example: Try to exit cleanly on SIGINT and SIGTERM.
949		1455
950	static void	1456	static void
951	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)	1457	sigint_cb (struct ev_loop *loop, struct ev_signal *w, int revents)
952	{	1458	{
953	ev_unloop (loop, EVUNLOOP_ALL);	1459	ev_unloop (loop, EVUNLOOP_ALL);
…		…
956	struct ev_signal signal_watcher;	1462	struct ev_signal signal_watcher;
957	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1463	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
958	ev_signal_start (loop, &sigint_cb);	1464	ev_signal_start (loop, &sigint_cb);
959		1465
960		1466
		1467	=head2 C<ev_stat> - did the file attributes just change?
		1468
		1469	This watches a filesystem path for attribute changes. That is, it calls
		1470	C<stat> regularly (or when the OS says it changed) and sees if it changed
		1471	compared to the last time, invoking the callback if it did.
		1472
		1473	The path does not need to exist: changing from "path exists" to "path does
		1474	not exist" is a status change like any other. The condition "path does
		1475	not exist" is signified by the C<st_nlink> field being zero (which is
		1476	otherwise always forced to be at least one) and all the other fields of
		1477	the stat buffer having unspecified contents.
		1478
		1479	The path I<should> be absolute and I<must not> end in a slash. If it is
		1480	relative and your working directory changes, the behaviour is undefined.
		1481
		1482	Since there is no standard to do this, the portable implementation simply
		1483	calls C<stat (2)> regularly on the path to see if it changed somehow. You
		1484	can specify a recommended polling interval for this case. If you specify
		1485	a polling interval of C<0> (highly recommended!) then a I<suitable,
		1486	unspecified default> value will be used (which you can expect to be around
		1487	five seconds, although this might change dynamically). Libev will also
		1488	impose a minimum interval which is currently around C<0.1>, but thats
		1489	usually overkill.
		1490
		1491	This watcher type is not meant for massive numbers of stat watchers,
		1492	as even with OS-supported change notifications, this can be
		1493	resource-intensive.
		1494
		1495	At the time of this writing, only the Linux inotify interface is
		1496	implemented (implementing kqueue support is left as an exercise for the
		1497	reader). Inotify will be used to give hints only and should not change the
		1498	semantics of C<ev_stat> watchers, which means that libev sometimes needs
		1499	to fall back to regular polling again even with inotify, but changes are
		1500	usually detected immediately, and if the file exists there will be no
		1501	polling.
		1502
		1503	=head3 Inotify
		1504
		1505	When C<inotify (7)> support has been compiled into libev (generally only
		1506	available on Linux) and present at runtime, it will be used to speed up
		1507	change detection where possible. The inotify descriptor will be created lazily
		1508	when the first C<ev_stat> watcher is being started.
		1509
		1510	Inotify presense does not change the semantics of C<ev_stat> watchers
		1511	except that changes might be detected earlier, and in some cases, to avoid
		1512	making regular C<stat> calls. Even in the presense of inotify support
		1513	there are many cases where libev has to resort to regular C<stat> polling.
		1514
		1515	(There is no support for kqueue, as apparently it cannot be used to
		1516	implement this functionality, due to the requirement of having a file
		1517	descriptor open on the object at all times).
		1518
		1519	=head3 The special problem of stat time resolution
		1520
		1521	The C<stat ()> syscall only supports full-second resolution portably, and
		1522	even on systems where the resolution is higher, many filesystems still
		1523	only support whole seconds.
		1524
		1525	That means that, if the time is the only thing that changes, you might
		1526	miss updates: on the first update, C<ev_stat> detects a change and calls
		1527	your callback, which does something. When there is another update within
		1528	the same second, C<ev_stat> will be unable to detect it.
		1529
		1530	The solution to this is to delay acting on a change for a second (or till
		1531	the next second boundary), using a roughly one-second delay C<ev_timer>
		1532	(C<ev_timer_set (w, 0., 1.01); ev_timer_again (loop, w)>). The C<.01>
		1533	is added to work around small timing inconsistencies of some operating
		1534	systems.
		1535
		1536	=head3 Watcher-Specific Functions and Data Members
		1537
		1538	=over 4
		1539
		1540	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
		1541
		1542	=item ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)
		1543
		1544	Configures the watcher to wait for status changes of the given
		1545	C<path>. The C<interval> is a hint on how quickly a change is expected to
		1546	be detected and should normally be specified as C<0> to let libev choose
		1547	a suitable value. The memory pointed to by C<path> must point to the same
		1548	path for as long as the watcher is active.
		1549
		1550	The callback will be receive C<EV_STAT> when a change was detected,
		1551	relative to the attributes at the time the watcher was started (or the
		1552	last change was detected).
		1553
		1554	=item ev_stat_stat (ev_stat *)
		1555
		1556	Updates the stat buffer immediately with new values. If you change the
		1557	watched path in your callback, you could call this fucntion to avoid
		1558	detecting this change (while introducing a race condition). Can also be
		1559	useful simply to find out the new values.
		1560
		1561	=item ev_statdata attr [read-only]
		1562
		1563	The most-recently detected attributes of the file. Although the type is of
		1564	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
		1565	suitable for your system. If the C<st_nlink> member is C<0>, then there
		1566	was some error while C<stat>ing the file.
		1567
		1568	=item ev_statdata prev [read-only]
		1569
		1570	The previous attributes of the file. The callback gets invoked whenever
		1571	C<prev> != C<attr>.
		1572
		1573	=item ev_tstamp interval [read-only]
		1574
		1575	The specified interval.
		1576
		1577	=item const char *path [read-only]
		1578
		1579	The filesystem path that is being watched.
		1580
		1581	=back
		1582
		1583	=head3 Examples
		1584
		1585	Example: Watch C</etc/passwd> for attribute changes.
		1586
		1587	static void
		1588	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
		1589	{
		1590	/* /etc/passwd changed in some way */
		1591	if (w->attr.st_nlink)
		1592	{
		1593	printf ("passwd current size %ld\n", (long)w->attr.st_size);
		1594	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
		1595	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
		1596	}
		1597	else
		1598	/* you shalt not abuse printf for puts */
		1599	puts ("wow, /etc/passwd is not there, expect problems. "
		1600	"if this is windows, they already arrived\n");
		1601	}
		1602
		1603	...
		1604	ev_stat passwd;
		1605
		1606	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
		1607	ev_stat_start (loop, &passwd);
		1608
		1609	Example: Like above, but additionally use a one-second delay so we do not
		1610	miss updates (however, frequent updates will delay processing, too, so
		1611	one might do the work both on C<ev_stat> callback invocation I<and> on
		1612	C<ev_timer> callback invocation).
		1613
		1614	static ev_stat passwd;
		1615	static ev_timer timer;
		1616
		1617	static void
		1618	timer_cb (EV_P_ ev_timer *w, int revents)
		1619	{
		1620	ev_timer_stop (EV_A_ w);
		1621
		1622	/* now it's one second after the most recent passwd change */
		1623	}
		1624
		1625	static void
		1626	stat_cb (EV_P_ ev_stat *w, int revents)
		1627	{
		1628	/* reset the one-second timer */
		1629	ev_timer_again (EV_A_ &timer);
		1630	}
		1631
		1632	...
