ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.8 by root, Mon Nov 12 08:20:02 2007 UTC vs.
Revision 1.255 by root, Tue Jul 14 19:11:31 2009 UTC

…		…
2		2
3	libev - a high performance full-featured event loop written in C	3	libev - a high performance full-featured event loop written in C
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
8		8
9	=head1 DESCRIPTION	9	=head2 EXAMPLE PROGRAM
		10
		11	// a single header file is required
		12	#include <ev.h>
		13
		14	#include <stdio.h> // for puts
		15
		16	// every watcher type has its own typedef'd struct
		17	// with the name ev_TYPE
		18	ev_io stdin_watcher;
		19	ev_timer timeout_watcher;
		20
		21	// all watcher callbacks have a similar signature
		22	// this callback is called when data is readable on stdin
		23	static void
		24	stdin_cb (EV_P_ ev_io *w, int revents)
		25	{
		26	puts ("stdin ready");
		27	// for one-shot events, one must manually stop the watcher
		28	// with its corresponding stop function.
		29	ev_io_stop (EV_A_ w);
		30
		31	// this causes all nested ev_loop's to stop iterating
		32	ev_unloop (EV_A_ EVUNLOOP_ALL);
		33	}
		34
		35	// another callback, this time for a time-out
		36	static void
		37	timeout_cb (EV_P_ ev_timer *w, int revents)
		38	{
		39	puts ("timeout");
		40	// this causes the innermost ev_loop to stop iterating
		41	ev_unloop (EV_A_ EVUNLOOP_ONE);
		42	}
		43
		44	int
		45	main (void)
		46	{
		47	// use the default event loop unless you have special needs
		48	struct ev_loop *loop = ev_default_loop (0);
		49
		50	// initialise an io watcher, then start it
		51	// this one will watch for stdin to become readable
		52	ev_io_init (&stdin_watcher, stdin_cb, /*STDIN_FILENO*/ 0, EV_READ);
		53	ev_io_start (loop, &stdin_watcher);
		54
		55	// initialise a timer watcher, then start it
		56	// simple non-repeating 5.5 second timeout
		57	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		58	ev_timer_start (loop, &timeout_watcher);
		59
		60	// now wait for events to arrive
		61	ev_loop (loop, 0);
		62
		63	// unloop was called, so exit
		64	return 0;
		65	}
		66
		67	=head1 ABOUT THIS DOCUMENT
		68
		69	This document documents the libev software package.
		70
		71	The newest version of this document is also available as an html-formatted
		72	web page you might find easier to navigate when reading it for the first
		73	time: L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod>.
		74
		75	While this document tries to be as complete as possible in documenting
		76	libev, its usage and the rationale behind its design, it is not a tutorial
		77	on event-based programming, nor will it introduce event-based programming
		78	with libev.
		79
		80	Familarity with event based programming techniques in general is assumed
		81	throughout this document.
		82
		83	=head1 ABOUT LIBEV
10		84
11	Libev is an event loop: you register interest in certain events (such as a	85	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	86	file descriptor being readable or a timeout occurring), and it will manage
13	these event sources and provide your program with events.	87	these event sources and provide your program with events.
14		88
15	To do this, it must take more or less complete control over your process	89	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	90	(or thread) by executing the I<event loop> handler, and will then
17	communicate events via a callback mechanism.	91	communicate events via a callback mechanism.
…		…
19	You register interest in certain events by registering so-called I<event	93	You register interest in certain events by registering so-called I<event
20	watchers>, which are relatively small C structures you initialise with the	94	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	95	details of the event, and then hand it over to libev by I<starting> the
22	watcher.	96	watcher.
23		97
24	=head1 FEATURES	98	=head2 FEATURES
25		99
26	Libev supports select, poll, the linux-specific epoll and the bsd-specific	100	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
27	kqueue mechanisms for file descriptor events, relative timers, absolute	101	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
28	timers with customised rescheduling, signal events, process status change	102	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
29	events (related to SIGCHLD), and event watchers dealing with the event	103	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
30	loop mechanism itself (idle, prepare and check watchers). It also is quite	104	with customised rescheduling (C<ev_periodic>), synchronous signals
		105	(C<ev_signal>), process status change events (C<ev_child>), and event
		106	watchers dealing with the event loop mechanism itself (C<ev_idle>,
		107	C<ev_embed>, C<ev_prepare> and C<ev_check> watchers) as well as
		108	file watchers (C<ev_stat>) and even limited support for fork events
		109	(C<ev_fork>).
		110
		111	It also is quite fast (see this
31	fast (see this L<benchmark|http://libev.schmorp.de/bench.html> comparing	112	L<benchmark|http://libev.schmorp.de/bench.html> comparing it to libevent
32	it to libevent for example).	113	for example).
33		114
34	=head1 CONVENTIONS	115	=head2 CONVENTIONS
35		116
36	Libev is very configurable. In this manual the default configuration	117	Libev is very configurable. In this manual the default (and most common)
37	will be described, which supports multiple event loops. For more info	118	configuration will be described, which supports multiple event loops. For
38	about various configuration options please have a look at the file	119	more info about various configuration options please have a look at
39	F<README.embed> in the libev distribution. If libev was configured without	120	B<EMBED> section in this manual. If libev was configured without support
40	support for multiple event loops, then all functions taking an initial	121	for multiple event loops, then all functions taking an initial argument of
41	argument of name C<loop> (which is always of type C<struct ev_loop *>)	122	name C<loop> (which is always of type C<ev_loop *>) will not have
42	will not have this argument.	123	this argument.
43		124
44	=head1 TIME AND OTHER GLOBAL FUNCTIONS	125	=head2 TIME REPRESENTATION
45		126
46	Libev represents time as a single floating point number, representing the	127	Libev represents time as a single floating point number, representing
47	(fractional) number of seconds since the (POSIX) epoch (somewhere near	128	the (fractional) number of seconds since the (POSIX) epoch (somewhere
48	the beginning of 1970, details are complicated, don't ask). This type is	129	near the beginning of 1970, details are complicated, don't ask). This
49	called C<ev_tstamp>, which is what you should use too. It usually aliases	130	type is called C<ev_tstamp>, which is what you should use too. It usually
50	to the double type in C.	131	aliases to the C<double> type in C. When you need to do any calculations
		132	on it, you should treat it as some floating point value. Unlike the name
		133	component C<stamp> might indicate, it is also used for time differences
		134	throughout libev.
		135
		136	=head1 ERROR HANDLING
		137
		138	Libev knows three classes of errors: operating system errors, usage errors
		139	and internal errors (bugs).
		140
		141	When libev catches an operating system error it cannot handle (for example
		142	a system call indicating a condition libev cannot fix), it calls the callback
		143	set via C<ev_set_syserr_cb>, which is supposed to fix the problem or
		144	abort. The default is to print a diagnostic message and to call C<abort
		145	()>.
		146
		147	When libev detects a usage error such as a negative timer interval, then
		148	it will print a diagnostic message and abort (via the C<assert> mechanism,
		149	so C<NDEBUG> will disable this checking): these are programming errors in
		150	the libev caller and need to be fixed there.
		151
		152	Libev also has a few internal error-checking C<assert>ions, and also has
		153	extensive consistency checking code. These do not trigger under normal
		154	circumstances, as they indicate either a bug in libev or worse.
		155
		156
		157	=head1 GLOBAL FUNCTIONS
		158
		159	These functions can be called anytime, even before initialising the
		160	library in any way.
51		161
52	=over 4	162	=over 4
53		163
54	=item ev_tstamp ev_time ()	164	=item ev_tstamp ev_time ()
55		165
56	Returns the current time as libev would use it.	166	Returns the current time as libev would use it. Please note that the
		167	C<ev_now> function is usually faster and also often returns the timestamp
		168	you actually want to know.
		169
		170	=item ev_sleep (ev_tstamp interval)
		171
		172	Sleep for the given interval: The current thread will be blocked until
		173	either it is interrupted or the given time interval has passed. Basically
		174	this is a sub-second-resolution C<sleep ()>.
57		175
58	=item int ev_version_major ()	176	=item int ev_version_major ()
59		177
60	=item int ev_version_minor ()	178	=item int ev_version_minor ()
61		179
62	You can find out the major and minor version numbers of the library	180	You can find out the major and minor ABI version numbers of the library
63	you linked against by calling the functions C<ev_version_major> and	181	you linked against by calling the functions C<ev_version_major> and
64	C<ev_version_minor>. If you want, you can compare against the global	182	C<ev_version_minor>. If you want, you can compare against the global
65	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the	183	symbols C<EV_VERSION_MAJOR> and C<EV_VERSION_MINOR>, which specify the
66	version of the library your program was compiled against.	184	version of the library your program was compiled against.
67		185
		186	These version numbers refer to the ABI version of the library, not the
		187	release version.
		188
68	Usually, its a good idea to terminate if the major versions mismatch,	189	Usually, it's a good idea to terminate if the major versions mismatch,
69	as this indicates an incompatible change. Minor versions are usually	190	as this indicates an incompatible change. Minor versions are usually
70	compatible to older versions, so a larger minor version alone is usually	191	compatible to older versions, so a larger minor version alone is usually
71	not a problem.	192	not a problem.
72		193
		194	Example: Make sure we haven't accidentally been linked against the wrong
		195	version.
		196
		197	assert (("libev version mismatch",
		198	ev_version_major () == EV_VERSION_MAJOR
		199	&& ev_version_minor () >= EV_VERSION_MINOR));
		200
		201	=item unsigned int ev_supported_backends ()
		202
		203	Return the set of all backends (i.e. their corresponding C<EV_BACKEND_*>
		204	value) compiled into this binary of libev (independent of their
		205	availability on the system you are running on). See C<ev_default_loop> for
		206	a description of the set values.
		207
		208	Example: make sure we have the epoll method, because yeah this is cool and
		209	a must have and can we have a torrent of it please!!!11
		210
		211	assert (("sorry, no epoll, no sex",
		212	ev_supported_backends () & EVBACKEND_EPOLL));
		213
		214	=item unsigned int ev_recommended_backends ()
		215
		216	Return the set of all backends compiled into this binary of libev and also
		217	recommended for this platform. This set is often smaller than the one
		218	returned by C<ev_supported_backends>, as for example kqueue is broken on
		219	most BSDs and will not be auto-detected unless you explicitly request it
		220	(assuming you know what you are doing). This is the set of backends that
		221	libev will probe for if you specify no backends explicitly.
		222
		223	=item unsigned int ev_embeddable_backends ()
		224
		225	Returns the set of backends that are embeddable in other event loops. This
		226	is the theoretical, all-platform, value. To find which backends
		227	might be supported on the current system, you would need to look at
		228	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
		229	recommended ones.
		230
		231	See the description of C<ev_embed> watchers for more info.
		232
73	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	233	=item ev_set_allocator (void *(*cb)(void *ptr, long size)) [NOT REENTRANT]
74		234
75	Sets the allocation function to use (the prototype is similar to the	235	Sets the allocation function to use (the prototype is similar - the
76	realloc C function, the semantics are identical). It is used to allocate	236	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
77	and free memory (no surprises here). If it returns zero when memory	237	used to allocate and free memory (no surprises here). If it returns zero
78	needs to be allocated, the library might abort or take some potentially	238	when memory needs to be allocated (C<size != 0>), the library might abort
79	destructive action. The default is your system realloc function.	239	or take some potentially destructive action.
		240
		241	Since some systems (at least OpenBSD and Darwin) fail to implement
		242	correct C<realloc> semantics, libev will use a wrapper around the system
		243	C<realloc> and C<free> functions by default.
80		244
81	You could override this function in high-availability programs to, say,	245	You could override this function in high-availability programs to, say,
82	free some memory if it cannot allocate memory, to use a special allocator,	246	free some memory if it cannot allocate memory, to use a special allocator,
83	or even to sleep a while and retry until some memory is available.	247	or even to sleep a while and retry until some memory is available.
84		248
		249	Example: Replace the libev allocator with one that waits a bit and then
		250	retries (example requires a standards-compliant C<realloc>).
		251
		252	static void *
		253	persistent_realloc (void *ptr, size_t size)
		254	{
		255	for (;;)
		256	{
		257	void *newptr = realloc (ptr, size);
		258
		259	if (newptr)
		260	return newptr;
		261
		262	sleep (60);
		263	}
		264	}
		265
		266	...
		267	ev_set_allocator (persistent_realloc);
		268
85	=item ev_set_syserr_cb (void (*cb)(const char *msg));	269	=item ev_set_syserr_cb (void (*cb)(const char *msg)); [NOT REENTRANT]
86		270
87	Set the callback function to call on a retryable syscall error (such	271	Set the callback function to call on a retryable system call error (such
88	as failed select, poll, epoll_wait). The message is a printable string	272	as failed select, poll, epoll_wait). The message is a printable string
89	indicating the system call or subsystem causing the problem. If this	273	indicating the system call or subsystem causing the problem. If this
90	callback is set, then libev will expect it to remedy the sitution, no	274	callback is set, then libev will expect it to remedy the situation, no
91	matter what, when it returns. That is, libev will generally retry the	275	matter what, when it returns. That is, libev will generally retry the
92	requested operation, or, if the condition doesn't go away, do bad stuff	276	requested operation, or, if the condition doesn't go away, do bad stuff
93	(such as abort).	277	(such as abort).
94		278
		279	Example: This is basically the same thing that libev does internally, too.
		280
		281	static void
		282	fatal_error (const char *msg)
		283	{
		284	perror (msg);
		285	abort ();
		286	}
		287
		288	...
		289	ev_set_syserr_cb (fatal_error);
		290
95	=back	291	=back
96		292
97	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	293	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
98		294
99	An event loop is described by a C<struct ev_loop *>. The library knows two	295	An event loop is described by a C<struct ev_loop *> (the C<struct>
100	types of such loops, the I<default> loop, which supports signals and child	296	is I<not> optional in this case, as there is also an C<ev_loop>
101	events, and dynamically created loops which do not.	297	I<function>).
102		298
103	If you use threads, a common model is to run the default event loop	299	The library knows two types of such loops, the I<default> loop, which
104	in your main thread (or in a separate thrad) and for each thread you	300	supports signals and child events, and dynamically created loops which do
105	create, you also create another event loop. Libev itself does no locking	301	not.
106	whatsoever, so if you mix calls to the same event loop in different
107	threads, make sure you lock (this is usually a bad idea, though, even if
108	done correctly, because its hideous and inefficient).
109		302
110	=over 4	303	=over 4
111		304
112	=item struct ev_loop *ev_default_loop (unsigned int flags)	305	=item struct ev_loop *ev_default_loop (unsigned int flags)
113		306
114	This will initialise the default event loop if it hasn't been initialised	307	This will initialise the default event loop if it hasn't been initialised
115	yet and return it. If the default loop could not be initialised, returns	308	yet and return it. If the default loop could not be initialised, returns
116	false. If it already was initialised it simply returns it (and ignores the	309	false. If it already was initialised it simply returns it (and ignores the
117	flags).	310	flags. If that is troubling you, check C<ev_backend ()> afterwards).
118		311
119	If you don't know what event loop to use, use the one returned from this	312	If you don't know what event loop to use, use the one returned from this
120	function.	313	function.
121		314
		315	Note that this function is I<not> thread-safe, so if you want to use it
		316	from multiple threads, you have to lock (note also that this is unlikely,
		317	as loops cannot be shared easily between threads anyway).
		318
		319	The default loop is the only loop that can handle C<ev_signal> and
		320	C<ev_child> watchers, and to do this, it always registers a handler
		321	for C<SIGCHLD>. If this is a problem for your application you can either
		322	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
		323	can simply overwrite the C<SIGCHLD> signal handler I<after> calling
		324	C<ev_default_init>.
		325
122	The flags argument can be used to specify special behaviour or specific	326	The flags argument can be used to specify special behaviour or specific
123	backends to use, and is usually specified as 0 (or EVFLAG_AUTO).	327	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
124		328
125	It supports the following flags:	329	The following flags are supported:
126		330
127	=over 4	331	=over 4
128		332
129	=item EVFLAG_AUTO	333	=item C<EVFLAG_AUTO>
130		334
131	The default flags value. Use this if you have no clue (its the right	335	The default flags value. Use this if you have no clue (it's the right
132	thing, believe me).	336	thing, believe me).
133		337
134	=item EVFLAG_NOENV	338	=item C<EVFLAG_NOENV>
135		339
136	If this flag bit is ored into the flag value (or the program runs setuid	340	If this flag bit is or'ed into the flag value (or the program runs setuid
137	or setgid) then libev will I<not> look at the environment variable	341	or setgid) then libev will I<not> look at the environment variable
138	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	342	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
139	override the flags completely if it is found in the environment. This is	343	override the flags completely if it is found in the environment. This is
140	useful to try out specific backends to test their performance, or to work	344	useful to try out specific backends to test their performance, or to work
141	around bugs.	345	around bugs.
142		346
		347	=item C<EVFLAG_FORKCHECK>
		348
		349	Instead of calling C<ev_default_fork> or C<ev_loop_fork> manually after
		350	a fork, you can also make libev check for a fork in each iteration by
		351	enabling this flag.
		352
		353	This works by calling C<getpid ()> on every iteration of the loop,
		354	and thus this might slow down your event loop if you do a lot of loop
		355	iterations and little real work, but is usually not noticeable (on my
		356	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
		357	without a system call and thus I<very> fast, but my GNU/Linux system also has
		358	C<pthread_atfork> which is even faster).
		359
		360	The big advantage of this flag is that you can forget about fork (and
		361	forget about forgetting to tell libev about forking) when you use this
		362	flag.
		363
		364	This flag setting cannot be overridden or specified in the C<LIBEV_FLAGS>
		365	environment variable.
		366
143	=item EVMETHOD_SELECT portable select backend	367	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
144		368
145	=item EVMETHOD_POLL poll backend (everywhere except windows)	369	This is your standard select(2) backend. Not I<completely> standard, as
		370	libev tries to roll its own fd_set with no limits on the number of fds,
		371	but if that fails, expect a fairly low limit on the number of fds when
		372	using this backend. It doesn't scale too well (O(highest_fd)), but its
		373	usually the fastest backend for a low number of (low-numbered :) fds.
146		374
147	=item EVMETHOD_EPOLL linux only	375	To get good performance out of this backend you need a high amount of
		376	parallelism (most of the file descriptors should be busy). If you are
		377	writing a server, you should C<accept ()> in a loop to accept as many
		378	connections as possible during one iteration. You might also want to have
		379	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		380	readiness notifications you get per iteration.
148		381
149	=item EVMETHOD_KQUEUE some bsds only	382	This backend maps C<EV_READ> to the C<readfds> set and C<EV_WRITE> to the
		383	C<writefds> set (and to work around Microsoft Windows bugs, also onto the
		384	C<exceptfds> set on that platform).
150		385
151	=item EVMETHOD_DEVPOLL solaris 8 only	386	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
152		387
153	=item EVMETHOD_PORT solaris 10 only	388	And this is your standard poll(2) backend. It's more complicated
		389	than select, but handles sparse fds better and has no artificial
		390	limit on the number of fds you can use (except it will slow down
		391	considerably with a lot of inactive fds). It scales similarly to select,
		392	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		393	performance tips.
154		394
155	If one or more of these are ored into the flags value, then only these	395	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
156	backends will be tried (in the reverse order as given here). If one are	396	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
157	specified, any backend will do.	397
		398	=item C<EVBACKEND_EPOLL> (value 4, Linux)
		399
		400	For few fds, this backend is a bit little slower than poll and select,
		401	but it scales phenomenally better. While poll and select usually scale
		402	like O(total_fds) where n is the total number of fds (or the highest fd),
		403	epoll scales either O(1) or O(active_fds).
		404
		405	The epoll mechanism deserves honorable mention as the most misdesigned
		406	of the more advanced event mechanisms: mere annoyances include silently
		407	dropping file descriptors, requiring a system call per change per file
		408	descriptor (and unnecessary guessing of parameters), problems with dup and
		409	so on. The biggest issue is fork races, however - if a program forks then
		410	I<both> parent and child process have to recreate the epoll set, which can
		411	take considerable time (one syscall per file descriptor) and is of course
		412	hard to detect.
		413
		414	Epoll is also notoriously buggy - embedding epoll fds I<should> work, but
		415	of course I<doesn't>, and epoll just loves to report events for totally
		416	I<different> file descriptors (even already closed ones, so one cannot
		417	even remove them from the set) than registered in the set (especially
		418	on SMP systems). Libev tries to counter these spurious notifications by
		419	employing an additional generation counter and comparing that against the
		420	events to filter out spurious ones, recreating the set when required.
		421
		422	While stopping, setting and starting an I/O watcher in the same iteration
		423	will result in some caching, there is still a system call per such
		424	incident (because the same I<file descriptor> could point to a different
		425	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
		426	file descriptors might not work very well if you register events for both
		427	file descriptors.
		428
		429	Best performance from this backend is achieved by not unregistering all
		430	watchers for a file descriptor until it has been closed, if possible,
		431	i.e. keep at least one watcher active per fd at all times. Stopping and
		432	starting a watcher (without re-setting it) also usually doesn't cause
		433	extra overhead. A fork can both result in spurious notifications as well
		434	as in libev having to destroy and recreate the epoll object, which can
		435	take considerable time and thus should be avoided.
		436
		437	All this means that, in practice, C<EVBACKEND_SELECT> can be as fast or
		438	faster than epoll for maybe up to a hundred file descriptors, depending on
		439	the usage. So sad.
		440
		441	While nominally embeddable in other event loops, this feature is broken in
		442	all kernel versions tested so far.
		443
		444	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		445	C<EVBACKEND_POLL>.
		446
		447	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
		448
		449	Kqueue deserves special mention, as at the time of this writing, it
		450	was broken on all BSDs except NetBSD (usually it doesn't work reliably
		451	with anything but sockets and pipes, except on Darwin, where of course
		452	it's completely useless). Unlike epoll, however, whose brokenness
		453	is by design, these kqueue bugs can (and eventually will) be fixed
		454	without API changes to existing programs. For this reason it's not being
		455	"auto-detected" unless you explicitly specify it in the flags (i.e. using
		456	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
		457	system like NetBSD.
		458
		459	You still can embed kqueue into a normal poll or select backend and use it
		460	only for sockets (after having made sure that sockets work with kqueue on
		461	the target platform). See C<ev_embed> watchers for more info.
		462
		463	It scales in the same way as the epoll backend, but the interface to the
		464	kernel is more efficient (which says nothing about its actual speed, of
		465	course). While stopping, setting and starting an I/O watcher does never
		466	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
		467	two event changes per incident. Support for C<fork ()> is very bad (but
		468	sane, unlike epoll) and it drops fds silently in similarly hard-to-detect
		469	cases
		470
		471	This backend usually performs well under most conditions.
		472
		473	While nominally embeddable in other event loops, this doesn't work
		474	everywhere, so you might need to test for this. And since it is broken
		475	almost everywhere, you should only use it when you have a lot of sockets
		476	(for which it usually works), by embedding it into another event loop
		477	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> (but C<poll> is of course
		478	also broken on OS X)) and, did I mention it, using it only for sockets.
		479
		480	This backend maps C<EV_READ> into an C<EVFILT_READ> kevent with
		481	C<NOTE_EOF>, and C<EV_WRITE> into an C<EVFILT_WRITE> kevent with
		482	C<NOTE_EOF>.
		483
		484	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
		485
		486	This is not implemented yet (and might never be, unless you send me an
		487	implementation). According to reports, C</dev/poll> only supports sockets
		488	and is not embeddable, which would limit the usefulness of this backend
		489	immensely.
		490
		491	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
		492
		493	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
		494	it's really slow, but it still scales very well (O(active_fds)).
