ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.30 by root, Fri Nov 23 04:36:03 2007 UTC vs.
Revision 1.249 by root, Wed Jul 8 04:29:31 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 REPRESENTATION	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
51		156
52	=head1 GLOBAL FUNCTIONS	157	=head1 GLOBAL FUNCTIONS
53		158
54	These functions can be called anytime, even before initialising the	159	These functions can be called anytime, even before initialising the
55	library in any way.	160	library in any way.
…		…
60		165
61	Returns the current time as libev would use it. Please note that the	166	Returns the current time as libev would use it. Please note that the
62	C<ev_now> function is usually faster and also often returns the timestamp	167	C<ev_now> function is usually faster and also often returns the timestamp
63	you actually want to know.	168	you actually want to know.
64		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 ()>.
		175
65	=item int ev_version_major ()	176	=item int ev_version_major ()
66		177
67	=item int ev_version_minor ()	178	=item int ev_version_minor ()
68		179
69	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
70	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
71	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
72	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
73	version of the library your program was compiled against.	184	version of the library your program was compiled against.
74		185
		186	These version numbers refer to the ABI version of the library, not the
		187	release version.
		188
75	Usually, it's a good idea to terminate if the major versions mismatch,	189	Usually, it's a good idea to terminate if the major versions mismatch,
76	as this indicates an incompatible change. Minor versions are usually	190	as this indicates an incompatible change. Minor versions are usually
77	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
78	not a problem.	192	not a problem.
79		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
80	=item ev_set_allocator (void *(*cb)(void *ptr, long size))	233	=item ev_set_allocator (void *(*cb)(void *ptr, long size)) [NOT REENTRANT]
81		234
82	Sets the allocation function to use (the prototype is similar to the	235	Sets the allocation function to use (the prototype is similar - the
83	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
84	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
85	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
86	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.
87		244
88	You could override this function in high-availability programs to, say,	245	You could override this function in high-availability programs to, say,
89	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,
90	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.
91		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
92	=item ev_set_syserr_cb (void (*cb)(const char *msg));	269	=item ev_set_syserr_cb (void (*cb)(const char *msg)); [NOT REENTRANT]
93		270
94	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
95	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
96	indicating the system call or subsystem causing the problem. If this	273	indicating the system call or subsystem causing the problem. If this
97	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
98	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
99	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
100	(such as abort).	277	(such as abort).
101		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
102	=back	291	=back
103		292
104	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP	293	=head1 FUNCTIONS CONTROLLING THE EVENT LOOP
105		294
106	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>
107	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>
108	events, and dynamically created loops which do not.	297	I<function>).
109		298
110	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
111	in your main thread (or in a separate thread) and for each thread you	300	supports signals and child events, and dynamically created loops which do
112	create, you also create another event loop. Libev itself does no locking	301	not.
113	whatsoever, so if you mix calls to the same event loop in different
114	threads, make sure you lock (this is usually a bad idea, though, even if
115	done correctly, because it's hideous and inefficient).
116		302
117	=over 4	303	=over 4
118		304
119	=item struct ev_loop *ev_default_loop (unsigned int flags)	305	=item struct ev_loop *ev_default_loop (unsigned int flags)
120		306
121	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
122	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
123	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
124	flags).	310	flags. If that is troubling you, check C<ev_backend ()> afterwards).
125		311
126	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
127	function.	313	function.
128		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
129	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
130	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>).
131		328
132	It supports the following flags:	329	The following flags are supported:
133		330
134	=over 4	331	=over 4
135		332
136	=item C<EVFLAG_AUTO>	333	=item C<EVFLAG_AUTO>
137		334
138	The default flags value. Use this if you have no clue (it's the right	335	The default flags value. Use this if you have no clue (it's the right
139	thing, believe me).	336	thing, believe me).
140		337
141	=item C<EVFLAG_NOENV>	338	=item C<EVFLAG_NOENV>
142		339
143	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
144	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
145	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will	342	C<LIBEV_FLAGS>. Otherwise (the default), this environment variable will
146	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
147	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
148	around bugs.	345	around bugs.
149		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
150	=item C<EVMETHOD_SELECT> (value 1, portable select backend)	367	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
151		368
152	This is your standard select(2) backend. Not I<completely> standard, as	369	This is your standard select(2) backend. Not I<completely> standard, as
153	libev tries to roll its own fd_set with no limits on the number of fds,	370	libev tries to roll its own fd_set with no limits on the number of fds,
154	but if that fails, expect a fairly low limit on the number of fds when	371	but if that fails, expect a fairly low limit on the number of fds when
155	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	372	using this backend. It doesn't scale too well (O(highest_fd)), but its
156	the fastest backend for a low number of fds.	373	usually the fastest backend for a low number of (low-numbered :) fds.
157		374
		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.
		381
		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).
		385
158	=item C<EVMETHOD_POLL> (value 2, poll backend, available everywhere except on windows)	386	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
159		387
160	And this is your standard poll(2) backend. It's more complicated than	388	And this is your standard poll(2) backend. It's more complicated
161	select, but handles sparse fds better and has no artificial limit on the	389	than select, but handles sparse fds better and has no artificial
162	number of fds you can use (except it will slow down considerably with a	390	limit on the number of fds you can use (except it will slow down
163	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	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.
164		394
		395	This backend maps C<EV_READ> to C<POLLIN | POLLERR | POLLHUP>, and
		396	C<EV_WRITE> to C<POLLOUT | POLLERR | POLLHUP>.
		397
165	=item C<EVMETHOD_EPOLL> (value 4, Linux)	398	=item C<EVBACKEND_EPOLL> (value 4, Linux)
166		399
167	For few fds, this backend is a bit little slower than poll and select,	400	For few fds, this backend is a bit little slower than poll and select,
168	but it scales phenomenally better. While poll and select usually scale like	401	but it scales phenomenally better. While poll and select usually scale
169	O(total_fds) where n is the total number of fds (or the highest fd), epoll scales	402	like O(total_fds) where n is the total number of fds (or the highest fd),
170	either O(1) or O(active_fds).	403	epoll scales either O(1) or O(active_fds).
171		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
172	While stopping and starting an I/O watcher in the same iteration will	422	While stopping, setting and starting an I/O watcher in the same iteration
173	result in some caching, there is still a syscall per such incident	423	will result in some caching, there is still a system call per such
174	(because the fd could point to a different file description now), so its	424	incident (because the same I<file descriptor> could point to a different
175	best to avoid that. Also, dup()ed file descriptors might not work very	425	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
176	well if you register events for both fds.	426	file descriptors might not work very well if you register events for both
		427	file descriptors.
177		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
178	=item C<EVMETHOD_KQUEUE> (value 8, most BSD clones)	447	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
179		448
180	Kqueue deserves special mention, as at the time of this writing, it	449	Kqueue deserves special mention, as at the time of this writing, it
181	was broken on all BSDs except NetBSD (usually it doesn't work with	450	was broken on all BSDs except NetBSD (usually it doesn't work reliably
182	anything but sockets and pipes, except on Darwin, where of course its	451	with anything but sockets and pipes, except on Darwin, where of course
183	completely useless). For this reason its not being "autodetected" unless	452	it's completely useless). Unlike epoll, however, whose brokenness
184	you explicitly specify the flags (i.e. you don't use EVFLAG_AUTO).	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.
185		462
186	It scales in the same way as the epoll backend, but the interface to the	463	It scales in the same way as the epoll backend, but the interface to the
187	kernel is more efficient (which says nothing about its actual speed, of	464	kernel is more efficient (which says nothing about its actual speed, of
188	course). While starting and stopping an I/O watcher does not cause an	465	course). While stopping, setting and starting an I/O watcher does never
189	extra syscall as with epoll, it still adds up to four event changes per	466	cause an extra system call as with C<EVBACKEND_EPOLL>, it still adds up to
190	incident, so its best to avoid that.	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
191		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
192	=item C<EVMETHOD_DEVPOLL> (value 16, Solaris 8)	484	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
193		485
194	This is not implemented yet (and might never be).	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.
195		490
196	=item C<EVMETHOD_PORT> (value 32, Solaris 10)	491	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
197		492
198	This uses the Solaris 10 port mechanism. As with everything on Solaris,	493	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
199	it's really slow, but it still scales very well (O(active_fds)).	494	it's really slow, but it still scales very well (O(active_fds)).
200		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
201	=item C<EVMETHOD_ALL>	513	=item C<EVBACKEND_ALL>
202		514
203	Try all backends (even potentially broken ones that wouldn't be tried	515	Try all backends (even potentially broken ones that wouldn't be tried
204	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	516	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
205	C<EVMETHOD_ALL & ~EVMETHOD_KQUEUE>.	517	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
		518
		519	It is definitely not recommended to use this flag.
206		520
207	=back	521	=back
208		522
209	If one or more of these are ored into the flags value, then only these	523	If one or more of these are or'ed into the flags value, then only these
210	backends will be tried (in the reverse order as given here). If none are	524	backends will be tried (in the reverse order as listed here). If none are
211	specified, most compiled-in backend will be tried, usually in reverse	525	specified, all backends in C<ev_recommended_backends ()> will be tried.
212	order of their flag values :)	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);
213		543
214	=item struct ev_loop *ev_loop_new (unsigned int flags)	544	=item struct ev_loop *ev_loop_new (unsigned int flags)
215		545
216	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
217	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
218	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
219	undefined behaviour (or a failed assertion if assertions are enabled).	549	undefined behaviour (or a failed assertion if assertions are enabled).
220		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
221	=item ev_default_destroy ()	561	=item ev_default_destroy ()
222		562
223	Destroys the default loop again (frees all memory and kernel state	563	Destroys the default loop again (frees all memory and kernel state
224	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
225	any way whatsoever, although you cannot 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>).
226		579
227	=item ev_loop_destroy (loop)	580	=item ev_loop_destroy (loop)
228		581
229	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
230	earlier call to C<ev_loop_new>.	583	earlier call to C<ev_loop_new>.
231		584
232	=item ev_default_fork ()	585	=item ev_default_fork ()
233		586
		587	This function sets a flag that causes subsequent C<ev_loop> iterations
234	This function reinitialises the kernel state for backends that have	588	to reinitialise the kernel state for backends that have one. Despite the
235	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
236	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
237	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.
238		593
239	You I<must> call this function in the child process after forking if and	594	On the other hand, you only need to call this function in the child
240	only if you want to use the event library in both processes. If you just	595	process if and only if you want to use the event library in the child. If
241	fork+exec, you don't have to call it.	596	you just fork+exec, you don't have to call it at all.
242		597
243	The function itself is quite fast and it's usually not a problem to call	598	The function itself is quite fast and it's usually not a problem to call
244	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
245	quite nicely into a call to C<pthread_atfork>:	600	quite nicely into a call to C<pthread_atfork>:
246		601
…		…
248		603
249	=item ev_loop_fork (loop)	604	=item ev_loop_fork (loop)
250		605
251	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
252	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
253	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.
254		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
255	=item unsigned int ev_method (loop)	638	=item unsigned int ev_backend (loop)
256		639
257	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
258	use.	641	use.
259		642
260	=item ev_tstamp ev_now (loop)	643	=item ev_tstamp ev_now (loop)
261		644
262	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
263	got events and started processing them. This timestamp does not change	646	received events and started processing them. This timestamp does not
264	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
265	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
266	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>).
267		688
268	=item ev_loop (loop, int flags)	689	=item ev_loop (loop, int flags)
269		690
270	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
271	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
272	events.	693	events.
273		694
274	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
275	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.
276		704
277	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
278	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
279	case there are no events and will return after one iteration of the loop.	707	process in case there are no events and will return after one iteration of
		708	the loop.
280		709
281	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
282	neccessary) and will handle those and any outstanding ones. It will block	711	necessary) and will handle those and any already outstanding ones. It
283	your process until at least one new event arrives, and will return after	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
284	one iteration of the loop.	715	iteration of the loop.
285		716
286	This flags value could be used to implement alternative looping	717	This is useful if you are waiting for some external event in conjunction
287	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
288	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.
289		721
290	Here are the gory details of what ev_loop does:	722	Here are the gory details of what C<ev_loop> does:
291		723
292	1. If there are no active watchers (reference count is zero), return.	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.
293	2. Queue and immediately call all prepare watchers.	727	- Queue and call all prepare watchers.
294	3. If we have been forked, recreate the kernel state.	728	- If we have been forked, detach and recreate the kernel state
		729	as to not disturb the other process.
295	4. Update the kernel state with all outstanding changes.	730	- Update the kernel state with all outstanding changes.
296	5. Update the "event loop time".	731	- Update the "event loop time" (ev_now ()).
297	6. Calculate for how long to block.	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.
298	7. Block the process, waiting for events.	736	- Block the process, waiting for any events.
		737	- Queue all outstanding I/O (fd) events.
299	8. Update the "event loop time" and do time jump handling.	738	- Update the "event loop time" (ev_now ()), and do time jump adjustments.
300	9. Queue all outstanding timers.	739	- Queue all expired timers.
301	10. Queue all outstanding periodics.	740	- Queue all expired periodics.
302	11. If no events are pending now, queue all idle watchers.	741	- Unless any events are pending now, queue all idle watchers.
303	12. Queue all check watchers.	742	- Queue all check watchers.
304	13. Call all queued watchers in reverse order (i.e. check watchers first).	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.
305	14. If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	746	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
306	was used, return, otherwise continue with step #1.	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!
307		757
308	=item ev_unloop (loop, how)	758	=item ev_unloop (loop, how)
309		759
310	Can be used to make a call to C<ev_loop> return early (but only after it	760	Can be used to make a call to C<ev_loop> return early (but only after it
311	has processed all outstanding events). The C<how> argument must be either	761	has processed all outstanding events). The C<how> argument must be either
312	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	762	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
313	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	763	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
314		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.
		768
315	=item ev_ref (loop)	769	=item ev_ref (loop)
316		770
317	=item ev_unref (loop)	771	=item ev_unref (loop)
318		772
319	Ref/unref can be used to add or remove a reference count on the event	773	Ref/unref can be used to add or remove a reference count on the event
320	loop: Every watcher keeps one reference, and as long as the reference	774	loop: Every watcher keeps one reference, and as long as the reference
321	count is nonzero, C<ev_loop> will not return on its own. If you have	775	count is nonzero, C<ev_loop> will not return on its own.
		776
322	a watcher you never unregister that should not keep C<ev_loop> from	777	If you have a watcher you never unregister that should not keep C<ev_loop>
323	returning, ev_unref() after starting, and ev_ref() before stopping it. For	778	from returning, call ev_unref() after starting, and ev_ref() before
		779	stopping it.
		780
324	example, libev itself uses this for its internal signal pipe: It is not	781	As an example, libev itself uses this for its internal signal pipe: It
325	visible to the libev user and should not keep C<ev_loop> from exiting if	782	is not visible to the libev user and should not keep C<ev_loop> from
326	no event watchers registered by it are active. It is also an excellent	783	exiting if no event watchers registered by it are active. It is also an
327	way to do this for generic recurring timers or from within third-party	784	excellent way to do this for generic recurring timers or from within
328	libraries. Just remember to I<unref after start> and I<ref before stop>.	785	third-party libraries. Just remember to I<unref after start> and I<ref
		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_loop_verify (loop)
		862
		863	This function only does something when C<EV_VERIFY> support has been
		864	compiled in, which is the default for non-minimal builds. It tries to go
		865	through all internal structures and checks them for validity. If anything
		866	is found to be inconsistent, it will print an error message to standard
		867	error and call C<abort ()>.
		868
		869	This can be used to catch bugs inside libev itself: under normal
		870	circumstances, this function will never abort as of course libev keeps its
		871	data structures consistent.
329		872
330	=back	873	=back
331		874
		875
332	=head1 ANATOMY OF A WATCHER	876	=head1 ANATOMY OF A WATCHER
		877
		878	In the following description, uppercase C<TYPE> in names stands for the
		879	watcher type, e.g. C<ev_TYPE_start> can mean C<ev_timer_start> for timer
		880	watchers and C<ev_io_start> for I/O watchers.
333		881
334	A watcher is a structure that you create and register to record your	882	A watcher is a structure that you create and register to record your
335	interest in some event. For instance, if you want to wait for STDIN to	883	interest in some event. For instance, if you want to wait for STDIN to
336	become readable, you would create an C<ev_io> watcher for that:	884	become readable, you would create an C<ev_io> watcher for that:
337		885
338	static void my_cb (struct ev_loop *loop, struct ev_io *w, int revents)	886	static void my_cb (struct ev_loop *loop, ev_io *w, int revents)
339	{	887	{
340	ev_io_stop (w);	888	ev_io_stop (w);
341	ev_unloop (loop, EVUNLOOP_ALL);	889	ev_unloop (loop, EVUNLOOP_ALL);
342	}	890	}
343		891
344	struct ev_loop *loop = ev_default_loop (0);	892	struct ev_loop *loop = ev_default_loop (0);
		893
345	struct ev_io stdin_watcher;	894	ev_io stdin_watcher;
		895
346	ev_init (&stdin_watcher, my_cb);	896	ev_init (&stdin_watcher, my_cb);
347	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);	897	ev_io_set (&stdin_watcher, STDIN_FILENO, EV_READ);
348	ev_io_start (loop, &stdin_watcher);	898	ev_io_start (loop, &stdin_watcher);
		899
349	ev_loop (loop, 0);	900	ev_loop (loop, 0);
350		901
351	As you can see, you are responsible for allocating the memory for your	902	As you can see, you are responsible for allocating the memory for your
352	watcher structures (and it is usually a bad idea to do this on the stack,	903	watcher structures (and it is I<usually> a bad idea to do this on the
353	although this can sometimes be quite valid).	904	stack).
		905
		906	Each watcher has an associated watcher structure (called C<struct ev_TYPE>
		907	or simply C<ev_TYPE>, as typedefs are provided for all watcher structs).