		1633	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		1634	ev_stat_start (loop, &passwd);
		1635	ev_timer_init (&timer, timer_cb, 0., 1.01);
		1636
		1637
961	=head2 C<ev_idle> - when you've got nothing better to do	1638	=head2 C<ev_idle> - when you've got nothing better to do...
962		1639
963	Idle watchers trigger events when there are no other events are pending	1640	Idle watchers trigger events when no other events of the same or higher
964	(prepare, check and other idle watchers do not count). That is, as long	1641	priority are pending (prepare, check and other idle watchers do not
965	as your process is busy handling sockets or timeouts (or even signals,	1642	count).
966	imagine) it will not be triggered. But when your process is idle all idle	1643
967	watchers are being called again and again, once per event loop iteration -	1644	That is, as long as your process is busy handling sockets or timeouts
		1645	(or even signals, imagine) of the same or higher priority it will not be
		1646	triggered. But when your process is idle (or only lower-priority watchers
		1647	are pending), the idle watchers are being called once per event loop
968	until stopped, that is, or your process receives more events and becomes	1648	iteration - until stopped, that is, or your process receives more events
969	busy.	1649	and becomes busy again with higher priority stuff.
970		1650
971	The most noteworthy effect is that as long as any idle watchers are	1651	The most noteworthy effect is that as long as any idle watchers are
972	active, the process will not block when waiting for new events.	1652	active, the process will not block when waiting for new events.
973		1653
974	Apart from keeping your process non-blocking (which is a useful	1654	Apart from keeping your process non-blocking (which is a useful
975	effect on its own sometimes), idle watchers are a good place to do	1655	effect on its own sometimes), idle watchers are a good place to do
976	"pseudo-background processing", or delay processing stuff to after the	1656	"pseudo-background processing", or delay processing stuff to after the
977	event loop has handled all outstanding events.	1657	event loop has handled all outstanding events.
978		1658
		1659	=head3 Watcher-Specific Functions and Data Members
		1660
979	=over 4	1661	=over 4
980		1662
981	=item ev_idle_init (ev_signal *, callback)	1663	=item ev_idle_init (ev_signal *, callback)
982		1664
983	Initialises and configures the idle watcher - it has no parameters of any	1665	Initialises and configures the idle watcher - it has no parameters of any
984	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1666	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
985	believe me.	1667	believe me.
986		1668
987	=back	1669	=back
988		1670
		1671	=head3 Examples
		1672
989	Example: dynamically allocate an C<ev_idle>, start it, and in the	1673	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
990	callback, free it. Alos, use no error checking, as usual.	1674	callback, free it. Also, use no error checking, as usual.
991		1675
992	static void	1676	static void
993	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)	1677	idle_cb (struct ev_loop *loop, struct ev_idle *w, int revents)
994	{	1678	{
995	free (w);	1679	free (w);
…		…
1000	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1684	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1001	ev_idle_init (idle_watcher, idle_cb);	1685	ev_idle_init (idle_watcher, idle_cb);
1002	ev_idle_start (loop, idle_cb);	1686	ev_idle_start (loop, idle_cb);
1003		1687
1004		1688
1005	=head2 C<ev_prepare> and C<ev_check> - customise your event loop	1689	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
1006		1690
1007	Prepare and check watchers are usually (but not always) used in tandem:	1691	Prepare and check watchers are usually (but not always) used in tandem:
1008	prepare watchers get invoked before the process blocks and check watchers	1692	prepare watchers get invoked before the process blocks and check watchers
1009	afterwards.	1693	afterwards.
1010		1694
		1695	You I<must not> call C<ev_loop> or similar functions that enter
		1696	the current event loop from either C<ev_prepare> or C<ev_check>
		1697	watchers. Other loops than the current one are fine, however. The
		1698	rationale behind this is that you do not need to check for recursion in
		1699	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
		1700	C<ev_check> so if you have one watcher of each kind they will always be
		1701	called in pairs bracketing the blocking call.
		1702
1011	Their main purpose is to integrate other event mechanisms into libev. This	1703	Their main purpose is to integrate other event mechanisms into libev and
1012	could be used, for example, to track variable changes, implement your own	1704	their use is somewhat advanced. This could be used, for example, to track
1013	watchers, integrate net-snmp or a coroutine library and lots more.	1705	variable changes, implement your own watchers, integrate net-snmp or a
		1706	coroutine library and lots more. They are also occasionally useful if
		1707	you cache some data and want to flush it before blocking (for example,
		1708	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
		1709	watcher).
1014		1710
1015	This is done by examining in each prepare call which file descriptors need	1711	This is done by examining in each prepare call which file descriptors need
1016	to be watched by the other library, registering C<ev_io> watchers for	1712	to be watched by the other library, registering C<ev_io> watchers for
1017	them and starting an C<ev_timer> watcher for any timeouts (many libraries	1713	them and starting an C<ev_timer> watcher for any timeouts (many libraries
1018	provide just this functionality). Then, in the check watcher you check for	1714	provide just this functionality). Then, in the check watcher you check for
…		…
1028	with priority higher than or equal to the event loop and one coroutine	1724	with priority higher than or equal to the event loop and one coroutine
1029	of lower priority, but only once, using idle watchers to keep the event	1725	of lower priority, but only once, using idle watchers to keep the event
1030	loop from blocking if lower-priority coroutines are active, thus mapping	1726	loop from blocking if lower-priority coroutines are active, thus mapping
1031	low-priority coroutines to idle/background tasks).	1727	low-priority coroutines to idle/background tasks).
1032		1728
		1729	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		1730	priority, to ensure that they are being run before any other watchers
		1731	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,
		1732	too) should not activate ("feed") events into libev. While libev fully
		1733	supports this, they will be called before other C<ev_check> watchers
		1734	did their job. As C<ev_check> watchers are often used to embed other
		1735	(non-libev) event loops those other event loops might be in an unusable
		1736	state until their C<ev_check> watcher ran (always remind yourself to
		1737	coexist peacefully with others).
		1738
		1739	=head3 Watcher-Specific Functions and Data Members
		1740
1033	=over 4	1741	=over 4
1034		1742
1035	=item ev_prepare_init (ev_prepare *, callback)	1743	=item ev_prepare_init (ev_prepare *, callback)
1036		1744
1037	=item ev_check_init (ev_check *, callback)	1745	=item ev_check_init (ev_check *, callback)
…		…
1040	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1748	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1041	macros, but using them is utterly, utterly and completely pointless.	1749	macros, but using them is utterly, utterly and completely pointless.