		495
		496	Please note that Solaris event ports can deliver a lot of spurious
		497	notifications, so you need to use non-blocking I/O or other means to avoid
		498	blocking when no data (or space) is available.
		499
		500	While this backend scales well, it requires one system call per active
		501	file descriptor per loop iteration. For small and medium numbers of file
		502	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		503	might perform better.
		504
		505	On the positive side, with the exception of the spurious readiness
		506	notifications, this backend actually performed fully to specification
		507	in all tests and is fully embeddable, which is a rare feat among the
		508	OS-specific backends (I vastly prefer correctness over speed hacks).
		509
		510	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
		511	C<EVBACKEND_POLL>.
		512
		513	=item C<EVBACKEND_ALL>
		514
		515	Try all backends (even potentially broken ones that wouldn't be tried
		516	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
		517	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
		518
		519	It is definitely not recommended to use this flag.
158		520
159	=back	521	=back
		522
		523	If one or more of these are or'ed into the flags value, then only these
		524	backends will be tried (in the reverse order as listed here). If none are
		525	specified, all backends in C<ev_recommended_backends ()> will be tried.
		526
		527	Example: This is the most typical usage.
		528
		529	if (!ev_default_loop (0))
		530	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
		531
		532	Example: Restrict libev to the select and poll backends, and do not allow
		533	environment settings to be taken into account:
		534
		535	ev_default_loop (EVBACKEND_POLL | EVBACKEND_SELECT | EVFLAG_NOENV);
		536
		537	Example: Use whatever libev has to offer, but make sure that kqueue is
		538	used if available (warning, breaks stuff, best use only with your own
		539	private event loop and only if you know the OS supports your types of
		540	fds):
		541
		542	ev_default_loop (ev_recommended_backends () | EVBACKEND_KQUEUE);
160		543
161	=item struct ev_loop *ev_loop_new (unsigned int flags)	544	=item struct ev_loop *ev_loop_new (unsigned int flags)
162		545
163	Similar to C<ev_default_loop>, but always creates a new event loop that is	546	Similar to C<ev_default_loop>, but always creates a new event loop that is
164	always distinct from the default loop. Unlike the default loop, it cannot	547	always distinct from the default loop. Unlike the default loop, it cannot
165	handle signal and child watchers, and attempts to do so will be greeted by	548	handle signal and child watchers, and attempts to do so will be greeted by
166	undefined behaviour (or a failed assertion if assertions are enabled).	549	undefined behaviour (or a failed assertion if assertions are enabled).
167		550
		551	Note that this function I<is> thread-safe, and the recommended way to use
		552	libev with threads is indeed to create one loop per thread, and using the
		553	default loop in the "main" or "initial" thread.
		554
		555	Example: Try to create a event loop that uses epoll and nothing else.
		556
		557	struct ev_loop *epoller = ev_loop_new (EVBACKEND_EPOLL | EVFLAG_NOENV);
		558	if (!epoller)
		559	fatal ("no epoll found here, maybe it hides under your chair");
		560
168	=item ev_default_destroy ()	561	=item ev_default_destroy ()
169		562
170	Destroys the default loop again (frees all memory and kernel state	563	Destroys the default loop again (frees all memory and kernel state
171	etc.). This stops all registered event watchers (by not touching them in	564	etc.). None of the active event watchers will be stopped in the normal
172	any way whatsoever, although you cnanot rely on this :).	565	sense, so e.g. C<ev_is_active> might still return true. It is your
		566	responsibility to either stop all watchers cleanly yourself I<before>
		567	calling this function, or cope with the fact afterwards (which is usually
		568	the easiest thing, you can just ignore the watchers and/or C<free ()> them
		569	for example).
		570
		571	Note that certain global state, such as signal state (and installed signal
		572	handlers), will not be freed by this function, and related watchers (such
		573	as signal and child watchers) would need to be stopped manually.
		574
		575	In general it is not advisable to call this function except in the
		576	rare occasion where you really need to free e.g. the signal handling
		577	pipe fds. If you need dynamically allocated loops it is better to use
		578	C<ev_loop_new> and C<ev_loop_destroy>).
173		579
174	=item ev_loop_destroy (loop)	580	=item ev_loop_destroy (loop)
175		581
176	Like C<ev_default_destroy>, but destroys an event loop created by an	582	Like C<ev_default_destroy>, but destroys an event loop created by an
177	earlier call to C<ev_loop_new>.	583	earlier call to C<ev_loop_new>.
178		584
179	=item ev_default_fork ()	585	=item ev_default_fork ()
180		586
		587	This function sets a flag that causes subsequent C<ev_loop> iterations
181	This function reinitialises the kernel state for backends that have	588	to reinitialise the kernel state for backends that have one. Despite the
182	one. Despite the name, you can call it anytime, but it makes most sense	589	name, you can call it anytime, but it makes most sense after forking, in
183	after forking, in either the parent or child process (or both, but that	590	the child process (or both child and parent, but that again makes little
184	again makes little sense).	591	sense). You I<must> call it in the child before using any of the libev
		592	functions, and it will only take effect at the next C<ev_loop> iteration.
185		593
186	You I<must> call this function after forking if and only if you want to	594	On the other hand, you only need to call this function in the child
187	use the event library in both processes. If you just fork+exec, you don't	595	process if and only if you want to use the event library in the child. If
188	have to call it.	596	you just fork+exec, you don't have to call it at all.
189		597
190	The function itself is quite fast and its usually not a problem to call	598	The function itself is quite fast and it's usually not a problem to call
191	it just in case after a fork. To make this easy, the function will fit in	599	it just in case after a fork. To make this easy, the function will fit in
192	quite nicely into a call to C<pthread_atfork>:	600	quite nicely into a call to C<pthread_atfork>:
193		601
194	pthread_atfork (0, 0, ev_default_fork);	602	pthread_atfork (0, 0, ev_default_fork);
195		603
196	=item ev_loop_fork (loop)	604	=item ev_loop_fork (loop)
197		605
198	Like C<ev_default_fork>, but acts on an event loop created by	606	Like C<ev_default_fork>, but acts on an event loop created by
199	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	607	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
200	after fork, and how you do this is entirely your own problem.	608	after fork that you want to re-use in the child, and how you do this is
		609	entirely your own problem.
201		610
		611	=item int ev_is_default_loop (loop)
		612
		613	Returns true when the given loop is, in fact, the default loop, and false
		614	otherwise.
		615
		616	=item unsigned int ev_loop_count (loop)
		617
		618	Returns the count of loop iterations for the loop, which is identical to
		619	the number of times libev did poll for new events. It starts at C<0> and
		620	happily wraps around with enough iterations.
		621
		622	This value can sometimes be useful as a generation counter of sorts (it
		623	"ticks" the number of loop iterations), as it roughly corresponds with
		624	C<ev_prepare> and C<ev_check> calls.
		625
		626	=item unsigned int ev_loop_depth (loop)
		627
		628	Returns the number of times C<ev_loop> was entered minus the number of
		629	times C<ev_loop> was exited, in other words, the recursion depth.
		630
		631	Outside C<ev_loop>, this number is zero. In a callback, this number is
		632	C<1>, unless C<ev_loop> was invoked recursively (or from another thread),
		633	in which case it is higher.
		634
		635	Leaving C<ev_loop> abnormally (setjmp/longjmp, cancelling the thread
		636	etc.), doesn't count as exit.
		637
202	=item unsigned int ev_method (loop)	638	=item unsigned int ev_backend (loop)
203		639
204	Returns one of the C<EVMETHOD_*> flags indicating the event backend in	640	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
205	use.	641	use.
206		642
207	=item ev_tstamp = ev_now (loop)	643	=item ev_tstamp ev_now (loop)
208		644
209	Returns the current "event loop time", which is the time the event loop	645	Returns the current "event loop time", which is the time the event loop
210	got events and started processing them. This timestamp does not change	646	received events and started processing them. This timestamp does not
211	as long as callbacks are being processed, and this is also the base time	647	change as long as callbacks are being processed, and this is also the base
212	used for relative timers. You can treat it as the timestamp of the event	648	time used for relative timers. You can treat it as the timestamp of the
213	occuring (or more correctly, the mainloop finding out about it).	649	event occurring (or more correctly, libev finding out about it).
		650
		651	=item ev_now_update (loop)
		652
		653	Establishes the current time by querying the kernel, updating the time
		654	returned by C<ev_now ()> in the progress. This is a costly operation and
		655	is usually done automatically within C<ev_loop ()>.
		656
		657	This function is rarely useful, but when some event callback runs for a
		658	very long time without entering the event loop, updating libev's idea of
		659	the current time is a good idea.
		660
		661	See also L<The special problem of time updates> in the C<ev_timer> section.
		662
		663	=item ev_suspend (loop)
		664
		665	=item ev_resume (loop)
		666
		667	These two functions suspend and resume a loop, for use when the loop is
		668	not used for a while and timeouts should not be processed.
		669
		670	A typical use case would be an interactive program such as a game: When
		671	the user presses C<^Z> to suspend the game and resumes it an hour later it
		672	would be best to handle timeouts as if no time had actually passed while
		673	the program was suspended. This can be achieved by calling C<ev_suspend>
		674	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		675	C<ev_resume> directly afterwards to resume timer processing.
		676
		677	Effectively, all C<ev_timer> watchers will be delayed by the time spend
		678	between C<ev_suspend> and C<ev_resume>, and all C<ev_periodic> watchers
		679	will be rescheduled (that is, they will lose any events that would have
		680	occured while suspended).
		681
		682	After calling C<ev_suspend> you B<must not> call I<any> function on the
		683	given loop other than C<ev_resume>, and you B<must not> call C<ev_resume>
		684	without a previous call to C<ev_suspend>.
		685
		686	Calling C<ev_suspend>/C<ev_resume> has the side effect of updating the
		687	event loop time (see C<ev_now_update>).
214		688
215	=item ev_loop (loop, int flags)	689	=item ev_loop (loop, int flags)
216		690
217	Finally, this is it, the event handler. This function usually is called	691	Finally, this is it, the event handler. This function usually is called
218	after you initialised all your watchers and you want to start handling	692	after you initialised all your watchers and you want to start handling
219	events.	693	events.
220		694
221	If the flags argument is specified as 0, it will not return until either	695	If the flags argument is specified as C<0>, it will not return until
222	no event watchers are active anymore or C<ev_unloop> was called.	696	either no event watchers are active anymore or C<ev_unloop> was called.
		697
		698	Please note that an explicit C<ev_unloop> is usually better than
		699	relying on all watchers to be stopped when deciding when a program has
		700	finished (especially in interactive programs), but having a program
		701	that automatically loops as long as it has to and no longer by virtue
		702	of relying on its watchers stopping correctly, that is truly a thing of
		703	beauty.
223		704
224	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle	705	A flags value of C<EVLOOP_NONBLOCK> will look for new events, will handle
225	those events and any outstanding ones, but will not block your process in	706	those events and any already outstanding ones, but will not block your
226	case there are no events.	707	process in case there are no events and will return after one iteration of
		708	the loop.
227		709
228	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if	710	A flags value of C<EVLOOP_ONESHOT> will look for new events (waiting if
229	neccessary) and will handle those and any outstanding ones. It will block	711	necessary) and will handle those and any already outstanding ones. It
230	your process until at least one new event arrives.	712	will block your process until at least one new event arrives (which could
		713	be an event internal to libev itself, so there is no guarantee that a
		714	user-registered callback will be called), and will return after one
		715	iteration of the loop.
231		716
232	This flags value could be used to implement alternative looping	717	This is useful if you are waiting for some external event in conjunction
233	constructs, but the C<prepare> and C<check> watchers provide a better and	718	with something not expressible using other libev watchers (i.e. "roll your
234	more generic mechanism.	719	own C<ev_loop>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
		720	usually a better approach for this kind of thing.
		721
		722	Here are the gory details of what C<ev_loop> does:
		723
		724	- Before the first iteration, call any pending watchers.
		725	* If EVFLAG_FORKCHECK was used, check for a fork.
		726	- If a fork was detected (by any means), queue and call all fork watchers.
		727	- Queue and call all prepare watchers.
		728	- If we have been forked, detach and recreate the kernel state
		729	as to not disturb the other process.
		730	- Update the kernel state with all outstanding changes.
		731	- Update the "event loop time" (ev_now ()).
		732	- Calculate for how long to sleep or block, if at all
		733	(active idle watchers, EVLOOP_NONBLOCK or not having
		734	any active watchers at all will result in not sleeping).
		735	- Sleep if the I/O and timer collect interval say so.
		736	- Block the process, waiting for any events.
		737	- Queue all outstanding I/O (fd) events.
		738	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
		739	- Queue all expired timers.
		740	- Queue all expired periodics.
		741	- Unless any events are pending now, queue all idle watchers.
		742	- Queue all check watchers.
		743	- Call all queued watchers in reverse order (i.e. check watchers first).
		744	Signals and child watchers are implemented as I/O watchers, and will
		745	be handled here by queueing them when their watcher gets executed.
		746	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
		747	were used, or there are no active watchers, return, otherwise
		748	continue with step *.
		749
		750	Example: Queue some jobs and then loop until no events are outstanding
		751	anymore.
		752
		753	... queue jobs here, make sure they register event watchers as long
		754	... as they still have work to do (even an idle watcher will do..)
		755	ev_loop (my_loop, 0);
		756	... jobs done or somebody called unloop. yeah!
235		757
236	=item ev_unloop (loop, how)	758	=item ev_unloop (loop, how)
237		759
238	Can be used to make a call to C<ev_loop> return early. The C<how> argument	760	Can be used to make a call to C<ev_loop> return early (but only after it
		761	has processed all outstanding events). The C<how> argument must be either
239	must be either C<EVUNLOOP_ONCE>, which will make the innermost C<ev_loop>	762	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
240	call return, or C<EVUNLOOP_ALL>, which will make all nested C<ev_loop>	763	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
241	calls return.	764
		765	This "unloop state" will be cleared when entering C<ev_loop> again.
		766
		767	It is safe to call C<ev_unloop> from otuside any C<ev_loop> calls.
242		768
243	=item ev_ref (loop)	769	=item ev_ref (loop)
244		770
245	=item ev_unref (loop)	771	=item ev_unref (loop)
246		772
247	Ref/unref can be used to add or remove a refcount on the event loop: Every	773	Ref/unref can be used to add or remove a reference count on the event
248	watcher keeps one reference. If you have a long-runing watcher you never	774	loop: Every watcher keeps one reference, and as long as the reference
249	unregister that should not keep ev_loop from running, ev_unref() after	775	count is nonzero, C<ev_loop> will not return on its own.
250	starting, and ev_ref() before stopping it. Libev itself uses this for	776
251	example for its internal signal pipe: It is not visible to you as a user	777	If you have a watcher you never unregister that should not keep C<ev_loop>
252	and should not keep C<ev_loop> from exiting if the work is done. It is	778	from returning, call ev_unref() after starting, and ev_ref() before
		779	stopping it.
		780
		781	As an example, libev itself uses this for its internal signal pipe: It
		782	is not visible to the libev user and should not keep C<ev_loop> from
		783	exiting if no event watchers registered by it are active. It is also an
253	also an excellent way to do this for generic recurring timers or from	784	excellent way to do this for generic recurring timers or from within
254	within third-party libraries. Just remember to unref after start and ref	785	third-party libraries. Just remember to I<unref after start> and I<ref
255	before stop.	786	before stop> (but only if the watcher wasn't active before, or was active
		787	before, respectively. Note also that libev might stop watchers itself
		788	(e.g. non-repeating timers) in which case you have to C<ev_ref>
		789	in the callback).
		790
		791	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
		792	running when nothing else is active.
		793
		794	ev_signal exitsig;
		795	ev_signal_init (&exitsig, sig_cb, SIGINT);
		796	ev_signal_start (loop, &exitsig);
		797	evf_unref (loop);
		798
		799	Example: For some weird reason, unregister the above signal handler again.
		800
		801	ev_ref (loop);
		802	ev_signal_stop (loop, &exitsig);
		803
		804	=item ev_set_io_collect_interval (loop, ev_tstamp interval)
		805
		806	=item ev_set_timeout_collect_interval (loop, ev_tstamp interval)
		807
		808	These advanced functions influence the time that libev will spend waiting
		809	for events. Both time intervals are by default C<0>, meaning that libev
		810	will try to invoke timer/periodic callbacks and I/O callbacks with minimum
		811	latency.
		812
		813	Setting these to a higher value (the C<interval> I<must> be >= C<0>)
		814	allows libev to delay invocation of I/O and timer/periodic callbacks
		815	to increase efficiency of loop iterations (or to increase power-saving
		816	opportunities).
		817
		818	The idea is that sometimes your program runs just fast enough to handle
		819	one (or very few) event(s) per loop iteration. While this makes the
		820	program responsive, it also wastes a lot of CPU time to poll for new
		821	events, especially with backends like C<select ()> which have a high
		822	overhead for the actual polling but can deliver many events at once.
		823
		824	By setting a higher I<io collect interval> you allow libev to spend more
		825	time collecting I/O events, so you can handle more events per iteration,
		826	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
		827	C<ev_timer>) will be not affected. Setting this to a non-null value will
		828	introduce an additional C<ev_sleep ()> call into most loop iterations. The
		829	sleep time ensures that libev will not poll for I/O events more often then
		830	once per this interval, on average.
		831
		832	Likewise, by setting a higher I<timeout collect interval> you allow libev
		833	to spend more time collecting timeouts, at the expense of increased
		834	latency/jitter/inexactness (the watcher callback will be called
		835	later). C<ev_io> watchers will not be affected. Setting this to a non-null
		836	value will not introduce any overhead in libev.
		837
		838	Many (busy) programs can usually benefit by setting the I/O collect
		839	interval to a value near C<0.1> or so, which is often enough for
		840	interactive servers (of course not for games), likewise for timeouts. It
		841	usually doesn't make much sense to set it to a lower value than C<0.01>,
		842	as this approaches the timing granularity of most systems. Note that if
		843	you do transactions with the outside world and you can't increase the
		844	parallelity, then this setting will limit your transaction rate (if you
		845	need to poll once per transaction and the I/O collect interval is 0.01,
		846	then you can't do more than 100 transations per second).
		847
		848	Setting the I<timeout collect interval> can improve the opportunity for
		849	saving power, as the program will "bundle" timer callback invocations that
		850	are "near" in time together, by delaying some, thus reducing the number of
		851	times the process sleeps and wakes up again. Another useful technique to
		852	reduce iterations/wake-ups is to use C<ev_periodic> watchers and make sure
		853	they fire on, say, one-second boundaries only.
		854
		855	Example: we only need 0.1s timeout granularity, and we wish not to poll
		856	more often than 100 times per second:
		857
		858	ev_set_timeout_collect_interval (EV_DEFAULT_UC_ 0.1);
		859	ev_set_io_collect_interval (EV_DEFAULT_UC_ 0.01);
		860
		861	=item ev_invoke_pending (loop)
		862
		863	This call will simply invoke all pending watchers while resetting their
		864	pending state. Normally, C<ev_loop> does this automatically when required,
		865	but when overriding the invoke callback this call comes handy.
		866
		867	=item ev_set_invoke_pending_cb (loop, void (*invoke_pending_cb)(EV_P))
		868
		869	This overrides the invoke pending functionality of the loop: Instead of
		870	invoking all pending watchers when there are any, C<ev_loop> will call
		871	this callback instead. This is useful, for example, when you want to
		872	invoke the actual watchers inside another context (another thread etc.).
		873
		874	If you want to reset the callback, use C<ev_invoke_pending> as new
		875	callback.
		876
		877	=item ev_set_loop_release_cb (loop, void (*release)(EV_P), void (*acquire)(EV_P))
		878
		879	Sometimes you want to share the same loop between multiple threads. This
		880	can be done relatively simply by putting mutex_lock/unlock calls around
		881	each call to a libev function.
		882
		883	However, C<ev_loop> can run an indefinite time, so it is not feasible to
		884	wait for it to return. One way around this is to wake up the loop via
		885	C<ev_unloop> and C<av_async_send>, another way is to set these I<release>
		886	and I<acquire> callbacks on the loop.
		887
		888	When set, then C<release> will be called just before the thread is
		889	suspended waiting for new events, and C<acquire> is called just
		890	afterwards.
		891
		892	Ideally, C<release> will just call your mutex_unlock function, and
		893	C<acquire> will just call the mutex_lock function again.
		894
		895	While event loop modifications are allowed between invocations of
		896	C<release> and C<acquire> (that's their only purpose after all), no
		897	modifications done will affect the event loop, i.e. adding watchers will
		898	have no effect on the set of file descriptors being watched, or the time
		899	waited. USe an C<ev_async> watcher to wake up C<ev_loop> when you want it
		900	to take note of any changes you made.
		901
		902	In theory, threads executing C<ev_loop> will be async-cancel safe between
		903	invocations of C<release> and C<acquire>.
		904
		905	See also the locking example in the C<THREADS> section later in this
		906	document.
		907
		908	=item ev_set_userdata (loop, void *data)
		909
		910	=item ev_userdata (loop)
		911
		912	Set and retrieve a single C<void *> associated with a loop. When
		913	C<ev_set_userdata> has never been called, then C<ev_userdata> returns
		914	C<0.>
		915
		916	These two functions can be used to associate arbitrary data with a loop,
		917	and are intended solely for the C<invoke_pending_cb>, C<release> and
		918	C<acquire> callbacks described above, but of course can be (ab-)used for
		919	any other purpose as well.
		920
		921	=item ev_loop_verify (loop)
		922
		923	This function only does something when C<EV_VERIFY> support has been
		924	compiled in, which is the default for non-minimal builds. It tries to go
		925	through all internal structures and checks them for validity. If anything
		926	is found to be inconsistent, it will print an error message to standard
		927	error and call C<abort ()>.
		928
		929	This can be used to catch bugs inside libev itself: under normal
		930	circumstances, this function will never abort as of course libev keeps its
		931	data structures consistent.
256		932
257	=back	933	=back
258		934
		935
259	=head1 ANATOMY OF A WATCHER	936	=head1 ANATOMY OF A WATCHER
		937
		938	In the following description, uppercase C<TYPE> in names stands for the
		939	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		940	watchers and C<ev_io_start> for I/O watchers.
260		941
261	A watcher is a structure that you create and register to record your	942	A watcher is a structure that you create and register to record your
262	interest in some event. For instance, if you want to wait for STDIN to	943	interest in some event. For instance, if you want to wait for STDIN to
263	become readable, you would create an ev_io watcher for that:	944	become readable, you would create an C<ev_io> watcher for that:
264		945
265	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	946	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
266	{	947	{
267	ev_io_stop (w);	948	ev_io_stop (w);
268	ev_unloop (loop, EVUNLOOP_ALL);	949	ev_unloop (loop, EVUNLOOP_ALL);
269	}	950	}
270		951
271	struct ev_loop *loop = ev_default_loop (0);	952	struct ev_loop *loop = ev_default_loop (0);
		953
272	struct ev_io stdin_watcher;	954	ev_io stdin_watcher;
		955
273	ev_init (&stdin_watcher, my_cb);	956	ev_init (&stdin_watcher, my_cb);
274	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	957	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
275	ev_io_start (loop, &stdin_watcher);	958	ev_io_start (loop, &stdin_watcher);
		959
276	ev_loop (loop, 0);	960	ev_loop (loop, 0);
277		961
278	As you can see, you are responsible for allocating the memory for your	962	As you can see, you are responsible for allocating the memory for your
279	watcher structures (and it is usually a bad idea to do this on the stack,	963	watcher structures (and it is I<usually> a bad idea to do this on the
280	although this can sometimes be quite valid).	964	stack).