354		908
355	Each watcher structure must be initialised by a call to C<ev_init	909	Each watcher structure must be initialised by a call to C<ev_init
356	(watcher *, callback)>, which expects a callback to be provided. This	910	(watcher *, callback)>, which expects a callback to be provided. This
357	callback gets invoked each time the event occurs (or, in the case of io	911	callback gets invoked each time the event occurs (or, in the case of I/O
358	watchers, each time the event loop detects that the file descriptor given	912	watchers, each time the event loop detects that the file descriptor given
359	is readable and/or writable).	913	is readable and/or writable).
360		914
361	Each watcher type has its own C<< ev_<type>_set (watcher *, ...) >> macro	915	Each watcher type further has its own C<< ev_TYPE_set (watcher *, ...) >>
362	with arguments specific to this watcher type. There is also a macro	916	macro to configure it, with arguments specific to the watcher type. There
363	to combine initialisation and setting in one call: C<< ev_<type>_init	917	is also a macro to combine initialisation and setting in one call: C<<
364	(watcher *, callback, ...) >>.	918	ev_TYPE_init (watcher *, callback, ...) >>.
365		919
366	To make the watcher actually watch out for events, you have to start it	920	To make the watcher actually watch out for events, you have to start it
367	with a watcher-specific start function (C<< ev_<type>_start (loop, watcher	921	with a watcher-specific start function (C<< ev_TYPE_start (loop, watcher
368	*) >>), and you can stop watching for events at any time by calling the	922	*) >>), and you can stop watching for events at any time by calling the
369	corresponding stop function (C<< ev_<type>_stop (loop, watcher *) >>.	923	corresponding stop function (C<< ev_TYPE_stop (loop, watcher *) >>.
370		924
371	As long as your watcher is active (has been started but not stopped) you	925	As long as your watcher is active (has been started but not stopped) you
372	must not touch the values stored in it. Most specifically you must never	926	must not touch the values stored in it. Most specifically you must never
373	reinitialise it or call its set method.	927	reinitialise it or call its C<ev_TYPE_set> macro.
374
375	You can check whether an event is active by calling the C<ev_is_active
376	(watcher *)> macro. To see whether an event is outstanding (but the
377	callback for it has not been called yet) you can use the C<ev_is_pending
378	(watcher *)> macro.
379		928
380	Each and every callback receives the event loop pointer as first, the	929	Each and every callback receives the event loop pointer as first, the
381	registered watcher structure as second, and a bitset of received events as	930	registered watcher structure as second, and a bitset of received events as
382	third argument.	931	third argument.
383		932
…		…
407	The signal specified in the C<ev_signal> watcher has been received by a thread.	956	The signal specified in the C<ev_signal> watcher has been received by a thread.
408		957
409	=item C<EV_CHILD>	958	=item C<EV_CHILD>
410		959
411	The pid specified in the C<ev_child> watcher has received a status change.	960	The pid specified in the C<ev_child> watcher has received a status change.
		961
		962	=item C<EV_STAT>
		963
		964	The path specified in the C<ev_stat> watcher changed its attributes somehow.
412		965
413	=item C<EV_IDLE>	966	=item C<EV_IDLE>
414		967
415	The C<ev_idle> watcher has determined that you have nothing better to do.	968	The C<ev_idle> watcher has determined that you have nothing better to do.
416		969
…		…
424	received events. Callbacks of both watcher types can start and stop as	977	received events. Callbacks of both watcher types can start and stop as
425	many watchers as they want, and all of them will be taken into account	978	many watchers as they want, and all of them will be taken into account
426	(for example, a C<ev_prepare> watcher might start an idle watcher to keep	979	(for example, a C<ev_prepare> watcher might start an idle watcher to keep
427	C<ev_loop> from blocking).	980	C<ev_loop> from blocking).
428		981
		982	=item C<EV_EMBED>
		983
		984	The embedded event loop specified in the C<ev_embed> watcher needs attention.
		985
		986	=item C<EV_FORK>
		987
		988	The event loop has been resumed in the child process after fork (see
		989	C<ev_fork>).
		990
		991	=item C<EV_ASYNC>
		992
		993	The given async watcher has been asynchronously notified (see C<ev_async>).
		994
		995	=item C<EV_CUSTOM>
		996
		997	Not ever sent (or otherwise used) by libev itself, but can be freely used
		998	by libev users to signal watchers (e.g. via C<ev_feed_event>).
		999
429	=item C<EV_ERROR>	1000	=item C<EV_ERROR>
430		1001
431	An unspecified error has occured, the watcher has been stopped. This might	1002	An unspecified error has occurred, the watcher has been stopped. This might
432	happen because the watcher could not be properly started because libev	1003	happen because the watcher could not be properly started because libev
433	ran out of memory, a file descriptor was found to be closed or any other	1004	ran out of memory, a file descriptor was found to be closed or any other
		1005	problem. Libev considers these application bugs.
		1006
434	problem. You best act on it by reporting the problem and somehow coping	1007	You best act on it by reporting the problem and somehow coping with the
435	with the watcher being stopped.	1008	watcher being stopped. Note that well-written programs should not receive
		1009	an error ever, so when your watcher receives it, this usually indicates a
		1010	bug in your program.
436		1011
437	Libev will usually signal a few "dummy" events together with an error,	1012	Libev will usually signal a few "dummy" events together with an error, for
438	for example it might indicate that a fd is readable or writable, and if	1013	example it might indicate that a fd is readable or writable, and if your
439	your callbacks is well-written it can just attempt the operation and cope	1014	callbacks is well-written it can just attempt the operation and cope with
440	with the error from read() or write(). This will not work in multithreaded	1015	the error from read() or write(). This will not work in multi-threaded
441	programs, though, so beware.	1016	programs, though, as the fd could already be closed and reused for another
		1017	thing, so beware.
442		1018
443	=back	1019	=back
444		1020
		1021	=head2 GENERIC WATCHER FUNCTIONS
		1022
		1023	=over 4
		1024
		1025	=item C<ev_init> (ev_TYPE *watcher, callback)
		1026
		1027	This macro initialises the generic portion of a watcher. The contents
		1028	of the watcher object can be arbitrary (so C<malloc> will do). Only
		1029	the generic parts of the watcher are initialised, you I<need> to call
		1030	the type-specific C<ev_TYPE_set> macro afterwards to initialise the
		1031	type-specific parts. For each type there is also a C<ev_TYPE_init> macro
		1032	which rolls both calls into one.
		1033
		1034	You can reinitialise a watcher at any time as long as it has been stopped
		1035	(or never started) and there are no pending events outstanding.
		1036
		1037	The callback is always of type C<void (*)(struct ev_loop *loop, ev_TYPE *watcher,
		1038	int revents)>.
		1039
		1040	Example: Initialise an C<ev_io> watcher in two steps.
		1041
		1042	ev_io w;
		1043	ev_init (&w, my_cb);
		1044	ev_io_set (&w, STDIN_FILENO, EV_READ);
		1045
		1046	=item C<ev_TYPE_set> (ev_TYPE *, [args])
		1047
		1048	This macro initialises the type-specific parts of a watcher. You need to
		1049	call C<ev_init> at least once before you call this macro, but you can
		1050	call C<ev_TYPE_set> any number of times. You must not, however, call this
		1051	macro on a watcher that is active (it can be pending, however, which is a
		1052	difference to the C<ev_init> macro).
		1053
		1054	Although some watcher types do not have type-specific arguments
		1055	(e.g. C<ev_prepare>) you still need to call its C<set> macro.
		1056
		1057	See C<ev_init>, above, for an example.
		1058
		1059	=item C<ev_TYPE_init> (ev_TYPE *watcher, callback, [args])
		1060
		1061	This convenience macro rolls both C<ev_init> and C<ev_TYPE_set> macro
		1062	calls into a single call. This is the most convenient method to initialise
		1063	a watcher. The same limitations apply, of course.
		1064
		1065	Example: Initialise and set an C<ev_io> watcher in one step.
		1066
		1067	ev_io_init (&w, my_cb, STDIN_FILENO, EV_READ);
		1068
		1069	=item C<ev_TYPE_start> (loop *, ev_TYPE *watcher)
		1070
		1071	Starts (activates) the given watcher. Only active watchers will receive
		1072	events. If the watcher is already active nothing will happen.
		1073
		1074	Example: Start the C<ev_io> watcher that is being abused as example in this
		1075	whole section.
		1076
		1077	ev_io_start (EV_DEFAULT_UC, &w);
		1078
		1079	=item C<ev_TYPE_stop> (loop *, ev_TYPE *watcher)
		1080
		1081	Stops the given watcher if active, and clears the pending status (whether
		1082	the watcher was active or not).
		1083
		1084	It is possible that stopped watchers are pending - for example,
		1085	non-repeating timers are being stopped when they become pending - but
		1086	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
		1087	pending. If you want to free or reuse the memory used by the watcher it is
		1088	therefore a good idea to always call its C<ev_TYPE_stop> function.
		1089
		1090	=item bool ev_is_active (ev_TYPE *watcher)
		1091
		1092	Returns a true value iff the watcher is active (i.e. it has been started
		1093	and not yet been stopped). As long as a watcher is active you must not modify
		1094	it.
		1095
		1096	=item bool ev_is_pending (ev_TYPE *watcher)
		1097
		1098	Returns a true value iff the watcher is pending, (i.e. it has outstanding
		1099	events but its callback has not yet been invoked). As long as a watcher
		1100	is pending (but not active) you must not call an init function on it (but
		1101	C<ev_TYPE_set> is safe), you must not change its priority, and you must
		1102	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		1103	it).
		1104
		1105	=item callback ev_cb (ev_TYPE *watcher)
		1106
		1107	Returns the callback currently set on the watcher.
		1108
		1109	=item ev_cb_set (ev_TYPE *watcher, callback)
		1110
		1111	Change the callback. You can change the callback at virtually any time
		1112	(modulo threads).
		1113
		1114	=item ev_set_priority (ev_TYPE *watcher, priority)
		1115
		1116	=item int ev_priority (ev_TYPE *watcher)
		1117
		1118	Set and query the priority of the watcher. The priority is a small
		1119	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		1120	(default: C<-2>). Pending watchers with higher priority will be invoked
		1121	before watchers with lower priority, but priority will not keep watchers
		1122	from being executed (except for C<ev_idle> watchers).
		1123
		1124	If you need to suppress invocation when higher priority events are pending
		1125	you need to look at C<ev_idle> watchers, which provide this functionality.
		1126
		1127	You I<must not> change the priority of a watcher as long as it is active or
		1128	pending.
		1129
		1130	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		1131	fine, as long as you do not mind that the priority value you query might
		1132	or might not have been clamped to the valid range.
		1133
		1134	The default priority used by watchers when no priority has been set is
		1135	always C<0>, which is supposed to not be too high and not be too low :).
		1136
		1137	See L<WATCHER PRIORITY MODELS>, below, for a more thorough treatment of
		1138	priorities.
		1139
		1140	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		1141
		1142	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		1143	C<loop> nor C<revents> need to be valid as long as the watcher callback
		1144	can deal with that fact, as both are simply passed through to the
		1145	callback.
		1146
		1147	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		1148
		1149	If the watcher is pending, this function clears its pending status and
		1150	returns its C<revents> bitset (as if its callback was invoked). If the
		1151	watcher isn't pending it does nothing and returns C<0>.
		1152
		1153	Sometimes it can be useful to "poll" a watcher instead of waiting for its
		1154	callback to be invoked, which can be accomplished with this function.
		1155
		1156	=back
		1157
		1158
445	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1159	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
446		1160
447	Each watcher has, by default, a member C<void *data> that you can change	1161	Each watcher has, by default, a member C<void *data> that you can change
448	and read at any time, libev will completely ignore it. This can be used	1162	and read at any time: libev will completely ignore it. This can be used
449	to associate arbitrary data with your watcher. If you need more data and	1163	to associate arbitrary data with your watcher. If you need more data and
450	don't want to allocate memory and store a pointer to it in that data	1164	don't want to allocate memory and store a pointer to it in that data
451	member, you can also "subclass" the watcher type and provide your own	1165	member, you can also "subclass" the watcher type and provide your own
452	data:	1166	data:
453		1167
454	struct my_io	1168	struct my_io
455	{	1169	{
456	struct ev_io io;	1170	ev_io io;
457	int otherfd;	1171	int otherfd;
458	void *somedata;	1172	void *somedata;
459	struct whatever *mostinteresting;	1173	struct whatever *mostinteresting;
460	}	1174	};
		1175
		1176	...
		1177	struct my_io w;
		1178	ev_io_init (&w.io, my_cb, fd, EV_READ);
461		1179
462	And since your callback will be called with a pointer to the watcher, you	1180	And since your callback will be called with a pointer to the watcher, you
463	can cast it back to your own type:	1181	can cast it back to your own type:
464		1182
465	static void my_cb (struct ev_loop *loop, struct ev_io *w_, int revents)	1183	static void my_cb (struct ev_loop *loop, ev_io *w_, int revents)
466	{	1184	{
467	struct my_io *w = (struct my_io *)w_;	1185	struct my_io *w = (struct my_io *)w_;
468	...	1186	...
469	}	1187	}
470		1188
471	More interesting and less C-conformant ways of catsing your callback type	1189	More interesting and less C-conformant ways of casting your callback type
472	have been omitted....	1190	instead have been omitted.
		1191
		1192	Another common scenario is to use some data structure with multiple
		1193	embedded watchers:
		1194
		1195	struct my_biggy
		1196	{
		1197	int some_data;
		1198	ev_timer t1;
		1199	ev_timer t2;
		1200	}
		1201
		1202	In this case getting the pointer to C<my_biggy> is a bit more
		1203	complicated: Either you store the address of your C<my_biggy> struct
		1204	in the C<data> member of the watcher (for woozies), or you need to use
		1205	some pointer arithmetic using C<offsetof> inside your watchers (for real
		1206	programmers):
		1207
		1208	#include <stddef.h>
		1209
		1210	static void
		1211	t1_cb (EV_P_ ev_timer *w, int revents)
		1212	{
		1213	struct my_biggy big = (struct my_biggy *)
		1214	(((char *)w) - offsetof (struct my_biggy, t1));
		1215	}
		1216
		1217	static void
		1218	t2_cb (EV_P_ ev_timer *w, int revents)
		1219	{
		1220	struct my_biggy big = (struct my_biggy *)
		1221	(((char *)w) - offsetof (struct my_biggy, t2));
		1222	}
		1223
		1224	=head2 WATCHER PRIORITY MODELS
		1225
		1226	Many event loops support I<watcher priorities>, which are usually small
		1227	integers that influence the ordering of event callback invocation
		1228	between watchers in some way, all else being equal.
		1229
		1230	In libev, Watcher priorities can be set using C<ev_set_priority>. See its
		1231	description for the more technical details such as the actual priority
		1232	range.
		1233
		1234	There are two common ways how these these priorities are being interpreted
		1235	by event loops:
		1236
		1237	In the more common lock-out model, higher priorities "lock out" invocation
		1238	of lower priority watchers, which means as long as higher priority
		1239	watchers receive events, lower priority watchers are not being invoked.
		1240
		1241	The less common only-for-ordering model uses priorities solely to order
		1242	callback invocation within a single event loop iteration: Higher priority
		1243	watchers are invoked before lower priority ones, but they all get invoked
		1244	before polling for new events.
		1245
		1246	Libev uses the second (only-for-ordering) model for all its watchers
		1247	except for idle watchers (which use the lock-out model).
		1248
		1249	The rationale behind this is that implementing the lock-out model for
		1250	watchers is not well supported by most kernel interfaces, and most event
		1251	libraries will just poll for the same events again and again as long as
		1252	their callbacks have not been executed, which is very inefficient in the
		1253	common case of one high-priority watcher locking out a mass of lower
		1254	priority ones.
		1255
		1256	Static (ordering) priorities are most useful when you have two or more
		1257	watchers handling the same resource: a typical usage example is having an
		1258	C<ev_io> watcher to receive data, and an associated C<ev_timer> to handle
		1259	timeouts. Under load, data might be received while the program handles
		1260	other jobs, but since timers normally get invoked first, the timeout
		1261	handler will be executed before checking for data. In that case, giving
		1262	the timer a lower priority than the I/O watcher ensures that I/O will be
		1263	handled first even under adverse conditions (which is usually, but not
		1264	always, what you want).
		1265
		1266	Since idle watchers use the "lock-out" model, meaning that idle watchers
		1267	will only be executed when no same or higher priority watchers have
		1268	received events, they can be used to implement the "lock-out" model when
		1269	required.
		1270
		1271	For example, to emulate how many other event libraries handle priorities,
		1272	you can associate an C<ev_idle> watcher to each such watcher, and in
		1273	the normal watcher callback, you just start the idle watcher. The real
		1274	processing is done in the idle watcher callback. This causes libev to
		1275	continously poll and process kernel event data for the watcher, but when
		1276	the lock-out case is known to be rare (which in turn is rare :), this is
		1277	workable.
		1278
		1279	Usually, however, the lock-out model implemented that way will perform
		1280	miserably under the type of load it was designed to handle. In that case,
		1281	it might be preferable to stop the real watcher before starting the
		1282	idle watcher, so the kernel will not have to process the event in case
		1283	the actual processing will be delayed for considerable time.
		1284
		1285	Here is an example of an I/O watcher that should run at a strictly lower
		1286	priority than the default, and which should only process data when no
		1287	other events are pending:
		1288
		1289	ev_idle idle; // actual processing watcher
		1290	ev_io io; // actual event watcher
		1291
		1292	static void
		1293	io_cb (EV_P_ ev_io *w, int revents)
		1294	{
		1295	// stop the I/O watcher, we received the event, but
		1296	// are not yet ready to handle it.
		1297	ev_io_stop (EV_A_ w);
		1298
		1299	// start the idle watcher to ahndle the actual event.
		1300	// it will not be executed as long as other watchers
		1301	// with the default priority are receiving events.