1042		1750
1043	=back	1751	=back
1044		1752
1045	Example: *TODO*.	1753	=head3 Examples
		1754
		1755	There are a number of principal ways to embed other event loops or modules
		1756	into libev. Here are some ideas on how to include libadns into libev
		1757	(there is a Perl module named C<EV::ADNS> that does this, which you could
		1758	use for an actually working example. Another Perl module named C<EV::Glib>
		1759	embeds a Glib main context into libev, and finally, C<Glib::EV> embeds EV
		1760	into the Glib event loop).
		1761
		1762	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
		1763	and in a check watcher, destroy them and call into libadns. What follows
		1764	is pseudo-code only of course. This requires you to either use a low
		1765	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		1766	the callbacks for the IO/timeout watchers might not have been called yet.
		1767
		1768	static ev_io iow [nfd];
		1769	static ev_timer tw;
		1770
		1771	static void
		1772	io_cb (ev_loop *loop, ev_io *w, int revents)
		1773	{
		1774	}
		1775
		1776	// create io watchers for each fd and a timer before blocking
		1777	static void
		1778	adns_prepare_cb (ev_loop *loop, ev_prepare *w, int revents)
		1779	{
		1780	int timeout = 3600000;
		1781	struct pollfd fds [nfd];
		1782	// actual code will need to loop here and realloc etc.
		1783	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
		1784
		1785	/* the callback is illegal, but won't be called as we stop during check */
		1786	ev_timer_init (&tw, 0, timeout * 1e-3);
		1787	ev_timer_start (loop, &tw);
		1788
		1789	// create one ev_io per pollfd
		1790	for (int i = 0; i < nfd; ++i)
		1791	{
		1792	ev_io_init (iow + i, io_cb, fds [i].fd,
		1793	((fds [i].events & POLLIN ? EV_READ : 0)
		1794	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
		1795
		1796	fds [i].revents = 0;
		1797	ev_io_start (loop, iow + i);
		1798	}
		1799	}
		1800
		1801	// stop all watchers after blocking
		1802	static void
		1803	adns_check_cb (ev_loop *loop, ev_check *w, int revents)
		1804	{
		1805	ev_timer_stop (loop, &tw);
		1806
		1807	for (int i = 0; i < nfd; ++i)
		1808	{
		1809	// set the relevant poll flags
		1810	// could also call adns_processreadable etc. here
		1811	struct pollfd *fd = fds + i;
		1812	int revents = ev_clear_pending (iow + i);
		1813	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		1814	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		1815
		1816	// now stop the watcher
		1817	ev_io_stop (loop, iow + i);
		1818	}
		1819
		1820	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
		1821	}
		1822
		1823	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		1824	in the prepare watcher and would dispose of the check watcher.
		1825
		1826	Method 3: If the module to be embedded supports explicit event
		1827	notification (adns does), you can also make use of the actual watcher
		1828	callbacks, and only destroy/create the watchers in the prepare watcher.
		1829
		1830	static void
		1831	timer_cb (EV_P_ ev_timer *w, int revents)
		1832	{
		1833	adns_state ads = (adns_state)w->data;
		1834	update_now (EV_A);
		1835
		1836	adns_processtimeouts (ads, &tv_now);
		1837	}
		1838
		1839	static void
		1840	io_cb (EV_P_ ev_io *w, int revents)
		1841	{
		1842	adns_state ads = (adns_state)w->data;
		1843	update_now (EV_A);
		1844
		1845	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		1846	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		1847	}
		1848
		1849	// do not ever call adns_afterpoll
		1850
		1851	Method 4: Do not use a prepare or check watcher because the module you
		1852	want to embed is too inflexible to support it. Instead, youc na override
		1853	their poll function. The drawback with this solution is that the main
		1854	loop is now no longer controllable by EV. The C<Glib::EV> module does
		1855	this.
		1856
		1857	static gint
		1858	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		1859	{
		1860	int got_events = 0;
		1861
		1862	for (n = 0; n < nfds; ++n)
		1863	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		1864
		1865	if (timeout >= 0)
		1866	// create/start timer
		1867
		1868	// poll
		1869	ev_loop (EV_A_ 0);
		1870
		1871	// stop timer again
		1872	if (timeout >= 0)
		1873	ev_timer_stop (EV_A_ &to);
		1874
		1875	// stop io watchers again - their callbacks should have set
		1876	for (n = 0; n < nfds; ++n)
		1877	ev_io_stop (EV_A_ iow [n]);
		1878
		1879	return got_events;
		1880	}
		1881
		1882
		1883	=head2 C<ev_embed> - when one backend isn't enough...
		1884
		1885	This is a rather advanced watcher type that lets you embed one event loop
		1886	into another (currently only C<ev_io> events are supported in the embedded
		1887	loop, other types of watchers might be handled in a delayed or incorrect
		1888	fashion and must not be used).
		1889
		1890	There are primarily two reasons you would want that: work around bugs and
		1891	prioritise I/O.
		1892
		1893	As an example for a bug workaround, the kqueue backend might only support
		1894	sockets on some platform, so it is unusable as generic backend, but you
		1895	still want to make use of it because you have many sockets and it scales
		1896	so nicely. In this case, you would create a kqueue-based loop and embed it
		1897	into your default loop (which might use e.g. poll). Overall operation will
		1898	be a bit slower because first libev has to poll and then call kevent, but
		1899	at least you can use both at what they are best.
		1900
		1901	As for prioritising I/O: rarely you have the case where some fds have
		1902	to be watched and handled very quickly (with low latency), and even
		1903	priorities and idle watchers might have too much overhead. In this case
		1904	you would put all the high priority stuff in one loop and all the rest in
		1905	a second one, and embed the second one in the first.
		1906
		1907	As long as the watcher is active, the callback will be invoked every time
		1908	there might be events pending in the embedded loop. The callback must then
		1909	call C<ev_embed_sweep (mainloop, watcher)> to make a single sweep and invoke
		1910	their callbacks (you could also start an idle watcher to give the embedded
		1911	loop strictly lower priority for example). You can also set the callback
		1912	to C<0>, in which case the embed watcher will automatically execute the
		1913	embedded loop sweep.
		1914
		1915	As long as the watcher is started it will automatically handle events. The
		1916	callback will be invoked whenever some events have been handled. You can
		1917	set the callback to C<0> to avoid having to specify one if you are not
		1918	interested in that.
		1919
		1920	Also, there have not currently been made special provisions for forking:
		1921	when you fork, you not only have to call C<ev_loop_fork> on both loops,
		1922	but you will also have to stop and restart any C<ev_embed> watchers
		1923	yourself.
		1924
		1925	Unfortunately, not all backends are embeddable, only the ones returned by
		1926	C<ev_embeddable_backends> are, which, unfortunately, does not include any
		1927	portable one.