		965
		966	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		967	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
281		968
282	Each watcher structure must be initialised by a call to C<ev_init	969	Each watcher structure must be initialised by a call to C<ev_init
283	(watcher *, callback)>, which expects a callback to be provided. This	970	(watcher *, callback)>, which expects a callback to be provided. This
284	callback gets invoked each time the event occurs (or, in the case of io	971	callback gets invoked each time the event occurs (or, in the case of I/O
285	watchers, each time the event loop detects that the file descriptor given	972	watchers, each time the event loop detects that the file descriptor given
286	is readable and/or writable).	973	is readable and/or writable).
287		974
288	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	975	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
289	with arguments specific to this watcher type. There is also a macro	976	macro to configure it, with arguments specific to the watcher type. There
290	to combine initialisation and setting in one call: C<< ev_<type>_init	977	is also a macro to combine initialisation and setting in one call: C<<
291	(watcher *, callback, ...) >>.	978	ev_TYPE_init (watcher *, callback, ...) >>.
292		979
293	To make the watcher actually watch out for events, you have to start it	980	To make the watcher actually watch out for events, you have to start it
294	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	981	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
295	*) >>), and you can stop watching for events at any time by calling the	982	*) >>), and you can stop watching for events at any time by calling the
296	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	983	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
297		984
298	As long as your watcher is active (has been started but not stopped) you	985	As long as your watcher is active (has been started but not stopped) you
299	must not touch the values stored in it. Most specifically you must never	986	must not touch the values stored in it. Most specifically you must never
300	reinitialise it or call its set method.	987	reinitialise it or call its C<ev_TYPE_set> macro.
301
302	You cna check whether an event is active by calling the C<ev_is_active
303	(watcher *)> macro. To see whether an event is outstanding (but the
304	callback for it has not been called yet) you cna use the C<ev_is_pending
305	(watcher *)> macro.
306		988
307	Each and every callback receives the event loop pointer as first, the	989	Each and every callback receives the event loop pointer as first, the
308	registered watcher structure as second, and a bitset of received events as	990	registered watcher structure as second, and a bitset of received events as
309	third argument.	991	third argument.
310		992
311	The rceeived events usually include a single bit per event type received	993	The received events usually include a single bit per event type received
312	(you can receive multiple events at the same time). The possible bit masks	994	(you can receive multiple events at the same time). The possible bit masks
313	are:	995	are:
314		996
315	=over 4	997	=over 4
316		998
317	=item EV_READ	999	=item C<EV_READ>
318		1000
319	=item EV_WRITE	1001	=item C<EV_WRITE>
320		1002
321	The file descriptor in the ev_io watcher has become readable and/or	1003	The file descriptor in the C<ev_io> watcher has become readable and/or
322	writable.	1004	writable.
323		1005
324	=item EV_TIMEOUT	1006	=item C<EV_TIMEOUT>
325		1007
326	The ev_timer watcher has timed out.	1008	The C<ev_timer> watcher has timed out.
327		1009
328	=item EV_PERIODIC	1010	=item C<EV_PERIODIC>
329		1011
330	The ev_periodic watcher has timed out.	1012	The C<ev_periodic> watcher has timed out.
331		1013
332	=item EV_SIGNAL	1014	=item C<EV_SIGNAL>
333		1015
334	The signal specified in the ev_signal watcher has been received by a thread.	1016	The signal specified in the C<ev_signal> watcher has been received by a thread.
335		1017
336	=item EV_CHILD	1018	=item C<EV_CHILD>
337		1019
338	The pid specified in the ev_child watcher has received a status change.	1020	The pid specified in the C<ev_child> watcher has received a status change.
339		1021
		1022	=item C<EV_STAT>
		1023
		1024	The path specified in the C<ev_stat> watcher changed its attributes somehow.
		1025
340	=item EV_IDLE	1026	=item C<EV_IDLE>
341		1027
342	The ev_idle watcher has determined that you have nothing better to do.	1028	The C<ev_idle> watcher has determined that you have nothing better to do.
343		1029
344	=item EV_PREPARE	1030	=item C<EV_PREPARE>
345		1031
346	=item EV_CHECK	1032	=item C<EV_CHECK>
347		1033
348	All ev_prepare watchers are invoked just I<before> C<ev_loop> starts	1034	All C<ev_prepare> watchers are invoked just I<before> C<ev_loop> starts
349	to gather new events, and all ev_check watchers are invoked just after	1035	to gather new events, and all C<ev_check> watchers are invoked just after
350	C<ev_loop> has gathered them, but before it invokes any callbacks for any	1036	C<ev_loop> has gathered them, but before it invokes any callbacks for any
351	received events. Callbacks of both watcher types can start and stop as	1037	received events. Callbacks of both watcher types can start and stop as
352	many watchers as they want, and all of them will be taken into account	1038	many watchers as they want, and all of them will be taken into account
353	(for example, a ev_prepare watcher might start an idle watcher to keep	1039	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
354	C<ev_loop> from blocking).	1040	C<ev_loop> from blocking).
355		1041
		1042	=item C<EV_EMBED>
		1043
		1044	The embedded event loop specified in the C<ev_embed> watcher needs attention.
		1045
		1046	=item C<EV_FORK>
		1047
		1048	The event loop has been resumed in the child process after fork (see
		1049	C<ev_fork>).
		1050
		1051	=item C<EV_ASYNC>
		1052
		1053	The given async watcher has been asynchronously notified (see C<ev_async>).
		1054
		1055	=item C<EV_CUSTOM>
		1056
		1057	Not ever sent (or otherwise used) by libev itself, but can be freely used
		1058	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		1059
356	=item EV_ERROR	1060	=item C<EV_ERROR>
357		1061
358	An unspecified error has occured, the watcher has been stopped. This might	1062	An unspecified error has occurred, the watcher has been stopped. This might
359	happen because the watcher could not be properly started because libev	1063	happen because the watcher could not be properly started because libev
360	ran out of memory, a file descriptor was found to be closed or any other	1064	ran out of memory, a file descriptor was found to be closed or any other
		1065	problem. Libev considers these application bugs.
		1066
361	problem. You best act on it by reporting the problem and somehow coping	1067	You best act on it by reporting the problem and somehow coping with the
362	with the watcher being stopped.	1068	watcher being stopped. Note that well-written programs should not receive
		1069	an error ever, so when your watcher receives it, this usually indicates a
		1070	bug in your program.
363		1071
364	Libev will usually signal a few "dummy" events together with an error,	1072	Libev will usually signal a few "dummy" events together with an error, for
365	for example it might indicate that a fd is readable or writable, and if	1073	example it might indicate that a fd is readable or writable, and if your
366	your callbacks is well-written it can just attempt the operation and cope	1074	callbacks is well-written it can just attempt the operation and cope with
367	with the error from read() or write(). This will not work in multithreaded	1075	the error from read() or write(). This will not work in multi-threaded
368	programs, though, so beware.	1076	programs, though, as the fd could already be closed and reused for another
		1077	thing, so beware.
369		1078
370	=back	1079	=back
371		1080
		1081	=head2 GENERIC WATCHER FUNCTIONS
		1082
		1083	=over 4
		1084
		1085	=item C<ev_init> (ev_TYPE *watcher, callback)
		1086
		1087	This macro initialises the generic portion of a watcher. The contents
		1088	of the watcher object can be arbitrary (so C<malloc> will do). Only
		1089	the generic parts of the watcher are initialised, you I<need> to call
		1090	the type-specific C<ev_TYPE_set> macro afterwards to initialise the
		1091	type-specific parts. For each type there is also a C<ev_TYPE_init> macro
		1092	which rolls both calls into one.
		1093
		1094	You can reinitialise a watcher at any time as long as it has been stopped
		1095	(or never started) and there are no pending events outstanding.
		1096
		1097	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
		1098	int revents)>.
		1099
		1100	Example: Initialise an C<ev_io> watcher in two steps.
		1101
		1102	ev_io w;
		1103	ev_init (&w, my_cb);
		1104	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1105
		1106	=item C<ev_TYPE_set> (ev_TYPE *, [args])
		1107
		1108	This macro initialises the type-specific parts of a watcher. You need to
		1109	call C<ev_init> at least once before you call this macro, but you can
		1110	call C<ev_TYPE_set> any number of times. You must not, however, call this
		1111	macro on a watcher that is active (it can be pending, however, which is a
		1112	difference to the C<ev_init> macro).
		1113
		1114	Although some watcher types do not have type-specific arguments
		1115	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
		1116
		1117	See C<ev_init>, above, for an example.
		1118
		1119	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
		1120
		1121	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
		1122	calls into a single call. This is the most convenient method to initialise
		1123	a watcher. The same limitations apply, of course.
		1124
		1125	Example: Initialise and set an C<ev_io> watcher in one step.
		1126
		1127	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1128
		1129	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
		1130
		1131	Starts (activates) the given watcher. Only active watchers will receive
		1132	events. If the watcher is already active nothing will happen.
		1133
		1134	Example: Start the C<ev_io> watcher that is being abused as example in this
		1135	whole section.
		1136
		1137	ev_io_start (EV_DEFAULT_UC, &w);
		1138
		1139	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
		1140
		1141	Stops the given watcher if active, and clears the pending status (whether
		1142	the watcher was active or not).
		1143
		1144	It is possible that stopped watchers are pending - for example,
		1145	non-repeating timers are being stopped when they become pending - but
		1146	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
		1147	pending. If you want to free or reuse the memory used by the watcher it is
		1148	therefore a good idea to always call its C<ev_TYPE_stop> function.
		1149
		1150	=item bool ev_is_active (ev_TYPE *watcher)
		1151
		1152	Returns a true value iff the watcher is active (i.e. it has been started
		1153	and not yet been stopped). As long as a watcher is active you must not modify
		1154	it.
		1155
		1156	=item bool ev_is_pending (ev_TYPE *watcher)
		1157
		1158	Returns a true value iff the watcher is pending, (i.e. it has outstanding
		1159	events but its callback has not yet been invoked). As long as a watcher
		1160	is pending (but not active) you must not call an init function on it (but
		1161	C<ev_TYPE_set> is safe), you must not change its priority, and you must
		1162	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		1163	it).
		1164
		1165	=item callback ev_cb (ev_TYPE *watcher)
		1166
		1167	Returns the callback currently set on the watcher.
		1168
		1169	=item ev_cb_set (ev_TYPE *watcher, callback)
		1170
		1171	Change the callback. You can change the callback at virtually any time
		1172	(modulo threads).
		1173
		1174	=item ev_set_priority (ev_TYPE *watcher, priority)
		1175
		1176	=item int ev_priority (ev_TYPE *watcher)
		1177
		1178	Set and query the priority of the watcher. The priority is a small
		1179	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		1180	(default: C<-2>). Pending watchers with higher priority will be invoked
		1181	before watchers with lower priority, but priority will not keep watchers
		1182	from being executed (except for C<ev_idle> watchers).
		1183
		1184	If you need to suppress invocation when higher priority events are pending
		1185	you need to look at C<ev_idle> watchers, which provide this functionality.
		1186
		1187	You I<must not> change the priority of a watcher as long as it is active or
		1188	pending.
		1189
		1190	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1191	fine, as long as you do not mind that the priority value you query might
		1192	or might not have been clamped to the valid range.
		1193
		1194	The default priority used by watchers when no priority has been set is
		1195	always C<0>, which is supposed to not be too high and not be too low :).
		1196
		1197	See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
		1198	priorities.
		1199
		1200	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		1201
		1202	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		1203	C<loop> nor C<revents> need to be valid as long as the watcher callback
		1204	can deal with that fact, as both are simply passed through to the
		1205	callback.
		1206
		1207	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		1208
		1209	If the watcher is pending, this function clears its pending status and
		1210	returns its C<revents> bitset (as if its callback was invoked). If the
		1211	watcher isn't pending it does nothing and returns C<0>.
		1212
		1213	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1214	callback to be invoked, which can be accomplished with this function.
		1215
		1216	=back
		1217
		1218
372	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1219	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
373		1220
374	Each watcher has, by default, a member C<void *data> that you can change	1221	Each watcher has, by default, a member C<void *data> that you can change
375	and read at any time, libev will completely ignore it. This cna be used	1222	and read at any time: libev will completely ignore it. This can be used
376	to associate arbitrary data with your watcher. If you need more data and	1223	to associate arbitrary data with your watcher. If you need more data and
377	don't want to allocate memory and store a pointer to it in that data	1224	don't want to allocate memory and store a pointer to it in that data
378	member, you can also "subclass" the watcher type and provide your own	1225	member, you can also "subclass" the watcher type and provide your own
379	data:	1226	data:
380		1227
381	struct my_io	1228	struct my_io
382	{	1229	{
383	struct ev_io io;	1230	ev_io io;
384	int otherfd;	1231	int otherfd;
385	void *somedata;	1232	void *somedata;
386	struct whatever *mostinteresting;	1233	struct whatever *mostinteresting;
387	}	1234	};
		1235
		1236	...
		1237	struct my_io w;
		1238	ev_io_init (&w.io, my_cb, fd, EV_READ);
388		1239
389	And since your callback will be called with a pointer to the watcher, you	1240	And since your callback will be called with a pointer to the watcher, you
390	can cast it back to your own type:	1241	can cast it back to your own type:
391		1242
392	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1243	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
393	{	1244	{
394	struct my_io *w = (struct my_io *)w_;	1245	struct my_io *w = (struct my_io *)w_;
395	...	1246	...
396	}	1247	}
397		1248
398	More interesting and less C-conformant ways of catsing your callback type	1249	More interesting and less C-conformant ways of casting your callback type
399	have been omitted....	1250	instead have been omitted.
		1251
		1252	Another common scenario is to use some data structure with multiple
		1253	embedded watchers:
		1254
		1255	struct my_biggy
		1256	{
		1257	int some_data;
		1258	ev_timer t1;
		1259	ev_timer t2;
		1260	}
		1261
		1262	In this case getting the pointer to C<my_biggy> is a bit more
		1263	complicated: Either you store the address of your C<my_biggy> struct
		1264	in the C<data> member of the watcher (for woozies), or you need to use
		1265	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1266	programmers):
		1267
		1268	#include <stddef.h>
		1269
		1270	static void
		1271	t1_cb (EV_P_ ev_timer *w, int revents)
		1272	{
		1273	struct my_biggy big = (struct my_biggy *)
		1274	(((char *)w) - offsetof (struct my_biggy, t1));
		1275	}
		1276
		1277	static void
		1278	t2_cb (EV_P_ ev_timer *w, int revents)
		1279	{
		1280	struct my_biggy big = (struct my_biggy *)
		1281	(((char *)w) - offsetof (struct my_biggy, t2));
		1282	}
		1283
		1284	=head2 WATCHER PRIORITY MODELS
		1285
		1286	Many event loops support I<watcher priorities>, which are usually small
		1287	integers that influence the ordering of event callback invocation
		1288	between watchers in some way, all else being equal.
		1289
		1290	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1291	description for the more technical details such as the actual priority
		1292	range.
		1293
		1294	There are two common ways how these these priorities are being interpreted
		1295	by event loops:
		1296
		1297	In the more common lock-out model, higher priorities "lock out" invocation
		1298	of lower priority watchers, which means as long as higher priority
		1299	watchers receive events, lower priority watchers are not being invoked.
		1300
		1301	The less common only-for-ordering model uses priorities solely to order
		1302	callback invocation within a single event loop iteration: Higher priority
		1303	watchers are invoked before lower priority ones, but they all get invoked
		1304	before polling for new events.
		1305
		1306	Libev uses the second (only-for-ordering) model for all its watchers
		1307	except for idle watchers (which use the lock-out model).
		1308
		1309	The rationale behind this is that implementing the lock-out model for
		1310	watchers is not well supported by most kernel interfaces, and most event
		1311	libraries will just poll for the same events again and again as long as
		1312	their callbacks have not been executed, which is very inefficient in the
		1313	common case of one high-priority watcher locking out a mass of lower
		1314	priority ones.
		1315
		1316	Static (ordering) priorities are most useful when you have two or more
		1317	watchers handling the same resource: a typical usage example is having an
		1318	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1319	timeouts. Under load, data might be received while the program handles
		1320	other jobs, but since timers normally get invoked first, the timeout
		1321	handler will be executed before checking for data. In that case, giving
		1322	the timer a lower priority than the I/O watcher ensures that I/O will be
		1323	handled first even under adverse conditions (which is usually, but not
		1324	always, what you want).
		1325
		1326	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1327	will only be executed when no same or higher priority watchers have
		1328	received events, they can be used to implement the "lock-out" model when
		1329	required.
		1330
		1331	For example, to emulate how many other event libraries handle priorities,
		1332	you can associate an C<ev_idle> watcher to each such watcher, and in
		1333	the normal watcher callback, you just start the idle watcher. The real
		1334	processing is done in the idle watcher callback. This causes libev to
		1335	continously poll and process kernel event data for the watcher, but when
		1336	the lock-out case is known to be rare (which in turn is rare :), this is
		1337	workable.
		1338
		1339	Usually, however, the lock-out model implemented that way will perform
		1340	miserably under the type of load it was designed to handle. In that case,
		1341	it might be preferable to stop the real watcher before starting the
		1342	idle watcher, so the kernel will not have to process the event in case
		1343	the actual processing will be delayed for considerable time.
		1344
		1345	Here is an example of an I/O watcher that should run at a strictly lower
		1346	priority than the default, and which should only process data when no
		1347	other events are pending:
		1348
		1349	ev_idle idle; // actual processing watcher
		1350	ev_io io; // actual event watcher
		1351
		1352	static void
		1353	io_cb (EV_P_ ev_io *w, int revents)
		1354	{
		1355	// stop the I/O watcher, we received the event, but
		1356	// are not yet ready to handle it.
		1357	ev_io_stop (EV_A_ w);
		1358
		1359	// start the idle watcher to ahndle the actual event.
		1360	// it will not be executed as long as other watchers
		1361	// with the default priority are receiving events.
		1362	ev_idle_start (EV_A_ &idle);
		1363	}
		1364
		1365	static void
		1366	idle_cb (EV_P_ ev_idle *w, int revents)
		1367	{
		1368	// actual processing
		1369	read (STDIN_FILENO, ...);
		1370
		1371	// have to start the I/O watcher again, as
		1372	// we have handled the event
		1373	ev_io_start (EV_P_ &io);
		1374	}
		1375
		1376	// initialisation
		1377	ev_idle_init (&idle, idle_cb);
		1378	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1379	ev_io_start (EV_DEFAULT_ &io);
		1380
		1381	In the "real" world, it might also be beneficial to start a timer, so that
		1382	low-priority connections can not be locked out forever under load. This
		1383	enables your program to keep a lower latency for important connections
		1384	during short periods of high load, while not completely locking out less
		1385	important ones.
400		1386
401		1387
402	=head1 WATCHER TYPES	1388	=head1 WATCHER TYPES
403		1389
404	This section describes each watcher in detail, but will not repeat	1390	This section describes each watcher in detail, but will not repeat
405	information given in the last section.	1391	information given in the last section. Any initialisation/set macros,
		1392	functions and members specific to the watcher type are explained.
406		1393
		1394	Members are additionally marked with either I<[read-only]>, meaning that,
		1395	while the watcher is active, you can look at the member and expect some
		1396	sensible content, but you must not modify it (you can modify it while the
		1397	watcher is stopped to your hearts content), or I<[read-write]>, which
		1398	means you can expect it to have some sensible content while the watcher
		1399	is active, but you can also modify it. Modifying it may not do something
		1400	sensible or take immediate effect (or do anything at all), but libev will
		1401	not crash or malfunction in any way.
		1402
		1403
407	=head2 struct ev_io - is my file descriptor readable or writable	1404	=head2 C<ev_io> - is this file descriptor readable or writable?
408		1405
409	I/O watchers check whether a file descriptor is readable or writable	1406	I/O watchers check whether a file descriptor is readable or writable
410	in each iteration of the event loop (This behaviour is called	1407	in each iteration of the event loop, or, more precisely, when reading
411	level-triggering because you keep receiving events as long as the	1408	would not block the process and writing would at least be able to write
412	condition persists. Remember you cna stop the watcher if you don't want to	1409	some data. This behaviour is called level-triggering because you keep
413	act on the event and neither want to receive future events).	1410	receiving events as long as the condition persists. Remember you can stop
		1411	the watcher if you don't want to act on the event and neither want to
		1412	receive future events.
414		1413
415	In general you can register as many read and/or write event watchers oer	1414	In general you can register as many read and/or write event watchers per
416	fd as you want (as long as you don't confuse yourself). Setting all file	1415	fd as you want (as long as you don't confuse yourself). Setting all file
417	descriptors to non-blocking mode is also usually a good idea (but not	1416	descriptors to non-blocking mode is also usually a good idea (but not
418	required if you know what you are doing).	1417	required if you know what you are doing).
419		1418
420	You have to be careful with dup'ed file descriptors, though. Some backends	1419	If you cannot use non-blocking mode, then force the use of a
421	(the linux epoll backend is a notable example) cannot handle dup'ed file	1420	known-to-be-good backend (at the time of this writing, this includes only
422	descriptors correctly if you register interest in two or more fds pointing	1421	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
423	to the same file/socket etc. description.	1422	descriptors for which non-blocking operation makes no sense (such as
		1423	files) - libev doesn't guarentee any specific behaviour in that case.
424		1424
425	If you must do this, then force the use of a known-to-be-good backend	1425	Another thing you have to watch out for is that it is quite easy to
426	(at the time of this writing, this includes only EVMETHOD_SELECT and	1426	receive "spurious" readiness notifications, that is your callback might
427	EVMETHOD_POLL).	1427	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
		1428	because there is no data. Not only are some backends known to create a
		1429	lot of those (for example Solaris ports), it is very easy to get into
		1430	this situation even with a relatively standard program structure. Thus
		1431	it is best to always use non-blocking I/O: An extra C<read>(2) returning
		1432	C<EAGAIN> is far preferable to a program hanging until some data arrives.
		1433
		1434	If you cannot run the fd in non-blocking mode (for example you should
		1435	not play around with an Xlib connection), then you have to separately
		1436	re-test whether a file descriptor is really ready with a known-to-be good
		1437	interface such as poll (fortunately in our Xlib example, Xlib already
		1438	does this on its own, so its quite safe to use). Some people additionally
		1439	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1440	indefinitely.
		1441
		1442	But really, best use non-blocking mode.
		1443
		1444	=head3 The special problem of disappearing file descriptors
		1445
		1446	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1447	descriptor (either due to calling C<close> explicitly or any other means,
		1448	such as C<dup2>). The reason is that you register interest in some file
		1449	descriptor, but when it goes away, the operating system will silently drop
		1450	this interest. If another file descriptor with the same number then is
		1451	registered with libev, there is no efficient way to see that this is, in
		1452	fact, a different file descriptor.
		1453
		1454	To avoid having to explicitly tell libev about such cases, libev follows
		1455	the following policy: Each time C<ev_io_set> is being called, libev
		1456	will assume that this is potentially a new file descriptor, otherwise
		1457	it is assumed that the file descriptor stays the same. That means that
		1458	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1459	descriptor even if the file descriptor number itself did not change.
		1460
		1461	This is how one would do it normally anyway, the important point is that
		1462	the libev application should not optimise around libev but should leave
		1463	optimisations to libev.
		1464
		1465	=head3 The special problem of dup'ed file descriptors
		1466
		1467	Some backends (e.g. epoll), cannot register events for file descriptors,
		1468	but only events for the underlying file descriptions. That means when you
		1469	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1470	events for them, only one file descriptor might actually receive events.
		1471
		1472	There is no workaround possible except not registering events
		1473	for potentially C<dup ()>'ed file descriptors, or to resort to
		1474	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1475
		1476	=head3 The special problem of fork
		1477
		1478	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1479	useless behaviour. Libev fully supports fork, but needs to be told about
		1480	it in the child.
		1481
		1482	To support fork in your programs, you either have to call
		1483	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1484	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1485	C<EVBACKEND_POLL>.