		1302	ev_idle_start (EV_A_ &idle);
		1303	}
		1304
		1305	static void
		1306	idle_cb (EV_P_ ev_idle *w, int revents)
		1307	{
		1308	// actual processing
		1309	read (STDIN_FILENO, ...);
		1310
		1311	// have to start the I/O watcher again, as
		1312	// we have handled the event
		1313	ev_io_start (EV_P_ &io);
		1314	}
		1315
		1316	// initialisation
		1317	ev_idle_init (&idle, idle_cb);
		1318	ev_io_init (&io, io_cb, STDIN_FILENO, EV_READ);
		1319	ev_io_start (EV_DEFAULT_ &io);
		1320
		1321	In the "real" world, it might also be beneficial to start a timer, so that
		1322	low-priority connections can not be locked out forever under load. This
		1323	enables your program to keep a lower latency for important connections
		1324	during short periods of high load, while not completely locking out less
		1325	important ones.
473		1326
474		1327
475	=head1 WATCHER TYPES	1328	=head1 WATCHER TYPES
476		1329
477	This section describes each watcher in detail, but will not repeat	1330	This section describes each watcher in detail, but will not repeat
478	information given in the last section.	1331	information given in the last section. Any initialisation/set macros,
		1332	functions and members specific to the watcher type are explained.
479		1333
		1334	Members are additionally marked with either I<[read-only]>, meaning that,
		1335	while the watcher is active, you can look at the member and expect some
		1336	sensible content, but you must not modify it (you can modify it while the
		1337	watcher is stopped to your hearts content), or I<[read-write]>, which
		1338	means you can expect it to have some sensible content while the watcher
		1339	is active, but you can also modify it. Modifying it may not do something
		1340	sensible or take immediate effect (or do anything at all), but libev will
		1341	not crash or malfunction in any way.
		1342
		1343
480	=head2 C<ev_io> - is this file descriptor readable or writable	1344	=head2 C<ev_io> - is this file descriptor readable or writable?
481		1345
482	I/O watchers check whether a file descriptor is readable or writable	1346	I/O watchers check whether a file descriptor is readable or writable
483	in each iteration of the event loop (This behaviour is called	1347	in each iteration of the event loop, or, more precisely, when reading
484	level-triggering because you keep receiving events as long as the	1348	would not block the process and writing would at least be able to write
485	condition persists. Remember you can stop the watcher if you don't want to	1349	some data. This behaviour is called level-triggering because you keep
486	act on the event and neither want to receive future events).	1350	receiving events as long as the condition persists. Remember you can stop
		1351	the watcher if you don't want to act on the event and neither want to
		1352	receive future events.
487		1353
488	In general you can register as many read and/or write event watchers per	1354	In general you can register as many read and/or write event watchers per
489	fd as you want (as long as you don't confuse yourself). Setting all file	1355	fd as you want (as long as you don't confuse yourself). Setting all file
490	descriptors to non-blocking mode is also usually a good idea (but not	1356	descriptors to non-blocking mode is also usually a good idea (but not
491	required if you know what you are doing).	1357	required if you know what you are doing).
492		1358
493	You have to be careful with dup'ed file descriptors, though. Some backends	1359	If you cannot use non-blocking mode, then force the use of a
494	(the linux epoll backend is a notable example) cannot handle dup'ed file	1360	known-to-be-good backend (at the time of this writing, this includes only
495	descriptors correctly if you register interest in two or more fds pointing	1361	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
496	to the same underlying file/socket etc. description (that is, they share	1362	descriptors for which non-blocking operation makes no sense (such as
497	the same underlying "file open").	1363	files) - libev doesn't guarentee any specific behaviour in that case.
498		1364
499	If you must do this, then force the use of a known-to-be-good backend	1365	Another thing you have to watch out for is that it is quite easy to
500	(at the time of this writing, this includes only EVMETHOD_SELECT and	1366	receive "spurious" readiness notifications, that is your callback might
501	EVMETHOD_POLL).	1367	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
		1368	because there is no data. Not only are some backends known to create a
		1369	lot of those (for example Solaris ports), it is very easy to get into
		1370	this situation even with a relatively standard program structure. Thus
		1371	it is best to always use non-blocking I/O: An extra C<read>(2) returning
		1372	C<EAGAIN> is far preferable to a program hanging until some data arrives.
		1373
		1374	If you cannot run the fd in non-blocking mode (for example you should
		1375	not play around with an Xlib connection), then you have to separately
		1376	re-test whether a file descriptor is really ready with a known-to-be good
		1377	interface such as poll (fortunately in our Xlib example, Xlib already
		1378	does this on its own, so its quite safe to use). Some people additionally
		1379	use C<SIGALRM> and an interval timer, just to be sure you won't block
		1380	indefinitely.
		1381
		1382	But really, best use non-blocking mode.
		1383
		1384	=head3 The special problem of disappearing file descriptors
		1385
		1386	Some backends (e.g. kqueue, epoll) need to be told about closing a file
		1387	descriptor (either due to calling C<close> explicitly or any other means,
		1388	such as C<dup2>). The reason is that you register interest in some file
		1389	descriptor, but when it goes away, the operating system will silently drop
		1390	this interest. If another file descriptor with the same number then is
		1391	registered with libev, there is no efficient way to see that this is, in
		1392	fact, a different file descriptor.
		1393
		1394	To avoid having to explicitly tell libev about such cases, libev follows
		1395	the following policy: Each time C<ev_io_set> is being called, libev
		1396	will assume that this is potentially a new file descriptor, otherwise
		1397	it is assumed that the file descriptor stays the same. That means that
		1398	you I<have> to call C<ev_io_set> (or C<ev_io_init>) when you change the
		1399	descriptor even if the file descriptor number itself did not change.
		1400
		1401	This is how one would do it normally anyway, the important point is that
		1402	the libev application should not optimise around libev but should leave
		1403	optimisations to libev.
		1404
		1405	=head3 The special problem of dup'ed file descriptors
		1406
		1407	Some backends (e.g. epoll), cannot register events for file descriptors,
		1408	but only events for the underlying file descriptions. That means when you
		1409	have C<dup ()>'ed file descriptors or weirder constellations, and register
		1410	events for them, only one file descriptor might actually receive events.
		1411
		1412	There is no workaround possible except not registering events
		1413	for potentially C<dup ()>'ed file descriptors, or to resort to
		1414	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
		1415
		1416	=head3 The special problem of fork
		1417
		1418	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
		1419	useless behaviour. Libev fully supports fork, but needs to be told about
		1420	it in the child.
		1421
		1422	To support fork in your programs, you either have to call
		1423	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,
		1424	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or
		1425	C<EVBACKEND_POLL>.
		1426
		1427	=head3 The special problem of SIGPIPE
		1428
		1429	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
		1430	when writing to a pipe whose other end has been closed, your program gets
		1431	sent a SIGPIPE, which, by default, aborts your program. For most programs
		1432	this is sensible behaviour, for daemons, this is usually undesirable.
		1433
		1434	So when you encounter spurious, unexplained daemon exits, make sure you
		1435	ignore SIGPIPE (and maybe make sure you log the exit status of your daemon
		1436	somewhere, as that would have given you a big clue).
		1437
		1438
		1439	=head3 Watcher-Specific Functions
502		1440
503	=over 4	1441	=over 4
504		1442
505	=item ev_io_init (ev_io *, callback, int fd, int events)	1443	=item ev_io_init (ev_io *, callback, int fd, int events)
506		1444
507	=item ev_io_set (ev_io *, int fd, int events)	1445	=item ev_io_set (ev_io *, int fd, int events)
508		1446
509	Configures an C<ev_io> watcher. The fd is the file descriptor to rceeive	1447	Configures an C<ev_io> watcher. The C<fd> is the file descriptor to
510	events for and events is either C<EV_READ>, C<EV_WRITE> or C<EV_READ |	1448	receive events for and C<events> is either C<EV_READ>, C<EV_WRITE> or
511	EV_WRITE> to receive the given events.	1449	C<EV_READ | EV_WRITE>, to express the desire to receive the given events.
		1450
		1451	=item int fd [read-only]
		1452
		1453	The file descriptor being watched.
		1454
		1455	=item int events [read-only]
		1456
		1457	The events being watched.
512		1458
513	=back	1459	=back
514		1460
		1461	=head3 Examples
		1462
		1463	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
		1464	readable, but only once. Since it is likely line-buffered, you could
		1465	attempt to read a whole line in the callback.
		1466
		1467	static void
		1468	stdin_readable_cb (struct ev_loop *loop, ev_io *w, int revents)
		1469	{
		1470	ev_io_stop (loop, w);
		1471	.. read from stdin here (or from w->fd) and handle any I/O errors
		1472	}
		1473
		1474	...
		1475	struct ev_loop *loop = ev_default_init (0);
		1476	ev_io stdin_readable;
		1477	ev_io_init (&stdin_readable, stdin_readable_cb, STDIN_FILENO, EV_READ);
		1478	ev_io_start (loop, &stdin_readable);
		1479	ev_loop (loop, 0);
		1480
		1481
515	=head2 C<ev_timer> - relative and optionally recurring timeouts	1482	=head2 C<ev_timer> - relative and optionally repeating timeouts
516		1483
517	Timer watchers are simple relative timers that generate an event after a	1484	Timer watchers are simple relative timers that generate an event after a
518	given time, and optionally repeating in regular intervals after that.	1485	given time, and optionally repeating in regular intervals after that.
519		1486
520	The timers are based on real time, that is, if you register an event that	1487	The timers are based on real time, that is, if you register an event that
521	times out after an hour and you reset your system clock to last years	1488	times out after an hour and you reset your system clock to January last
522	time, it will still time out after (roughly) and hour. "Roughly" because	1489	year, it will still time out after (roughly) one hour. "Roughly" because
523	detecting time jumps is hard, and some inaccuracies are unavoidable (the	1490	detecting time jumps is hard, and some inaccuracies are unavoidable (the
524	monotonic clock option helps a lot here).	1491	monotonic clock option helps a lot here).
		1492
		1493	The callback is guaranteed to be invoked only I<after> its timeout has
		1494	passed (not I<at>, so on systems with very low-resolution clocks this
		1495	might introduce a small delay). If multiple timers become ready during the
		1496	same loop iteration then the ones with earlier time-out values are invoked
		1497	before ones of the same priority with later time-out values (but this is
		1498	no longer true when a callback calls C<ev_loop> recursively).
		1499
		1500	=head3 Be smart about timeouts
		1501
		1502	Many real-world problems involve some kind of timeout, usually for error
		1503	recovery. A typical example is an HTTP request - if the other side hangs,
		1504	you want to raise some error after a while.
		1505
		1506	What follows are some ways to handle this problem, from obvious and
		1507	inefficient to smart and efficient.
		1508
		1509	In the following, a 60 second activity timeout is assumed - a timeout that
		1510	gets reset to 60 seconds each time there is activity (e.g. each time some
		1511	data or other life sign was received).
		1512
		1513	=over 4
		1514
		1515	=item 1. Use a timer and stop, reinitialise and start it on activity.
		1516
		1517	This is the most obvious, but not the most simple way: In the beginning,
		1518	start the watcher:
		1519
		1520	ev_timer_init (timer, callback, 60., 0.);
		1521	ev_timer_start (loop, timer);
		1522
		1523	Then, each time there is some activity, C<ev_timer_stop> it, initialise it
		1524	and start it again:
		1525
		1526	ev_timer_stop (loop, timer);
		1527	ev_timer_set (timer, 60., 0.);
		1528	ev_timer_start (loop, timer);
		1529
		1530	This is relatively simple to implement, but means that each time there is
		1531	some activity, libev will first have to remove the timer from its internal
		1532	data structure and then add it again. Libev tries to be fast, but it's
		1533	still not a constant-time operation.
		1534
		1535	=item 2. Use a timer and re-start it with C<ev_timer_again> inactivity.
		1536
		1537	This is the easiest way, and involves using C<ev_timer_again> instead of
		1538	C<ev_timer_start>.
		1539
		1540	To implement this, configure an C<ev_timer> with a C<repeat> value
		1541	of C<60> and then call C<ev_timer_again> at start and each time you
		1542	successfully read or write some data. If you go into an idle state where
		1543	you do not expect data to travel on the socket, you can C<ev_timer_stop>
		1544	the timer, and C<ev_timer_again> will automatically restart it if need be.
		1545
		1546	That means you can ignore both the C<ev_timer_start> function and the
		1547	C<after> argument to C<ev_timer_set>, and only ever use the C<repeat>
		1548	member and C<ev_timer_again>.
		1549
		1550	At start:
		1551
		1552	ev_init (timer, callback);
		1553	timer->repeat = 60.;
		1554	ev_timer_again (loop, timer);
		1555
		1556	Each time there is some activity:
		1557
		1558	ev_timer_again (loop, timer);
		1559
		1560	It is even possible to change the time-out on the fly, regardless of
		1561	whether the watcher is active or not:
		1562
		1563	timer->repeat = 30.;
		1564	ev_timer_again (loop, timer);
		1565
		1566	This is slightly more efficient then stopping/starting the timer each time
		1567	you want to modify its timeout value, as libev does not have to completely
		1568	remove and re-insert the timer from/into its internal data structure.
		1569
		1570	It is, however, even simpler than the "obvious" way to do it.
		1571
		1572	=item 3. Let the timer time out, but then re-arm it as required.
		1573
		1574	This method is more tricky, but usually most efficient: Most timeouts are
		1575	relatively long compared to the intervals between other activity - in
		1576	our example, within 60 seconds, there are usually many I/O events with
		1577	associated activity resets.
		1578
		1579	In this case, it would be more efficient to leave the C<ev_timer> alone,
		1580	but remember the time of last activity, and check for a real timeout only
		1581	within the callback:
		1582
		1583	ev_tstamp last_activity; // time of last activity
		1584
		1585	static void
		1586	callback (EV_P_ ev_timer *w, int revents)
		1587	{
		1588	ev_tstamp now = ev_now (EV_A);
		1589	ev_tstamp timeout = last_activity + 60.;
		1590
		1591	// if last_activity + 60. is older than now, we did time out
		1592	if (timeout < now)
		1593	{
		1594	// timeout occured, take action
		1595	}
		1596	else
		1597	{
		1598	// callback was invoked, but there was some activity, re-arm
		1599	// the watcher to fire in last_activity + 60, which is
		1600	// guaranteed to be in the future, so "again" is positive:
		1601	w->repeat = timeout - now;
		1602	ev_timer_again (EV_A_ w);
		1603	}
		1604	}
		1605
		1606	To summarise the callback: first calculate the real timeout (defined
		1607	as "60 seconds after the last activity"), then check if that time has
		1608	been reached, which means something I<did>, in fact, time out. Otherwise
		1609	the callback was invoked too early (C<timeout> is in the future), so
		1610	re-schedule the timer to fire at that future time, to see if maybe we have
		1611	a timeout then.
		1612
		1613	Note how C<ev_timer_again> is used, taking advantage of the
		1614	C<ev_timer_again> optimisation when the timer is already running.
		1615
		1616	This scheme causes more callback invocations (about one every 60 seconds
		1617	minus half the average time between activity), but virtually no calls to
		1618	libev to change the timeout.
		1619
		1620	To start the timer, simply initialise the watcher and set C<last_activity>
		1621	to the current time (meaning we just have some activity :), then call the
		1622	callback, which will "do the right thing" and start the timer:
		1623
		1624	ev_init (timer, callback);
		1625	last_activity = ev_now (loop);
		1626	callback (loop, timer, EV_TIMEOUT);
		1627
		1628	And when there is some activity, simply store the current time in
		1629	C<last_activity>, no libev calls at all:
		1630
		1631	last_actiivty = ev_now (loop);
		1632
		1633	This technique is slightly more complex, but in most cases where the
		1634	time-out is unlikely to be triggered, much more efficient.
		1635
		1636	Changing the timeout is trivial as well (if it isn't hard-coded in the
		1637	callback :) - just change the timeout and invoke the callback, which will
		1638	fix things for you.
		1639
		1640	=item 4. Wee, just use a double-linked list for your timeouts.
		1641
		1642	If there is not one request, but many thousands (millions...), all
		1643	employing some kind of timeout with the same timeout value, then one can
		1644	do even better:
		1645
		1646	When starting the timeout, calculate the timeout value and put the timeout
		1647	at the I<end> of the list.
		1648
		1649	Then use an C<ev_timer> to fire when the timeout at the I<beginning> of
		1650	the list is expected to fire (for example, using the technique #3).
		1651
		1652	When there is some activity, remove the timer from the list, recalculate
		1653	the timeout, append it to the end of the list again, and make sure to
		1654	update the C<ev_timer> if it was taken from the beginning of the list.
		1655
		1656	This way, one can manage an unlimited number of timeouts in O(1) time for
		1657	starting, stopping and updating the timers, at the expense of a major
		1658	complication, and having to use a constant timeout. The constant timeout
		1659	ensures that the list stays sorted.
		1660
		1661	=back
		1662
		1663	So which method the best?
		1664
		1665	Method #2 is a simple no-brain-required solution that is adequate in most
		1666	situations. Method #3 requires a bit more thinking, but handles many cases
		1667	better, and isn't very complicated either. In most case, choosing either
		1668	one is fine, with #3 being better in typical situations.
		1669
		1670	Method #1 is almost always a bad idea, and buys you nothing. Method #4 is
		1671	rather complicated, but extremely efficient, something that really pays
		1672	off after the first million or so of active timers, i.e. it's usually
		1673	overkill :)
		1674
		1675	=head3 The special problem of time updates
		1676
		1677	Establishing the current time is a costly operation (it usually takes at
		1678	least two system calls): EV therefore updates its idea of the current
		1679	time only before and after C<ev_loop> collects new events, which causes a
		1680	growing difference between C<ev_now ()> and C<ev_time ()> when handling
		1681	lots of events in one iteration.