		1928
		1929	So when you want to use this feature you will always have to be prepared
		1930	that you cannot get an embeddable loop. The recommended way to get around
		1931	this is to have a separate variables for your embeddable loop, try to
		1932	create it, and if that fails, use the normal loop for everything.
		1933
		1934	=head3 Watcher-Specific Functions and Data Members
		1935
		1936	=over 4
		1937
		1938	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
		1939
		1940	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)
		1941
		1942	Configures the watcher to embed the given loop, which must be
		1943	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		1944	invoked automatically, otherwise it is the responsibility of the callback
		1945	to invoke it (it will continue to be called until the sweep has been done,
		1946	if you do not want thta, you need to temporarily stop the embed watcher).
		1947
		1948	=item ev_embed_sweep (loop, ev_embed *)
		1949
		1950	Make a single, non-blocking sweep over the embedded loop. This works
		1951	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		1952	apropriate way for embedded loops.
		1953
		1954	=item struct ev_loop *other [read-only]
		1955
		1956	The embedded event loop.
		1957
		1958	=back
		1959
		1960	=head3 Examples
		1961
		1962	Example: Try to get an embeddable event loop and embed it into the default
		1963	event loop. If that is not possible, use the default loop. The default
		1964	loop is stored in C<loop_hi>, while the mebeddable loop is stored in
		1965	C<loop_lo> (which is C<loop_hi> in the acse no embeddable loop can be
		1966	used).
		1967
		1968	struct ev_loop *loop_hi = ev_default_init (0);
		1969	struct ev_loop *loop_lo = 0;
		1970	struct ev_embed embed;
		1971
		1972	// see if there is a chance of getting one that works
		1973	// (remember that a flags value of 0 means autodetection)
		1974	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		1975	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		1976	: 0;
		1977
		1978	// if we got one, then embed it, otherwise default to loop_hi
		1979	if (loop_lo)
		1980	{
		1981	ev_embed_init (&embed, 0, loop_lo);
		1982	ev_embed_start (loop_hi, &embed);
		1983	}
		1984	else
		1985	loop_lo = loop_hi;
		1986
		1987	Example: Check if kqueue is available but not recommended and create
		1988	a kqueue backend for use with sockets (which usually work with any
		1989	kqueue implementation). Store the kqueue/socket-only event loop in
		1990	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		1991
		1992	struct ev_loop *loop = ev_default_init (0);
		1993	struct ev_loop *loop_socket = 0;
		1994	struct ev_embed embed;
		1995
		1996	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		1997	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		1998	{
		1999	ev_embed_init (&embed, 0, loop_socket);
		2000	ev_embed_start (loop, &embed);
		2001	}
		2002
		2003	if (!loop_socket)
		2004	loop_socket = loop;
		2005
		2006	// now use loop_socket for all sockets, and loop for everything else
		2007
		2008
		2009	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
		2010
		2011	Fork watchers are called when a C<fork ()> was detected (usually because
		2012	whoever is a good citizen cared to tell libev about it by calling
		2013	C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the
		2014	event loop blocks next and before C<ev_check> watchers are being called,
		2015	and only in the child after the fork. If whoever good citizen calling
		2016	C<ev_default_fork> cheats and calls it in the wrong process, the fork
		2017	handlers will be invoked, too, of course.
		2018
		2019	=head3 Watcher-Specific Functions and Data Members
		2020
		2021	=over 4
		2022
		2023	=item ev_fork_init (ev_signal *, callback)
		2024
		2025	Initialises and configures the fork watcher - it has no parameters of any
		2026	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
		2027	believe me.
		2028
		2029	=back
1046		2030
1047		2031
1048	=head1 OTHER FUNCTIONS	2032	=head1 OTHER FUNCTIONS
1049		2033
1050	There are some other functions of possible interest. Described. Here. Now.	2034	There are some other functions of possible interest. Described. Here. Now.
…		…
1081	/* stdin might have data for us, joy! */;	2065	/* stdin might have data for us, joy! */;
1082	}	2066	}
1083		2067
1084	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2068	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
1085		2069
1086	=item ev_feed_event (loop, watcher, int events)	2070	=item ev_feed_event (ev_loop *, watcher *, int revents)
1087		2071
1088	Feeds the given event set into the event loop, as if the specified event	2072	Feeds the given event set into the event loop, as if the specified event
1089	had happened for the specified watcher (which must be a pointer to an	2073	had happened for the specified watcher (which must be a pointer to an
1090	initialised but not necessarily started event watcher).	2074	initialised but not necessarily started event watcher).
1091		2075
1092	=item ev_feed_fd_event (loop, int fd, int revents)	2076	=item ev_feed_fd_event (ev_loop *, int fd, int revents)
1093		2077
1094	Feed an event on the given fd, as if a file descriptor backend detected	2078	Feed an event on the given fd, as if a file descriptor backend detected
1095	the given events it.	2079	the given events it.
1096		2080
1097	=item ev_feed_signal_event (loop, int signum)	2081	=item ev_feed_signal_event (ev_loop *loop, int signum)
1098		2082
1099	Feed an event as if the given signal occured (loop must be the default loop!).	2083	Feed an event as if the given signal occured (C<loop> must be the default
		2084	loop!).
1100		2085
1101	=back	2086	=back
1102		2087
1103		2088
1104	=head1 LIBEVENT EMULATION	2089	=head1 LIBEVENT EMULATION
…		…
1128		2113
1129	=back	2114	=back
1130		2115
1131	=head1 C++ SUPPORT	2116	=head1 C++ SUPPORT
1132		2117
1133	TBD.	2118	Libev comes with some simplistic wrapper classes for C++ that mainly allow
		2119	you to use some convinience methods to start/stop watchers and also change
		2120	the callback model to a model using method callbacks on objects.
		2121
		2122	To use it,
		2123
		2124	#include <ev++.h>
		2125
		2126	This automatically includes F<ev.h> and puts all of its definitions (many
		2127	of them macros) into the global namespace. All C++ specific things are
		2128	put into the C<ev> namespace. It should support all the same embedding
		2129	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
		2130
		2131	Care has been taken to keep the overhead low. The only data member the C++
		2132	classes add (compared to plain C-style watchers) is the event loop pointer
		2133	that the watcher is associated with (or no additional members at all if
		2134	you disable C<EV_MULTIPLICITY> when embedding libev).
		2135
		2136	Currently, functions, and static and non-static member functions can be
		2137	used as callbacks. Other types should be easy to add as long as they only
		2138	need one additional pointer for context. If you need support for other
		2139	types of functors please contact the author (preferably after implementing
		2140	it).
		2141
		2142	Here is a list of things available in the C<ev> namespace:
		2143
		2144	=over 4
		2145
		2146	=item C<ev::READ>, C<ev::WRITE> etc.
		2147
		2148	These are just enum values with the same values as the C<EV_READ> etc.
		2149	macros from F<ev.h>.