		1486
		1487	=head3 The special problem of SIGPIPE
		1488
		1489	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1490	when writing to a pipe whose other end has been closed, your program gets
		1491	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1492	this is sensible behaviour, for daemons, this is usually undesirable.
		1493
		1494	So when you encounter spurious, unexplained daemon exits, make sure you
		1495	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1496	somewhere, as that would have given you a big clue).
		1497
		1498
		1499	=head3 Watcher-Specific Functions
428		1500
429	=over 4	1501	=over 4
430		1502
431	=item ev_io_init (ev_io *, callback, int fd, int events)	1503	=item ev_io_init (ev_io *, callback, int fd, int events)
432		1504
433	=item ev_io_set (ev_io *, int fd, int events)	1505	=item ev_io_set (ev_io *, int fd, int events)
434		1506
435	Configures an ev_io watcher. The fd is the file descriptor to rceeive	1507	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
436	events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ |	1508	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
437	EV_WRITE> to receive the given events.	1509	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
		1510
		1511	=item int fd [read-only]
		1512
		1513	The file descriptor being watched.
		1514
		1515	=item int events [read-only]
		1516
		1517	The events being watched.
438		1518
439	=back	1519	=back
440		1520
		1521	=head3 Examples
		1522
		1523	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
		1524	readable, but only once. Since it is likely line-buffered, you could
		1525	attempt to read a whole line in the callback.
		1526
		1527	static void
		1528	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
		1529	{
		1530	ev_io_stop (loop, w);
		1531	.. read from stdin here (or from w->fd) and handle any I/O errors
		1532	}
		1533
		1534	...
		1535	struct ev_loop *loop = ev_default_init (0);
		1536	ev_io stdin_readable;
		1537	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
		1538	ev_io_start (loop, &stdin_readable);
		1539	ev_loop (loop, 0);
		1540
		1541
441	=head2 struct ev_timer - relative and optionally recurring timeouts	1542	=head2 C<ev_timer> - relative and optionally repeating timeouts
442		1543
443	Timer watchers are simple relative timers that generate an event after a	1544	Timer watchers are simple relative timers that generate an event after a
444	given time, and optionally repeating in regular intervals after that.	1545	given time, and optionally repeating in regular intervals after that.
445		1546
446	The timers are based on real time, that is, if you register an event that	1547	The timers are based on real time, that is, if you register an event that
447	times out after an hour and youreset your system clock to last years	1548	times out after an hour and you reset your system clock to January last
448	time, it will still time out after (roughly) and hour. "Roughly" because	1549	year, it will still time out after (roughly) one hour. "Roughly" because
449	detecting time jumps is hard, and soem inaccuracies are unavoidable (the	1550	detecting time jumps is hard, and some inaccuracies are unavoidable (the
450	monotonic clock option helps a lot here).	1551	monotonic clock option helps a lot here).
451		1552
		1553	The callback is guaranteed to be invoked only I<after> its timeout has
		1554	passed (not I<at>, so on systems with very low-resolution clocks this
		1555	might introduce a small delay). If multiple timers become ready during the
		1556	same loop iteration then the ones with earlier time-out values are invoked
		1557	before ones of the same priority with later time-out values (but this is
		1558	no longer true when a callback calls C<ev_loop> recursively).
		1559
		1560	=head3 Be smart about timeouts
		1561
		1562	Many real-world problems involve some kind of timeout, usually for error
		1563	recovery. A typical example is an HTTP request - if the other side hangs,
		1564	you want to raise some error after a while.
		1565
		1566	What follows are some ways to handle this problem, from obvious and
		1567	inefficient to smart and efficient.
		1568
		1569	In the following, a 60 second activity timeout is assumed - a timeout that
		1570	gets reset to 60 seconds each time there is activity (e.g. each time some
		1571	data or other life sign was received).
		1572
452	=over 4	1573	=over 4
453		1574
		1575	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1576
		1577	This is the most obvious, but not the most simple way: In the beginning,
		1578	start the watcher:
		1579
		1580	ev_timer_init (timer, callback, 60., 0.);
		1581	ev_timer_start (loop, timer);
		1582
		1583	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1584	and start it again:
		1585
		1586	ev_timer_stop (loop, timer);
		1587	ev_timer_set (timer, 60., 0.);
		1588	ev_timer_start (loop, timer);
		1589
		1590	This is relatively simple to implement, but means that each time there is
		1591	some activity, libev will first have to remove the timer from its internal
		1592	data structure and then add it again. Libev tries to be fast, but it's
		1593	still not a constant-time operation.
		1594
		1595	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1596
		1597	This is the easiest way, and involves using C<ev_timer_again> instead of
		1598	C<ev_timer_start>.
		1599
		1600	To implement this, configure an C<ev_timer> with a C<repeat> value
		1601	of C<60> and then call C<ev_timer_again> at start and each time you
		1602	successfully read or write some data. If you go into an idle state where
		1603	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1604	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1605
		1606	That means you can ignore both the C<ev_timer_start> function and the
		1607	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1608	member and C<ev_timer_again>.
		1609
		1610	At start:
		1611
		1612	ev_init (timer, callback);
		1613	timer->repeat = 60.;
		1614	ev_timer_again (loop, timer);
		1615
		1616	Each time there is some activity:
		1617
		1618	ev_timer_again (loop, timer);
		1619
		1620	It is even possible to change the time-out on the fly, regardless of
		1621	whether the watcher is active or not:
		1622
		1623	timer->repeat = 30.;
		1624	ev_timer_again (loop, timer);
		1625
		1626	This is slightly more efficient then stopping/starting the timer each time
		1627	you want to modify its timeout value, as libev does not have to completely
		1628	remove and re-insert the timer from/into its internal data structure.
		1629
		1630	It is, however, even simpler than the "obvious" way to do it.
		1631
		1632	=item 3. Let the timer time out, but then re-arm it as required.
		1633
		1634	This method is more tricky, but usually most efficient: Most timeouts are
		1635	relatively long compared to the intervals between other activity - in
		1636	our example, within 60 seconds, there are usually many I/O events with
		1637	associated activity resets.
		1638
		1639	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1640	but remember the time of last activity, and check for a real timeout only
		1641	within the callback:
		1642
		1643	ev_tstamp last_activity; // time of last activity
		1644
		1645	static void
		1646	callback (EV_P_ ev_timer *w, int revents)
		1647	{
		1648	ev_tstamp now = ev_now (EV_A);
		1649	ev_tstamp timeout = last_activity + 60.;
		1650
		1651	// if last_activity + 60. is older than now, we did time out
		1652	if (timeout < now)
		1653	{
		1654	// timeout occured, take action
		1655	}
		1656	else
		1657	{
		1658	// callback was invoked, but there was some activity, re-arm
		1659	// the watcher to fire in last_activity + 60, which is
		1660	// guaranteed to be in the future, so "again" is positive:
		1661	w->repeat = timeout - now;
		1662	ev_timer_again (EV_A_ w);
		1663	}
		1664	}
		1665
		1666	To summarise the callback: first calculate the real timeout (defined
		1667	as "60 seconds after the last activity"), then check if that time has
		1668	been reached, which means something I<did>, in fact, time out. Otherwise
		1669	the callback was invoked too early (C<timeout> is in the future), so
		1670	re-schedule the timer to fire at that future time, to see if maybe we have
		1671	a timeout then.
		1672
		1673	Note how C<ev_timer_again> is used, taking advantage of the
		1674	C<ev_timer_again> optimisation when the timer is already running.
		1675
		1676	This scheme causes more callback invocations (about one every 60 seconds
		1677	minus half the average time between activity), but virtually no calls to
		1678	libev to change the timeout.
		1679
		1680	To start the timer, simply initialise the watcher and set C<last_activity>
		1681	to the current time (meaning we just have some activity :), then call the
		1682	callback, which will "do the right thing" and start the timer:
		1683
		1684	ev_init (timer, callback);
		1685	last_activity = ev_now (loop);
		1686	callback (loop, timer, EV_TIMEOUT);
		1687
		1688	And when there is some activity, simply store the current time in
		1689	C<last_activity>, no libev calls at all:
		1690
		1691	last_actiivty = ev_now (loop);
		1692
		1693	This technique is slightly more complex, but in most cases where the
		1694	time-out is unlikely to be triggered, much more efficient.
		1695
		1696	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1697	callback :) - just change the timeout and invoke the callback, which will
		1698	fix things for you.
		1699
		1700	=item 4. Wee, just use a double-linked list for your timeouts.
		1701
		1702	If there is not one request, but many thousands (millions...), all
		1703	employing some kind of timeout with the same timeout value, then one can
		1704	do even better:
		1705
		1706	When starting the timeout, calculate the timeout value and put the timeout
		1707	at the I<end> of the list.
		1708
		1709	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1710	the list is expected to fire (for example, using the technique #3).
		1711
		1712	When there is some activity, remove the timer from the list, recalculate
		1713	the timeout, append it to the end of the list again, and make sure to
		1714	update the C<ev_timer> if it was taken from the beginning of the list.
		1715
		1716	This way, one can manage an unlimited number of timeouts in O(1) time for
		1717	starting, stopping and updating the timers, at the expense of a major
		1718	complication, and having to use a constant timeout. The constant timeout
		1719	ensures that the list stays sorted.
		1720
		1721	=back
		1722
		1723	So which method the best?
		1724
		1725	Method #2 is a simple no-brain-required solution that is adequate in most
		1726	situations. Method #3 requires a bit more thinking, but handles many cases
		1727	better, and isn't very complicated either. In most case, choosing either
		1728	one is fine, with #3 being better in typical situations.
		1729
		1730	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1731	rather complicated, but extremely efficient, something that really pays
		1732	off after the first million or so of active timers, i.e. it's usually
		1733	overkill :)
		1734
		1735	=head3 The special problem of time updates
		1736
		1737	Establishing the current time is a costly operation (it usually takes at
		1738	least two system calls): EV therefore updates its idea of the current
		1739	time only before and after C<ev_loop> collects new events, which causes a
		1740	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1741	lots of events in one iteration.
		1742
		1743	The relative timeouts are calculated relative to the C<ev_now ()>
		1744	time. This is usually the right thing as this timestamp refers to the time
		1745	of the event triggering whatever timeout you are modifying/starting. If
		1746	you suspect event processing to be delayed and you I<need> to base the
		1747	timeout on the current time, use something like this to adjust for this:
		1748
		1749	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
		1750
		1751	If the event loop is suspended for a long time, you can also force an
		1752	update of the time returned by C<ev_now ()> by calling C<ev_now_update
		1753	()>.
		1754
		1755	=head3 Watcher-Specific Functions and Data Members
		1756
		1757	=over 4
		1758
454	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1759	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
455		1760
456	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1761	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
457		1762
458	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1763	Configure the timer to trigger after C<after> seconds. If C<repeat>
459	C<0.>, then it will automatically be stopped. If it is positive, then the	1764	is C<0.>, then it will automatically be stopped once the timeout is
460	timer will automatically be configured to trigger again C<repeat> seconds	1765	reached. If it is positive, then the timer will automatically be
461	later, again, and again, until stopped manually.	1766	configured to trigger again C<repeat> seconds later, again, and again,
		1767	until stopped manually.
462		1768
463	The timer itself will do a best-effort at avoiding drift, that is, if you	1769	The timer itself will do a best-effort at avoiding drift, that is, if
464	configure a timer to trigger every 10 seconds, then it will trigger at	1770	you configure a timer to trigger every 10 seconds, then it will normally
465	exactly 10 second intervals. If, however, your program cannot keep up with	1771	trigger at exactly 10 second intervals. If, however, your program cannot
466	the timer (ecause it takes longer than those 10 seconds to do stuff) the	1772	keep up with the timer (because it takes longer than those 10 seconds to
467	timer will not fire more than once per event loop iteration.	1773	do stuff) the timer will not fire more than once per event loop iteration.
468		1774
469	=item ev_timer_again (loop)	1775	=item ev_timer_again (loop, ev_timer *)
470		1776
471	This will act as if the timer timed out and restart it again if it is	1777	This will act as if the timer timed out and restart it again if it is
472	repeating. The exact semantics are:	1778	repeating. The exact semantics are:
473		1779
		1780	If the timer is pending, its pending status is cleared.
		1781
474	If the timer is started but nonrepeating, stop it.	1782	If the timer is started but non-repeating, stop it (as if it timed out).
475		1783
476	If the timer is repeating, either start it if necessary (with the repeat	1784	If the timer is repeating, either start it if necessary (with the
477	value), or reset the running timer to the repeat value.	1785	C<repeat> value), or reset the running timer to the C<repeat> value.
478		1786
479	This sounds a bit complicated, but here is a useful and typical	1787	This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
480	example: Imagine you have a tcp connection and you want a so-called idle	1788	usage example.
481	timeout, that is, you want to be called when there have been, say, 60	1789
482	seconds of inactivity on the socket. The easiest way to do this is to	1790	=item ev_tstamp repeat [read-write]
483	configure an ev_timer with after=repeat=60 and calling ev_timer_again each	1791
484	time you successfully read or write some data. If you go into an idle	1792	The current C<repeat> value. Will be used each time the watcher times out
485	state where you do not expect data to travel on the socket, you can stop	1793	or C<ev_timer_again> is called, and determines the next timeout (if any),
486	the timer, and again will automatically restart it if need be.	1794	which is also when any modifications are taken into account.
487		1795
488	=back	1796	=back
489		1797
		1798	=head3 Examples
		1799
		1800	Example: Create a timer that fires after 60 seconds.
		1801
		1802	static void
		1803	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
		1804	{
		1805	.. one minute over, w is actually stopped right here
		1806	}
		1807
		1808	ev_timer mytimer;
		1809	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
		1810	ev_timer_start (loop, &mytimer);
		1811
		1812	Example: Create a timeout timer that times out after 10 seconds of
		1813	inactivity.
		1814
		1815	static void
		1816	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
		1817	{
		1818	.. ten seconds without any activity
		1819	}
		1820
		1821	ev_timer mytimer;
		1822	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
		1823	ev_timer_again (&mytimer); /* start timer */
		1824	ev_loop (loop, 0);
		1825
		1826	// and in some piece of code that gets executed on any "activity":
		1827	// reset the timeout to start ticking again at 10 seconds
		1828	ev_timer_again (&mytimer);
		1829
		1830
490	=head2 ev_periodic - to cron or not to cron it	1831	=head2 C<ev_periodic> - to cron or not to cron?
491		1832
492	Periodic watchers are also timers of a kind, but they are very versatile	1833	Periodic watchers are also timers of a kind, but they are very versatile
493	(and unfortunately a bit complex).	1834	(and unfortunately a bit complex).
494		1835
495	Unlike ev_timer's, they are not based on real time (or relative time)	1836	Unlike C<ev_timer>, periodic watchers are not based on real time (or
496	but on wallclock time (absolute time). You can tell a periodic watcher	1837	relative time, the physical time that passes) but on wall clock time
497	to trigger "at" some specific point in time. For example, if you tell a	1838	(absolute time, the thing you can read on your calender or clock). The
498	periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now ()	1839	difference is that wall clock time can run faster or slower than real
499	+ 10.>) and then reset your system clock to the last year, then it will	1840	time, and time jumps are not uncommon (e.g. when you adjust your
500	take a year to trigger the event (unlike an ev_timer, which would trigger	1841	wrist-watch).
501	roughly 10 seconds later and of course not if you reset your system time
502	again).
503		1842
504	They can also be used to implement vastly more complex timers, such as	1843	You can tell a periodic watcher to trigger after some specific point
		1844	in time: for example, if you tell a periodic watcher to trigger "in 10
		1845	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		1846	not a delay) and then reset your system clock to January of the previous
		1847	year, then it will take a year or more to trigger the event (unlike an
		1848	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		1849	it, as it uses a relative timeout).
		1850
		1851	C<ev_periodic> watchers can also be used to implement vastly more complex
505	triggering an event on eahc midnight, local time.	1852	timers, such as triggering an event on each "midnight, local time", or
		1853	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		1854	those cannot react to time jumps.
		1855
		1856	As with timers, the callback is guaranteed to be invoked only when the
		1857	point in time where it is supposed to trigger has passed. If multiple
		1858	timers become ready during the same loop iteration then the ones with
		1859	earlier time-out values are invoked before ones with later time-out values
		1860	(but this is no longer true when a callback calls C<ev_loop> recursively).
		1861
		1862	=head3 Watcher-Specific Functions and Data Members
506		1863
507	=over 4	1864	=over 4
508		1865
509	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1866	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
510		1867
511	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1868	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
512		1869
513	Lots of arguments, lets sort it out... There are basically three modes of	1870	Lots of arguments, let's sort it out... There are basically three modes of
514	operation, and we will explain them from simplest to complex:	1871	operation, and we will explain them from simplest to most complex:
515
516		1872
517	=over 4	1873	=over 4
518		1874
519	=item * absolute timer (interval = reschedule_cb = 0)	1875	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
520		1876
521	In this configuration the watcher triggers an event at the wallclock time	1877	In this configuration the watcher triggers an event after the wall clock
522	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1878	time C<offset> has passed. It will not repeat and will not adjust when a
523	that is, if it is to be run at January 1st 2011 then it will run when the	1879	time jump occurs, that is, if it is to be run at January 1st 2011 then it
524	system time reaches or surpasses this time.	1880	will be stopped and invoked when the system clock reaches or surpasses
		1881	this point in time.
525		1882
526	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1883	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
527		1884
528	In this mode the watcher will always be scheduled to time out at the next	1885	In this mode the watcher will always be scheduled to time out at the next
529	C<at + N * interval> time (for some integer N) and then repeat, regardless	1886	C<offset + N * interval> time (for some integer N, which can also be
530	of any time jumps.	1887	negative) and then repeat, regardless of any time jumps. The C<offset>
		1888	argument is merely an offset into the C<interval> periods.
531		1889
532	This can be used to create timers that do not drift with respect to system	1890	This can be used to create timers that do not drift with respect to the
533	time:	1891	system clock, for example, here is an C<ev_periodic> that triggers each
		1892	hour, on the hour (with respect to UTC):
534		1893
535	ev_periodic_set (&periodic, 0., 3600., 0);	1894	ev_periodic_set (&periodic, 0., 3600., 0);
536		1895
537	This doesn't mean there will always be 3600 seconds in between triggers,	1896	This doesn't mean there will always be 3600 seconds in between triggers,
538	but only that the the callback will be called when the system time shows a	1897	but only that the callback will be called when the system time shows a
539	full hour (UTC), or more correct, when the system time is evenly divisible	1898	full hour (UTC), or more correctly, when the system time is evenly divisible
540	by 3600.	1899	by 3600.
541		1900
542	Another way to think about it (for the mathematically inclined) is that	1901	Another way to think about it (for the mathematically inclined) is that
543	ev_periodic will try to run the callback in this mode at the next possible	1902	C<ev_periodic> will try to run the callback in this mode at the next possible
544	time where C<time = at (mod interval)>, regardless of any time jumps.	1903	time where C<time = offset (mod interval)>, regardless of any time jumps.
545		1904
		1905	For numerical stability it is preferable that the C<offset> value is near
		1906	C<ev_now ()> (the current time), but there is no range requirement for
		1907	this value, and in fact is often specified as zero.
		1908
		1909	Note also that there is an upper limit to how often a timer can fire (CPU
		1910	speed for example), so if C<interval> is very small then timing stability
		1911	will of course deteriorate. Libev itself tries to be exact to be about one
		1912	millisecond (if the OS supports it and the machine is fast enough).
		1913
546	=item * manual reschedule mode (reschedule_cb = callback)	1914	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
547		1915
548	In this mode the values for C<interval> and C<at> are both being	1916	In this mode the values for C<interval> and C<offset> are both being
549	ignored. Instead, each time the periodic watcher gets scheduled, the	1917	ignored. Instead, each time the periodic watcher gets scheduled, the
550	reschedule callback will be called with the watcher as first, and the	1918	reschedule callback will be called with the watcher as first, and the
551	current time as second argument.	1919	current time as second argument.
552		1920
553	NOTE: I<This callback MUST NOT stop or destroy the periodic or any other	1921	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
554	periodic watcher, ever, or make any event loop modificstions>. If you need	1922	or make ANY other event loop modifications whatsoever, unless explicitly
555	to stop it, return 1e30 (or so, fudge fudge) and stop it afterwards.	1923	allowed by documentation here>.
556		1924
		1925	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1926	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1927	only event loop modification you are allowed to do).
		1928
557	Its prototype is c<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1929	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
558	ev_tstamp now)>, e.g.:	1930	*w, ev_tstamp now)>, e.g.:
559		1931
		1932	static ev_tstamp
560	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1933	my_rescheduler (ev_periodic *w, ev_tstamp now)
561	{	1934	{
562	return now + 60.;	1935	return now + 60.;
563	}	1936	}
564		1937
565	It must return the next time to trigger, based on the passed time value	1938	It must return the next time to trigger, based on the passed time value
566	(that is, the lowest time value larger than to the second argument). It	1939	(that is, the lowest time value larger than to the second argument). It
567	will usually be called just before the callback will be triggered, but	1940	will usually be called just before the callback will be triggered, but
568	might be called at other times, too.	1941	might be called at other times, too.
569		1942
		1943	NOTE: I<< This callback must always return a time that is higher than or
		1944	equal to the passed C<now> value >>.
		1945
570	This can be used to create very complex timers, such as a timer that	1946	This can be used to create very complex timers, such as a timer that
571	triggers on each midnight, local time. To do this, you would calculate the	1947	triggers on "next midnight, local time". To do this, you would calculate the
572	next midnight after C<now> and return the timestamp value for this. How you do this	1948	next midnight after C<now> and return the timestamp value for this. How
573	is, again, up to you (but it is not trivial).	1949	you do this is, again, up to you (but it is not trivial, which is the main
		1950	reason I omitted it as an example).
574		1951
575	=back	1952	=back
576		1953
577	=item ev_periodic_again (loop, ev_periodic *)	1954	=item ev_periodic_again (loop, ev_periodic *)
578		1955
579	Simply stops and restarts the periodic watcher again. This is only useful	1956	Simply stops and restarts the periodic watcher again. This is only useful
580	when you changed some parameters or the reschedule callback would return	1957	when you changed some parameters or the reschedule callback would return
581	a different time than the last time it was called (e.g. in a crond like	1958	a different time than the last time it was called (e.g. in a crond like
582	program when the crontabs have changed).	1959	program when the crontabs have changed).
583		1960
		1961	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1962
		1963	When active, returns the absolute time that the watcher is supposed
		1964	to trigger next. This is not the same as the C<offset> argument to
		1965	C<ev_periodic_set>, but indeed works even in interval and manual
		1966	rescheduling modes.
		1967
		1968	=item ev_tstamp offset [read-write]
		1969
		1970	When repeating, this contains the offset value, otherwise this is the
		1971	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		1972	although libev might modify this value for better numerical stability).
		1973
		1974	Can be modified any time, but changes only take effect when the periodic
		1975	timer fires or C<ev_periodic_again> is being called.
		1976
		1977	=item ev_tstamp interval [read-write]
		1978
		1979	The current interval value. Can be modified any time, but changes only
		1980	take effect when the periodic timer fires or C<ev_periodic_again> is being
		1981	called.
		1982
		1983	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
		1984
		1985	The current reschedule callback, or C<0>, if this functionality is
		1986	switched off. Can be changed any time, but changes only take effect when
		1987	the periodic timer fires or C<ev_periodic_again> is being called.
		1988
584	=back	1989	=back
585		1990
		1991	=head3 Examples
		1992
		1993	Example: Call a callback every hour, or, more precisely, whenever the
		1994	system time is divisible by 3600. The callback invocation times have
		1995	potentially a lot of jitter, but good long-term stability.