525		1682
526	The relative timeouts are calculated relative to the C<ev_now ()>	1683	The relative timeouts are calculated relative to the C<ev_now ()>
527	time. This is usually the right thing as this timestamp refers to the time	1684	time. This is usually the right thing as this timestamp refers to the time
528	of the event triggering whatever timeout you are modifying/starting. If	1685	of the event triggering whatever timeout you are modifying/starting. If
529	you suspect event processing to be delayed and you I<need> to base the timeout	1686	you suspect event processing to be delayed and you I<need> to base the
530	on the current time, use something like this to adjust for this:	1687	timeout on the current time, use something like this to adjust for this:
531		1688
532	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);	1689	ev_timer_set (&timer, after + ev_now () - ev_time (), 0.);
533		1690
534	The callback is guarenteed to be invoked only when its timeout has passed,	1691	If the event loop is suspended for a long time, you can also force an
535	but if multiple timers become ready during the same loop iteration then	1692	update of the time returned by C<ev_now ()> by calling C<ev_now_update
536	order of execution is undefined.	1693	()>.
		1694
		1695	=head3 Watcher-Specific Functions and Data Members
537		1696
538	=over 4	1697	=over 4
539		1698
540	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)	1699	=item ev_timer_init (ev_timer *, callback, ev_tstamp after, ev_tstamp repeat)
541		1700
542	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)	1701	=item ev_timer_set (ev_timer *, ev_tstamp after, ev_tstamp repeat)
543		1702
544	Configure the timer to trigger after C<after> seconds. If C<repeat> is	1703	Configure the timer to trigger after C<after> seconds. If C<repeat>
545	C<0.>, then it will automatically be stopped. If it is positive, then the	1704	is C<0.>, then it will automatically be stopped once the timeout is
546	timer will automatically be configured to trigger again C<repeat> seconds	1705	reached. If it is positive, then the timer will automatically be
547	later, again, and again, until stopped manually.	1706	configured to trigger again C<repeat> seconds later, again, and again,
		1707	until stopped manually.
548		1708
549	The timer itself will do a best-effort at avoiding drift, that is, if you	1709	The timer itself will do a best-effort at avoiding drift, that is, if
550	configure a timer to trigger every 10 seconds, then it will trigger at	1710	you configure a timer to trigger every 10 seconds, then it will normally
551	exactly 10 second intervals. If, however, your program cannot keep up with	1711	trigger at exactly 10 second intervals. If, however, your program cannot
552	the timer (because it takes longer than those 10 seconds to do stuff) the	1712	keep up with the timer (because it takes longer than those 10 seconds to
553	timer will not fire more than once per event loop iteration.	1713	do stuff) the timer will not fire more than once per event loop iteration.
554		1714
555	=item ev_timer_again (loop)	1715	=item ev_timer_again (loop, ev_timer *)
556		1716
557	This will act as if the timer timed out and restart it again if it is	1717	This will act as if the timer timed out and restart it again if it is
558	repeating. The exact semantics are:	1718	repeating. The exact semantics are:
559		1719
		1720	If the timer is pending, its pending status is cleared.
		1721
560	If the timer is started but nonrepeating, stop it.	1722	If the timer is started but non-repeating, stop it (as if it timed out).
561		1723
562	If the timer is repeating, either start it if necessary (with the repeat	1724	If the timer is repeating, either start it if necessary (with the
563	value), or reset the running timer to the repeat value.	1725	C<repeat> value), or reset the running timer to the C<repeat> value.
564		1726
565	This sounds a bit complicated, but here is a useful and typical	1727	This sounds a bit complicated, see L<Be smart about timeouts>, above, for a
566	example: Imagine you have a tcp connection and you want a so-called idle	1728	usage example.
567	timeout, that is, you want to be called when there have been, say, 60	1729
568	seconds of inactivity on the socket. The easiest way to do this is to	1730	=item ev_tstamp repeat [read-write]
569	configure an C<ev_timer> with after=repeat=60 and calling ev_timer_again each	1731
570	time you successfully read or write some data. If you go into an idle	1732	The current C<repeat> value. Will be used each time the watcher times out
571	state where you do not expect data to travel on the socket, you can stop	1733	or C<ev_timer_again> is called, and determines the next timeout (if any),
572	the timer, and again will automatically restart it if need be.	1734	which is also when any modifications are taken into account.
573		1735
574	=back	1736	=back
575		1737
		1738	=head3 Examples
		1739
		1740	Example: Create a timer that fires after 60 seconds.
		1741
		1742	static void
		1743	one_minute_cb (struct ev_loop *loop, ev_timer *w, int revents)
		1744	{
		1745	.. one minute over, w is actually stopped right here
		1746	}
		1747
		1748	ev_timer mytimer;
		1749	ev_timer_init (&mytimer, one_minute_cb, 60., 0.);
		1750	ev_timer_start (loop, &mytimer);
		1751
		1752	Example: Create a timeout timer that times out after 10 seconds of
		1753	inactivity.
		1754
		1755	static void
		1756	timeout_cb (struct ev_loop *loop, ev_timer *w, int revents)
		1757	{
		1758	.. ten seconds without any activity
		1759	}
		1760
		1761	ev_timer mytimer;
		1762	ev_timer_init (&mytimer, timeout_cb, 0., 10.); /* note, only repeat used */
		1763	ev_timer_again (&mytimer); /* start timer */
		1764	ev_loop (loop, 0);
		1765
		1766	// and in some piece of code that gets executed on any "activity":
		1767	// reset the timeout to start ticking again at 10 seconds
		1768	ev_timer_again (&mytimer);
		1769
		1770
576	=head2 C<ev_periodic> - to cron or not to cron	1771	=head2 C<ev_periodic> - to cron or not to cron?
577		1772
578	Periodic watchers are also timers of a kind, but they are very versatile	1773	Periodic watchers are also timers of a kind, but they are very versatile
579	(and unfortunately a bit complex).	1774	(and unfortunately a bit complex).
580		1775
581	Unlike C<ev_timer>'s, they are not based on real time (or relative time)	1776	Unlike C<ev_timer>, periodic watchers are not based on real time (or
582	but on wallclock time (absolute time). You can tell a periodic watcher	1777	relative time, the physical time that passes) but on wall clock time
583	to trigger "at" some specific point in time. For example, if you tell a	1778	(absolute time, the thing you can read on your calender or clock). The
584	periodic watcher to trigger in 10 seconds (by specifiying e.g. c<ev_now ()	1779	difference is that wall clock time can run faster or slower than real
585	+ 10.>) and then reset your system clock to the last year, then it will	1780	time, and time jumps are not uncommon (e.g. when you adjust your
586	take a year to trigger the event (unlike an C<ev_timer>, which would trigger	1781	wrist-watch).
587	roughly 10 seconds later and of course not if you reset your system time
588	again).
589		1782
590	They can also be used to implement vastly more complex timers, such as	1783	You can tell a periodic watcher to trigger after some specific point
		1784	in time: for example, if you tell a periodic watcher to trigger "in 10
		1785	seconds" (by specifying e.g. C<ev_now () + 10.>, that is, an absolute time
		1786	not a delay) and then reset your system clock to January of the previous
		1787	year, then it will take a year or more to trigger the event (unlike an
		1788	C<ev_timer>, which would still trigger roughly 10 seconds after starting
		1789	it, as it uses a relative timeout).
		1790
		1791	C<ev_periodic> watchers can also be used to implement vastly more complex
591	triggering an event on eahc midnight, local time.	1792	timers, such as triggering an event on each "midnight, local time", or
		1793	other complicated rules. This cannot be done with C<ev_timer> watchers, as
		1794	those cannot react to time jumps.
592		1795
593	As with timers, the callback is guarenteed to be invoked only when the	1796	As with timers, the callback is guaranteed to be invoked only when the
594	time (C<at>) has been passed, but if multiple periodic timers become ready	1797	point in time where it is supposed to trigger has passed. If multiple
595	during the same loop iteration then order of execution is undefined.	1798	timers become ready during the same loop iteration then the ones with
		1799	earlier time-out values are invoked before ones with later time-out values
		1800	(but this is no longer true when a callback calls C<ev_loop> recursively).
		1801
		1802	=head3 Watcher-Specific Functions and Data Members
596		1803
597	=over 4	1804	=over 4
598		1805
599	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp at, ev_tstamp interval, reschedule_cb)	1806	=item ev_periodic_init (ev_periodic *, callback, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
600		1807
601	=item ev_periodic_set (ev_periodic *, ev_tstamp after, ev_tstamp repeat, reschedule_cb)	1808	=item ev_periodic_set (ev_periodic *, ev_tstamp offset, ev_tstamp interval, reschedule_cb)
602		1809
603	Lots of arguments, lets sort it out... There are basically three modes of	1810	Lots of arguments, let's sort it out... There are basically three modes of
604	operation, and we will explain them from simplest to complex:	1811	operation, and we will explain them from simplest to most complex:
605		1812
606	=over 4	1813	=over 4
607		1814
608	=item * absolute timer (interval = reschedule_cb = 0)	1815	=item * absolute timer (offset = absolute time, interval = 0, reschedule_cb = 0)
609		1816
610	In this configuration the watcher triggers an event at the wallclock time	1817	In this configuration the watcher triggers an event after the wall clock
611	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1818	time C<offset> has passed. It will not repeat and will not adjust when a
612	that is, if it is to be run at January 1st 2011 then it will run when the	1819	time jump occurs, that is, if it is to be run at January 1st 2011 then it
613	system time reaches or surpasses this time.	1820	will be stopped and invoked when the system clock reaches or surpasses
		1821	this point in time.
614		1822
615	=item * non-repeating interval timer (interval > 0, reschedule_cb = 0)	1823	=item * repeating interval timer (offset = offset within interval, interval > 0, reschedule_cb = 0)
616		1824
617	In this mode the watcher will always be scheduled to time out at the next	1825	In this mode the watcher will always be scheduled to time out at the next
618	C<at + N * interval> time (for some integer N) and then repeat, regardless	1826	C<offset + N * interval> time (for some integer N, which can also be
619	of any time jumps.	1827	negative) and then repeat, regardless of any time jumps. The C<offset>
		1828	argument is merely an offset into the C<interval> periods.
620		1829
621	This can be used to create timers that do not drift with respect to system	1830	This can be used to create timers that do not drift with respect to the
622	time:	1831	system clock, for example, here is an C<ev_periodic> that triggers each
		1832	hour, on the hour (with respect to UTC):
623		1833
624	ev_periodic_set (&periodic, 0., 3600., 0);	1834	ev_periodic_set (&periodic, 0., 3600., 0);
625		1835
626	This doesn't mean there will always be 3600 seconds in between triggers,	1836	This doesn't mean there will always be 3600 seconds in between triggers,
627	but only that the the callback will be called when the system time shows a	1837	but only that the callback will be called when the system time shows a
628	full hour (UTC), or more correctly, when the system time is evenly divisible	1838	full hour (UTC), or more correctly, when the system time is evenly divisible
629	by 3600.	1839	by 3600.
630		1840
631	Another way to think about it (for the mathematically inclined) is that	1841	Another way to think about it (for the mathematically inclined) is that
632	C<ev_periodic> will try to run the callback in this mode at the next possible	1842	C<ev_periodic> will try to run the callback in this mode at the next possible
633	time where C<time = at (mod interval)>, regardless of any time jumps.	1843	time where C<time = offset (mod interval)>, regardless of any time jumps.
634		1844
		1845	For numerical stability it is preferable that the C<offset> value is near
		1846	C<ev_now ()> (the current time), but there is no range requirement for
		1847	this value, and in fact is often specified as zero.
		1848
		1849	Note also that there is an upper limit to how often a timer can fire (CPU
		1850	speed for example), so if C<interval> is very small then timing stability
		1851	will of course deteriorate. Libev itself tries to be exact to be about one
		1852	millisecond (if the OS supports it and the machine is fast enough).
		1853
635	=item * manual reschedule mode (reschedule_cb = callback)	1854	=item * manual reschedule mode (offset ignored, interval ignored, reschedule_cb = callback)
636		1855
637	In this mode the values for C<interval> and C<at> are both being	1856	In this mode the values for C<interval> and C<offset> are both being
638	ignored. Instead, each time the periodic watcher gets scheduled, the	1857	ignored. Instead, each time the periodic watcher gets scheduled, the
639	reschedule callback will be called with the watcher as first, and the	1858	reschedule callback will be called with the watcher as first, and the
640	current time as second argument.	1859	current time as second argument.
641		1860
642	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher,	1861	NOTE: I<This callback MUST NOT stop or destroy any periodic watcher, ever,
643	ever, or make any event loop modifications>. If you need to stop it,	1862	or make ANY other event loop modifications whatsoever, unless explicitly
644	return C<now + 1e30> (or so, fudge fudge) and stop it afterwards (e.g. by	1863	allowed by documentation here>.
645	starting a prepare watcher).
646		1864
		1865	If you need to stop it, return C<now + 1e30> (or so, fudge fudge) and stop
		1866	it afterwards (e.g. by starting an C<ev_prepare> watcher, which is the
		1867	only event loop modification you are allowed to do).
		1868
647	Its prototype is C<ev_tstamp (*reschedule_cb)(struct ev_periodic *w,	1869	The callback prototype is C<ev_tstamp (*reschedule_cb)(ev_periodic
648	ev_tstamp now)>, e.g.:	1870	*w, ev_tstamp now)>, e.g.:
649		1871
		1872	static ev_tstamp
650	static ev_tstamp my_rescheduler (struct ev_periodic *w, ev_tstamp now)	1873	my_rescheduler (ev_periodic *w, ev_tstamp now)
651	{	1874	{
652	return now + 60.;	1875	return now + 60.;
653	}	1876	}
654		1877
655	It must return the next time to trigger, based on the passed time value	1878	It must return the next time to trigger, based on the passed time value
656	(that is, the lowest time value larger than to the second argument). It	1879	(that is, the lowest time value larger than to the second argument). It
657	will usually be called just before the callback will be triggered, but	1880	will usually be called just before the callback will be triggered, but
658	might be called at other times, too.	1881	might be called at other times, too.
659		1882
660	NOTE: I<< This callback must always return a time that is later than the	1883	NOTE: I<< This callback must always return a time that is higher than or
661	passed C<now> value >>. Not even C<now> itself will do, it I<must> be larger.	1884	equal to the passed C<now> value >>.
662		1885
663	This can be used to create very complex timers, such as a timer that	1886	This can be used to create very complex timers, such as a timer that
664	triggers on each midnight, local time. To do this, you would calculate the	1887	triggers on "next midnight, local time". To do this, you would calculate the
665	next midnight after C<now> and return the timestamp value for this. How	1888	next midnight after C<now> and return the timestamp value for this. How
666	you do this is, again, up to you (but it is not trivial, which is the main	1889	you do this is, again, up to you (but it is not trivial, which is the main
667	reason I omitted it as an example).	1890	reason I omitted it as an example).
668		1891
669	=back	1892	=back
…		…
673	Simply stops and restarts the periodic watcher again. This is only useful	1896	Simply stops and restarts the periodic watcher again. This is only useful
674	when you changed some parameters or the reschedule callback would return	1897	when you changed some parameters or the reschedule callback would return
675	a different time than the last time it was called (e.g. in a crond like	1898	a different time than the last time it was called (e.g. in a crond like
676	program when the crontabs have changed).	1899	program when the crontabs have changed).
677		1900
		1901	=item ev_tstamp ev_periodic_at (ev_periodic *)
		1902
		1903	When active, returns the absolute time that the watcher is supposed
		1904	to trigger next. This is not the same as the C<offset> argument to
		1905	C<ev_periodic_set>, but indeed works even in interval and manual
		1906	rescheduling modes.
		1907
		1908	=item ev_tstamp offset [read-write]
		1909
		1910	When repeating, this contains the offset value, otherwise this is the
		1911	absolute point in time (the C<offset> value passed to C<ev_periodic_set>,
		1912	although libev might modify this value for better numerical stability).
		1913
		1914	Can be modified any time, but changes only take effect when the periodic
		1915	timer fires or C<ev_periodic_again> is being called.
		1916
		1917	=item ev_tstamp interval [read-write]
		1918
		1919	The current interval value. Can be modified any time, but changes only
		1920	take effect when the periodic timer fires or C<ev_periodic_again> is being
		1921	called.
		1922
		1923	=item ev_tstamp (*reschedule_cb)(ev_periodic *w, ev_tstamp now) [read-write]
		1924
		1925	The current reschedule callback, or C<0>, if this functionality is
		1926	switched off. Can be changed any time, but changes only take effect when
		1927	the periodic timer fires or C<ev_periodic_again> is being called.
		1928
678	=back	1929	=back
679		1930
		1931	=head3 Examples
		1932
		1933	Example: Call a callback every hour, or, more precisely, whenever the
		1934	system time is divisible by 3600. The callback invocation times have
		1935	potentially a lot of jitter, but good long-term stability.
		1936
		1937	static void
		1938	clock_cb (struct ev_loop *loop, ev_io *w, int revents)
		1939	{
		1940	... its now a full hour (UTC, or TAI or whatever your clock follows)
		1941	}
		1942
		1943	ev_periodic hourly_tick;
		1944	ev_periodic_init (&hourly_tick, clock_cb, 0., 3600., 0);
		1945	ev_periodic_start (loop, &hourly_tick);
		1946
		1947	Example: The same as above, but use a reschedule callback to do it:
		1948
		1949	#include <math.h>
		1950
		1951	static ev_tstamp
		1952	my_scheduler_cb (ev_periodic *w, ev_tstamp now)
		1953	{
		1954	return now + (3600. - fmod (now, 3600.));
		1955	}
		1956
		1957	ev_periodic_init (&hourly_tick, clock_cb, 0., 0., my_scheduler_cb);
		1958
		1959	Example: Call a callback every hour, starting now:
		1960
		1961	ev_periodic hourly_tick;
		1962	ev_periodic_init (&hourly_tick, clock_cb,
		1963	fmod (ev_now (loop), 3600.), 3600., 0);
		1964	ev_periodic_start (loop, &hourly_tick);
		1965
		1966
680	=head2 C<ev_signal> - signal me when a signal gets signalled	1967	=head2 C<ev_signal> - signal me when a signal gets signalled!
681		1968
682	Signal watchers will trigger an event when the process receives a specific	1969	Signal watchers will trigger an event when the process receives a specific
683	signal one or more times. Even though signals are very asynchronous, libev	1970	signal one or more times. Even though signals are very asynchronous, libev
684	will try it's best to deliver signals synchronously, i.e. as part of the	1971	will try it's best to deliver signals synchronously, i.e. as part of the
685	normal event processing, like any other event.	1972	normal event processing, like any other event.