		2150
		2151	=item C<ev::tstamp>, C<ev::now>
		2152
		2153	Aliases to the same types/functions as with the C<ev_> prefix.
		2154
		2155	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.
		2156
		2157	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of
		2158	the same name in the C<ev> namespace, with the exception of C<ev_signal>
		2159	which is called C<ev::sig> to avoid clashes with the C<signal> macro
		2160	defines by many implementations.
		2161
		2162	All of those classes have these methods:
		2163
		2164	=over 4
		2165
		2166	=item ev::TYPE::TYPE ()
		2167
		2168	=item ev::TYPE::TYPE (struct ev_loop *)
		2169
		2170	=item ev::TYPE::~TYPE
		2171
		2172	The constructor (optionally) takes an event loop to associate the watcher
		2173	with. If it is omitted, it will use C<EV_DEFAULT>.
		2174
		2175	The constructor calls C<ev_init> for you, which means you have to call the
		2176	C<set> method before starting it.
		2177
		2178	It will not set a callback, however: You have to call the templated C<set>
		2179	method to set a callback before you can start the watcher.
		2180
		2181	(The reason why you have to use a method is a limitation in C++ which does
		2182	not allow explicit template arguments for constructors).
		2183
		2184	The destructor automatically stops the watcher if it is active.
		2185
		2186	=item w->set<class, &class::method> (object *)
		2187
		2188	This method sets the callback method to call. The method has to have a
		2189	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		2190	first argument and the C<revents> as second. The object must be given as
		2191	parameter and is stored in the C<data> member of the watcher.
		2192
		2193	This method synthesizes efficient thunking code to call your method from
		2194	the C callback that libev requires. If your compiler can inline your
		2195	callback (i.e. it is visible to it at the place of the C<set> call and
		2196	your compiler is good :), then the method will be fully inlined into the
		2197	thunking function, making it as fast as a direct C callback.
		2198
		2199	Example: simple class declaration and watcher initialisation
		2200
		2201	struct myclass
		2202	{
		2203	void io_cb (ev::io &w, int revents) { }
		2204	}
		2205
		2206	myclass obj;
		2207	ev::io iow;
		2208	iow.set <myclass, &myclass::io_cb> (&obj);
		2209
		2210	=item w->set<function> (void *data = 0)
		2211
		2212	Also sets a callback, but uses a static method or plain function as
		2213	callback. The optional C<data> argument will be stored in the watcher's
		2214	C<data> member and is free for you to use.
		2215
		2216	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		2217
		2218	See the method-C<set> above for more details.
		2219
		2220	Example:
		2221
		2222	static void io_cb (ev::io &w, int revents) { }
		2223	iow.set <io_cb> ();
		2224
		2225	=item w->set (struct ev_loop *)
		2226
		2227	Associates a different C<struct ev_loop> with this watcher. You can only
		2228	do this when the watcher is inactive (and not pending either).
		2229
		2230	=item w->set ([args])
		2231
		2232	Basically the same as C<ev_TYPE_set>, with the same args. Must be
		2233	called at least once. Unlike the C counterpart, an active watcher gets
		2234	automatically stopped and restarted when reconfiguring it with this
		2235	method.
		2236
		2237	=item w->start ()
		2238
		2239	Starts the watcher. Note that there is no C<loop> argument, as the
		2240	constructor already stores the event loop.
		2241
		2242	=item w->stop ()
		2243
		2244	Stops the watcher if it is active. Again, no C<loop> argument.
		2245
		2246	=item w->again () (C<ev::timer>, C<ev::periodic> only)
		2247
		2248	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
		2249	C<ev_TYPE_again> function.
		2250
		2251	=item w->sweep () (C<ev::embed> only)
		2252
		2253	Invokes C<ev_embed_sweep>.
		2254
		2255	=item w->update () (C<ev::stat> only)
		2256
		2257	Invokes C<ev_stat_stat>.
		2258
		2259	=back
		2260
		2261	=back
		2262
		2263	Example: Define a class with an IO and idle watcher, start one of them in
		2264	the constructor.
		2265
		2266	class myclass
		2267	{
		2268	ev_io io; void io_cb (ev::io &w, int revents);
		2269	ev_idle idle void idle_cb (ev::idle &w, int revents);
		2270
		2271	myclass ();
		2272	}
		2273
		2274	myclass::myclass (int fd)
		2275	{
		2276	io .set <myclass, &myclass::io_cb > (this);
		2277	idle.set <myclass, &myclass::idle_cb> (this);
		2278
		2279	io.start (fd, ev::READ);
		2280	}
		2281
		2282
		2283	=head1 MACRO MAGIC
		2284
		2285	Libev can be compiled with a variety of options, the most fundamantal
		2286	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
		2287	functions and callbacks have an initial C<struct ev_loop *> argument.
		2288
		2289	To make it easier to write programs that cope with either variant, the
		2290	following macros are defined:
		2291
		2292	=over 4
		2293
		2294	=item C<EV_A>, C<EV_A_>
		2295
		2296	This provides the loop I<argument> for functions, if one is required ("ev
		2297	loop argument"). The C<EV_A> form is used when this is the sole argument,
		2298	C<EV_A_> is used when other arguments are following. Example:
		2299
		2300	ev_unref (EV_A);
		2301	ev_timer_add (EV_A_ watcher);
		2302	ev_loop (EV_A_ 0);
		2303
		2304	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
		2305	which is often provided by the following macro.
		2306
		2307	=item C<EV_P>, C<EV_P_>
		2308
		2309	This provides the loop I<parameter> for functions, if one is required ("ev
		2310	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
		2311	C<EV_P_> is used when other parameters are following. Example:
		2312
		2313	// this is how ev_unref is being declared
		2314	static void ev_unref (EV_P);
		2315
		2316	// this is how you can declare your typical callback
		2317	static void cb (EV_P_ ev_timer *w, int revents)
		2318
		2319	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
		2320	suitable for use with C<EV_A>.
		2321
		2322	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
		2323
		2324	Similar to the other two macros, this gives you the value of the default
		2325	loop, if multiple loops are supported ("ev loop default").
		2326
		2327	=back
		2328
		2329	Example: Declare and initialise a check watcher, utilising the above
		2330	macros so it will work regardless of whether multiple loops are supported
		2331	or not.
		2332
		2333	static void
		2334	check_cb (EV_P_ ev_timer *w, int revents)
		2335	{
		2336	ev_check_stop (EV_A_ w);
		2337	}
		2338
		2339	ev_check check;
		2340	ev_check_init (&check, check_cb);
		2341	ev_check_start (EV_DEFAULT_ &check);
		2342	ev_loop (EV_DEFAULT_ 0);
		2343
		2344	=head1 EMBEDDING
		2345
		2346	Libev can (and often is) directly embedded into host
		2347	applications. Examples of applications that embed it include the Deliantra
		2348	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
		2349	and rxvt-unicode.