		1996
		1997	static void
		1998	clock_cb (struct ev_loop *loop, ev_io *w, int revents)
		1999	{
		2000	... its now a full hour (UTC, or TAI or whatever your clock follows)
		2001	}
		2002
		2003	ev_periodic hourly_tick;
		2004	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
		2005	ev_periodic_start (loop, &hourly_tick);
		2006
		2007	Example: The same as above, but use a reschedule callback to do it:
		2008
		2009	#include <math.h>
		2010
		2011	static ev_tstamp
		2012	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
		2013	{
		2014	return now + (3600. - fmod (now, 3600.));
		2015	}
		2016
		2017	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
		2018
		2019	Example: Call a callback every hour, starting now:
		2020
		2021	ev_periodic hourly_tick;
		2022	ev_periodic_init (&hourly_tick, clock_cb,
		2023	fmod (ev_now (loop), 3600.), 3600., 0);
		2024	ev_periodic_start (loop, &hourly_tick);
		2025
		2026
586	=head2 ev_signal - signal me when a signal gets signalled	2027	=head2 C<ev_signal> - signal me when a signal gets signalled!
587		2028
588	Signal watchers will trigger an event when the process receives a specific	2029	Signal watchers will trigger an event when the process receives a specific
589	signal one or more times. Even though signals are very asynchronous, libev	2030	signal one or more times. Even though signals are very asynchronous, libev
590	will try its best to deliver signals synchronously, i.e. as part of the	2031	will try it's best to deliver signals synchronously, i.e. as part of the
591	normal event processing, like any other event.	2032	normal event processing, like any other event.
592		2033
		2034	If you want signals asynchronously, just use C<sigaction> as you would
		2035	do without libev and forget about sharing the signal. You can even use
		2036	C<ev_async> from a signal handler to synchronously wake up an event loop.
		2037
593	You cna configure as many watchers as you like per signal. Only when the	2038	You can configure as many watchers as you like per signal. Only when the
594	first watcher gets started will libev actually register a signal watcher	2039	first watcher gets started will libev actually register a signal handler
595	with the kernel (thus it coexists with your own signal handlers as long	2040	with the kernel (thus it coexists with your own signal handlers as long as
596	as you don't register any with libev). Similarly, when the last signal	2041	you don't register any with libev for the same signal). Similarly, when
597	watcher for a signal is stopped libev will reset the signal handler to	2042	the last signal watcher for a signal is stopped, libev will reset the
598	SIG_DFL (regardless of what it was set to before).	2043	signal handler to SIG_DFL (regardless of what it was set to before).
		2044
		2045	If possible and supported, libev will install its handlers with
		2046	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
		2047	interrupted. If you have a problem with system calls getting interrupted by
		2048	signals you can block all signals in an C<ev_check> watcher and unblock
		2049	them in an C<ev_prepare> watcher.
		2050
		2051	=head3 Watcher-Specific Functions and Data Members
599		2052
600	=over 4	2053	=over 4
601		2054
602	=item ev_signal_init (ev_signal *, callback, int signum)	2055	=item ev_signal_init (ev_signal *, callback, int signum)
603		2056
604	=item ev_signal_set (ev_signal *, int signum)	2057	=item ev_signal_set (ev_signal *, int signum)
605		2058
606	Configures the watcher to trigger on the given signal number (usually one	2059	Configures the watcher to trigger on the given signal number (usually one
607	of the C<SIGxxx> constants).	2060	of the C<SIGxxx> constants).
608		2061
		2062	=item int signum [read-only]
		2063
		2064	The signal the watcher watches out for.
		2065
609	=back	2066	=back
610		2067
		2068	=head3 Examples
		2069
		2070	Example: Try to exit cleanly on SIGINT.
		2071
		2072	static void
		2073	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
		2074	{
		2075	ev_unloop (loop, EVUNLOOP_ALL);
		2076	}
		2077
		2078	ev_signal signal_watcher;
		2079	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		2080	ev_signal_start (loop, &signal_watcher);
		2081
		2082
611	=head2 ev_child - wait for pid status changes	2083	=head2 C<ev_child> - watch out for process status changes
612		2084
613	Child watchers trigger when your process receives a SIGCHLD in response to	2085	Child watchers trigger when your process receives a SIGCHLD in response to
614	some child status changes (most typically when a child of yours dies).	2086	some child status changes (most typically when a child of yours dies or
		2087	exits). It is permissible to install a child watcher I<after> the child
		2088	has been forked (which implies it might have already exited), as long
		2089	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2090	forking and then immediately registering a watcher for the child is fine,
		2091	but forking and registering a watcher a few event loop iterations later or
		2092	in the next callback invocation is not.
		2093
		2094	Only the default event loop is capable of handling signals, and therefore
		2095	you can only register child watchers in the default event loop.
		2096
		2097	Due to some design glitches inside libev, child watchers will always be
		2098	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2099	libev)
		2100
		2101	=head3 Process Interaction
		2102
		2103	Libev grabs C<SIGCHLD> as soon as the default event loop is
		2104	initialised. This is necessary to guarantee proper behaviour even if
		2105	the first child watcher is started after the child exits. The occurrence
		2106	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		2107	synchronously as part of the event loop processing. Libev always reaps all
		2108	children, even ones not watched.
		2109
		2110	=head3 Overriding the Built-In Processing
		2111
		2112	Libev offers no special support for overriding the built-in child
		2113	processing, but if your application collides with libev's default child
		2114	handler, you can override it easily by installing your own handler for
		2115	C<SIGCHLD> after initialising the default loop, and making sure the
		2116	default loop never gets destroyed. You are encouraged, however, to use an
		2117	event-based approach to child reaping and thus use libev's support for
		2118	that, so other libev users can use C<ev_child> watchers freely.
		2119
		2120	=head3 Stopping the Child Watcher
		2121
		2122	Currently, the child watcher never gets stopped, even when the
		2123	child terminates, so normally one needs to stop the watcher in the
		2124	callback. Future versions of libev might stop the watcher automatically
		2125	when a child exit is detected.
		2126
		2127	=head3 Watcher-Specific Functions and Data Members
615		2128
616	=over 4	2129	=over 4
617		2130
618	=item ev_child_init (ev_child *, callback, int pid)	2131	=item ev_child_init (ev_child *, callback, int pid, int trace)
619		2132
620	=item ev_child_set (ev_child *, int pid)	2133	=item ev_child_set (ev_child *, int pid, int trace)
621		2134
622	Configures the watcher to wait for status changes of process C<pid> (or	2135	Configures the watcher to wait for status changes of process C<pid> (or
623	I<any> process if C<pid> is specified as C<0>). The callback can look	2136	I<any> process if C<pid> is specified as C<0>). The callback can look
624	at the C<rstatus> member of the C<ev_child> watcher structure to see	2137	at the C<rstatus> member of the C<ev_child> watcher structure to see
625	the status word (use the macros from C<sys/wait.h>). The C<rpid> member	2138	the status word (use the macros from C<sys/wait.h> and see your systems
626	contains the pid of the process causing the status change.	2139	C<waitpid> documentation). The C<rpid> member contains the pid of the
		2140	process causing the status change. C<trace> must be either C<0> (only
		2141	activate the watcher when the process terminates) or C<1> (additionally
		2142	activate the watcher when the process is stopped or continued).
		2143
		2144	=item int pid [read-only]
		2145
		2146	The process id this watcher watches out for, or C<0>, meaning any process id.
		2147
		2148	=item int rpid [read-write]
		2149
		2150	The process id that detected a status change.
		2151
		2152	=item int rstatus [read-write]
		2153
		2154	The process exit/trace status caused by C<rpid> (see your systems
		2155	C<waitpid> and C<sys/wait.h> documentation for details).
627		2156
628	=back	2157	=back
629		2158
		2159	=head3 Examples
		2160
		2161	Example: C<fork()> a new process and install a child handler to wait for
		2162	its completion.
		2163
		2164	ev_child cw;
		2165
		2166	static void
		2167	child_cb (EV_P_ ev_child *w, int revents)
		2168	{
		2169	ev_child_stop (EV_A_ w);
		2170	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
		2171	}
		2172
		2173	pid_t pid = fork ();
		2174
		2175	if (pid < 0)
		2176	// error
		2177	else if (pid == 0)
		2178	{
		2179	// the forked child executes here
		2180	exit (1);
		2181	}
		2182	else
		2183	{
		2184	ev_child_init (&cw, child_cb, pid, 0);
		2185	ev_child_start (EV_DEFAULT_ &cw);
		2186	}
		2187
		2188
		2189	=head2 C<ev_stat> - did the file attributes just change?
		2190
		2191	This watches a file system path for attribute changes. That is, it calls
		2192	C<stat> on that path in regular intervals (or when the OS says it changed)
		2193	and sees if it changed compared to the last time, invoking the callback if
		2194	it did.
		2195
		2196	The path does not need to exist: changing from "path exists" to "path does
		2197	not exist" is a status change like any other. The condition "path does not
		2198	exist" (or more correctly "path cannot be stat'ed") is signified by the
		2199	C<st_nlink> field being zero (which is otherwise always forced to be at
		2200	least one) and all the other fields of the stat buffer having unspecified
		2201	contents.
		2202
		2203	The path I<must not> end in a slash or contain special components such as
		2204	C<.> or C<..>. The path I<should> be absolute: If it is relative and
		2205	your working directory changes, then the behaviour is undefined.
		2206
		2207	Since there is no portable change notification interface available, the
		2208	portable implementation simply calls C<stat(2)> regularly on the path
		2209	to see if it changed somehow. You can specify a recommended polling
		2210	interval for this case. If you specify a polling interval of C<0> (highly
		2211	recommended!) then a I<suitable, unspecified default> value will be used
		2212	(which you can expect to be around five seconds, although this might
		2213	change dynamically). Libev will also impose a minimum interval which is
		2214	currently around C<0.1>, but that's usually overkill.
		2215
		2216	This watcher type is not meant for massive numbers of stat watchers,
		2217	as even with OS-supported change notifications, this can be
		2218	resource-intensive.
		2219
		2220	At the time of this writing, the only OS-specific interface implemented
		2221	is the Linux inotify interface (implementing kqueue support is left as an
		2222	exercise for the reader. Note, however, that the author sees no way of
		2223	implementing C<ev_stat> semantics with kqueue, except as a hint).
		2224
		2225	=head3 ABI Issues (Largefile Support)
		2226
		2227	Libev by default (unless the user overrides this) uses the default
		2228	compilation environment, which means that on systems with large file
		2229	support disabled by default, you get the 32 bit version of the stat
		2230	structure. When using the library from programs that change the ABI to
		2231	use 64 bit file offsets the programs will fail. In that case you have to
		2232	compile libev with the same flags to get binary compatibility. This is
		2233	obviously the case with any flags that change the ABI, but the problem is
		2234	most noticeably displayed with ev_stat and large file support.
		2235
		2236	The solution for this is to lobby your distribution maker to make large
		2237	file interfaces available by default (as e.g. FreeBSD does) and not
		2238	optional. Libev cannot simply switch on large file support because it has
		2239	to exchange stat structures with application programs compiled using the
		2240	default compilation environment.
		2241
		2242	=head3 Inotify and Kqueue
		2243
		2244	When C<inotify (7)> support has been compiled into libev and present at
		2245	runtime, it will be used to speed up change detection where possible. The
		2246	inotify descriptor will be created lazily when the first C<ev_stat>
		2247	watcher is being started.
		2248
		2249	Inotify presence does not change the semantics of C<ev_stat> watchers
		2250	except that changes might be detected earlier, and in some cases, to avoid
		2251	making regular C<stat> calls. Even in the presence of inotify support
		2252	there are many cases where libev has to resort to regular C<stat> polling,
		2253	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2254	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2255	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2256	xfs are fully working) libev usually gets away without polling.
		2257
		2258	There is no support for kqueue, as apparently it cannot be used to
		2259	implement this functionality, due to the requirement of having a file
		2260	descriptor open on the object at all times, and detecting renames, unlinks
		2261	etc. is difficult.
		2262
		2263	=head3 C<stat ()> is a synchronous operation
		2264
		2265	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2266	the process. The exception are C<ev_stat> watchers - those call C<stat
		2267	()>, which is a synchronous operation.
		2268
		2269	For local paths, this usually doesn't matter: unless the system is very
		2270	busy or the intervals between stat's are large, a stat call will be fast,
		2271	as the path data is usually in memory already (except when starting the
		2272	watcher).
		2273
		2274	For networked file systems, calling C<stat ()> can block an indefinite
		2275	time due to network issues, and even under good conditions, a stat call
		2276	often takes multiple milliseconds.
		2277
		2278	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2279	paths, although this is fully supported by libev.
		2280
		2281	=head3 The special problem of stat time resolution
		2282
		2283	The C<stat ()> system call only supports full-second resolution portably,
		2284	and even on systems where the resolution is higher, most file systems
		2285	still only support whole seconds.
		2286
		2287	That means that, if the time is the only thing that changes, you can
		2288	easily miss updates: on the first update, C<ev_stat> detects a change and
		2289	calls your callback, which does something. When there is another update
		2290	within the same second, C<ev_stat> will be unable to detect unless the
		2291	stat data does change in other ways (e.g. file size).
		2292
		2293	The solution to this is to delay acting on a change for slightly more
		2294	than a second (or till slightly after the next full second boundary), using
		2295	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		2296	ev_timer_again (loop, w)>).
		2297
		2298	The C<.02> offset is added to work around small timing inconsistencies
		2299	of some operating systems (where the second counter of the current time
		2300	might be be delayed. One such system is the Linux kernel, where a call to
		2301	C<gettimeofday> might return a timestamp with a full second later than
		2302	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		2303	update file times then there will be a small window where the kernel uses
		2304	the previous second to update file times but libev might already execute
		2305	the timer callback).
		2306
		2307	=head3 Watcher-Specific Functions and Data Members
		2308
		2309	=over 4
		2310
		2311	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
		2312
		2313	=item ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)
		2314
		2315	Configures the watcher to wait for status changes of the given
		2316	C<path>. The C<interval> is a hint on how quickly a change is expected to
		2317	be detected and should normally be specified as C<0> to let libev choose
		2318	a suitable value. The memory pointed to by C<path> must point to the same
		2319	path for as long as the watcher is active.
		2320
		2321	The callback will receive an C<EV_STAT> event when a change was detected,
		2322	relative to the attributes at the time the watcher was started (or the
		2323	last change was detected).
		2324
		2325	=item ev_stat_stat (loop, ev_stat *)
		2326
		2327	Updates the stat buffer immediately with new values. If you change the
		2328	watched path in your callback, you could call this function to avoid
		2329	detecting this change (while introducing a race condition if you are not
		2330	the only one changing the path). Can also be useful simply to find out the
		2331	new values.
		2332
		2333	=item ev_statdata attr [read-only]
		2334
		2335	The most-recently detected attributes of the file. Although the type is
		2336	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
		2337	suitable for your system, but you can only rely on the POSIX-standardised
		2338	members to be present. If the C<st_nlink> member is C<0>, then there was
		2339	some error while C<stat>ing the file.
		2340
		2341	=item ev_statdata prev [read-only]
		2342
		2343	The previous attributes of the file. The callback gets invoked whenever
		2344	C<prev> != C<attr>, or, more precisely, one or more of these members
		2345	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		2346	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
		2347
		2348	=item ev_tstamp interval [read-only]
		2349
		2350	The specified interval.
		2351
		2352	=item const char *path [read-only]
		2353
		2354	The file system path that is being watched.
		2355
		2356	=back
		2357
		2358	=head3 Examples
		2359
		2360	Example: Watch C</etc/passwd> for attribute changes.
		2361
		2362	static void
		2363	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
		2364	{
		2365	/* /etc/passwd changed in some way */
		2366	if (w->attr.st_nlink)
		2367	{
		2368	printf ("passwd current size %ld\n", (long)w->attr.st_size);
		2369	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
		2370	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
		2371	}
		2372	else
		2373	/* you shalt not abuse printf for puts */
		2374	puts ("wow, /etc/passwd is not there, expect problems. "
		2375	"if this is windows, they already arrived\n");
		2376	}
		2377
		2378	...
		2379	ev_stat passwd;
		2380
		2381	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
		2382	ev_stat_start (loop, &passwd);
		2383
		2384	Example: Like above, but additionally use a one-second delay so we do not
		2385	miss updates (however, frequent updates will delay processing, too, so
		2386	one might do the work both on C<ev_stat> callback invocation I<and> on
		2387	C<ev_timer> callback invocation).
		2388
		2389	static ev_stat passwd;
		2390	static ev_timer timer;
		2391
		2392	static void
		2393	timer_cb (EV_P_ ev_timer *w, int revents)
		2394	{
		2395	ev_timer_stop (EV_A_ w);
		2396
		2397	/* now it's one second after the most recent passwd change */
		2398	}
		2399
		2400	static void
		2401	stat_cb (EV_P_ ev_stat *w, int revents)
		2402	{
		2403	/* reset the one-second timer */
		2404	ev_timer_again (EV_A_ &timer);
		2405	}
		2406
		2407	...
		2408	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		2409	ev_stat_start (loop, &passwd);
		2410	ev_timer_init (&timer, timer_cb, 0., 1.02);
		2411
		2412
630	=head2 ev_idle - when you've got nothing better to do	2413	=head2 C<ev_idle> - when you've got nothing better to do...
631		2414
632	Idle watchers trigger events when there are no other I/O or timer (or	2415	Idle watchers trigger events when no other events of the same or higher
633	periodic) events pending. That is, as long as your process is busy	2416	priority are pending (prepare, check and other idle watchers do not count
634	handling sockets or timeouts it will not be called. But when your process	2417	as receiving "events").
635	is idle all idle watchers are being called again and again - until	2418
		2419	That is, as long as your process is busy handling sockets or timeouts
		2420	(or even signals, imagine) of the same or higher priority it will not be
		2421	triggered. But when your process is idle (or only lower-priority watchers
		2422	are pending), the idle watchers are being called once per event loop
636	stopped, that is, or your process receives more events.	2423	iteration - until stopped, that is, or your process receives more events
		2424	and becomes busy again with higher priority stuff.
637		2425
638	The most noteworthy effect is that as long as any idle watchers are	2426	The most noteworthy effect is that as long as any idle watchers are
639	active, the process will not block when waiting for new events.	2427	active, the process will not block when waiting for new events.
640		2428
641	Apart from keeping your process non-blocking (which is a useful	2429	Apart from keeping your process non-blocking (which is a useful
642	effect on its own sometimes), idle watchers are a good place to do	2430	effect on its own sometimes), idle watchers are a good place to do
643	"pseudo-background processing", or delay processing stuff to after the	2431	"pseudo-background processing", or delay processing stuff to after the
644	event loop has handled all outstanding events.	2432	event loop has handled all outstanding events.
645		2433
		2434	=head3 Watcher-Specific Functions and Data Members
		2435
646	=over 4	2436	=over 4
647		2437
648	=item ev_idle_init (ev_signal *, callback)	2438	=item ev_idle_init (ev_idle *, callback)
649		2439
650	Initialises and configures the idle watcher - it has no parameters of any	2440	Initialises and configures the idle watcher - it has no parameters of any
651	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2441	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
652	believe me.	2442	believe me.
653		2443
654	=back	2444	=back
655		2445
656	=head2 prepare and check - your hooks into the event loop	2446	=head3 Examples
657		2447
		2448	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
		2449	callback, free it. Also, use no error checking, as usual.
		2450
		2451	static void
		2452	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
		2453	{
		2454	free (w);
		2455	// now do something you wanted to do when the program has
		2456	// no longer anything immediate to do.
		2457	}
		2458
		2459	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
		2460	ev_idle_init (idle_watcher, idle_cb);
		2461	ev_idle_start (loop, idle_watcher);
		2462
		2463
		2464	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
		2465
658	Prepare and check watchers usually (but not always) are used in	2466	Prepare and check watchers are usually (but not always) used in pairs:
659	tandom. Prepare watchers get invoked before the process blocks and check	2467	prepare watchers get invoked before the process blocks and check watchers
660	watchers afterwards.	2468	afterwards.
661		2469
		2470	You I<must not> call C<ev_loop> or similar functions that enter
		2471	the current event loop from either C<ev_prepare> or C<ev_check>
		2472	watchers. Other loops than the current one are fine, however. The
		2473	rationale behind this is that you do not need to check for recursion in
		2474	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
		2475	C<ev_check> so if you have one watcher of each kind they will always be
		2476	called in pairs bracketing the blocking call.
		2477
662	Their main purpose is to integrate other event mechanisms into libev. This	2478	Their main purpose is to integrate other event mechanisms into libev and
663	could be used, for example, to track variable changes, implement your own	2479	their use is somewhat advanced. They could be used, for example, to track
664	watchers, integrate net-snmp or a coroutine library and lots more.	2480	variable changes, implement your own watchers, integrate net-snmp or a
		2481	coroutine library and lots more. They are also occasionally useful if
		2482	you cache some data and want to flush it before blocking (for example,
		2483	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
		2484	watcher).
665		2485
666	This is done by examining in each prepare call which file descriptors need	2486	This is done by examining in each prepare call which file descriptors
667	to be watched by the other library, registering ev_io watchers for them	2487	need to be watched by the other library, registering C<ev_io> watchers
668	and starting an ev_timer watcher for any timeouts (many libraries provide	2488	for them and starting an C<ev_timer> watcher for any timeouts (many
669	just this functionality). Then, in the check watcher you check for any	2489	libraries provide exactly this functionality). Then, in the check watcher,
670	events that occured (by making your callbacks set soem flags for example)	2490	you check for any events that occurred (by checking the pending status
671	and call back into the library.	2491	of all watchers and stopping them) and call back into the library. The
		2492	I/O and timer callbacks will never actually be called (but must be valid
		2493	nevertheless, because you never know, you know?).
672		2494
673	As another example, the perl Coro module uses these hooks to integrate	2495	As another example, the Perl Coro module uses these hooks to integrate
674	coroutines into libev programs, by yielding to other active coroutines	2496	coroutines into libev programs, by yielding to other active coroutines
675	during each prepare and only letting the process block if no coroutines	2497	during each prepare and only letting the process block if no coroutines
676	are ready to run.	2498	are ready to run (it's actually more complicated: it only runs coroutines
		2499	with priority higher than or equal to the event loop and one coroutine
		2500	of lower priority, but only once, using idle watchers to keep the event
		2501	loop from blocking if lower-priority coroutines are active, thus mapping
		2502	low-priority coroutines to idle/background tasks).
		2503
		2504	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		2505	priority, to ensure that they are being run before any other watchers
		2506	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2507
		2508	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
		2509	activate ("feed") events into libev. While libev fully supports this, they
		2510	might get executed before other C<ev_check> watchers did their job. As
		2511	C<ev_check> watchers are often used to embed other (non-libev) event
		2512	loops those other event loops might be in an unusable state until their
		2513	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
		2514	others).
		2515
		2516	=head3 Watcher-Specific Functions and Data Members
677		2517
678	=over 4	2518	=over 4
679		2519
680	=item ev_prepare_init (ev_prepare *, callback)	2520	=item ev_prepare_init (ev_prepare *, callback)
681		2521
682	=item ev_check_init (ev_check *, callback)	2522	=item ev_check_init (ev_check *, callback)
683		2523
684	Initialises and configures the prepare or check watcher - they have no	2524	Initialises and configures the prepare or check watcher - they have no
685	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2525	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
686	macros, but using them is utterly, utterly pointless.	2526	macros, but using them is utterly, utterly, utterly and completely
		2527	pointless.
687		2528
688	=back	2529	=back
689		2530
		2531	=head3 Examples
		2532
		2533	There are a number of principal ways to embed other event loops or modules
		2534	into libev. Here are some ideas on how to include libadns into libev
		2535	(there is a Perl module named C<EV::ADNS> that does this, which you could
		2536	use as a working example. Another Perl module named C<EV::Glib> embeds a
		2537	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		2538	Glib event loop).