686		1973
		1974	If you want signals asynchronously, just use C<sigaction> as you would
		1975	do without libev and forget about sharing the signal. You can even use
		1976	C<ev_async> from a signal handler to synchronously wake up an event loop.
		1977
687	You can configure as many watchers as you like per signal. Only when the	1978	You can configure as many watchers as you like per signal. Only when the
688	first watcher gets started will libev actually register a signal watcher	1979	first watcher gets started will libev actually register a signal handler
689	with the kernel (thus it coexists with your own signal handlers as long	1980	with the kernel (thus it coexists with your own signal handlers as long as
690	as you don't register any with libev). Similarly, when the last signal	1981	you don't register any with libev for the same signal). Similarly, when
691	watcher for a signal is stopped libev will reset the signal handler to	1982	the last signal watcher for a signal is stopped, libev will reset the
692	SIG_DFL (regardless of what it was set to before).	1983	signal handler to SIG_DFL (regardless of what it was set to before).
		1984
		1985	If possible and supported, libev will install its handlers with
		1986	C<SA_RESTART> behaviour enabled, so system calls should not be unduly
		1987	interrupted. If you have a problem with system calls getting interrupted by
		1988	signals you can block all signals in an C<ev_check> watcher and unblock
		1989	them in an C<ev_prepare> watcher.
		1990
		1991	=head3 Watcher-Specific Functions and Data Members
693		1992
694	=over 4	1993	=over 4
695		1994
696	=item ev_signal_init (ev_signal *, callback, int signum)	1995	=item ev_signal_init (ev_signal *, callback, int signum)
697		1996
698	=item ev_signal_set (ev_signal *, int signum)	1997	=item ev_signal_set (ev_signal *, int signum)
699		1998
700	Configures the watcher to trigger on the given signal number (usually one	1999	Configures the watcher to trigger on the given signal number (usually one
701	of the C<SIGxxx> constants).	2000	of the C<SIGxxx> constants).
702		2001
		2002	=item int signum [read-only]
		2003
		2004	The signal the watcher watches out for.
		2005
703	=back	2006	=back
704		2007
		2008	=head3 Examples
		2009
		2010	Example: Try to exit cleanly on SIGINT.
		2011
		2012	static void
		2013	sigint_cb (struct ev_loop *loop, ev_signal *w, int revents)
		2014	{
		2015	ev_unloop (loop, EVUNLOOP_ALL);
		2016	}
		2017
		2018	ev_signal signal_watcher;
		2019	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		2020	ev_signal_start (loop, &signal_watcher);
		2021
		2022
705	=head2 C<ev_child> - wait for pid status changes	2023	=head2 C<ev_child> - watch out for process status changes
706		2024
707	Child watchers trigger when your process receives a SIGCHLD in response to	2025	Child watchers trigger when your process receives a SIGCHLD in response to
708	some child status changes (most typically when a child of yours dies).	2026	some child status changes (most typically when a child of yours dies or
		2027	exits). It is permissible to install a child watcher I<after> the child
		2028	has been forked (which implies it might have already exited), as long
		2029	as the event loop isn't entered (or is continued from a watcher), i.e.,
		2030	forking and then immediately registering a watcher for the child is fine,
		2031	but forking and registering a watcher a few event loop iterations later or
		2032	in the next callback invocation is not.
		2033
		2034	Only the default event loop is capable of handling signals, and therefore
		2035	you can only register child watchers in the default event loop.
		2036
		2037	Due to some design glitches inside libev, child watchers will always be
		2038	handled at maximum priority (their priority is set to C<EV_MAXPRI> by
		2039	libev)
		2040
		2041	=head3 Process Interaction
		2042
		2043	Libev grabs C<SIGCHLD> as soon as the default event loop is
		2044	initialised. This is necessary to guarantee proper behaviour even if
		2045	the first child watcher is started after the child exits. The occurrence
		2046	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		2047	synchronously as part of the event loop processing. Libev always reaps all
		2048	children, even ones not watched.
		2049
		2050	=head3 Overriding the Built-In Processing
		2051
		2052	Libev offers no special support for overriding the built-in child
		2053	processing, but if your application collides with libev's default child
		2054	handler, you can override it easily by installing your own handler for
		2055	C<SIGCHLD> after initialising the default loop, and making sure the
		2056	default loop never gets destroyed. You are encouraged, however, to use an
		2057	event-based approach to child reaping and thus use libev's support for
		2058	that, so other libev users can use C<ev_child> watchers freely.
		2059
		2060	=head3 Stopping the Child Watcher
		2061
		2062	Currently, the child watcher never gets stopped, even when the
		2063	child terminates, so normally one needs to stop the watcher in the
		2064	callback. Future versions of libev might stop the watcher automatically
		2065	when a child exit is detected.
		2066
		2067	=head3 Watcher-Specific Functions and Data Members
709		2068
710	=over 4	2069	=over 4
711		2070
712	=item ev_child_init (ev_child *, callback, int pid)	2071	=item ev_child_init (ev_child *, callback, int pid, int trace)
713		2072
714	=item ev_child_set (ev_child *, int pid)	2073	=item ev_child_set (ev_child *, int pid, int trace)
715		2074
716	Configures the watcher to wait for status changes of process C<pid> (or	2075	Configures the watcher to wait for status changes of process C<pid> (or
717	I<any> process if C<pid> is specified as C<0>). The callback can look	2076	I<any> process if C<pid> is specified as C<0>). The callback can look
718	at the C<rstatus> member of the C<ev_child> watcher structure to see	2077	at the C<rstatus> member of the C<ev_child> watcher structure to see
719	the status word (use the macros from C<sys/wait.h> and see your systems	2078	the status word (use the macros from C<sys/wait.h> and see your systems
720	C<waitpid> documentation). The C<rpid> member contains the pid of the	2079	C<waitpid> documentation). The C<rpid> member contains the pid of the
721	process causing the status change.	2080	process causing the status change. C<trace> must be either C<0> (only
		2081	activate the watcher when the process terminates) or C<1> (additionally
		2082	activate the watcher when the process is stopped or continued).
		2083
		2084	=item int pid [read-only]
		2085
		2086	The process id this watcher watches out for, or C<0>, meaning any process id.
		2087
		2088	=item int rpid [read-write]
		2089
		2090	The process id that detected a status change.
		2091
		2092	=item int rstatus [read-write]
		2093
		2094	The process exit/trace status caused by C<rpid> (see your systems
		2095	C<waitpid> and C<sys/wait.h> documentation for details).
722		2096
723	=back	2097	=back
724		2098
		2099	=head3 Examples
		2100
		2101	Example: C<fork()> a new process and install a child handler to wait for
		2102	its completion.
		2103
		2104	ev_child cw;
		2105
		2106	static void
		2107	child_cb (EV_P_ ev_child *w, int revents)
		2108	{
		2109	ev_child_stop (EV_A_ w);
		2110	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
		2111	}
		2112
		2113	pid_t pid = fork ();
		2114
		2115	if (pid < 0)
		2116	// error
		2117	else if (pid == 0)
		2118	{
		2119	// the forked child executes here
		2120	exit (1);
		2121	}
		2122	else
		2123	{
		2124	ev_child_init (&cw, child_cb, pid, 0);
		2125	ev_child_start (EV_DEFAULT_ &cw);
		2126	}
		2127
		2128
		2129	=head2 C<ev_stat> - did the file attributes just change?
		2130
		2131	This watches a file system path for attribute changes. That is, it calls
		2132	C<stat> on that path in regular intervals (or when the OS says it changed)
		2133	and sees if it changed compared to the last time, invoking the callback if
		2134	it did.
		2135
		2136	The path does not need to exist: changing from "path exists" to "path does
		2137	not exist" is a status change like any other. The condition "path does not
		2138	exist" (or more correctly "path cannot be stat'ed") is signified by the
		2139	C<st_nlink> field being zero (which is otherwise always forced to be at
		2140	least one) and all the other fields of the stat buffer having unspecified
		2141	contents.
		2142
		2143	The path I<must not> end in a slash or contain special components such as
		2144	C<.> or C<..>. The path I<should> be absolute: If it is relative and
		2145	your working directory changes, then the behaviour is undefined.
		2146
		2147	Since there is no portable change notification interface available, the
		2148	portable implementation simply calls C<stat(2)> regularly on the path
		2149	to see if it changed somehow. You can specify a recommended polling
		2150	interval for this case. If you specify a polling interval of C<0> (highly
		2151	recommended!) then a I<suitable, unspecified default> value will be used
		2152	(which you can expect to be around five seconds, although this might
		2153	change dynamically). Libev will also impose a minimum interval which is
		2154	currently around C<0.1>, but that's usually overkill.
		2155
		2156	This watcher type is not meant for massive numbers of stat watchers,
		2157	as even with OS-supported change notifications, this can be
		2158	resource-intensive.
		2159
		2160	At the time of this writing, the only OS-specific interface implemented
		2161	is the Linux inotify interface (implementing kqueue support is left as an
		2162	exercise for the reader. Note, however, that the author sees no way of
		2163	implementing C<ev_stat> semantics with kqueue, except as a hint).
		2164
		2165	=head3 ABI Issues (Largefile Support)
		2166
		2167	Libev by default (unless the user overrides this) uses the default
		2168	compilation environment, which means that on systems with large file
		2169	support disabled by default, you get the 32 bit version of the stat
		2170	structure. When using the library from programs that change the ABI to
		2171	use 64 bit file offsets the programs will fail. In that case you have to
		2172	compile libev with the same flags to get binary compatibility. This is
		2173	obviously the case with any flags that change the ABI, but the problem is
		2174	most noticeably displayed with ev_stat and large file support.
		2175
		2176	The solution for this is to lobby your distribution maker to make large
		2177	file interfaces available by default (as e.g. FreeBSD does) and not
		2178	optional. Libev cannot simply switch on large file support because it has
		2179	to exchange stat structures with application programs compiled using the
		2180	default compilation environment.
		2181
		2182	=head3 Inotify and Kqueue
		2183
		2184	When C<inotify (7)> support has been compiled into libev and present at
		2185	runtime, it will be used to speed up change detection where possible. The
		2186	inotify descriptor will be created lazily when the first C<ev_stat>
		2187	watcher is being started.
		2188
		2189	Inotify presence does not change the semantics of C<ev_stat> watchers
		2190	except that changes might be detected earlier, and in some cases, to avoid
		2191	making regular C<stat> calls. Even in the presence of inotify support
		2192	there are many cases where libev has to resort to regular C<stat> polling,
		2193	but as long as kernel 2.6.25 or newer is used (2.6.24 and older have too
		2194	many bugs), the path exists (i.e. stat succeeds), and the path resides on
		2195	a local filesystem (libev currently assumes only ext2/3, jfs, reiserfs and
		2196	xfs are fully working) libev usually gets away without polling.
		2197
		2198	There is no support for kqueue, as apparently it cannot be used to
		2199	implement this functionality, due to the requirement of having a file
		2200	descriptor open on the object at all times, and detecting renames, unlinks
		2201	etc. is difficult.
		2202
		2203	=head3 C<stat ()> is a synchronous operation
		2204
		2205	Libev doesn't normally do any kind of I/O itself, and so is not blocking
		2206	the process. The exception are C<ev_stat> watchers - those call C<stat
		2207	()>, which is a synchronous operation.
		2208
		2209	For local paths, this usually doesn't matter: unless the system is very
		2210	busy or the intervals between stat's are large, a stat call will be fast,
		2211	as the path data is usually in memory already (except when starting the
		2212	watcher).
		2213
		2214	For networked file systems, calling C<stat ()> can block an indefinite
		2215	time due to network issues, and even under good conditions, a stat call
		2216	often takes multiple milliseconds.
		2217
		2218	Therefore, it is best to avoid using C<ev_stat> watchers on networked
		2219	paths, although this is fully supported by libev.
		2220
		2221	=head3 The special problem of stat time resolution
		2222
		2223	The C<stat ()> system call only supports full-second resolution portably,
		2224	and even on systems where the resolution is higher, most file systems
		2225	still only support whole seconds.
		2226
		2227	That means that, if the time is the only thing that changes, you can
		2228	easily miss updates: on the first update, C<ev_stat> detects a change and
		2229	calls your callback, which does something. When there is another update
		2230	within the same second, C<ev_stat> will be unable to detect unless the
		2231	stat data does change in other ways (e.g. file size).
		2232
		2233	The solution to this is to delay acting on a change for slightly more
		2234	than a second (or till slightly after the next full second boundary), using
		2235	a roughly one-second-delay C<ev_timer> (e.g. C<ev_timer_set (w, 0., 1.02);
		2236	ev_timer_again (loop, w)>).
		2237
		2238	The C<.02> offset is added to work around small timing inconsistencies
		2239	of some operating systems (where the second counter of the current time
		2240	might be be delayed. One such system is the Linux kernel, where a call to
		2241	C<gettimeofday> might return a timestamp with a full second later than
		2242	a subsequent C<time> call - if the equivalent of C<time ()> is used to
		2243	update file times then there will be a small window where the kernel uses
		2244	the previous second to update file times but libev might already execute
		2245	the timer callback).
		2246
		2247	=head3 Watcher-Specific Functions and Data Members
		2248
		2249	=over 4
		2250
		2251	=item ev_stat_init (ev_stat *, callback, const char *path, ev_tstamp interval)
		2252
		2253	=item ev_stat_set (ev_stat *, const char *path, ev_tstamp interval)
		2254
		2255	Configures the watcher to wait for status changes of the given
		2256	C<path>. The C<interval> is a hint on how quickly a change is expected to
		2257	be detected and should normally be specified as C<0> to let libev choose
		2258	a suitable value. The memory pointed to by C<path> must point to the same
		2259	path for as long as the watcher is active.
		2260
		2261	The callback will receive an C<EV_STAT> event when a change was detected,
		2262	relative to the attributes at the time the watcher was started (or the
		2263	last change was detected).
		2264
		2265	=item ev_stat_stat (loop, ev_stat *)
		2266
		2267	Updates the stat buffer immediately with new values. If you change the
		2268	watched path in your callback, you could call this function to avoid
		2269	detecting this change (while introducing a race condition if you are not
		2270	the only one changing the path). Can also be useful simply to find out the
		2271	new values.
		2272
		2273	=item ev_statdata attr [read-only]
		2274
		2275	The most-recently detected attributes of the file. Although the type is
		2276	C<ev_statdata>, this is usually the (or one of the) C<struct stat> types
		2277	suitable for your system, but you can only rely on the POSIX-standardised
		2278	members to be present. If the C<st_nlink> member is C<0>, then there was
		2279	some error while C<stat>ing the file.
		2280
		2281	=item ev_statdata prev [read-only]
		2282
		2283	The previous attributes of the file. The callback gets invoked whenever
		2284	C<prev> != C<attr>, or, more precisely, one or more of these members
		2285	differ: C<st_dev>, C<st_ino>, C<st_mode>, C<st_nlink>, C<st_uid>,
		2286	C<st_gid>, C<st_rdev>, C<st_size>, C<st_atime>, C<st_mtime>, C<st_ctime>.
		2287
		2288	=item ev_tstamp interval [read-only]
		2289
		2290	The specified interval.
		2291
		2292	=item const char *path [read-only]
		2293
		2294	The file system path that is being watched.
		2295
		2296	=back
		2297
		2298	=head3 Examples
		2299
		2300	Example: Watch C</etc/passwd> for attribute changes.
		2301
		2302	static void
		2303	passwd_cb (struct ev_loop *loop, ev_stat *w, int revents)
		2304	{
		2305	/* /etc/passwd changed in some way */
		2306	if (w->attr.st_nlink)
		2307	{
		2308	printf ("passwd current size %ld\n", (long)w->attr.st_size);
		2309	printf ("passwd current atime %ld\n", (long)w->attr.st_mtime);
		2310	printf ("passwd current mtime %ld\n", (long)w->attr.st_mtime);
		2311	}
		2312	else
		2313	/* you shalt not abuse printf for puts */
		2314	puts ("wow, /etc/passwd is not there, expect problems. "
		2315	"if this is windows, they already arrived\n");
		2316	}
		2317
		2318	...
		2319	ev_stat passwd;
		2320
		2321	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
		2322	ev_stat_start (loop, &passwd);
		2323
		2324	Example: Like above, but additionally use a one-second delay so we do not
		2325	miss updates (however, frequent updates will delay processing, too, so
		2326	one might do the work both on C<ev_stat> callback invocation I<and> on
		2327	C<ev_timer> callback invocation).
		2328
		2329	static ev_stat passwd;
		2330	static ev_timer timer;
		2331
		2332	static void
		2333	timer_cb (EV_P_ ev_timer *w, int revents)
		2334	{
		2335	ev_timer_stop (EV_A_ w);
		2336
		2337	/* now it's one second after the most recent passwd change */
		2338	}
		2339
		2340	static void
		2341	stat_cb (EV_P_ ev_stat *w, int revents)
		2342	{
		2343	/* reset the one-second timer */
		2344	ev_timer_again (EV_A_ &timer);
		2345	}
		2346
		2347	...
		2348	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		2349	ev_stat_start (loop, &passwd);
		2350	ev_timer_init (&timer, timer_cb, 0., 1.02);
		2351
		2352
725	=head2 C<ev_idle> - when you've got nothing better to do	2353	=head2 C<ev_idle> - when you've got nothing better to do...
726		2354
727	Idle watchers trigger events when there are no other events are pending	2355	Idle watchers trigger events when no other events of the same or higher
728	(prepare, check and other idle watchers do not count). That is, as long	2356	priority are pending (prepare, check and other idle watchers do not count
729	as your process is busy handling sockets or timeouts (or even signals,	2357	as receiving "events").
730	imagine) it will not be triggered. But when your process is idle all idle	2358
731	watchers are being called again and again, once per event loop iteration -	2359	That is, as long as your process is busy handling sockets or timeouts
		2360	(or even signals, imagine) of the same or higher priority it will not be
		2361	triggered. But when your process is idle (or only lower-priority watchers
		2362	are pending), the idle watchers are being called once per event loop
732	until stopped, that is, or your process receives more events and becomes	2363	iteration - until stopped, that is, or your process receives more events
733	busy.	2364	and becomes busy again with higher priority stuff.