		2350
		2351	The goal is to enable you to just copy the necessary files into your
		2352	source directory without having to change even a single line in them, so
		2353	you can easily upgrade by simply copying (or having a checked-out copy of
		2354	libev somewhere in your source tree).
		2355
		2356	=head2 FILESETS
		2357
		2358	Depending on what features you need you need to include one or more sets of files
		2359	in your app.
		2360
		2361	=head3 CORE EVENT LOOP
		2362
		2363	To include only the libev core (all the C<ev_*> functions), with manual
		2364	configuration (no autoconf):
		2365
		2366	#define EV_STANDALONE 1
		2367	#include "ev.c"
		2368
		2369	This will automatically include F<ev.h>, too, and should be done in a
		2370	single C source file only to provide the function implementations. To use
		2371	it, do the same for F<ev.h> in all files wishing to use this API (best
		2372	done by writing a wrapper around F<ev.h> that you can include instead and
		2373	where you can put other configuration options):
		2374
		2375	#define EV_STANDALONE 1
		2376	#include "ev.h"
		2377
		2378	Both header files and implementation files can be compiled with a C++
		2379	compiler (at least, thats a stated goal, and breakage will be treated
		2380	as a bug).
		2381
		2382	You need the following files in your source tree, or in a directory
		2383	in your include path (e.g. in libev/ when using -Ilibev):
		2384
		2385	ev.h
		2386	ev.c
		2387	ev_vars.h
		2388	ev_wrap.h
		2389
		2390	ev_win32.c required on win32 platforms only
		2391
		2392	ev_select.c only when select backend is enabled (which is enabled by default)
		2393	ev_poll.c only when poll backend is enabled (disabled by default)
		2394	ev_epoll.c only when the epoll backend is enabled (disabled by default)
		2395	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
		2396	ev_port.c only when the solaris port backend is enabled (disabled by default)
		2397
		2398	F<ev.c> includes the backend files directly when enabled, so you only need
		2399	to compile this single file.
		2400
		2401	=head3 LIBEVENT COMPATIBILITY API
		2402
		2403	To include the libevent compatibility API, also include:
		2404
		2405	#include "event.c"
		2406
		2407	in the file including F<ev.c>, and:
		2408
		2409	#include "event.h"
		2410
		2411	in the files that want to use the libevent API. This also includes F<ev.h>.
		2412
		2413	You need the following additional files for this:
		2414
		2415	event.h
		2416	event.c
		2417
		2418	=head3 AUTOCONF SUPPORT
		2419
		2420	Instead of using C<EV_STANDALONE=1> and providing your config in
		2421	whatever way you want, you can also C<m4_include([libev.m4])> in your
		2422	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
		2423	include F<config.h> and configure itself accordingly.
		2424
		2425	For this of course you need the m4 file:
		2426
		2427	libev.m4
		2428
		2429	=head2 PREPROCESSOR SYMBOLS/MACROS
		2430
		2431	Libev can be configured via a variety of preprocessor symbols you have to define
		2432	before including any of its files. The default is not to build for multiplicity
		2433	and only include the select backend.
		2434
		2435	=over 4
		2436
		2437	=item EV_STANDALONE
		2438
		2439	Must always be C<1> if you do not use autoconf configuration, which
		2440	keeps libev from including F<config.h>, and it also defines dummy
		2441	implementations for some libevent functions (such as logging, which is not
		2442	supported). It will also not define any of the structs usually found in
		2443	F<event.h> that are not directly supported by the libev core alone.
		2444
		2445	=item EV_USE_MONOTONIC
		2446
		2447	If defined to be C<1>, libev will try to detect the availability of the
		2448	monotonic clock option at both compiletime and runtime. Otherwise no use
		2449	of the monotonic clock option will be attempted. If you enable this, you
		2450	usually have to link against librt or something similar. Enabling it when
		2451	the functionality isn't available is safe, though, although you have
		2452	to make sure you link against any libraries where the C<clock_gettime>
		2453	function is hiding in (often F<-lrt>).
		2454
		2455	=item EV_USE_REALTIME
		2456
		2457	If defined to be C<1>, libev will try to detect the availability of the
		2458	realtime clock option at compiletime (and assume its availability at
		2459	runtime if successful). Otherwise no use of the realtime clock option will
		2460	be attempted. This effectively replaces C<gettimeofday> by C<clock_get
		2461	(CLOCK_REALTIME, ...)> and will not normally affect correctness. See the
		2462	note about libraries in the description of C<EV_USE_MONOTONIC>, though.
		2463
		2464	=item EV_USE_NANOSLEEP
		2465
		2466	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		2467	and will use it for delays. Otherwise it will use C<select ()>.
		2468
		2469	=item EV_USE_SELECT
		2470
		2471	If undefined or defined to be C<1>, libev will compile in support for the
		2472	C<select>(2) backend. No attempt at autodetection will be done: if no
		2473	other method takes over, select will be it. Otherwise the select backend
		2474	will not be compiled in.
		2475
		2476	=item EV_SELECT_USE_FD_SET
		2477
		2478	If defined to C<1>, then the select backend will use the system C<fd_set>
		2479	structure. This is useful if libev doesn't compile due to a missing
		2480	C<NFDBITS> or C<fd_mask> definition or it misguesses the bitset layout on
		2481	exotic systems. This usually limits the range of file descriptors to some
		2482	low limit such as 1024 or might have other limitations (winsocket only
		2483	allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation, might
		2484	influence the size of the C<fd_set> used.
		2485
		2486	=item EV_SELECT_IS_WINSOCKET
		2487
		2488	When defined to C<1>, the select backend will assume that
		2489	select/socket/connect etc. don't understand file descriptors but
		2490	wants osf handles on win32 (this is the case when the select to
		2491	be used is the winsock select). This means that it will call
		2492	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
		2493	it is assumed that all these functions actually work on fds, even
		2494	on win32. Should not be defined on non-win32 platforms.
		2495
		2496	=item EV_FD_TO_WIN32_HANDLE
		2497
		2498	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		2499	file descriptors to socket handles. When not defining this symbol (the
		2500	default), then libev will call C<_get_osfhandle>, which is usually
		2501	correct. In some cases, programs use their own file descriptor management,
		2502	in which case they can provide this function to map fds to socket handles.
		2503
		2504	=item EV_USE_POLL
		2505
		2506	If defined to be C<1>, libev will compile in support for the C<poll>(2)
		2507	backend. Otherwise it will be enabled on non-win32 platforms. It
		2508	takes precedence over select.
		2509
		2510	=item EV_USE_EPOLL
		2511
		2512	If defined to be C<1>, libev will compile in support for the Linux
		2513	C<epoll>(7) backend. Its availability will be detected at runtime,
		2514	otherwise another method will be used as fallback. This is the
		2515	preferred backend for GNU/Linux systems.