		2539
		2540	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
		2541	and in a check watcher, destroy them and call into libadns. What follows
		2542	is pseudo-code only of course. This requires you to either use a low
		2543	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		2544	the callbacks for the IO/timeout watchers might not have been called yet.
		2545
		2546	static ev_io iow [nfd];
		2547	static ev_timer tw;
		2548
		2549	static void
		2550	io_cb (struct ev_loop *loop, ev_io *w, int revents)
		2551	{
		2552	}
		2553
		2554	// create io watchers for each fd and a timer before blocking
		2555	static void
		2556	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
		2557	{
		2558	int timeout = 3600000;
		2559	struct pollfd fds [nfd];
		2560	// actual code will need to loop here and realloc etc.
		2561	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
		2562
		2563	/* the callback is illegal, but won't be called as we stop during check */
		2564	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
		2565	ev_timer_start (loop, &tw);
		2566
		2567	// create one ev_io per pollfd
		2568	for (int i = 0; i < nfd; ++i)
		2569	{
		2570	ev_io_init (iow + i, io_cb, fds [i].fd,
		2571	((fds [i].events & POLLIN ? EV_READ : 0)
		2572	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
		2573
		2574	fds [i].revents = 0;
		2575	ev_io_start (loop, iow + i);
		2576	}
		2577	}
		2578
		2579	// stop all watchers after blocking
		2580	static void
		2581	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
		2582	{
		2583	ev_timer_stop (loop, &tw);
		2584
		2585	for (int i = 0; i < nfd; ++i)
		2586	{
		2587	// set the relevant poll flags
		2588	// could also call adns_processreadable etc. here
		2589	struct pollfd *fd = fds + i;
		2590	int revents = ev_clear_pending (iow + i);
		2591	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		2592	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		2593
		2594	// now stop the watcher
		2595	ev_io_stop (loop, iow + i);
		2596	}
		2597
		2598	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
		2599	}
		2600
		2601	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		2602	in the prepare watcher and would dispose of the check watcher.
		2603
		2604	Method 3: If the module to be embedded supports explicit event
		2605	notification (libadns does), you can also make use of the actual watcher
		2606	callbacks, and only destroy/create the watchers in the prepare watcher.
		2607
		2608	static void
		2609	timer_cb (EV_P_ ev_timer *w, int revents)
		2610	{
		2611	adns_state ads = (adns_state)w->data;
		2612	update_now (EV_A);
		2613
		2614	adns_processtimeouts (ads, &tv_now);
		2615	}
		2616
		2617	static void
		2618	io_cb (EV_P_ ev_io *w, int revents)
		2619	{
		2620	adns_state ads = (adns_state)w->data;
		2621	update_now (EV_A);
		2622
		2623	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		2624	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		2625	}
		2626
		2627	// do not ever call adns_afterpoll
		2628
		2629	Method 4: Do not use a prepare or check watcher because the module you
		2630	want to embed is not flexible enough to support it. Instead, you can
		2631	override their poll function. The drawback with this solution is that the
		2632	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
		2633	this approach, effectively embedding EV as a client into the horrible
		2634	libglib event loop.
		2635
		2636	static gint
		2637	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		2638	{
		2639	int got_events = 0;
		2640
		2641	for (n = 0; n < nfds; ++n)
		2642	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		2643
		2644	if (timeout >= 0)
		2645	// create/start timer
		2646
		2647	// poll
		2648	ev_loop (EV_A_ 0);
		2649
		2650	// stop timer again
		2651	if (timeout >= 0)
		2652	ev_timer_stop (EV_A_ &to);
		2653
		2654	// stop io watchers again - their callbacks should have set
		2655	for (n = 0; n < nfds; ++n)
		2656	ev_io_stop (EV_A_ iow [n]);
		2657
		2658	return got_events;
		2659	}
		2660
		2661
		2662	=head2 C<ev_embed> - when one backend isn't enough...
		2663
		2664	This is a rather advanced watcher type that lets you embed one event loop
		2665	into another (currently only C<ev_io> events are supported in the embedded
		2666	loop, other types of watchers might be handled in a delayed or incorrect
		2667	fashion and must not be used).
		2668
		2669	There are primarily two reasons you would want that: work around bugs and
		2670	prioritise I/O.
		2671
		2672	As an example for a bug workaround, the kqueue backend might only support
		2673	sockets on some platform, so it is unusable as generic backend, but you
		2674	still want to make use of it because you have many sockets and it scales
		2675	so nicely. In this case, you would create a kqueue-based loop and embed
		2676	it into your default loop (which might use e.g. poll). Overall operation
		2677	will be a bit slower because first libev has to call C<poll> and then
		2678	C<kevent>, but at least you can use both mechanisms for what they are
		2679	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
		2680
		2681	As for prioritising I/O: under rare circumstances you have the case where
		2682	some fds have to be watched and handled very quickly (with low latency),
		2683	and even priorities and idle watchers might have too much overhead. In
		2684	this case you would put all the high priority stuff in one loop and all
		2685	the rest in a second one, and embed the second one in the first.
		2686
		2687	As long as the watcher is active, the callback will be invoked every
		2688	time there might be events pending in the embedded loop. The callback
		2689	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
		2690	sweep and invoke their callbacks (the callback doesn't need to invoke the
		2691	C<ev_embed_sweep> function directly, it could also start an idle watcher
		2692	to give the embedded loop strictly lower priority for example).
		2693
		2694	You can also set the callback to C<0>, in which case the embed watcher
		2695	will automatically execute the embedded loop sweep whenever necessary.
		2696
		2697	Fork detection will be handled transparently while the C<ev_embed> watcher
		2698	is active, i.e., the embedded loop will automatically be forked when the
		2699	embedding loop forks. In other cases, the user is responsible for calling
		2700	C<ev_loop_fork> on the embedded loop.
		2701
		2702	Unfortunately, not all backends are embeddable: only the ones returned by
		2703	C<ev_embeddable_backends> are, which, unfortunately, does not include any
		2704	portable one.
		2705
		2706	So when you want to use this feature you will always have to be prepared
		2707	that you cannot get an embeddable loop. The recommended way to get around
		2708	this is to have a separate variables for your embeddable loop, try to
		2709	create it, and if that fails, use the normal loop for everything.
		2710
		2711	=head3 C<ev_embed> and fork
		2712
		2713	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2714	automatically be applied to the embedded loop as well, so no special
		2715	fork handling is required in that case. When the watcher is not running,
		2716	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2717	as applicable.
		2718
		2719	=head3 Watcher-Specific Functions and Data Members
		2720
		2721	=over 4
		2722
		2723	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
		2724
		2725	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)
		2726
		2727	Configures the watcher to embed the given loop, which must be
		2728	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		2729	invoked automatically, otherwise it is the responsibility of the callback
		2730	to invoke it (it will continue to be called until the sweep has been done,
		2731	if you do not want that, you need to temporarily stop the embed watcher).
		2732
		2733	=item ev_embed_sweep (loop, ev_embed *)
		2734
		2735	Make a single, non-blocking sweep over the embedded loop. This works
		2736	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		2737	appropriate way for embedded loops.
		2738
		2739	=item struct ev_loop *other [read-only]
		2740
		2741	The embedded event loop.
		2742
		2743	=back
		2744
		2745	=head3 Examples
		2746
		2747	Example: Try to get an embeddable event loop and embed it into the default
		2748	event loop. If that is not possible, use the default loop. The default
		2749	loop is stored in C<loop_hi>, while the embeddable loop is stored in
		2750	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
		2751	used).
		2752
		2753	struct ev_loop *loop_hi = ev_default_init (0);
		2754	struct ev_loop *loop_lo = 0;
		2755	ev_embed embed;
		2756
		2757	// see if there is a chance of getting one that works
		2758	// (remember that a flags value of 0 means autodetection)
		2759	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		2760	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		2761	: 0;
		2762
		2763	// if we got one, then embed it, otherwise default to loop_hi
		2764	if (loop_lo)
		2765	{
		2766	ev_embed_init (&embed, 0, loop_lo);
		2767	ev_embed_start (loop_hi, &embed);
		2768	}
		2769	else
		2770	loop_lo = loop_hi;
		2771
		2772	Example: Check if kqueue is available but not recommended and create
		2773	a kqueue backend for use with sockets (which usually work with any
		2774	kqueue implementation). Store the kqueue/socket-only event loop in
		2775	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		2776
		2777	struct ev_loop *loop = ev_default_init (0);
		2778	struct ev_loop *loop_socket = 0;
		2779	ev_embed embed;
		2780
		2781	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2782	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2783	{
		2784	ev_embed_init (&embed, 0, loop_socket);
		2785	ev_embed_start (loop, &embed);
		2786	}
		2787
		2788	if (!loop_socket)
		2789	loop_socket = loop;
		2790
		2791	// now use loop_socket for all sockets, and loop for everything else
		2792
		2793
		2794	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
		2795
		2796	Fork watchers are called when a C<fork ()> was detected (usually because
		2797	whoever is a good citizen cared to tell libev about it by calling
		2798	C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the
		2799	event loop blocks next and before C<ev_check> watchers are being called,
		2800	and only in the child after the fork. If whoever good citizen calling
		2801	C<ev_default_fork> cheats and calls it in the wrong process, the fork
		2802	handlers will be invoked, too, of course.
		2803
		2804	=head3 The special problem of life after fork - how is it possible?
		2805
		2806	Most uses of C<fork()> consist of forking, then some simple calls to ste
		2807	up/change the process environment, followed by a call to C<exec()>. This
		2808	sequence should be handled by libev without any problems.
		2809
		2810	This changes when the application actually wants to do event handling
		2811	in the child, or both parent in child, in effect "continuing" after the
		2812	fork.
		2813
		2814	The default mode of operation (for libev, with application help to detect
		2815	forks) is to duplicate all the state in the child, as would be expected
		2816	when I<either> the parent I<or> the child process continues.
		2817
		2818	When both processes want to continue using libev, then this is usually the
		2819	wrong result. In that case, usually one process (typically the parent) is
		2820	supposed to continue with all watchers in place as before, while the other
		2821	process typically wants to start fresh, i.e. without any active watchers.
		2822
		2823	The cleanest and most efficient way to achieve that with libev is to
		2824	simply create a new event loop, which of course will be "empty", and
		2825	use that for new watchers. This has the advantage of not touching more
		2826	memory than necessary, and thus avoiding the copy-on-write, and the
		2827	disadvantage of having to use multiple event loops (which do not support
		2828	signal watchers).
		2829
		2830	When this is not possible, or you want to use the default loop for
		2831	other reasons, then in the process that wants to start "fresh", call
		2832	C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying
		2833	the default loop will "orphan" (not stop) all registered watchers, so you
		2834	have to be careful not to execute code that modifies those watchers. Note
		2835	also that in that case, you have to re-register any signal watchers.
		2836
		2837	=head3 Watcher-Specific Functions and Data Members
		2838
		2839	=over 4
		2840
		2841	=item ev_fork_init (ev_signal *, callback)
		2842
		2843	Initialises and configures the fork watcher - it has no parameters of any
		2844	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
		2845	believe me.
		2846
		2847	=back
		2848
		2849
		2850	=head2 C<ev_async> - how to wake up another event loop
		2851
		2852	In general, you cannot use an C<ev_loop> from multiple threads or other
		2853	asynchronous sources such as signal handlers (as opposed to multiple event
		2854	loops - those are of course safe to use in different threads).
		2855
		2856	Sometimes, however, you need to wake up another event loop you do not
		2857	control, for example because it belongs to another thread. This is what
		2858	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2859	can signal it by calling C<ev_async_send>, which is thread- and signal
		2860	safe.
		2861
		2862	This functionality is very similar to C<ev_signal> watchers, as signals,
		2863	too, are asynchronous in nature, and signals, too, will be compressed
		2864	(i.e. the number of callback invocations may be less than the number of
		2865	C<ev_async_sent> calls).
		2866
		2867	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2868	just the default loop.
		2869
		2870	=head3 Queueing
		2871
		2872	C<ev_async> does not support queueing of data in any way. The reason
		2873	is that the author does not know of a simple (or any) algorithm for a
		2874	multiple-writer-single-reader queue that works in all cases and doesn't
		2875	need elaborate support such as pthreads.
		2876
		2877	That means that if you want to queue data, you have to provide your own
		2878	queue. But at least I can tell you how to implement locking around your
		2879	queue:
		2880
		2881	=over 4
		2882
		2883	=item queueing from a signal handler context
		2884
		2885	To implement race-free queueing, you simply add to the queue in the signal
		2886	handler but you block the signal handler in the watcher callback. Here is
		2887	an example that does that for some fictitious SIGUSR1 handler:
		2888
		2889	static ev_async mysig;
		2890
		2891	static void
		2892	sigusr1_handler (void)
		2893	{
		2894	sometype data;
		2895
		2896	// no locking etc.
		2897	queue_put (data);
		2898	ev_async_send (EV_DEFAULT_ &mysig);
		2899	}
		2900
		2901	static void
		2902	mysig_cb (EV_P_ ev_async *w, int revents)
		2903	{
		2904	sometype data;
		2905	sigset_t block, prev;
		2906
		2907	sigemptyset (&block);
		2908	sigaddset (&block, SIGUSR1);
		2909	sigprocmask (SIG_BLOCK, &block, &prev);
		2910
		2911	while (queue_get (&data))
		2912	process (data);
		2913
		2914	if (sigismember (&prev, SIGUSR1)
		2915	sigprocmask (SIG_UNBLOCK, &block, 0);
		2916	}
		2917
		2918	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2919	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2920	either...).
		2921
		2922	=item queueing from a thread context
		2923
		2924	The strategy for threads is different, as you cannot (easily) block
		2925	threads but you can easily preempt them, so to queue safely you need to
		2926	employ a traditional mutex lock, such as in this pthread example:
		2927
		2928	static ev_async mysig;
		2929	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2930
		2931	static void
		2932	otherthread (void)
		2933	{
		2934	// only need to lock the actual queueing operation
		2935	pthread_mutex_lock (&mymutex);
		2936	queue_put (data);
		2937	pthread_mutex_unlock (&mymutex);
		2938
		2939	ev_async_send (EV_DEFAULT_ &mysig);
		2940	}
		2941
		2942	static void
		2943	mysig_cb (EV_P_ ev_async *w, int revents)
		2944	{
		2945	pthread_mutex_lock (&mymutex);
		2946
		2947	while (queue_get (&data))
		2948	process (data);
		2949
		2950	pthread_mutex_unlock (&mymutex);
		2951	}
		2952
		2953	=back
		2954
		2955
		2956	=head3 Watcher-Specific Functions and Data Members
		2957
		2958	=over 4
		2959
		2960	=item ev_async_init (ev_async *, callback)
		2961
		2962	Initialises and configures the async watcher - it has no parameters of any
		2963	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
		2964	trust me.
		2965
		2966	=item ev_async_send (loop, ev_async *)
		2967
		2968	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2969	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2970	C<ev_feed_event>, this call is safe to do from other threads, signal or
		2971	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
		2972	section below on what exactly this means).
		2973
		2974	Note that, as with other watchers in libev, multiple events might get
		2975	compressed into a single callback invocation (another way to look at this
		2976	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
		2977	reset when the event loop detects that).
		2978
		2979	This call incurs the overhead of a system call only once per event loop
		2980	iteration, so while the overhead might be noticeable, it doesn't apply to
		2981	repeated calls to C<ev_async_send> for the same event loop.
		2982
		2983	=item bool = ev_async_pending (ev_async *)
		2984
		2985	Returns a non-zero value when C<ev_async_send> has been called on the
		2986	watcher but the event has not yet been processed (or even noted) by the
		2987	event loop.
		2988
		2989	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2990	the loop iterates next and checks for the watcher to have become active,
		2991	it will reset the flag again. C<ev_async_pending> can be used to very
		2992	quickly check whether invoking the loop might be a good idea.
		2993
		2994	Not that this does I<not> check whether the watcher itself is pending,
		2995	only whether it has been requested to make this watcher pending: there
		2996	is a time window between the event loop checking and resetting the async
		2997	notification, and the callback being invoked.
		2998
		2999	=back
		3000
		3001
690	=head1 OTHER FUNCTIONS	3002	=head1 OTHER FUNCTIONS
691		3003
692	There are some other fucntions of possible interest. Described. Here. Now.	3004	There are some other functions of possible interest. Described. Here. Now.
693		3005
694	=over 4	3006	=over 4
695		3007
696	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	3008	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
697		3009
698	This function combines a simple timer and an I/O watcher, calls your	3010	This function combines a simple timer and an I/O watcher, calls your
699	callback on whichever event happens first and automatically stop both	3011	callback on whichever event happens first and automatically stops both
700	watchers. This is useful if you want to wait for a single event on an fd	3012	watchers. This is useful if you want to wait for a single event on an fd
701	or timeout without havign to allocate/configure/start/stop/free one or	3013	or timeout without having to allocate/configure/start/stop/free one or
702	more watchers yourself.	3014	more watchers yourself.
703		3015
704	If C<fd> is less than 0, then no I/O watcher will be started and events is	3016	If C<fd> is less than 0, then no I/O watcher will be started and the
705	ignored. Otherwise, an ev_io watcher for the given C<fd> and C<events> set	3017	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
706	will be craeted and started.	3018	the given C<fd> and C<events> set will be created and started.
707		3019
708	If C<timeout> is less than 0, then no timeout watcher will be	3020	If C<timeout> is less than 0, then no timeout watcher will be
709	started. Otherwise an ev_timer watcher with after = C<timeout> (and repeat	3021	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
710	= 0) will be started.	3022	repeat = 0) will be started. C<0> is a valid timeout.
711		3023
712	The callback has the type C<void (*cb)(int revents, void *arg)> and	3024	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
713	gets passed an events set (normally a combination of EV_ERROR, EV_READ,	3025	passed an C<revents> set like normal event callbacks (a combination of
714	EV_WRITE or EV_TIMEOUT) and the C<arg> value passed to C<ev_once>:	3026	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
		3027	value passed to C<ev_once>. Note that it is possible to receive I<both>
		3028	a timeout and an io event at the same time - you probably should give io
		3029	events precedence.
715		3030
		3031	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		3032
716	static void stdin_ready (int revents, void *arg)	3033	static void stdin_ready (int revents, void *arg)
717	{	3034	{
718	if (revents & EV_TIMEOUT)
719	/* doh, nothing entered */
720	else if (revents & EV_READ)	3035	if (revents & EV_READ)
721	/* stdin might have data for us, joy! */	3036	/* stdin might have data for us, joy! */;
		3037	else if (revents & EV_TIMEOUT)
		3038	/* doh, nothing entered */;
722	}	3039	}
723		3040
724	ev_once (STDIN_FILENO, EV_READm 10., stdin_ready, 0);	3041	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
725		3042
726	=item ev_feed_event (loop, watcher, int events)	3043	=item ev_feed_event (struct ev_loop *, watcher *, int revents)
727		3044
728	Feeds the given event set into the event loop, as if the specified event	3045	Feeds the given event set into the event loop, as if the specified event
729	has happened for the specified watcher (which must be a pointer to an	3046	had happened for the specified watcher (which must be a pointer to an
730	initialised but not necessarily active event watcher).	3047	initialised but not necessarily started event watcher).
731		3048
732	=item ev_feed_fd_event (loop, int fd, int revents)	3049	=item ev_feed_fd_event (struct ev_loop *, int fd, int revents)
733		3050
734	Feed an event on the given fd, as if a file descriptor backend detected it.	3051	Feed an event on the given fd, as if a file descriptor backend detected
		3052	the given events it.
735		3053
736	=item ev_feed_signal_event (loop, int signum)	3054	=item ev_feed_signal_event (struct ev_loop *loop, int signum)
737		3055
738	Feed an event as if the given signal occured (loop must be the default loop!).	3056	Feed an event as if the given signal occurred (C<loop> must be the default
		3057	loop!).
739		3058
740	=back	3059	=back
741		3060
		3061
		3062	=head1 LIBEVENT EMULATION
		3063
		3064	Libev offers a compatibility emulation layer for libevent. It cannot
		3065	emulate the internals of libevent, so here are some usage hints:
		3066
		3067	=over 4
		3068
		3069	=item * Use it by including <event.h>, as usual.
		3070
		3071	=item * The following members are fully supported: ev_base, ev_callback,
		3072	ev_arg, ev_fd, ev_res, ev_events.
		3073
		3074	=item * Avoid using ev_flags and the EVLIST_*-macros, while it is
		3075	maintained by libev, it does not work exactly the same way as in libevent (consider
		3076	it a private API).
		3077
		3078	=item * Priorities are not currently supported. Initialising priorities
		3079	will fail and all watchers will have the same priority, even though there
		3080	is an ev_pri field.
		3081
		3082	=item * In libevent, the last base created gets the signals, in libev, the
		3083	first base created (== the default loop) gets the signals.
		3084
		3085	=item * Other members are not supported.
		3086
		3087	=item * The libev emulation is I<not> ABI compatible to libevent, you need
		3088	to use the libev header file and library.
		3089
		3090	=back
		3091
		3092	=head1 C++ SUPPORT
		3093
		3094	Libev comes with some simplistic wrapper classes for C++ that mainly allow
		3095	you to use some convenience methods to start/stop watchers and also change
		3096	the callback model to a model using method callbacks on objects.
		3097
		3098	To use it,
		3099
		3100	#include <ev++.h>
		3101
		3102	This automatically includes F<ev.h> and puts all of its definitions (many
		3103	of them macros) into the global namespace. All C++ specific things are
		3104	put into the C<ev> namespace. It should support all the same embedding
		3105	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
		3106
		3107	Care has been taken to keep the overhead low. The only data member the C++
		3108	classes add (compared to plain C-style watchers) is the event loop pointer
		3109	that the watcher is associated with (or no additional members at all if
		3110	you disable C<EV_MULTIPLICITY> when embedding libev).
		3111
		3112	Currently, functions, and static and non-static member functions can be
		3113	used as callbacks. Other types should be easy to add as long as they only
		3114	need one additional pointer for context. If you need support for other
		3115	types of functors please contact the author (preferably after implementing
		3116	it).
		3117
		3118	Here is a list of things available in the C<ev> namespace:
		3119
		3120	=over 4
		3121
		3122	=item C<ev::READ>, C<ev::WRITE> etc.
		3123
		3124	These are just enum values with the same values as the C<EV_READ> etc.
		3125	macros from F<ev.h>.
		3126
		3127	=item C<ev::tstamp>, C<ev::now>
		3128
		3129	Aliases to the same types/functions as with the C<ev_> prefix.
		3130
		3131	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.
		3132
		3133	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of
		3134	the same name in the C<ev> namespace, with the exception of C<ev_signal>
		3135	which is called C<ev::sig> to avoid clashes with the C<signal> macro
		3136	defines by many implementations.
		3137
		3138	All of those classes have these methods:
		3139
		3140	=over 4
		3141
		3142	=item ev::TYPE::TYPE ()
		3143
		3144	=item ev::TYPE::TYPE (struct ev_loop *)
		3145
		3146	=item ev::TYPE::~TYPE
		3147
		3148	The constructor (optionally) takes an event loop to associate the watcher
		3149	with. If it is omitted, it will use C<EV_DEFAULT>.
		3150
		3151	The constructor calls C<ev_init> for you, which means you have to call the
		3152	C<set> method before starting it.
		3153
		3154	It will not set a callback, however: You have to call the templated C<set>
		3155	method to set a callback before you can start the watcher.
		3156
		3157	(The reason why you have to use a method is a limitation in C++ which does
		3158	not allow explicit template arguments for constructors).