734		2365
735	The most noteworthy effect is that as long as any idle watchers are	2366	The most noteworthy effect is that as long as any idle watchers are
736	active, the process will not block when waiting for new events.	2367	active, the process will not block when waiting for new events.
737		2368
738	Apart from keeping your process non-blocking (which is a useful	2369	Apart from keeping your process non-blocking (which is a useful
739	effect on its own sometimes), idle watchers are a good place to do	2370	effect on its own sometimes), idle watchers are a good place to do
740	"pseudo-background processing", or delay processing stuff to after the	2371	"pseudo-background processing", or delay processing stuff to after the
741	event loop has handled all outstanding events.	2372	event loop has handled all outstanding events.
742		2373
		2374	=head3 Watcher-Specific Functions and Data Members
		2375
743	=over 4	2376	=over 4
744		2377
745	=item ev_idle_init (ev_signal *, callback)	2378	=item ev_idle_init (ev_idle *, callback)
746		2379
747	Initialises and configures the idle watcher - it has no parameters of any	2380	Initialises and configures the idle watcher - it has no parameters of any
748	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	2381	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
749	believe me.	2382	believe me.
750		2383
751	=back	2384	=back
752		2385
		2386	=head3 Examples
		2387
		2388	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
		2389	callback, free it. Also, use no error checking, as usual.
		2390
		2391	static void
		2392	idle_cb (struct ev_loop *loop, ev_idle *w, int revents)
		2393	{
		2394	free (w);
		2395	// now do something you wanted to do when the program has
		2396	// no longer anything immediate to do.
		2397	}
		2398
		2399	ev_idle *idle_watcher = malloc (sizeof (ev_idle));
		2400	ev_idle_init (idle_watcher, idle_cb);
		2401	ev_idle_start (loop, idle_watcher);
		2402
		2403
753	=head2 C<ev_prepare> and C<ev_check> - customise your event loop	2404	=head2 C<ev_prepare> and C<ev_check> - customise your event loop!
754		2405
755	Prepare and check watchers are usually (but not always) used in tandem:	2406	Prepare and check watchers are usually (but not always) used in pairs:
756	prepare watchers get invoked before the process blocks and check watchers	2407	prepare watchers get invoked before the process blocks and check watchers
757	afterwards.	2408	afterwards.
758		2409
		2410	You I<must not> call C<ev_loop> or similar functions that enter
		2411	the current event loop from either C<ev_prepare> or C<ev_check>
		2412	watchers. Other loops than the current one are fine, however. The
		2413	rationale behind this is that you do not need to check for recursion in
		2414	those watchers, i.e. the sequence will always be C<ev_prepare>, blocking,
		2415	C<ev_check> so if you have one watcher of each kind they will always be
		2416	called in pairs bracketing the blocking call.
		2417
759	Their main purpose is to integrate other event mechanisms into libev. This	2418	Their main purpose is to integrate other event mechanisms into libev and
760	could be used, for example, to track variable changes, implement your own	2419	their use is somewhat advanced. They could be used, for example, to track
761	watchers, integrate net-snmp or a coroutine library and lots more.	2420	variable changes, implement your own watchers, integrate net-snmp or a
		2421	coroutine library and lots more. They are also occasionally useful if
		2422	you cache some data and want to flush it before blocking (for example,
		2423	in X programs you might want to do an C<XFlush ()> in an C<ev_prepare>
		2424	watcher).
762		2425
763	This is done by examining in each prepare call which file descriptors need	2426	This is done by examining in each prepare call which file descriptors
764	to be watched by the other library, registering C<ev_io> watchers for	2427	need to be watched by the other library, registering C<ev_io> watchers
765	them and starting an C<ev_timer> watcher for any timeouts (many libraries	2428	for them and starting an C<ev_timer> watcher for any timeouts (many
766	provide just this functionality). Then, in the check watcher you check for	2429	libraries provide exactly this functionality). Then, in the check watcher,
767	any events that occured (by checking the pending status of all watchers	2430	you check for any events that occurred (by checking the pending status
768	and stopping them) and call back into the library. The I/O and timer	2431	of all watchers and stopping them) and call back into the library. The
769	callbacks will never actually be called (but must be valid nevertheless,	2432	I/O and timer callbacks will never actually be called (but must be valid
770	because you never know, you know?).	2433	nevertheless, because you never know, you know?).
771		2434
772	As another example, the Perl Coro module uses these hooks to integrate	2435	As another example, the Perl Coro module uses these hooks to integrate
773	coroutines into libev programs, by yielding to other active coroutines	2436	coroutines into libev programs, by yielding to other active coroutines
774	during each prepare and only letting the process block if no coroutines	2437	during each prepare and only letting the process block if no coroutines
775	are ready to run (it's actually more complicated: it only runs coroutines	2438	are ready to run (it's actually more complicated: it only runs coroutines
776	with priority higher than or equal to the event loop and one coroutine	2439	with priority higher than or equal to the event loop and one coroutine
777	of lower priority, but only once, using idle watchers to keep the event	2440	of lower priority, but only once, using idle watchers to keep the event
778	loop from blocking if lower-priority coroutines are active, thus mapping	2441	loop from blocking if lower-priority coroutines are active, thus mapping
779	low-priority coroutines to idle/background tasks).	2442	low-priority coroutines to idle/background tasks).
780		2443
		2444	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
		2445	priority, to ensure that they are being run before any other watchers
		2446	after the poll (this doesn't matter for C<ev_prepare> watchers).
		2447
		2448	Also, C<ev_check> watchers (and C<ev_prepare> watchers, too) should not
		2449	activate ("feed") events into libev. While libev fully supports this, they
		2450	might get executed before other C<ev_check> watchers did their job. As
		2451	C<ev_check> watchers are often used to embed other (non-libev) event
		2452	loops those other event loops might be in an unusable state until their
		2453	C<ev_check> watcher ran (always remind yourself to coexist peacefully with
		2454	others).
		2455
		2456	=head3 Watcher-Specific Functions and Data Members
		2457
781	=over 4	2458	=over 4
782		2459
783	=item ev_prepare_init (ev_prepare *, callback)	2460	=item ev_prepare_init (ev_prepare *, callback)
784		2461
785	=item ev_check_init (ev_check *, callback)	2462	=item ev_check_init (ev_check *, callback)
786		2463
787	Initialises and configures the prepare or check watcher - they have no	2464	Initialises and configures the prepare or check watcher - they have no
788	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	2465	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
789	macros, but using them is utterly, utterly and completely pointless.	2466	macros, but using them is utterly, utterly, utterly and completely
		2467	pointless.
790		2468
791	=back	2469	=back
792		2470
		2471	=head3 Examples
		2472
		2473	There are a number of principal ways to embed other event loops or modules
		2474	into libev. Here are some ideas on how to include libadns into libev
		2475	(there is a Perl module named C<EV::ADNS> that does this, which you could
		2476	use as a working example. Another Perl module named C<EV::Glib> embeds a
		2477	Glib main context into libev, and finally, C<Glib::EV> embeds EV into the
		2478	Glib event loop).
		2479
		2480	Method 1: Add IO watchers and a timeout watcher in a prepare handler,
		2481	and in a check watcher, destroy them and call into libadns. What follows
		2482	is pseudo-code only of course. This requires you to either use a low
		2483	priority for the check watcher or use C<ev_clear_pending> explicitly, as
		2484	the callbacks for the IO/timeout watchers might not have been called yet.
		2485
		2486	static ev_io iow [nfd];
		2487	static ev_timer tw;
		2488
		2489	static void
		2490	io_cb (struct ev_loop *loop, ev_io *w, int revents)
		2491	{
		2492	}
		2493
		2494	// create io watchers for each fd and a timer before blocking
		2495	static void
		2496	adns_prepare_cb (struct ev_loop *loop, ev_prepare *w, int revents)
		2497	{
		2498	int timeout = 3600000;
		2499	struct pollfd fds [nfd];
		2500	// actual code will need to loop here and realloc etc.
		2501	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
		2502
		2503	/* the callback is illegal, but won't be called as we stop during check */
		2504	ev_timer_init (&tw, 0, timeout * 1e-3, 0.);
		2505	ev_timer_start (loop, &tw);
		2506
		2507	// create one ev_io per pollfd
		2508	for (int i = 0; i < nfd; ++i)
		2509	{
		2510	ev_io_init (iow + i, io_cb, fds [i].fd,
		2511	((fds [i].events & POLLIN ? EV_READ : 0)
		2512	| (fds [i].events & POLLOUT ? EV_WRITE : 0)));
		2513
		2514	fds [i].revents = 0;
		2515	ev_io_start (loop, iow + i);
		2516	}
		2517	}
		2518
		2519	// stop all watchers after blocking
		2520	static void
		2521	adns_check_cb (struct ev_loop *loop, ev_check *w, int revents)
		2522	{
		2523	ev_timer_stop (loop, &tw);
		2524
		2525	for (int i = 0; i < nfd; ++i)
		2526	{
		2527	// set the relevant poll flags
		2528	// could also call adns_processreadable etc. here
		2529	struct pollfd *fd = fds + i;
		2530	int revents = ev_clear_pending (iow + i);
		2531	if (revents & EV_READ ) fd->revents |= fd->events & POLLIN;
		2532	if (revents & EV_WRITE) fd->revents |= fd->events & POLLOUT;
		2533
		2534	// now stop the watcher
		2535	ev_io_stop (loop, iow + i);
		2536	}
		2537
		2538	adns_afterpoll (adns, fds, nfd, timeval_from (ev_now (loop));
		2539	}
		2540
		2541	Method 2: This would be just like method 1, but you run C<adns_afterpoll>
		2542	in the prepare watcher and would dispose of the check watcher.
		2543
		2544	Method 3: If the module to be embedded supports explicit event
		2545	notification (libadns does), you can also make use of the actual watcher
		2546	callbacks, and only destroy/create the watchers in the prepare watcher.
		2547
		2548	static void
		2549	timer_cb (EV_P_ ev_timer *w, int revents)
		2550	{
		2551	adns_state ads = (adns_state)w->data;
		2552	update_now (EV_A);
		2553
		2554	adns_processtimeouts (ads, &tv_now);
		2555	}
		2556
		2557	static void
		2558	io_cb (EV_P_ ev_io *w, int revents)
		2559	{
		2560	adns_state ads = (adns_state)w->data;
		2561	update_now (EV_A);
		2562
		2563	if (revents & EV_READ ) adns_processreadable (ads, w->fd, &tv_now);
		2564	if (revents & EV_WRITE) adns_processwriteable (ads, w->fd, &tv_now);
		2565	}
		2566
		2567	// do not ever call adns_afterpoll
		2568
		2569	Method 4: Do not use a prepare or check watcher because the module you
		2570	want to embed is not flexible enough to support it. Instead, you can
		2571	override their poll function. The drawback with this solution is that the
		2572	main loop is now no longer controllable by EV. The C<Glib::EV> module uses
		2573	this approach, effectively embedding EV as a client into the horrible
		2574	libglib event loop.
		2575
		2576	static gint
		2577	event_poll_func (GPollFD *fds, guint nfds, gint timeout)
		2578	{
		2579	int got_events = 0;
		2580
		2581	for (n = 0; n < nfds; ++n)
		2582	// create/start io watcher that sets the relevant bits in fds[n] and increment got_events
		2583
		2584	if (timeout >= 0)
		2585	// create/start timer
		2586
		2587	// poll
		2588	ev_loop (EV_A_ 0);
		2589
		2590	// stop timer again
		2591	if (timeout >= 0)
		2592	ev_timer_stop (EV_A_ &to);
		2593
		2594	// stop io watchers again - their callbacks should have set
		2595	for (n = 0; n < nfds; ++n)
		2596	ev_io_stop (EV_A_ iow [n]);
		2597
		2598	return got_events;
		2599	}
		2600
		2601
		2602	=head2 C<ev_embed> - when one backend isn't enough...
		2603
		2604	This is a rather advanced watcher type that lets you embed one event loop
		2605	into another (currently only C<ev_io> events are supported in the embedded
		2606	loop, other types of watchers might be handled in a delayed or incorrect
		2607	fashion and must not be used).
		2608
		2609	There are primarily two reasons you would want that: work around bugs and
		2610	prioritise I/O.
		2611
		2612	As an example for a bug workaround, the kqueue backend might only support
		2613	sockets on some platform, so it is unusable as generic backend, but you
		2614	still want to make use of it because you have many sockets and it scales
		2615	so nicely. In this case, you would create a kqueue-based loop and embed
		2616	it into your default loop (which might use e.g. poll). Overall operation
		2617	will be a bit slower because first libev has to call C<poll> and then
		2618	C<kevent>, but at least you can use both mechanisms for what they are
		2619	best: C<kqueue> for scalable sockets and C<poll> if you want it to work :)
		2620
		2621	As for prioritising I/O: under rare circumstances you have the case where
		2622	some fds have to be watched and handled very quickly (with low latency),
		2623	and even priorities and idle watchers might have too much overhead. In
		2624	this case you would put all the high priority stuff in one loop and all
		2625	the rest in a second one, and embed the second one in the first.
		2626
		2627	As long as the watcher is active, the callback will be invoked every
		2628	time there might be events pending in the embedded loop. The callback
		2629	must then call C<ev_embed_sweep (mainloop, watcher)> to make a single
		2630	sweep and invoke their callbacks (the callback doesn't need to invoke the
		2631	C<ev_embed_sweep> function directly, it could also start an idle watcher
		2632	to give the embedded loop strictly lower priority for example).
		2633
		2634	You can also set the callback to C<0>, in which case the embed watcher
		2635	will automatically execute the embedded loop sweep whenever necessary.
		2636
		2637	Fork detection will be handled transparently while the C<ev_embed> watcher
		2638	is active, i.e., the embedded loop will automatically be forked when the
		2639	embedding loop forks. In other cases, the user is responsible for calling
		2640	C<ev_loop_fork> on the embedded loop.
		2641
		2642	Unfortunately, not all backends are embeddable: only the ones returned by
		2643	C<ev_embeddable_backends> are, which, unfortunately, does not include any
		2644	portable one.
		2645
		2646	So when you want to use this feature you will always have to be prepared
		2647	that you cannot get an embeddable loop. The recommended way to get around
		2648	this is to have a separate variables for your embeddable loop, try to
		2649	create it, and if that fails, use the normal loop for everything.
		2650
		2651	=head3 C<ev_embed> and fork
		2652
		2653	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2654	automatically be applied to the embedded loop as well, so no special
		2655	fork handling is required in that case. When the watcher is not running,
		2656	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2657	as applicable.
		2658
		2659	=head3 Watcher-Specific Functions and Data Members
		2660
		2661	=over 4
		2662
		2663	=item ev_embed_init (ev_embed *, callback, struct ev_loop *embedded_loop)
		2664
		2665	=item ev_embed_set (ev_embed *, callback, struct ev_loop *embedded_loop)
		2666
		2667	Configures the watcher to embed the given loop, which must be
		2668	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		2669	invoked automatically, otherwise it is the responsibility of the callback
		2670	to invoke it (it will continue to be called until the sweep has been done,
		2671	if you do not want that, you need to temporarily stop the embed watcher).
		2672
		2673	=item ev_embed_sweep (loop, ev_embed *)
		2674
		2675	Make a single, non-blocking sweep over the embedded loop. This works
		2676	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		2677	appropriate way for embedded loops.
		2678
		2679	=item struct ev_loop *other [read-only]
		2680
		2681	The embedded event loop.
		2682
		2683	=back
		2684
		2685	=head3 Examples
		2686
		2687	Example: Try to get an embeddable event loop and embed it into the default
		2688	event loop. If that is not possible, use the default loop. The default
		2689	loop is stored in C<loop_hi>, while the embeddable loop is stored in
		2690	C<loop_lo> (which is C<loop_hi> in the case no embeddable loop can be
		2691	used).
		2692
		2693	struct ev_loop *loop_hi = ev_default_init (0);
		2694	struct ev_loop *loop_lo = 0;
		2695	ev_embed embed;
		2696
		2697	// see if there is a chance of getting one that works
		2698	// (remember that a flags value of 0 means autodetection)
		2699	loop_lo = ev_embeddable_backends () & ev_recommended_backends ()
		2700	? ev_loop_new (ev_embeddable_backends () & ev_recommended_backends ())
		2701	: 0;
		2702
		2703	// if we got one, then embed it, otherwise default to loop_hi
		2704	if (loop_lo)
		2705	{
		2706	ev_embed_init (&embed, 0, loop_lo);
		2707	ev_embed_start (loop_hi, &embed);
		2708	}
		2709	else
		2710	loop_lo = loop_hi;
		2711
		2712	Example: Check if kqueue is available but not recommended and create
		2713	a kqueue backend for use with sockets (which usually work with any
		2714	kqueue implementation). Store the kqueue/socket-only event loop in
		2715	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
		2716
		2717	struct ev_loop *loop = ev_default_init (0);
		2718	struct ev_loop *loop_socket = 0;
		2719	ev_embed embed;
		2720
		2721	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2722	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2723	{
		2724	ev_embed_init (&embed, 0, loop_socket);
		2725	ev_embed_start (loop, &embed);
		2726	}
		2727
		2728	if (!loop_socket)
		2729	loop_socket = loop;
		2730
		2731	// now use loop_socket for all sockets, and loop for everything else
		2732
		2733
		2734	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
		2735
		2736	Fork watchers are called when a C<fork ()> was detected (usually because
		2737	whoever is a good citizen cared to tell libev about it by calling
		2738	C<ev_default_fork> or C<ev_loop_fork>). The invocation is done before the
		2739	event loop blocks next and before C<ev_check> watchers are being called,
		2740	and only in the child after the fork. If whoever good citizen calling
		2741	C<ev_default_fork> cheats and calls it in the wrong process, the fork
		2742	handlers will be invoked, too, of course.