		2516
		2517	=item EV_USE_KQUEUE
		2518
		2519	If defined to be C<1>, libev will compile in support for the BSD style
		2520	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
		2521	otherwise another method will be used as fallback. This is the preferred
		2522	backend for BSD and BSD-like systems, although on most BSDs kqueue only
		2523	supports some types of fds correctly (the only platform we found that
		2524	supports ptys for example was NetBSD), so kqueue might be compiled in, but
		2525	not be used unless explicitly requested. The best way to use it is to find
		2526	out whether kqueue supports your type of fd properly and use an embedded
		2527	kqueue loop.
		2528
		2529	=item EV_USE_PORT
		2530
		2531	If defined to be C<1>, libev will compile in support for the Solaris
		2532	10 port style backend. Its availability will be detected at runtime,
		2533	otherwise another method will be used as fallback. This is the preferred
		2534	backend for Solaris 10 systems.
		2535
		2536	=item EV_USE_DEVPOLL
		2537
		2538	reserved for future expansion, works like the USE symbols above.
		2539
		2540	=item EV_USE_INOTIFY
		2541
		2542	If defined to be C<1>, libev will compile in support for the Linux inotify
		2543	interface to speed up C<ev_stat> watchers. Its actual availability will
		2544	be detected at runtime.
		2545
		2546	=item EV_H
		2547
		2548	The name of the F<ev.h> header file used to include it. The default if
		2549	undefined is C<"ev.h"> in F<event.h> and F<ev.c>. This can be used to
		2550	virtually rename the F<ev.h> header file in case of conflicts.
		2551
		2552	=item EV_CONFIG_H
		2553
		2554	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
		2555	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
		2556	C<EV_H>, above.
		2557
		2558	=item EV_EVENT_H
		2559
		2560	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
		2561	of how the F<event.h> header can be found, the dfeault is C<"event.h">.
		2562
		2563	=item EV_PROTOTYPES
		2564
		2565	If defined to be C<0>, then F<ev.h> will not define any function
		2566	prototypes, but still define all the structs and other symbols. This is
		2567	occasionally useful if you want to provide your own wrapper functions
		2568	around libev functions.
		2569
		2570	=item EV_MULTIPLICITY
		2571
		2572	If undefined or defined to C<1>, then all event-loop-specific functions
		2573	will have the C<struct ev_loop *> as first argument, and you can create
		2574	additional independent event loops. Otherwise there will be no support
		2575	for multiple event loops and there is no first event loop pointer
		2576	argument. Instead, all functions act on the single default loop.
		2577
		2578	=item EV_MINPRI
		2579
		2580	=item EV_MAXPRI
		2581
		2582	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		2583	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		2584	provide for more priorities by overriding those symbols (usually defined
		2585	to be C<-2> and C<2>, respectively).
		2586
		2587	When doing priority-based operations, libev usually has to linearly search
		2588	all the priorities, so having many of them (hundreds) uses a lot of space
		2589	and time, so using the defaults of five priorities (-2 .. +2) is usually
		2590	fine.
		2591
		2592	If your embedding app does not need any priorities, defining these both to
		2593	C<0> will save some memory and cpu.
		2594
		2595	=item EV_PERIODIC_ENABLE
		2596
		2597	If undefined or defined to be C<1>, then periodic timers are supported. If
		2598	defined to be C<0>, then they are not. Disabling them saves a few kB of
		2599	code.
		2600
		2601	=item EV_IDLE_ENABLE
		2602
		2603	If undefined or defined to be C<1>, then idle watchers are supported. If
		2604	defined to be C<0>, then they are not. Disabling them saves a few kB of
		2605	code.
		2606
		2607	=item EV_EMBED_ENABLE
		2608
		2609	If undefined or defined to be C<1>, then embed watchers are supported. If
		2610	defined to be C<0>, then they are not.
		2611
		2612	=item EV_STAT_ENABLE
		2613
		2614	If undefined or defined to be C<1>, then stat watchers are supported. If
		2615	defined to be C<0>, then they are not.
		2616
		2617	=item EV_FORK_ENABLE
		2618
		2619	If undefined or defined to be C<1>, then fork watchers are supported. If
		2620	defined to be C<0>, then they are not.
		2621
		2622	=item EV_MINIMAL
		2623
		2624	If you need to shave off some kilobytes of code at the expense of some
		2625	speed, define this symbol to C<1>. Currently only used for gcc to override
		2626	some inlining decisions, saves roughly 30% codesize of amd64.
		2627
		2628	=item EV_PID_HASHSIZE
		2629
		2630	C<ev_child> watchers use a small hash table to distribute workload by
		2631	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
		2632	than enough. If you need to manage thousands of children you might want to
		2633	increase this value (I<must> be a power of two).
		2634
		2635	=item EV_INOTIFY_HASHSIZE
		2636
		2637	C<ev_stat> watchers use a small hash table to distribute workload by
		2638	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
		2639	usually more than enough. If you need to manage thousands of C<ev_stat>
		2640	watchers you might want to increase this value (I<must> be a power of
		2641	two).
		2642
		2643	=item EV_COMMON
		2644
		2645	By default, all watchers have a C<void *data> member. By redefining
		2646	this macro to a something else you can include more and other types of
		2647	members. You have to define it each time you include one of the files,
		2648	though, and it must be identical each time.
		2649
		2650	For example, the perl EV module uses something like this:
		2651
		2652	#define EV_COMMON \
		2653	SV *self; /* contains this struct */ \
		2654	SV *cb_sv, *fh /* note no trailing ";" */
		2655
		2656	=item EV_CB_DECLARE (type)
		2657
		2658	=item EV_CB_INVOKE (watcher, revents)
		2659
		2660	=item ev_set_cb (ev, cb)
		2661
		2662	Can be used to change the callback member declaration in each watcher,
		2663	and the way callbacks are invoked and set. Must expand to a struct member
		2664	definition and a statement, respectively. See the F<ev.h> header file for
		2665	their default definitions. One possible use for overriding these is to
		2666	avoid the C<struct ev_loop *> as first argument in all cases, or to use
		2667	method calls instead of plain function calls in C++.
		2668
		2669	=head2 EXPORTED API SYMBOLS
		2670
		2671	If you need to re-export the API (e.g. via a dll) and you need a list of
		2672	exported symbols, you can use the provided F<Symbol.*> files which list
		2673	all public symbols, one per line:
		2674
		2675	Symbols.ev for libev proper
		2676	Symbols.event for the libevent emulation
		2677
		2678	This can also be used to rename all public symbols to avoid clashes with
		2679	multiple versions of libev linked together (which is obviously bad in
		2680	itself, but sometimes it is inconvinient to avoid this).
		2681
		2682	A sed command like this will create wrapper C<#define>'s that you need to
		2683	include before including F<ev.h>:
		2684
		2685	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		2686
		2687	This would create a file F<wrap.h> which essentially looks like this:
		2688
		2689	#define ev_backend myprefix_ev_backend
		2690	#define ev_check_start myprefix_ev_check_start
		2691	#define ev_check_stop myprefix_ev_check_stop
		2692	...