		3159
		3160	The destructor automatically stops the watcher if it is active.
		3161
		3162	=item w->set<class, &class::method> (object *)
		3163
		3164	This method sets the callback method to call. The method has to have a
		3165	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		3166	first argument and the C<revents> as second. The object must be given as
		3167	parameter and is stored in the C<data> member of the watcher.
		3168
		3169	This method synthesizes efficient thunking code to call your method from
		3170	the C callback that libev requires. If your compiler can inline your
		3171	callback (i.e. it is visible to it at the place of the C<set> call and
		3172	your compiler is good :), then the method will be fully inlined into the
		3173	thunking function, making it as fast as a direct C callback.
		3174
		3175	Example: simple class declaration and watcher initialisation
		3176
		3177	struct myclass
		3178	{
		3179	void io_cb (ev::io &w, int revents) { }
		3180	}
		3181
		3182	myclass obj;
		3183	ev::io iow;
		3184	iow.set <myclass, &myclass::io_cb> (&obj);
		3185
		3186	=item w->set (object *)
		3187
		3188	This is an B<experimental> feature that might go away in a future version.
		3189
		3190	This is a variation of a method callback - leaving out the method to call
		3191	will default the method to C<operator ()>, which makes it possible to use
		3192	functor objects without having to manually specify the C<operator ()> all
		3193	the time. Incidentally, you can then also leave out the template argument
		3194	list.
		3195
		3196	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3197	int revents)>.
		3198
		3199	See the method-C<set> above for more details.
		3200
		3201	Example: use a functor object as callback.
		3202
		3203	struct myfunctor
		3204	{
		3205	void operator() (ev::io &w, int revents)
		3206	{
		3207	...
		3208	}
		3209	}
		3210
		3211	myfunctor f;
		3212
		3213	ev::io w;
		3214	w.set (&f);
		3215
		3216	=item w->set<function> (void *data = 0)
		3217
		3218	Also sets a callback, but uses a static method or plain function as
		3219	callback. The optional C<data> argument will be stored in the watcher's
		3220	C<data> member and is free for you to use.
		3221
		3222	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		3223
		3224	See the method-C<set> above for more details.
		3225
		3226	Example: Use a plain function as callback.
		3227
		3228	static void io_cb (ev::io &w, int revents) { }
		3229	iow.set <io_cb> ();
		3230
		3231	=item w->set (struct ev_loop *)
		3232
		3233	Associates a different C<struct ev_loop> with this watcher. You can only
		3234	do this when the watcher is inactive (and not pending either).
		3235
		3236	=item w->set ([arguments])
		3237
		3238	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be
		3239	called at least once. Unlike the C counterpart, an active watcher gets
		3240	automatically stopped and restarted when reconfiguring it with this
		3241	method.
		3242
		3243	=item w->start ()
		3244
		3245	Starts the watcher. Note that there is no C<loop> argument, as the
		3246	constructor already stores the event loop.
		3247
		3248	=item w->stop ()
		3249
		3250	Stops the watcher if it is active. Again, no C<loop> argument.
		3251
		3252	=item w->again () (C<ev::timer>, C<ev::periodic> only)
		3253
		3254	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
		3255	C<ev_TYPE_again> function.
		3256
		3257	=item w->sweep () (C<ev::embed> only)
		3258
		3259	Invokes C<ev_embed_sweep>.
		3260
		3261	=item w->update () (C<ev::stat> only)
		3262
		3263	Invokes C<ev_stat_stat>.
		3264
		3265	=back
		3266
		3267	=back
		3268
		3269	Example: Define a class with an IO and idle watcher, start one of them in
		3270	the constructor.
		3271
		3272	class myclass
		3273	{
		3274	ev::io io ; void io_cb (ev::io &w, int revents);
		3275	ev::idle idle; void idle_cb (ev::idle &w, int revents);
		3276
		3277	myclass (int fd)
		3278	{
		3279	io .set <myclass, &myclass::io_cb > (this);
		3280	idle.set <myclass, &myclass::idle_cb> (this);
		3281
		3282	io.start (fd, ev::READ);
		3283	}
		3284	};
		3285
		3286
		3287	=head1 OTHER LANGUAGE BINDINGS
		3288
		3289	Libev does not offer other language bindings itself, but bindings for a
		3290	number of languages exist in the form of third-party packages. If you know
		3291	any interesting language binding in addition to the ones listed here, drop
		3292	me a note.
		3293
		3294	=over 4
		3295
		3296	=item Perl
		3297
		3298	The EV module implements the full libev API and is actually used to test
		3299	libev. EV is developed together with libev. Apart from the EV core module,
		3300	there are additional modules that implement libev-compatible interfaces
		3301	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
		3302	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3303	and C<EV::Glib>).
		3304
		3305	It can be found and installed via CPAN, its homepage is at
		3306	L<http://software.schmorp.de/pkg/EV>.
		3307
		3308	=item Python
		3309
		3310	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		3311	seems to be quite complete and well-documented.
		3312
		3313	=item Ruby
		3314
		3315	Tony Arcieri has written a ruby extension that offers access to a subset
		3316	of the libev API and adds file handle abstractions, asynchronous DNS and
		3317	more on top of it. It can be found via gem servers. Its homepage is at
		3318	L<http://rev.rubyforge.org/>.
		3319
		3320	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		3321	makes rev work even on mingw.
		3322
		3323	=item Haskell
		3324
		3325	A haskell binding to libev is available at
		3326	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		3327
		3328	=item D
		3329
		3330	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		3331	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		3332
		3333	=item Ocaml
		3334
		3335	Erkki Seppala has written Ocaml bindings for libev, to be found at
		3336	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		3337
		3338	=back
		3339
		3340
		3341	=head1 MACRO MAGIC
		3342
		3343	Libev can be compiled with a variety of options, the most fundamental
		3344	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
		3345	functions and callbacks have an initial C<struct ev_loop *> argument.
		3346
		3347	To make it easier to write programs that cope with either variant, the
		3348	following macros are defined:
		3349
		3350	=over 4
		3351
		3352	=item C<EV_A>, C<EV_A_>
		3353
		3354	This provides the loop I<argument> for functions, if one is required ("ev
		3355	loop argument"). The C<EV_A> form is used when this is the sole argument,
		3356	C<EV_A_> is used when other arguments are following. Example:
		3357
		3358	ev_unref (EV_A);
		3359	ev_timer_add (EV_A_ watcher);
		3360	ev_loop (EV_A_ 0);
		3361
		3362	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
		3363	which is often provided by the following macro.
		3364
		3365	=item C<EV_P>, C<EV_P_>
		3366
		3367	This provides the loop I<parameter> for functions, if one is required ("ev
		3368	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
		3369	C<EV_P_> is used when other parameters are following. Example:
		3370
		3371	// this is how ev_unref is being declared
		3372	static void ev_unref (EV_P);
		3373
		3374	// this is how you can declare your typical callback
		3375	static void cb (EV_P_ ev_timer *w, int revents)
		3376
		3377	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
		3378	suitable for use with C<EV_A>.
		3379
		3380	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
		3381
		3382	Similar to the other two macros, this gives you the value of the default
		3383	loop, if multiple loops are supported ("ev loop default").
		3384
		3385	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		3386
		3387	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		3388	default loop has been initialised (C<UC> == unchecked). Their behaviour
		3389	is undefined when the default loop has not been initialised by a previous
		3390	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		3391
		3392	It is often prudent to use C<EV_DEFAULT> when initialising the first
		3393	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
		3394
		3395	=back
		3396
		3397	Example: Declare and initialise a check watcher, utilising the above
		3398	macros so it will work regardless of whether multiple loops are supported
		3399	or not.
		3400
		3401	static void
		3402	check_cb (EV_P_ ev_timer *w, int revents)
		3403	{
		3404	ev_check_stop (EV_A_ w);
		3405	}
		3406
		3407	ev_check check;
		3408	ev_check_init (&check, check_cb);
		3409	ev_check_start (EV_DEFAULT_ &check);
		3410	ev_loop (EV_DEFAULT_ 0);
		3411
		3412	=head1 EMBEDDING
		3413
		3414	Libev can (and often is) directly embedded into host
		3415	applications. Examples of applications that embed it include the Deliantra
		3416	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
		3417	and rxvt-unicode.
		3418
		3419	The goal is to enable you to just copy the necessary files into your
		3420	source directory without having to change even a single line in them, so
		3421	you can easily upgrade by simply copying (or having a checked-out copy of
		3422	libev somewhere in your source tree).
		3423
		3424	=head2 FILESETS
		3425
		3426	Depending on what features you need you need to include one or more sets of files
		3427	in your application.
		3428
		3429	=head3 CORE EVENT LOOP
		3430
		3431	To include only the libev core (all the C<ev_*> functions), with manual
		3432	configuration (no autoconf):
		3433
		3434	#define EV_STANDALONE 1
		3435	#include "ev.c"
		3436
		3437	This will automatically include F<ev.h>, too, and should be done in a
		3438	single C source file only to provide the function implementations. To use
		3439	it, do the same for F<ev.h> in all files wishing to use this API (best
		3440	done by writing a wrapper around F<ev.h> that you can include instead and
		3441	where you can put other configuration options):
		3442
		3443	#define EV_STANDALONE 1
		3444	#include "ev.h"
		3445
		3446	Both header files and implementation files can be compiled with a C++
		3447	compiler (at least, that's a stated goal, and breakage will be treated
		3448	as a bug).
		3449
		3450	You need the following files in your source tree, or in a directory
		3451	in your include path (e.g. in libev/ when using -Ilibev):
		3452
		3453	ev.h
		3454	ev.c
		3455	ev_vars.h
		3456	ev_wrap.h
		3457
		3458	ev_win32.c required on win32 platforms only
		3459
		3460	ev_select.c only when select backend is enabled (which is enabled by default)
		3461	ev_poll.c only when poll backend is enabled (disabled by default)
		3462	ev_epoll.c only when the epoll backend is enabled (disabled by default)
		3463	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
		3464	ev_port.c only when the solaris port backend is enabled (disabled by default)
		3465
		3466	F<ev.c> includes the backend files directly when enabled, so you only need
		3467	to compile this single file.
		3468
		3469	=head3 LIBEVENT COMPATIBILITY API
		3470
		3471	To include the libevent compatibility API, also include:
		3472
		3473	#include "event.c"
		3474
		3475	in the file including F<ev.c>, and:
		3476
		3477	#include "event.h"
		3478
		3479	in the files that want to use the libevent API. This also includes F<ev.h>.
		3480
		3481	You need the following additional files for this:
		3482
		3483	event.h
		3484	event.c
		3485
		3486	=head3 AUTOCONF SUPPORT
		3487
		3488	Instead of using C<EV_STANDALONE=1> and providing your configuration in
		3489	whatever way you want, you can also C<m4_include([libev.m4])> in your
		3490	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
		3491	include F<config.h> and configure itself accordingly.
		3492
		3493	For this of course you need the m4 file:
		3494
		3495	libev.m4
		3496
		3497	=head2 PREPROCESSOR SYMBOLS/MACROS
		3498
		3499	Libev can be configured via a variety of preprocessor symbols you have to
		3500	define before including any of its files. The default in the absence of
		3501	autoconf is documented for every option.
		3502
		3503	=over 4
		3504
		3505	=item EV_STANDALONE
		3506
		3507	Must always be C<1> if you do not use autoconf configuration, which
		3508	keeps libev from including F<config.h>, and it also defines dummy
		3509	implementations for some libevent functions (such as logging, which is not
		3510	supported). It will also not define any of the structs usually found in
		3511	F<event.h> that are not directly supported by the libev core alone.
		3512
		3513	In stanbdalone mode, libev will still try to automatically deduce the
		3514	configuration, but has to be more conservative.
		3515
		3516	=item EV_USE_MONOTONIC
		3517
		3518	If defined to be C<1>, libev will try to detect the availability of the
		3519	monotonic clock option at both compile time and runtime. Otherwise no
		3520	use of the monotonic clock option will be attempted. If you enable this,
		3521	you usually have to link against librt or something similar. Enabling it
		3522	when the functionality isn't available is safe, though, although you have
		3523	to make sure you link against any libraries where the C<clock_gettime>
		3524	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
		3525
		3526	=item EV_USE_REALTIME
		3527
		3528	If defined to be C<1>, libev will try to detect the availability of the
		3529	real-time clock option at compile time (and assume its availability
		3530	at runtime if successful). Otherwise no use of the real-time clock
		3531	option will be attempted. This effectively replaces C<gettimeofday>
		3532	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
		3533	correctness. See the note about libraries in the description of
		3534	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		3535	C<EV_USE_CLOCK_SYSCALL>.
		3536
		3537	=item EV_USE_CLOCK_SYSCALL
		3538
		3539	If defined to be C<1>, libev will try to use a direct syscall instead
		3540	of calling the system-provided C<clock_gettime> function. This option
		3541	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		3542	unconditionally pulls in C<libpthread>, slowing down single-threaded
		3543	programs needlessly. Using a direct syscall is slightly slower (in
		3544	theory), because no optimised vdso implementation can be used, but avoids
		3545	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		3546	higher, as it simplifies linking (no need for C<-lrt>).
		3547
		3548	=item EV_USE_NANOSLEEP
		3549
		3550	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		3551	and will use it for delays. Otherwise it will use C<select ()>.
		3552
		3553	=item EV_USE_EVENTFD
		3554
		3555	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		3556	available and will probe for kernel support at runtime. This will improve
		3557	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		3558	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		3559	2.7 or newer, otherwise disabled.
		3560
		3561	=item EV_USE_SELECT
		3562
		3563	If undefined or defined to be C<1>, libev will compile in support for the
		3564	C<select>(2) backend. No attempt at auto-detection will be done: if no
		3565	other method takes over, select will be it. Otherwise the select backend
		3566	will not be compiled in.
		3567
		3568	=item EV_SELECT_USE_FD_SET
		3569
		3570	If defined to C<1>, then the select backend will use the system C<fd_set>
		3571	structure. This is useful if libev doesn't compile due to a missing
		3572	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
		3573	on exotic systems. This usually limits the range of file descriptors to
		3574	some low limit such as 1024 or might have other limitations (winsocket
		3575	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
		3576	configures the maximum size of the C<fd_set>.
		3577
		3578	=item EV_SELECT_IS_WINSOCKET
		3579
		3580	When defined to C<1>, the select backend will assume that
		3581	select/socket/connect etc. don't understand file descriptors but
		3582	wants osf handles on win32 (this is the case when the select to
		3583	be used is the winsock select). This means that it will call
		3584	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
		3585	it is assumed that all these functions actually work on fds, even
		3586	on win32. Should not be defined on non-win32 platforms.
		3587
		3588	=item EV_FD_TO_WIN32_HANDLE
		3589
		3590	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		3591	file descriptors to socket handles. When not defining this symbol (the
		3592	default), then libev will call C<_get_osfhandle>, which is usually
		3593	correct. In some cases, programs use their own file descriptor management,
		3594	in which case they can provide this function to map fds to socket handles.
		3595
		3596	=item EV_USE_POLL
		3597
		3598	If defined to be C<1>, libev will compile in support for the C<poll>(2)
		3599	backend. Otherwise it will be enabled on non-win32 platforms. It
		3600	takes precedence over select.
		3601
		3602	=item EV_USE_EPOLL
		3603
		3604	If defined to be C<1>, libev will compile in support for the Linux
		3605	C<epoll>(7) backend. Its availability will be detected at runtime,
		3606	otherwise another method will be used as fallback. This is the preferred
		3607	backend for GNU/Linux systems. If undefined, it will be enabled if the
		3608	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		3609
		3610	=item EV_USE_KQUEUE
		3611
		3612	If defined to be C<1>, libev will compile in support for the BSD style
		3613	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
		3614	otherwise another method will be used as fallback. This is the preferred
		3615	backend for BSD and BSD-like systems, although on most BSDs kqueue only
		3616	supports some types of fds correctly (the only platform we found that
		3617	supports ptys for example was NetBSD), so kqueue might be compiled in, but
		3618	not be used unless explicitly requested. The best way to use it is to find
		3619	out whether kqueue supports your type of fd properly and use an embedded
		3620	kqueue loop.
		3621
		3622	=item EV_USE_PORT
		3623
		3624	If defined to be C<1>, libev will compile in support for the Solaris
		3625	10 port style backend. Its availability will be detected at runtime,
		3626	otherwise another method will be used as fallback. This is the preferred
		3627	backend for Solaris 10 systems.
		3628
		3629	=item EV_USE_DEVPOLL
		3630
		3631	Reserved for future expansion, works like the USE symbols above.
		3632
		3633	=item EV_USE_INOTIFY
		3634
		3635	If defined to be C<1>, libev will compile in support for the Linux inotify
		3636	interface to speed up C<ev_stat> watchers. Its actual availability will
		3637	be detected at runtime. If undefined, it will be enabled if the headers
		3638	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		3639
		3640	=item EV_ATOMIC_T
		3641
		3642	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		3643	access is atomic with respect to other threads or signal contexts. No such
		3644	type is easily found in the C language, so you can provide your own type
		3645	that you know is safe for your purposes. It is used both for signal handler "locking"
		3646	as well as for signal and thread safety in C<ev_async> watchers.
		3647
		3648	In the absence of this define, libev will use C<sig_atomic_t volatile>
		3649	(from F<signal.h>), which is usually good enough on most platforms.
		3650
		3651	=item EV_H
		3652
		3653	The name of the F<ev.h> header file used to include it. The default if
		3654	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
		3655	used to virtually rename the F<ev.h> header file in case of conflicts.
		3656
		3657	=item EV_CONFIG_H
		3658
		3659	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
		3660	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
		3661	C<EV_H>, above.
		3662
		3663	=item EV_EVENT_H
		3664
		3665	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
		3666	of how the F<event.h> header can be found, the default is C<"event.h">.
		3667
		3668	=item EV_PROTOTYPES
		3669
		3670	If defined to be C<0>, then F<ev.h> will not define any function
		3671	prototypes, but still define all the structs and other symbols. This is
		3672	occasionally useful if you want to provide your own wrapper functions
		3673	around libev functions.
		3674
		3675	=item EV_MULTIPLICITY
		3676
		3677	If undefined or defined to C<1>, then all event-loop-specific functions
		3678	will have the C<struct ev_loop *> as first argument, and you can create
		3679	additional independent event loops. Otherwise there will be no support
		3680	for multiple event loops and there is no first event loop pointer
		3681	argument. Instead, all functions act on the single default loop.
		3682
		3683	=item EV_MINPRI
		3684
		3685	=item EV_MAXPRI
		3686
		3687	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		3688	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		3689	provide for more priorities by overriding those symbols (usually defined
		3690	to be C<-2> and C<2>, respectively).
		3691
		3692	When doing priority-based operations, libev usually has to linearly search
		3693	all the priorities, so having many of them (hundreds) uses a lot of space
		3694	and time, so using the defaults of five priorities (-2 .. +2) is usually
		3695	fine.
		3696
		3697	If your embedding application does not need any priorities, defining these
		3698	both to C<0> will save some memory and CPU.
		3699
		3700	=item EV_PERIODIC_ENABLE
		3701
		3702	If undefined or defined to be C<1>, then periodic timers are supported. If
		3703	defined to be C<0>, then they are not. Disabling them saves a few kB of
		3704	code.
		3705
		3706	=item EV_IDLE_ENABLE
		3707
		3708	If undefined or defined to be C<1>, then idle watchers are supported. If
		3709	defined to be C<0>, then they are not. Disabling them saves a few kB of
		3710	code.
		3711
		3712	=item EV_EMBED_ENABLE
		3713
		3714	If undefined or defined to be C<1>, then embed watchers are supported. If
		3715	defined to be C<0>, then they are not. Embed watchers rely on most other
		3716	watcher types, which therefore must not be disabled.
		3717
		3718	=item EV_STAT_ENABLE
		3719
		3720	If undefined or defined to be C<1>, then stat watchers are supported. If
		3721	defined to be C<0>, then they are not.
		3722
		3723	=item EV_FORK_ENABLE
		3724
		3725	If undefined or defined to be C<1>, then fork watchers are supported. If
		3726	defined to be C<0>, then they are not.
		3727
		3728	=item EV_ASYNC_ENABLE
		3729
		3730	If undefined or defined to be C<1>, then async watchers are supported. If
		3731	defined to be C<0>, then they are not.
		3732
		3733	=item EV_MINIMAL
		3734
		3735	If you need to shave off some kilobytes of code at the expense of some
		3736	speed (but with the full API), define this symbol to C<1>. Currently this
		3737	is used to override some inlining decisions, saves roughly 30% code size
		3738	on amd64. It also selects a much smaller 2-heap for timer management over
		3739	the default 4-heap.
		3740
		3741	You can save even more by disabling watcher types you do not need
		3742	and setting C<EV_MAXPRI> == C<EV_MINPRI>. Also, disabling C<assert>
		3743	(C<-DNDEBUG>) will usually reduce code size a lot.
		3744
		3745	Defining C<EV_MINIMAL> to C<2> will additionally reduce the core API to
		3746	provide a bare-bones event library. See C<ev.h> for details on what parts
		3747	of the API are still available, and do not complain if this subset changes
		3748	over time.
		3749
		3750	=item EV_PID_HASHSIZE
		3751
		3752	C<ev_child> watchers use a small hash table to distribute workload by
		3753	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
		3754	than enough. If you need to manage thousands of children you might want to
		3755	increase this value (I<must> be a power of two).
		3756
		3757	=item EV_INOTIFY_HASHSIZE
		3758
		3759	C<ev_stat> watchers use a small hash table to distribute workload by
		3760	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
		3761	usually more than enough. If you need to manage thousands of C<ev_stat>
		3762	watchers you might want to increase this value (I<must> be a power of
		3763	two).
		3764
		3765	=item EV_USE_4HEAP
		3766
		3767	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3768	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		3769	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		3770	faster performance with many (thousands) of watchers.
		3771
		3772	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3773	(disabled).
		3774
		3775	=item EV_HEAP_CACHE_AT
		3776
		3777	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3778	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		3779	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3780	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3781	but avoids random read accesses on heap changes. This improves performance
		3782	noticeably with many (hundreds) of watchers.
		3783
		3784	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3785	(disabled).
		3786
		3787	=item EV_VERIFY
		3788
		3789	Controls how much internal verification (see C<ev_loop_verify ()>) will
		3790	be done: If set to C<0>, no internal verification code will be compiled
		3791	in. If set to C<1>, then verification code will be compiled in, but not
		3792	called. If set to C<2>, then the internal verification code will be
		3793	called once per loop, which can slow down libev. If set to C<3>, then the
		3794	verification code will be called very frequently, which will slow down
		3795	libev considerably.
		3796
		3797	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
		3798	C<0>.
		3799
		3800	=item EV_COMMON
		3801
		3802	By default, all watchers have a C<void *data> member. By redefining
		3803	this macro to a something else you can include more and other types of
		3804	members. You have to define it each time you include one of the files,
		3805	though, and it must be identical each time.
		3806
		3807	For example, the perl EV module uses something like this:
		3808
		3809	#define EV_COMMON \
		3810	SV *self; /* contains this struct */ \
		3811	SV *cb_sv, *fh /* note no trailing ";" */
		3812
		3813	=item EV_CB_DECLARE (type)
		3814
		3815	=item EV_CB_INVOKE (watcher, revents)
		3816
		3817	=item ev_set_cb (ev, cb)
		3818
		3819	Can be used to change the callback member declaration in each watcher,
		3820	and the way callbacks are invoked and set. Must expand to a struct member
		3821	definition and a statement, respectively. See the F<ev.h> header file for
		3822	their default definitions. One possible use for overriding these is to
		3823	avoid the C<struct ev_loop *> as first argument in all cases, or to use
		3824	method calls instead of plain function calls in C++.