		2743
		2744	=head3 The special problem of life after fork - how is it possible?
		2745
		2746	Most uses of C<fork()> consist of forking, then some simple calls to ste
		2747	up/change the process environment, followed by a call to C<exec()>. This
		2748	sequence should be handled by libev without any problems.
		2749
		2750	This changes when the application actually wants to do event handling
		2751	in the child, or both parent in child, in effect "continuing" after the
		2752	fork.
		2753
		2754	The default mode of operation (for libev, with application help to detect
		2755	forks) is to duplicate all the state in the child, as would be expected
		2756	when I<either> the parent I<or> the child process continues.
		2757
		2758	When both processes want to continue using libev, then this is usually the
		2759	wrong result. In that case, usually one process (typically the parent) is
		2760	supposed to continue with all watchers in place as before, while the other
		2761	process typically wants to start fresh, i.e. without any active watchers.
		2762
		2763	The cleanest and most efficient way to achieve that with libev is to
		2764	simply create a new event loop, which of course will be "empty", and
		2765	use that for new watchers. This has the advantage of not touching more
		2766	memory than necessary, and thus avoiding the copy-on-write, and the
		2767	disadvantage of having to use multiple event loops (which do not support
		2768	signal watchers).
		2769
		2770	When this is not possible, or you want to use the default loop for
		2771	other reasons, then in the process that wants to start "fresh", call
		2772	C<ev_default_destroy ()> followed by C<ev_default_loop (...)>. Destroying
		2773	the default loop will "orphan" (not stop) all registered watchers, so you
		2774	have to be careful not to execute code that modifies those watchers. Note
		2775	also that in that case, you have to re-register any signal watchers.
		2776
		2777	=head3 Watcher-Specific Functions and Data Members
		2778
		2779	=over 4
		2780
		2781	=item ev_fork_init (ev_signal *, callback)
		2782
		2783	Initialises and configures the fork watcher - it has no parameters of any
		2784	kind. There is a C<ev_fork_set> macro, but using it is utterly pointless,
		2785	believe me.
		2786
		2787	=back
		2788
		2789
		2790	=head2 C<ev_async> - how to wake up another event loop
		2791
		2792	In general, you cannot use an C<ev_loop> from multiple threads or other
		2793	asynchronous sources such as signal handlers (as opposed to multiple event
		2794	loops - those are of course safe to use in different threads).
		2795
		2796	Sometimes, however, you need to wake up another event loop you do not
		2797	control, for example because it belongs to another thread. This is what
		2798	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2799	can signal it by calling C<ev_async_send>, which is thread- and signal
		2800	safe.
		2801
		2802	This functionality is very similar to C<ev_signal> watchers, as signals,
		2803	too, are asynchronous in nature, and signals, too, will be compressed
		2804	(i.e. the number of callback invocations may be less than the number of
		2805	C<ev_async_sent> calls).
		2806
		2807	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2808	just the default loop.
		2809
		2810	=head3 Queueing
		2811
		2812	C<ev_async> does not support queueing of data in any way. The reason
		2813	is that the author does not know of a simple (or any) algorithm for a
		2814	multiple-writer-single-reader queue that works in all cases and doesn't
		2815	need elaborate support such as pthreads.
		2816
		2817	That means that if you want to queue data, you have to provide your own
		2818	queue. But at least I can tell you how to implement locking around your
		2819	queue:
		2820
		2821	=over 4
		2822
		2823	=item queueing from a signal handler context
		2824
		2825	To implement race-free queueing, you simply add to the queue in the signal
		2826	handler but you block the signal handler in the watcher callback. Here is
		2827	an example that does that for some fictitious SIGUSR1 handler:
		2828
		2829	static ev_async mysig;
		2830
		2831	static void
		2832	sigusr1_handler (void)
		2833	{
		2834	sometype data;
		2835
		2836	// no locking etc.
		2837	queue_put (data);
		2838	ev_async_send (EV_DEFAULT_ &mysig);
		2839	}
		2840
		2841	static void
		2842	mysig_cb (EV_P_ ev_async *w, int revents)
		2843	{
		2844	sometype data;
		2845	sigset_t block, prev;
		2846
		2847	sigemptyset (&block);
		2848	sigaddset (&block, SIGUSR1);
		2849	sigprocmask (SIG_BLOCK, &block, &prev);
		2850
		2851	while (queue_get (&data))
		2852	process (data);
		2853
		2854	if (sigismember (&prev, SIGUSR1)
		2855	sigprocmask (SIG_UNBLOCK, &block, 0);
		2856	}
		2857
		2858	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2859	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2860	either...).
		2861
		2862	=item queueing from a thread context
		2863
		2864	The strategy for threads is different, as you cannot (easily) block
		2865	threads but you can easily preempt them, so to queue safely you need to
		2866	employ a traditional mutex lock, such as in this pthread example:
		2867
		2868	static ev_async mysig;
		2869	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2870
		2871	static void
		2872	otherthread (void)
		2873	{
		2874	// only need to lock the actual queueing operation
		2875	pthread_mutex_lock (&mymutex);
		2876	queue_put (data);
		2877	pthread_mutex_unlock (&mymutex);
		2878
		2879	ev_async_send (EV_DEFAULT_ &mysig);
		2880	}
		2881
		2882	static void
		2883	mysig_cb (EV_P_ ev_async *w, int revents)
		2884	{
		2885	pthread_mutex_lock (&mymutex);
		2886
		2887	while (queue_get (&data))
		2888	process (data);
		2889
		2890	pthread_mutex_unlock (&mymutex);
		2891	}
		2892
		2893	=back
		2894
		2895
		2896	=head3 Watcher-Specific Functions and Data Members
		2897
		2898	=over 4
		2899
		2900	=item ev_async_init (ev_async *, callback)
		2901
		2902	Initialises and configures the async watcher - it has no parameters of any
		2903	kind. There is a C<ev_async_set> macro, but using it is utterly pointless,
		2904	trust me.
		2905
		2906	=item ev_async_send (loop, ev_async *)
		2907
		2908	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2909	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2910	C<ev_feed_event>, this call is safe to do from other threads, signal or
		2911	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding
		2912	section below on what exactly this means).
		2913
		2914	Note that, as with other watchers in libev, multiple events might get
		2915	compressed into a single callback invocation (another way to look at this
		2916	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
		2917	reset when the event loop detects that).
		2918
		2919	This call incurs the overhead of a system call only once per event loop
		2920	iteration, so while the overhead might be noticeable, it doesn't apply to
		2921	repeated calls to C<ev_async_send> for the same event loop.
		2922
		2923	=item bool = ev_async_pending (ev_async *)
		2924
		2925	Returns a non-zero value when C<ev_async_send> has been called on the
		2926	watcher but the event has not yet been processed (or even noted) by the
		2927	event loop.
		2928
		2929	C<ev_async_send> sets a flag in the watcher and wakes up the loop. When
		2930	the loop iterates next and checks for the watcher to have become active,
		2931	it will reset the flag again. C<ev_async_pending> can be used to very
		2932	quickly check whether invoking the loop might be a good idea.
		2933
		2934	Not that this does I<not> check whether the watcher itself is pending,
		2935	only whether it has been requested to make this watcher pending: there
		2936	is a time window between the event loop checking and resetting the async
		2937	notification, and the callback being invoked.
		2938
		2939	=back
		2940
		2941
793	=head1 OTHER FUNCTIONS	2942	=head1 OTHER FUNCTIONS
794		2943
795	There are some other functions of possible interest. Described. Here. Now.	2944	There are some other functions of possible interest. Described. Here. Now.
796		2945
797	=over 4	2946	=over 4
798		2947
799	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	2948	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
800		2949
801	This function combines a simple timer and an I/O watcher, calls your	2950	This function combines a simple timer and an I/O watcher, calls your
802	callback on whichever event happens first and automatically stop both	2951	callback on whichever event happens first and automatically stops both
803	watchers. This is useful if you want to wait for a single event on an fd	2952	watchers. This is useful if you want to wait for a single event on an fd
804	or timeout without having to allocate/configure/start/stop/free one or	2953	or timeout without having to allocate/configure/start/stop/free one or
805	more watchers yourself.	2954	more watchers yourself.
806		2955
807	If C<fd> is less than 0, then no I/O watcher will be started and events	2956	If C<fd> is less than 0, then no I/O watcher will be started and the
808	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2957	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
809	C<events> set will be craeted and started.	2958	the given C<fd> and C<events> set will be created and started.
810		2959
811	If C<timeout> is less than 0, then no timeout watcher will be	2960	If C<timeout> is less than 0, then no timeout watcher will be
812	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2961	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
813	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2962	repeat = 0) will be started. C<0> is a valid timeout.
814	dubious value.
815		2963
816	The callback has the type C<void (*cb)(int revents, void *arg)> and gets	2964	The callback has the type C<void (*cb)(int revents, void *arg)> and gets
817	passed an C<revents> set like normal event callbacks (a combination of	2965	passed an C<revents> set like normal event callbacks (a combination of
818	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2966	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
819	value passed to C<ev_once>:	2967	value passed to C<ev_once>. Note that it is possible to receive I<both>
		2968	a timeout and an io event at the same time - you probably should give io
		2969	events precedence.
820		2970
		2971	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
		2972
821	static void stdin_ready (int revents, void *arg)	2973	static void stdin_ready (int revents, void *arg)
822	{	2974	{
823	if (revents & EV_TIMEOUT)
824	/* doh, nothing entered */;
825	else if (revents & EV_READ)	2975	if (revents & EV_READ)
826	/* stdin might have data for us, joy! */;	2976	/* stdin might have data for us, joy! */;
		2977	else if (revents & EV_TIMEOUT)
		2978	/* doh, nothing entered */;
827	}	2979	}
828		2980
829	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2981	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
830		2982
831	=item ev_feed_event (loop, watcher, int events)	2983	=item ev_feed_event (struct ev_loop *, watcher *, int revents)
832		2984
833	Feeds the given event set into the event loop, as if the specified event	2985	Feeds the given event set into the event loop, as if the specified event
834	had happened for the specified watcher (which must be a pointer to an	2986	had happened for the specified watcher (which must be a pointer to an
835	initialised but not necessarily started event watcher).	2987	initialised but not necessarily started event watcher).
836		2988
837	=item ev_feed_fd_event (loop, int fd, int revents)	2989	=item ev_feed_fd_event (struct ev_loop *, int fd, int revents)
838		2990
839	Feed an event on the given fd, as if a file descriptor backend detected	2991	Feed an event on the given fd, as if a file descriptor backend detected
840	the given events it.	2992	the given events it.
841		2993
842	=item ev_feed_signal_event (loop, int signum)	2994	=item ev_feed_signal_event (struct ev_loop *loop, int signum)
843		2995
844	Feed an event as if the given signal occured (loop must be the default loop!).	2996	Feed an event as if the given signal occurred (C<loop> must be the default
		2997	loop!).
845		2998
846	=back	2999	=back
		3000
847		3001
848	=head1 LIBEVENT EMULATION	3002	=head1 LIBEVENT EMULATION
849		3003
850	Libev offers a compatibility emulation layer for libevent. It cannot	3004	Libev offers a compatibility emulation layer for libevent. It cannot
851	emulate the internals of libevent, so here are some usage hints:	3005	emulate the internals of libevent, so here are some usage hints:
…		…
863		3017
864	=item * Priorities are not currently supported. Initialising priorities	3018	=item * Priorities are not currently supported. Initialising priorities
865	will fail and all watchers will have the same priority, even though there	3019	will fail and all watchers will have the same priority, even though there
866	is an ev_pri field.	3020	is an ev_pri field.
867		3021
		3022	=item * In libevent, the last base created gets the signals, in libev, the
		3023	first base created (== the default loop) gets the signals.
		3024
868	=item * Other members are not supported.	3025	=item * Other members are not supported.
869		3026
870	=item * The libev emulation is I<not> ABI compatible to libevent, you need	3027	=item * The libev emulation is I<not> ABI compatible to libevent, you need
871	to use the libev header file and library.	3028	to use the libev header file and library.
872		3029
873	=back	3030	=back
874		3031
875	=head1 C++ SUPPORT	3032	=head1 C++ SUPPORT
876		3033
877	TBD.	3034	Libev comes with some simplistic wrapper classes for C++ that mainly allow
		3035	you to use some convenience methods to start/stop watchers and also change
		3036	the callback model to a model using method callbacks on objects.
		3037
		3038	To use it,
		3039
		3040	#include <ev++.h>
		3041
		3042	This automatically includes F<ev.h> and puts all of its definitions (many
		3043	of them macros) into the global namespace. All C++ specific things are
		3044	put into the C<ev> namespace. It should support all the same embedding
		3045	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
		3046
		3047	Care has been taken to keep the overhead low. The only data member the C++
		3048	classes add (compared to plain C-style watchers) is the event loop pointer
		3049	that the watcher is associated with (or no additional members at all if
		3050	you disable C<EV_MULTIPLICITY> when embedding libev).
		3051
		3052	Currently, functions, and static and non-static member functions can be
		3053	used as callbacks. Other types should be easy to add as long as they only
		3054	need one additional pointer for context. If you need support for other
		3055	types of functors please contact the author (preferably after implementing
		3056	it).
		3057
		3058	Here is a list of things available in the C<ev> namespace:
		3059
		3060	=over 4
		3061
		3062	=item C<ev::READ>, C<ev::WRITE> etc.
		3063
		3064	These are just enum values with the same values as the C<EV_READ> etc.
		3065	macros from F<ev.h>.
		3066
		3067	=item C<ev::tstamp>, C<ev::now>
		3068
		3069	Aliases to the same types/functions as with the C<ev_> prefix.
		3070
		3071	=item C<ev::io>, C<ev::timer>, C<ev::periodic>, C<ev::idle>, C<ev::sig> etc.
		3072
		3073	For each C<ev_TYPE> watcher in F<ev.h> there is a corresponding class of
		3074	the same name in the C<ev> namespace, with the exception of C<ev_signal>
		3075	which is called C<ev::sig> to avoid clashes with the C<signal> macro
		3076	defines by many implementations.
		3077
		3078	All of those classes have these methods:
		3079
		3080	=over 4
		3081
		3082	=item ev::TYPE::TYPE ()
		3083
		3084	=item ev::TYPE::TYPE (struct ev_loop *)
		3085
		3086	=item ev::TYPE::~TYPE
		3087
		3088	The constructor (optionally) takes an event loop to associate the watcher
		3089	with. If it is omitted, it will use C<EV_DEFAULT>.
		3090
		3091	The constructor calls C<ev_init> for you, which means you have to call the
		3092	C<set> method before starting it.
		3093
		3094	It will not set a callback, however: You have to call the templated C<set>
		3095	method to set a callback before you can start the watcher.
		3096
		3097	(The reason why you have to use a method is a limitation in C++ which does
		3098	not allow explicit template arguments for constructors).
		3099
		3100	The destructor automatically stops the watcher if it is active.
		3101
		3102	=item w->set<class, &class::method> (object *)
		3103
		3104	This method sets the callback method to call. The method has to have a
		3105	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		3106	first argument and the C<revents> as second. The object must be given as
		3107	parameter and is stored in the C<data> member of the watcher.
		3108
		3109	This method synthesizes efficient thunking code to call your method from
		3110	the C callback that libev requires. If your compiler can inline your
		3111	callback (i.e. it is visible to it at the place of the C<set> call and
		3112	your compiler is good :), then the method will be fully inlined into the
		3113	thunking function, making it as fast as a direct C callback.
		3114
		3115	Example: simple class declaration and watcher initialisation
		3116
		3117	struct myclass
		3118	{
		3119	void io_cb (ev::io &w, int revents) { }
		3120	}
		3121
		3122	myclass obj;
		3123	ev::io iow;
		3124	iow.set <myclass, &myclass::io_cb> (&obj);
		3125
		3126	=item w->set (object *)
		3127
		3128	This is an B<experimental> feature that might go away in a future version.
		3129
		3130	This is a variation of a method callback - leaving out the method to call
		3131	will default the method to C<operator ()>, which makes it possible to use
		3132	functor objects without having to manually specify the C<operator ()> all
		3133	the time. Incidentally, you can then also leave out the template argument
		3134	list.
		3135
		3136	The C<operator ()> method prototype must be C<void operator ()(watcher &w,
		3137	int revents)>.
		3138
		3139	See the method-C<set> above for more details.
		3140
		3141	Example: use a functor object as callback.
		3142
		3143	struct myfunctor
		3144	{
		3145	void operator() (ev::io &w, int revents)
		3146	{
		3147	...
		3148	}
		3149	}
		3150
		3151	myfunctor f;
		3152
		3153	ev::io w;
		3154	w.set (&f);
		3155
		3156	=item w->set<function> (void *data = 0)
		3157
		3158	Also sets a callback, but uses a static method or plain function as
		3159	callback. The optional C<data> argument will be stored in the watcher's
		3160	C<data> member and is free for you to use.
		3161
		3162	The prototype of the C<function> must be C<void (*)(ev::TYPE &w, int)>.
		3163
		3164	See the method-C<set> above for more details.
		3165
		3166	Example: Use a plain function as callback.
		3167
		3168	static void io_cb (ev::io &w, int revents) { }
		3169	iow.set <io_cb> ();
		3170
		3171	=item w->set (struct ev_loop *)
		3172
		3173	Associates a different C<struct ev_loop> with this watcher. You can only
		3174	do this when the watcher is inactive (and not pending either).
		3175
		3176	=item w->set ([arguments])
		3177
		3178	Basically the same as C<ev_TYPE_set>, with the same arguments. Must be
		3179	called at least once. Unlike the C counterpart, an active watcher gets
		3180	automatically stopped and restarted when reconfiguring it with this
		3181	method.
		3182
		3183	=item w->start ()
		3184
		3185	Starts the watcher. Note that there is no C<loop> argument, as the
		3186	constructor already stores the event loop.