		2693
		2694	=head2 EXAMPLES
		2695
		2696	For a real-world example of a program the includes libev
		2697	verbatim, you can have a look at the EV perl module
		2698	(L<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
		2699	the F<libev/> subdirectory and includes them in the F<EV/EVAPI.h> (public
		2700	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
		2701	will be compiled. It is pretty complex because it provides its own header
		2702	file.
		2703
		2704	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
		2705	that everybody includes and which overrides some configure choices:
		2706
		2707	#define EV_MINIMAL 1
		2708	#define EV_USE_POLL 0
		2709	#define EV_MULTIPLICITY 0
		2710	#define EV_PERIODIC_ENABLE 0
		2711	#define EV_STAT_ENABLE 0
		2712	#define EV_FORK_ENABLE 0
		2713	#define EV_CONFIG_H <config.h>
		2714	#define EV_MINPRI 0
		2715	#define EV_MAXPRI 0
		2716
		2717	#include "ev++.h"
		2718
		2719	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
		2720
		2721	#include "ev_cpp.h"
		2722	#include "ev.c"
		2723
		2724
		2725	=head1 COMPLEXITIES
		2726
		2727	In this section the complexities of (many of) the algorithms used inside
		2728	libev will be explained. For complexity discussions about backends see the
		2729	documentation for C<ev_default_init>.
		2730
		2731	All of the following are about amortised time: If an array needs to be
		2732	extended, libev needs to realloc and move the whole array, but this
		2733	happens asymptotically never with higher number of elements, so O(1) might
		2734	mean it might do a lengthy realloc operation in rare cases, but on average
		2735	it is much faster and asymptotically approaches constant time.
		2736
		2737	=over 4
		2738
		2739	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		2740
		2741	This means that, when you have a watcher that triggers in one hour and
		2742	there are 100 watchers that would trigger before that then inserting will
		2743	have to skip roughly seven (C<ld 100>) of these watchers.
		2744
		2745	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		2746
		2747	That means that changing a timer costs less than removing/adding them
		2748	as only the relative motion in the event queue has to be paid for.
		2749
		2750	=item Starting io/check/prepare/idle/signal/child watchers: O(1)
		2751
		2752	These just add the watcher into an array or at the head of a list.
		2753
		2754	=item Stopping check/prepare/idle watchers: O(1)
		2755
		2756	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		2757
		2758	These watchers are stored in lists then need to be walked to find the
		2759	correct watcher to remove. The lists are usually short (you don't usually
		2760	have many watchers waiting for the same fd or signal).
		2761
		2762	=item Finding the next timer in each loop iteration: O(1)
		2763
		2764	By virtue of using a binary heap, the next timer is always found at the
		2765	beginning of the storage array.
		2766
		2767	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		2768
		2769	A change means an I/O watcher gets started or stopped, which requires
		2770	libev to recalculate its status (and possibly tell the kernel, depending
		2771	on backend and wether C<ev_io_set> was used).
		2772
		2773	=item Activating one watcher (putting it into the pending state): O(1)
		2774
		2775	=item Priority handling: O(number_of_priorities)
		2776
		2777	Priorities are implemented by allocating some space for each
		2778	priority. When doing priority-based operations, libev usually has to
		2779	linearly search all the priorities, but starting/stopping and activating
		2780	watchers becomes O(1) w.r.t. prioritiy handling.
		2781
		2782	=back
		2783
		2784
		2785	=head1 Win32 platform limitations and workarounds
		2786
		2787	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		2788	requires, and its I/O model is fundamentally incompatible with the POSIX
		2789	model. Libev still offers limited functionality on this platform in
		2790	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		2791	descriptors. This only applies when using Win32 natively, not when using
		2792	e.g. cygwin.
		2793
		2794	There is no supported compilation method available on windows except
		2795	embedding it into other applications.
		2796
		2797	Due to the many, low, and arbitrary limits on the win32 platform and the
		2798	abysmal performance of winsockets, using a large number of sockets is not
		2799	recommended (and not reasonable). If your program needs to use more than
		2800	a hundred or so sockets, then likely it needs to use a totally different
		2801	implementation for windows, as libev offers the POSIX model, which cannot
		2802	be implemented efficiently on windows (microsoft monopoly games).
		2803
		2804	=over 4
		2805
		2806	=item The winsocket select function
		2807
		2808	The winsocket C<select> function doesn't follow POSIX in that it requires
		2809	socket I<handles> and not socket I<file descriptors>. This makes select
		2810	very inefficient, and also requires a mapping from file descriptors
		2811	to socket handles. See the discussion of the C<EV_SELECT_USE_FD_SET>,
		2812	C<EV_SELECT_IS_WINSOCKET> and C<EV_FD_TO_WIN32_HANDLE> preprocessor
		2813	symbols for more info.
		2814
		2815	The configuration for a "naked" win32 using the microsoft runtime
		2816	libraries and raw winsocket select is:
		2817
		2818	#define EV_USE_SELECT 1
		2819	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		2820
		2821	Note that winsockets handling of fd sets is O(n), so you can easily get a
		2822	complexity in the O(n²) range when using win32.
		2823
		2824	=item Limited number of file descriptors
		2825
		2826	Windows has numerous arbitrary (and low) limits on things. Early versions
		2827	of winsocket's select only supported waiting for a max. of C<64> handles
		2828	(probably owning to the fact that all windows kernels can only wait for
		2829	C<64> things at the same time internally; microsoft recommends spawning a
		2830	chain of threads and wait for 63 handles and the previous thread in each).
		2831
		2832	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		2833	to some high number (e.g. C<2048>) before compiling the winsocket select
		2834	call (which might be in libev or elsewhere, for example, perl does its own
		2835	select emulation on windows).
		2836
		2837	Another limit is the number of file descriptors in the microsoft runtime
		2838	libraries, which by default is C<64> (there must be a hidden I<64> fetish
		2839	or something like this inside microsoft). You can increase this by calling
		2840	C<_setmaxstdio>, which can increase this limit to C<2048> (another
		2841	arbitrary limit), but is broken in many versions of the microsoft runtime
		2842	libraries.
		2843
		2844	This might get you to about C<512> or C<2048> sockets (depending on
		2845	windows version and/or the phase of the moon). To get more, you need to
		2846	wrap all I/O functions and provide your own fd management, but the cost of
		2847	calling select (O(n²)) will likely make this unworkable.
		2848
		2849	=back
		2850
1134		2851
1135	=head1 AUTHOR	2852	=head1 AUTHOR
1136		2853
1137	Marc Lehmann <libev@schmorp.de>.	2854	Marc Lehmann <libev@schmorp.de>.
1138		2855

Diff Legend

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