		3825
		3826	=back
		3827
		3828	=head2 EXPORTED API SYMBOLS
		3829
		3830	If you need to re-export the API (e.g. via a DLL) and you need a list of
		3831	exported symbols, you can use the provided F<Symbol.*> files which list
		3832	all public symbols, one per line:
		3833
		3834	Symbols.ev for libev proper
		3835	Symbols.event for the libevent emulation
		3836
		3837	This can also be used to rename all public symbols to avoid clashes with
		3838	multiple versions of libev linked together (which is obviously bad in
		3839	itself, but sometimes it is inconvenient to avoid this).
		3840
		3841	A sed command like this will create wrapper C<#define>'s that you need to
		3842	include before including F<ev.h>:
		3843
		3844	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		3845
		3846	This would create a file F<wrap.h> which essentially looks like this:
		3847
		3848	#define ev_backend myprefix_ev_backend
		3849	#define ev_check_start myprefix_ev_check_start
		3850	#define ev_check_stop myprefix_ev_check_stop
		3851	...
		3852
		3853	=head2 EXAMPLES
		3854
		3855	For a real-world example of a program the includes libev
		3856	verbatim, you can have a look at the EV perl module
		3857	(L<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
		3858	the F<libev/> subdirectory and includes them in the F<EV/EVAPI.h> (public
		3859	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
		3860	will be compiled. It is pretty complex because it provides its own header
		3861	file.
		3862
		3863	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
		3864	that everybody includes and which overrides some configure choices:
		3865
		3866	#define EV_MINIMAL 1
		3867	#define EV_USE_POLL 0
		3868	#define EV_MULTIPLICITY 0
		3869	#define EV_PERIODIC_ENABLE 0
		3870	#define EV_STAT_ENABLE 0
		3871	#define EV_FORK_ENABLE 0
		3872	#define EV_CONFIG_H <config.h>
		3873	#define EV_MINPRI 0
		3874	#define EV_MAXPRI 0
		3875
		3876	#include "ev++.h"
		3877
		3878	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
		3879
		3880	#include "ev_cpp.h"
		3881	#include "ev.c"
		3882
		3883	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
		3884
		3885	=head2 THREADS AND COROUTINES
		3886
		3887	=head3 THREADS
		3888
		3889	All libev functions are reentrant and thread-safe unless explicitly
		3890	documented otherwise, but libev implements no locking itself. This means
		3891	that you can use as many loops as you want in parallel, as long as there
		3892	are no concurrent calls into any libev function with the same loop
		3893	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		3894	of course): libev guarantees that different event loops share no data
		3895	structures that need any locking.
		3896
		3897	Or to put it differently: calls with different loop parameters can be done
		3898	concurrently from multiple threads, calls with the same loop parameter
		3899	must be done serially (but can be done from different threads, as long as
		3900	only one thread ever is inside a call at any point in time, e.g. by using
		3901	a mutex per loop).
		3902
		3903	Specifically to support threads (and signal handlers), libev implements
		3904	so-called C<ev_async> watchers, which allow some limited form of
		3905	concurrency on the same event loop, namely waking it up "from the
		3906	outside".
		3907
		3908	If you want to know which design (one loop, locking, or multiple loops
		3909	without or something else still) is best for your problem, then I cannot
		3910	help you, but here is some generic advice:
		3911
		3912	=over 4
		3913
		3914	=item * most applications have a main thread: use the default libev loop
		3915	in that thread, or create a separate thread running only the default loop.
		3916
		3917	This helps integrating other libraries or software modules that use libev
		3918	themselves and don't care/know about threading.
		3919
		3920	=item * one loop per thread is usually a good model.
		3921
		3922	Doing this is almost never wrong, sometimes a better-performance model
		3923	exists, but it is always a good start.
		3924
		3925	=item * other models exist, such as the leader/follower pattern, where one
		3926	loop is handed through multiple threads in a kind of round-robin fashion.
		3927
		3928	Choosing a model is hard - look around, learn, know that usually you can do
		3929	better than you currently do :-)
		3930
		3931	=item * often you need to talk to some other thread which blocks in the
		3932	event loop.
		3933
		3934	C<ev_async> watchers can be used to wake them up from other threads safely
		3935	(or from signal contexts...).
		3936
		3937	An example use would be to communicate signals or other events that only
		3938	work in the default loop by registering the signal watcher with the
		3939	default loop and triggering an C<ev_async> watcher from the default loop
		3940	watcher callback into the event loop interested in the signal.
		3941
		3942	=back
		3943
		3944	=head4 THREAD LOCKING EXAMPLE
		3945
		3946	Here is a fictitious example of how to run an event loop in a different
		3947	thread than where callbacks are being invoked and watchers are
		3948	created/added/removed.
		3949
		3950	For a real-world example, see the C<EV::Loop::Async> perl module,
		3951	which uses exactly this technique (which is suited for many high-level
		3952	languages).
		3953
		3954	The example uses a pthread mutex to protect the loop data, a condition
		3955	variable to wait for callback invocations, an async watcher to notify the
		3956	event loop thread and an unspecified mechanism to wake up the main thread.
		3957
		3958	First, you need to associate some data with the event loop:
		3959
		3960	typedef struct {
		3961	mutex_t lock; /* global loop lock */
		3962	ev_async async_w;
		3963	thread_t tid;
		3964	cond_t invoke_cv;
		3965	} userdata;
		3966
		3967	void prepare_loop (EV_P)
		3968	{
		3969	// for simplicity, we use a static userdata struct.
		3970	static userdata u;
		3971
		3972	ev_async_init (&u->async_w, async_cb);
		3973	ev_async_start (EV_A_ &u->async_w);
		3974
		3975	pthread_mutex_init (&u->lock, 0);
		3976	pthread_cond_init (&u->invoke_cv, 0);
		3977
		3978	// now associate this with the loop
		3979	ev_set_userdata (EV_A_ u);
		3980	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		3981	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		3982
		3983	// then create the thread running ev_loop
		3984	pthread_create (&u->tid, 0, l_run, EV_A);
		3985	}
		3986
		3987	The callback for the C<ev_async> watcher does nothing: the watcher is used
		3988	solely to wake up the event loop so it takes notice of any new watchers
		3989	that might have been added:
		3990
		3991	static void
		3992	async_cb (EV_P_ ev_async *w, int revents)
		3993	{
		3994	// just used for the side effects
		3995	}
		3996
		3997	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		3998	protecting the loop data, respectively.
		3999
		4000	static void
		4001	l_release (EV_P)
		4002	{
		4003	userdata *u = ev_userdata (EV_A);
		4004	pthread_mutex_unlock (&u->lock);
		4005	}
		4006
		4007	static void
		4008	l_acquire (EV_P)
		4009	{
		4010	userdata *u = ev_userdata (EV_A);
		4011	pthread_mutex_lock (&u->lock);
		4012	}
		4013
		4014	The event loop thread first acquires the mutex, and then jumps straight
		4015	into C<ev_loop>:
		4016
		4017	void *
		4018	l_run (void *thr_arg)
		4019	{
		4020	struct ev_loop *loop = (struct ev_loop *)thr_arg;
		4021
		4022	l_acquire (EV_A);
		4023	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		4024	ev_loop (EV_A_ 0);
		4025	l_release (EV_A);
		4026
		4027	return 0;
		4028	}
		4029
		4030	Instead of invoking all pending watchers, the C<l_invoke> callback will
		4031	signal the main thread via some unspecified mechanism (signals? pipe
		4032	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		4033	have been called:
		4034
		4035	static void
		4036	l_invoke (EV_P)
		4037	{
		4038	userdata *u = ev_userdata (EV_A);
		4039
		4040	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		4041
		4042	pthread_cond_wait (&u->invoke_cv, &u->lock);
		4043	}
		4044
		4045	Now, whenever the main thread gets told to invoke pending watchers, it
		4046	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		4047	thread to continue:
		4048
		4049	static void
		4050	real_invoke_pending (EV_P)
		4051	{
		4052	userdata *u = ev_userdata (EV_A);
		4053
		4054	pthread_mutex_lock (&u->lock);
		4055	ev_invoke_pending (EV_A);
		4056	pthread_cond_signal (&u->invoke_cv);
		4057	pthread_mutex_unlock (&u->lock);
		4058	}
		4059
		4060	Whenever you want to start/stop a watcher or do other modifications to an
		4061	event loop, you will now have to lock:
		4062
		4063	ev_timer timeout_watcher;
		4064	userdata *u = ev_userdata (EV_A);
		4065
		4066	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		4067
		4068	pthread_mutex_lock (&u->lock);
		4069	ev_timer_start (EV_A_ &timeout_watcher);
		4070	ev_async_send (EV_A_ &u->async_w);
		4071	pthread_mutex_unlock (&u->lock);
		4072
		4073	Note that sending the C<ev_async> watcher is required because otherwise
		4074	an event loop currently blocking in the kernel will have no knowledge
		4075	about the newly added timer. By waking up the loop it will pick up any new
		4076	watchers in the next event loop iteration.
		4077
		4078	=head3 COROUTINES
		4079
		4080	Libev is very accommodating to coroutines ("cooperative threads"):
		4081	libev fully supports nesting calls to its functions from different
		4082	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		4083	different coroutines, and switch freely between both coroutines running
		4084	the loop, as long as you don't confuse yourself). The only exception is
		4085	that you must not do this from C<ev_periodic> reschedule callbacks.
		4086
		4087	Care has been taken to ensure that libev does not keep local state inside
		4088	C<ev_loop>, and other calls do not usually allow for coroutine switches as
		4089	they do not call any callbacks.
		4090
		4091	=head2 COMPILER WARNINGS
		4092
		4093	Depending on your compiler and compiler settings, you might get no or a
		4094	lot of warnings when compiling libev code. Some people are apparently
		4095	scared by this.
		4096
		4097	However, these are unavoidable for many reasons. For one, each compiler
		4098	has different warnings, and each user has different tastes regarding
		4099	warning options. "Warn-free" code therefore cannot be a goal except when
		4100	targeting a specific compiler and compiler-version.
		4101
		4102	Another reason is that some compiler warnings require elaborate
		4103	workarounds, or other changes to the code that make it less clear and less
		4104	maintainable.
		4105
		4106	And of course, some compiler warnings are just plain stupid, or simply
		4107	wrong (because they don't actually warn about the condition their message
		4108	seems to warn about). For example, certain older gcc versions had some
		4109	warnings that resulted an extreme number of false positives. These have
		4110	been fixed, but some people still insist on making code warn-free with
		4111	such buggy versions.
		4112
		4113	While libev is written to generate as few warnings as possible,
		4114	"warn-free" code is not a goal, and it is recommended not to build libev
		4115	with any compiler warnings enabled unless you are prepared to cope with
		4116	them (e.g. by ignoring them). Remember that warnings are just that:
		4117	warnings, not errors, or proof of bugs.
		4118
		4119
		4120	=head2 VALGRIND
		4121
		4122	Valgrind has a special section here because it is a popular tool that is
		4123	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		4124
		4125	If you think you found a bug (memory leak, uninitialised data access etc.)
		4126	in libev, then check twice: If valgrind reports something like:
		4127
		4128	==2274== definitely lost: 0 bytes in 0 blocks.
		4129	==2274== possibly lost: 0 bytes in 0 blocks.
		4130	==2274== still reachable: 256 bytes in 1 blocks.
		4131
		4132	Then there is no memory leak, just as memory accounted to global variables
		4133	is not a memleak - the memory is still being referenced, and didn't leak.
		4134
		4135	Similarly, under some circumstances, valgrind might report kernel bugs
		4136	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		4137	although an acceptable workaround has been found here), or it might be
		4138	confused.
		4139
		4140	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		4141	make it into some kind of religion.
		4142
		4143	If you are unsure about something, feel free to contact the mailing list
		4144	with the full valgrind report and an explanation on why you think this
		4145	is a bug in libev (best check the archives, too :). However, don't be
		4146	annoyed when you get a brisk "this is no bug" answer and take the chance
		4147	of learning how to interpret valgrind properly.
		4148
		4149	If you need, for some reason, empty reports from valgrind for your project
		4150	I suggest using suppression lists.
		4151
		4152
		4153	=head1 PORTABILITY NOTES
		4154
		4155	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		4156
		4157	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		4158	requires, and its I/O model is fundamentally incompatible with the POSIX
		4159	model. Libev still offers limited functionality on this platform in
		4160	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		4161	descriptors. This only applies when using Win32 natively, not when using
		4162	e.g. cygwin.
		4163
		4164	Lifting these limitations would basically require the full
		4165	re-implementation of the I/O system. If you are into these kinds of
		4166	things, then note that glib does exactly that for you in a very portable
		4167	way (note also that glib is the slowest event library known to man).
		4168
		4169	There is no supported compilation method available on windows except
		4170	embedding it into other applications.
		4171
		4172	Sensible signal handling is officially unsupported by Microsoft - libev
		4173	tries its best, but under most conditions, signals will simply not work.
		4174
		4175	Not a libev limitation but worth mentioning: windows apparently doesn't
		4176	accept large writes: instead of resulting in a partial write, windows will
		4177	either accept everything or return C<ENOBUFS> if the buffer is too large,
		4178	so make sure you only write small amounts into your sockets (less than a
		4179	megabyte seems safe, but this apparently depends on the amount of memory
		4180	available).
		4181
		4182	Due to the many, low, and arbitrary limits on the win32 platform and
		4183	the abysmal performance of winsockets, using a large number of sockets
		4184	is not recommended (and not reasonable). If your program needs to use
		4185	more than a hundred or so sockets, then likely it needs to use a totally
		4186	different implementation for windows, as libev offers the POSIX readiness
		4187	notification model, which cannot be implemented efficiently on windows
		4188	(due to Microsoft monopoly games).
		4189
		4190	A typical way to use libev under windows is to embed it (see the embedding
		4191	section for details) and use the following F<evwrap.h> header file instead
		4192	of F<ev.h>:
		4193
		4194	#define EV_STANDALONE /* keeps ev from requiring config.h */
		4195	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		4196
		4197	#include "ev.h"
		4198
		4199	And compile the following F<evwrap.c> file into your project (make sure
		4200	you do I<not> compile the F<ev.c> or any other embedded source files!):
		4201
		4202	#include "evwrap.h"
		4203	#include "ev.c"
		4204
		4205	=over 4
		4206
		4207	=item The winsocket select function
		4208
		4209	The winsocket C<select> function doesn't follow POSIX in that it
		4210	requires socket I<handles> and not socket I<file descriptors> (it is
		4211	also extremely buggy). This makes select very inefficient, and also
		4212	requires a mapping from file descriptors to socket handles (the Microsoft
		4213	C runtime provides the function C<_open_osfhandle> for this). See the
		4214	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		4215	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
		4216
		4217	The configuration for a "naked" win32 using the Microsoft runtime
		4218	libraries and raw winsocket select is:
		4219
		4220	#define EV_USE_SELECT 1
		4221	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		4222
		4223	Note that winsockets handling of fd sets is O(n), so you can easily get a
		4224	complexity in the O(n²) range when using win32.
		4225
		4226	=item Limited number of file descriptors
		4227
		4228	Windows has numerous arbitrary (and low) limits on things.
		4229
		4230	Early versions of winsocket's select only supported waiting for a maximum
		4231	of C<64> handles (probably owning to the fact that all windows kernels
		4232	can only wait for C<64> things at the same time internally; Microsoft
		4233	recommends spawning a chain of threads and wait for 63 handles and the
		4234	previous thread in each. Sounds great!).
		4235
		4236	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		4237	to some high number (e.g. C<2048>) before compiling the winsocket select
		4238	call (which might be in libev or elsewhere, for example, perl and many
		4239	other interpreters do their own select emulation on windows).
		4240
		4241	Another limit is the number of file descriptors in the Microsoft runtime
		4242	libraries, which by default is C<64> (there must be a hidden I<64>
		4243	fetish or something like this inside Microsoft). You can increase this
		4244	by calling C<_setmaxstdio>, which can increase this limit to C<2048>
		4245	(another arbitrary limit), but is broken in many versions of the Microsoft
		4246	runtime libraries. This might get you to about C<512> or C<2048> sockets
		4247	(depending on windows version and/or the phase of the moon). To get more,
		4248	you need to wrap all I/O functions and provide your own fd management, but
		4249	the cost of calling select (O(n²)) will likely make this unworkable.
		4250
		4251	=back
		4252
		4253	=head2 PORTABILITY REQUIREMENTS
		4254
		4255	In addition to a working ISO-C implementation and of course the
		4256	backend-specific APIs, libev relies on a few additional extensions:
		4257
		4258	=over 4
		4259
		4260	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		4261	calling conventions regardless of C<ev_watcher_type *>.
		4262
		4263	Libev assumes not only that all watcher pointers have the same internal
		4264	structure (guaranteed by POSIX but not by ISO C for example), but it also
		4265	assumes that the same (machine) code can be used to call any watcher
		4266	callback: The watcher callbacks have different type signatures, but libev
		4267	calls them using an C<ev_watcher *> internally.
		4268
		4269	=item C<sig_atomic_t volatile> must be thread-atomic as well
		4270
		4271	The type C<sig_atomic_t volatile> (or whatever is defined as
		4272	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		4273	threads. This is not part of the specification for C<sig_atomic_t>, but is
		4274	believed to be sufficiently portable.
		4275
		4276	=item C<sigprocmask> must work in a threaded environment
		4277
		4278	Libev uses C<sigprocmask> to temporarily block signals. This is not
		4279	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		4280	pthread implementations will either allow C<sigprocmask> in the "main
		4281	thread" or will block signals process-wide, both behaviours would
		4282	be compatible with libev. Interaction between C<sigprocmask> and
		4283	C<pthread_sigmask> could complicate things, however.
		4284
		4285	The most portable way to handle signals is to block signals in all threads
		4286	except the initial one, and run the default loop in the initial thread as
		4287	well.
		4288
		4289	=item C<long> must be large enough for common memory allocation sizes
		4290
		4291	To improve portability and simplify its API, libev uses C<long> internally
		4292	instead of C<size_t> when allocating its data structures. On non-POSIX
		4293	systems (Microsoft...) this might be unexpectedly low, but is still at
		4294	least 31 bits everywhere, which is enough for hundreds of millions of
		4295	watchers.
		4296
		4297	=item C<double> must hold a time value in seconds with enough accuracy
		4298
		4299	The type C<double> is used to represent timestamps. It is required to
		4300	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		4301	enough for at least into the year 4000. This requirement is fulfilled by
		4302	implementations implementing IEEE 754, which is basically all existing
		4303	ones. With IEEE 754 doubles, you get microsecond accuracy until at least
		4304	2200.
		4305
		4306	=back
		4307
		4308	If you know of other additional requirements drop me a note.
		4309
		4310
		4311	=head1 ALGORITHMIC COMPLEXITIES
		4312
		4313	In this section the complexities of (many of) the algorithms used inside
		4314	libev will be documented. For complexity discussions about backends see
		4315	the documentation for C<ev_default_init>.
		4316
		4317	All of the following are about amortised time: If an array needs to be
		4318	extended, libev needs to realloc and move the whole array, but this
		4319	happens asymptotically rarer with higher number of elements, so O(1) might
		4320	mean that libev does a lengthy realloc operation in rare cases, but on
		4321	average it is much faster and asymptotically approaches constant time.
		4322
		4323	=over 4
		4324
		4325	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		4326
		4327	This means that, when you have a watcher that triggers in one hour and
		4328	there are 100 watchers that would trigger before that, then inserting will
		4329	have to skip roughly seven (C<ld 100>) of these watchers.
		4330
		4331	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		4332
		4333	That means that changing a timer costs less than removing/adding them,
		4334	as only the relative motion in the event queue has to be paid for.
		4335
		4336	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
		4337
		4338	These just add the watcher into an array or at the head of a list.
		4339
		4340	=item Stopping check/prepare/idle/fork/async watchers: O(1)
		4341
		4342	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		4343
		4344	These watchers are stored in lists, so they need to be walked to find the
		4345	correct watcher to remove. The lists are usually short (you don't usually
		4346	have many watchers waiting for the same fd or signal: one is typical, two
		4347	is rare).
		4348
		4349	=item Finding the next timer in each loop iteration: O(1)
		4350
		4351	By virtue of using a binary or 4-heap, the next timer is always found at a
		4352	fixed position in the storage array.
		4353
		4354	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		4355
		4356	A change means an I/O watcher gets started or stopped, which requires
		4357	libev to recalculate its status (and possibly tell the kernel, depending
		4358	on backend and whether C<ev_io_set> was used).
		4359
		4360	=item Activating one watcher (putting it into the pending state): O(1)
		4361
		4362	=item Priority handling: O(number_of_priorities)
		4363
		4364	Priorities are implemented by allocating some space for each
		4365	priority. When doing priority-based operations, libev usually has to
		4366	linearly search all the priorities, but starting/stopping and activating
		4367	watchers becomes O(1) with respect to priority handling.
		4368
		4369	=item Sending an ev_async: O(1)
		4370
		4371	=item Processing ev_async_send: O(number_of_async_watchers)
		4372
		4373	=item Processing signals: O(max_signal_number)
		4374
		4375	Sending involves a system call I<iff> there were no other C<ev_async_send>
		4376	calls in the current loop iteration. Checking for async and signal events
		4377	involves iterating over all running async watchers or all signal numbers.
		4378
		4379	=back
		4380
		4381
		4382	=head1 GLOSSARY
		4383
		4384	=over 4
		4385
		4386	=item active
		4387
		4388	A watcher is active as long as it has been started (has been attached to
		4389	an event loop) but not yet stopped (disassociated from the event loop).
		4390
		4391	=item application
		4392
		4393	In this document, an application is whatever is using libev.
		4394
		4395	=item callback
		4396
		4397	The address of a function that is called when some event has been
		4398	detected. Callbacks are being passed the event loop, the watcher that
		4399	received the event, and the actual event bitset.
		4400
		4401	=item callback invocation
		4402
		4403	The act of calling the callback associated with a watcher.
		4404
		4405	=item event
		4406
		4407	A change of state of some external event, such as data now being available
		4408	for reading on a file descriptor, time having passed or simply not having
		4409	any other events happening anymore.
		4410
		4411	In libev, events are represented as single bits (such as C<EV_READ> or
		4412	C<EV_TIMEOUT>).
		4413
		4414	=item event library
		4415
		4416	A software package implementing an event model and loop.
		4417
		4418	=item event loop
		4419
		4420	An entity that handles and processes external events and converts them
		4421	into callback invocations.
		4422
		4423	=item event model
		4424
		4425	The model used to describe how an event loop handles and processes
		4426	watchers and events.
		4427
		4428	=item pending
		4429
		4430	A watcher is pending as soon as the corresponding event has been detected,
		4431	and stops being pending as soon as the watcher will be invoked or its
		4432	pending status is explicitly cleared by the application.
		4433
		4434	A watcher can be pending, but not active. Stopping a watcher also clears
		4435	its pending status.
		4436
		4437	=item real time
		4438
		4439	The physical time that is observed. It is apparently strictly monotonic :)
		4440
		4441	=item wall-clock time
		4442
		4443	The time and date as shown on clocks. Unlike real time, it can actually
		4444	be wrong and jump forwards and backwards, e.g. when the you adjust your
		4445	clock.
		4446
		4447	=item watcher
		4448
		4449	A data structure that describes interest in certain events. Watchers need
		4450	to be started (attached to an event loop) before they can receive events.
		4451
		4452	=item watcher invocation
		4453
		4454	The act of calling the callback associated with a watcher.
		4455
		4456	=back
		4457
742	=head1 AUTHOR	4458	=head1 AUTHOR
743		4459
744	Marc Lehmann <libev@schmorp.de>.	4460	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.
745		4461

Diff Legend

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