		3187
		3188	=item w->stop ()
		3189
		3190	Stops the watcher if it is active. Again, no C<loop> argument.
		3191
		3192	=item w->again () (C<ev::timer>, C<ev::periodic> only)
		3193
		3194	For C<ev::timer> and C<ev::periodic>, this invokes the corresponding
		3195	C<ev_TYPE_again> function.
		3196
		3197	=item w->sweep () (C<ev::embed> only)
		3198
		3199	Invokes C<ev_embed_sweep>.
		3200
		3201	=item w->update () (C<ev::stat> only)
		3202
		3203	Invokes C<ev_stat_stat>.
		3204
		3205	=back
		3206
		3207	=back
		3208
		3209	Example: Define a class with an IO and idle watcher, start one of them in
		3210	the constructor.
		3211
		3212	class myclass
		3213	{
		3214	ev::io io ; void io_cb (ev::io &w, int revents);
		3215	ev::idle idle; void idle_cb (ev::idle &w, int revents);
		3216
		3217	myclass (int fd)
		3218	{
		3219	io .set <myclass, &myclass::io_cb > (this);
		3220	idle.set <myclass, &myclass::idle_cb> (this);
		3221
		3222	io.start (fd, ev::READ);
		3223	}
		3224	};
		3225
		3226
		3227	=head1 OTHER LANGUAGE BINDINGS
		3228
		3229	Libev does not offer other language bindings itself, but bindings for a
		3230	number of languages exist in the form of third-party packages. If you know
		3231	any interesting language binding in addition to the ones listed here, drop
		3232	me a note.
		3233
		3234	=over 4
		3235
		3236	=item Perl
		3237
		3238	The EV module implements the full libev API and is actually used to test
		3239	libev. EV is developed together with libev. Apart from the EV core module,
		3240	there are additional modules that implement libev-compatible interfaces
		3241	to C<libadns> (C<EV::ADNS>, but C<AnyEvent::DNS> is preferred nowadays),
		3242	C<Net::SNMP> (C<Net::SNMP::EV>) and the C<libglib> event core (C<Glib::EV>
		3243	and C<EV::Glib>).
		3244
		3245	It can be found and installed via CPAN, its homepage is at
		3246	L<http://software.schmorp.de/pkg/EV>.
		3247
		3248	=item Python
		3249
		3250	Python bindings can be found at L<http://code.google.com/p/pyev/>. It
		3251	seems to be quite complete and well-documented.
		3252
		3253	=item Ruby
		3254
		3255	Tony Arcieri has written a ruby extension that offers access to a subset
		3256	of the libev API and adds file handle abstractions, asynchronous DNS and
		3257	more on top of it. It can be found via gem servers. Its homepage is at
		3258	L<http://rev.rubyforge.org/>.
		3259
		3260	Roger Pack reports that using the link order C<-lws2_32 -lmsvcrt-ruby-190>
		3261	makes rev work even on mingw.
		3262
		3263	=item Haskell
		3264
		3265	A haskell binding to libev is available at
		3266	L<http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hlibev>.
		3267
		3268	=item D
		3269
		3270	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		3271	be found at L<http://proj.llucax.com.ar/wiki/evd>.
		3272
		3273	=item Ocaml
		3274
		3275	Erkki Seppala has written Ocaml bindings for libev, to be found at
		3276	L<http://modeemi.cs.tut.fi/~flux/software/ocaml-ev/>.
		3277
		3278	=back
		3279
		3280
		3281	=head1 MACRO MAGIC
		3282
		3283	Libev can be compiled with a variety of options, the most fundamental
		3284	of which is C<EV_MULTIPLICITY>. This option determines whether (most)
		3285	functions and callbacks have an initial C<struct ev_loop *> argument.
		3286
		3287	To make it easier to write programs that cope with either variant, the
		3288	following macros are defined:
		3289
		3290	=over 4
		3291
		3292	=item C<EV_A>, C<EV_A_>
		3293
		3294	This provides the loop I<argument> for functions, if one is required ("ev
		3295	loop argument"). The C<EV_A> form is used when this is the sole argument,
		3296	C<EV_A_> is used when other arguments are following. Example:
		3297
		3298	ev_unref (EV_A);
		3299	ev_timer_add (EV_A_ watcher);
		3300	ev_loop (EV_A_ 0);
		3301
		3302	It assumes the variable C<loop> of type C<struct ev_loop *> is in scope,
		3303	which is often provided by the following macro.
		3304
		3305	=item C<EV_P>, C<EV_P_>
		3306
		3307	This provides the loop I<parameter> for functions, if one is required ("ev
		3308	loop parameter"). The C<EV_P> form is used when this is the sole parameter,
		3309	C<EV_P_> is used when other parameters are following. Example:
		3310
		3311	// this is how ev_unref is being declared
		3312	static void ev_unref (EV_P);
		3313
		3314	// this is how you can declare your typical callback
		3315	static void cb (EV_P_ ev_timer *w, int revents)
		3316
		3317	It declares a parameter C<loop> of type C<struct ev_loop *>, quite
		3318	suitable for use with C<EV_A>.
		3319
		3320	=item C<EV_DEFAULT>, C<EV_DEFAULT_>
		3321
		3322	Similar to the other two macros, this gives you the value of the default
		3323	loop, if multiple loops are supported ("ev loop default").
		3324
		3325	=item C<EV_DEFAULT_UC>, C<EV_DEFAULT_UC_>
		3326
		3327	Usage identical to C<EV_DEFAULT> and C<EV_DEFAULT_>, but requires that the
		3328	default loop has been initialised (C<UC> == unchecked). Their behaviour
		3329	is undefined when the default loop has not been initialised by a previous
		3330	execution of C<EV_DEFAULT>, C<EV_DEFAULT_> or C<ev_default_init (...)>.
		3331
		3332	It is often prudent to use C<EV_DEFAULT> when initialising the first
		3333	watcher in a function but use C<EV_DEFAULT_UC> afterwards.
		3334
		3335	=back
		3336
		3337	Example: Declare and initialise a check watcher, utilising the above
		3338	macros so it will work regardless of whether multiple loops are supported
		3339	or not.
		3340
		3341	static void
		3342	check_cb (EV_P_ ev_timer *w, int revents)
		3343	{
		3344	ev_check_stop (EV_A_ w);
		3345	}
		3346
		3347	ev_check check;
		3348	ev_check_init (&check, check_cb);
		3349	ev_check_start (EV_DEFAULT_ &check);
		3350	ev_loop (EV_DEFAULT_ 0);
		3351
		3352	=head1 EMBEDDING
		3353
		3354	Libev can (and often is) directly embedded into host
		3355	applications. Examples of applications that embed it include the Deliantra
		3356	Game Server, the EV perl module, the GNU Virtual Private Ethernet (gvpe)
		3357	and rxvt-unicode.
		3358
		3359	The goal is to enable you to just copy the necessary files into your
		3360	source directory without having to change even a single line in them, so
		3361	you can easily upgrade by simply copying (or having a checked-out copy of
		3362	libev somewhere in your source tree).
		3363
		3364	=head2 FILESETS
		3365
		3366	Depending on what features you need you need to include one or more sets of files
		3367	in your application.
		3368
		3369	=head3 CORE EVENT LOOP
		3370
		3371	To include only the libev core (all the C<ev_*> functions), with manual
		3372	configuration (no autoconf):
		3373
		3374	#define EV_STANDALONE 1
		3375	#include "ev.c"
		3376
		3377	This will automatically include F<ev.h>, too, and should be done in a
		3378	single C source file only to provide the function implementations. To use
		3379	it, do the same for F<ev.h> in all files wishing to use this API (best
		3380	done by writing a wrapper around F<ev.h> that you can include instead and
		3381	where you can put other configuration options):
		3382
		3383	#define EV_STANDALONE 1
		3384	#include "ev.h"
		3385
		3386	Both header files and implementation files can be compiled with a C++
		3387	compiler (at least, that's a stated goal, and breakage will be treated
		3388	as a bug).
		3389
		3390	You need the following files in your source tree, or in a directory
		3391	in your include path (e.g. in libev/ when using -Ilibev):
		3392
		3393	ev.h
		3394	ev.c
		3395	ev_vars.h
		3396	ev_wrap.h
		3397
		3398	ev_win32.c required on win32 platforms only
		3399
		3400	ev_select.c only when select backend is enabled (which is enabled by default)
		3401	ev_poll.c only when poll backend is enabled (disabled by default)
		3402	ev_epoll.c only when the epoll backend is enabled (disabled by default)
		3403	ev_kqueue.c only when the kqueue backend is enabled (disabled by default)
		3404	ev_port.c only when the solaris port backend is enabled (disabled by default)
		3405
		3406	F<ev.c> includes the backend files directly when enabled, so you only need
		3407	to compile this single file.
		3408
		3409	=head3 LIBEVENT COMPATIBILITY API
		3410
		3411	To include the libevent compatibility API, also include:
		3412
		3413	#include "event.c"
		3414
		3415	in the file including F<ev.c>, and:
		3416
		3417	#include "event.h"
		3418
		3419	in the files that want to use the libevent API. This also includes F<ev.h>.
		3420
		3421	You need the following additional files for this:
		3422
		3423	event.h
		3424	event.c
		3425
		3426	=head3 AUTOCONF SUPPORT
		3427
		3428	Instead of using C<EV_STANDALONE=1> and providing your configuration in
		3429	whatever way you want, you can also C<m4_include([libev.m4])> in your
		3430	F<configure.ac> and leave C<EV_STANDALONE> undefined. F<ev.c> will then
		3431	include F<config.h> and configure itself accordingly.
		3432
		3433	For this of course you need the m4 file:
		3434
		3435	libev.m4
		3436
		3437	=head2 PREPROCESSOR SYMBOLS/MACROS
		3438
		3439	Libev can be configured via a variety of preprocessor symbols you have to
		3440	define before including any of its files. The default in the absence of
		3441	autoconf is documented for every option.
		3442
		3443	=over 4
		3444
		3445	=item EV_STANDALONE
		3446
		3447	Must always be C<1> if you do not use autoconf configuration, which
		3448	keeps libev from including F<config.h>, and it also defines dummy
		3449	implementations for some libevent functions (such as logging, which is not
		3450	supported). It will also not define any of the structs usually found in
		3451	F<event.h> that are not directly supported by the libev core alone.
		3452
		3453	In stanbdalone mode, libev will still try to automatically deduce the
		3454	configuration, but has to be more conservative.
		3455
		3456	=item EV_USE_MONOTONIC
		3457
		3458	If defined to be C<1>, libev will try to detect the availability of the
		3459	monotonic clock option at both compile time and runtime. Otherwise no
		3460	use of the monotonic clock option will be attempted. If you enable this,
		3461	you usually have to link against librt or something similar. Enabling it
		3462	when the functionality isn't available is safe, though, although you have
		3463	to make sure you link against any libraries where the C<clock_gettime>
		3464	function is hiding in (often F<-lrt>). See also C<EV_USE_CLOCK_SYSCALL>.
		3465
		3466	=item EV_USE_REALTIME
		3467
		3468	If defined to be C<1>, libev will try to detect the availability of the
		3469	real-time clock option at compile time (and assume its availability
		3470	at runtime if successful). Otherwise no use of the real-time clock
		3471	option will be attempted. This effectively replaces C<gettimeofday>
		3472	by C<clock_get (CLOCK_REALTIME, ...)> and will not normally affect
		3473	correctness. See the note about libraries in the description of
		3474	C<EV_USE_MONOTONIC>, though. Defaults to the opposite value of
		3475	C<EV_USE_CLOCK_SYSCALL>.
		3476
		3477	=item EV_USE_CLOCK_SYSCALL
		3478
		3479	If defined to be C<1>, libev will try to use a direct syscall instead
		3480	of calling the system-provided C<clock_gettime> function. This option
		3481	exists because on GNU/Linux, C<clock_gettime> is in C<librt>, but C<librt>
		3482	unconditionally pulls in C<libpthread>, slowing down single-threaded
		3483	programs needlessly. Using a direct syscall is slightly slower (in
		3484	theory), because no optimised vdso implementation can be used, but avoids
		3485	the pthread dependency. Defaults to C<1> on GNU/Linux with glibc 2.x or
		3486	higher, as it simplifies linking (no need for C<-lrt>).
		3487
		3488	=item EV_USE_NANOSLEEP
		3489
		3490	If defined to be C<1>, libev will assume that C<nanosleep ()> is available
		3491	and will use it for delays. Otherwise it will use C<select ()>.
		3492
		3493	=item EV_USE_EVENTFD
		3494
		3495	If defined to be C<1>, then libev will assume that C<eventfd ()> is
		3496	available and will probe for kernel support at runtime. This will improve
		3497	C<ev_signal> and C<ev_async> performance and reduce resource consumption.
		3498	If undefined, it will be enabled if the headers indicate GNU/Linux + Glibc
		3499	2.7 or newer, otherwise disabled.
		3500
		3501	=item EV_USE_SELECT
		3502
		3503	If undefined or defined to be C<1>, libev will compile in support for the
		3504	C<select>(2) backend. No attempt at auto-detection will be done: if no
		3505	other method takes over, select will be it. Otherwise the select backend
		3506	will not be compiled in.
		3507
		3508	=item EV_SELECT_USE_FD_SET
		3509
		3510	If defined to C<1>, then the select backend will use the system C<fd_set>
		3511	structure. This is useful if libev doesn't compile due to a missing
		3512	C<NFDBITS> or C<fd_mask> definition or it mis-guesses the bitset layout
		3513	on exotic systems. This usually limits the range of file descriptors to
		3514	some low limit such as 1024 or might have other limitations (winsocket
		3515	only allows 64 sockets). The C<FD_SETSIZE> macro, set before compilation,
		3516	configures the maximum size of the C<fd_set>.
		3517
		3518	=item EV_SELECT_IS_WINSOCKET
		3519
		3520	When defined to C<1>, the select backend will assume that
		3521	select/socket/connect etc. don't understand file descriptors but
		3522	wants osf handles on win32 (this is the case when the select to
		3523	be used is the winsock select). This means that it will call
		3524	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
		3525	it is assumed that all these functions actually work on fds, even
		3526	on win32. Should not be defined on non-win32 platforms.
		3527
		3528	=item EV_FD_TO_WIN32_HANDLE
		3529
		3530	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		3531	file descriptors to socket handles. When not defining this symbol (the
		3532	default), then libev will call C<_get_osfhandle>, which is usually
		3533	correct. In some cases, programs use their own file descriptor management,
		3534	in which case they can provide this function to map fds to socket handles.
		3535
		3536	=item EV_USE_POLL
		3537
		3538	If defined to be C<1>, libev will compile in support for the C<poll>(2)
		3539	backend. Otherwise it will be enabled on non-win32 platforms. It
		3540	takes precedence over select.
		3541
		3542	=item EV_USE_EPOLL
		3543
		3544	If defined to be C<1>, libev will compile in support for the Linux
		3545	C<epoll>(7) backend. Its availability will be detected at runtime,
		3546	otherwise another method will be used as fallback. This is the preferred
		3547	backend for GNU/Linux systems. If undefined, it will be enabled if the
		3548	headers indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		3549
		3550	=item EV_USE_KQUEUE
		3551
		3552	If defined to be C<1>, libev will compile in support for the BSD style
		3553	C<kqueue>(2) backend. Its actual availability will be detected at runtime,
		3554	otherwise another method will be used as fallback. This is the preferred
		3555	backend for BSD and BSD-like systems, although on most BSDs kqueue only
		3556	supports some types of fds correctly (the only platform we found that
		3557	supports ptys for example was NetBSD), so kqueue might be compiled in, but
		3558	not be used unless explicitly requested. The best way to use it is to find
		3559	out whether kqueue supports your type of fd properly and use an embedded
		3560	kqueue loop.
		3561
		3562	=item EV_USE_PORT
		3563
		3564	If defined to be C<1>, libev will compile in support for the Solaris
		3565	10 port style backend. Its availability will be detected at runtime,
		3566	otherwise another method will be used as fallback. This is the preferred
		3567	backend for Solaris 10 systems.
		3568
		3569	=item EV_USE_DEVPOLL
		3570
		3571	Reserved for future expansion, works like the USE symbols above.
		3572
		3573	=item EV_USE_INOTIFY
		3574
		3575	If defined to be C<1>, libev will compile in support for the Linux inotify
		3576	interface to speed up C<ev_stat> watchers. Its actual availability will
		3577	be detected at runtime. If undefined, it will be enabled if the headers
		3578	indicate GNU/Linux + Glibc 2.4 or newer, otherwise disabled.
		3579
		3580	=item EV_ATOMIC_T
		3581
		3582	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		3583	access is atomic with respect to other threads or signal contexts. No such
		3584	type is easily found in the C language, so you can provide your own type
		3585	that you know is safe for your purposes. It is used both for signal handler "locking"
		3586	as well as for signal and thread safety in C<ev_async> watchers.
		3587
		3588	In the absence of this define, libev will use C<sig_atomic_t volatile>
		3589	(from F<signal.h>), which is usually good enough on most platforms.
		3590
		3591	=item EV_H
		3592
		3593	The name of the F<ev.h> header file used to include it. The default if
		3594	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
		3595	used to virtually rename the F<ev.h> header file in case of conflicts.
		3596
		3597	=item EV_CONFIG_H
		3598
		3599	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
		3600	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
		3601	C<EV_H>, above.
		3602
		3603	=item EV_EVENT_H
		3604
		3605	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
		3606	of how the F<event.h> header can be found, the default is C<"event.h">.
		3607
		3608	=item EV_PROTOTYPES
		3609
		3610	If defined to be C<0>, then F<ev.h> will not define any function
		3611	prototypes, but still define all the structs and other symbols. This is
		3612	occasionally useful if you want to provide your own wrapper functions
		3613	around libev functions.
		3614
		3615	=item EV_MULTIPLICITY
		3616
		3617	If undefined or defined to C<1>, then all event-loop-specific functions
		3618	will have the C<struct ev_loop *> as first argument, and you can create
		3619	additional independent event loops. Otherwise there will be no support
		3620	for multiple event loops and there is no first event loop pointer
		3621	argument. Instead, all functions act on the single default loop.
		3622
		3623	=item EV_MINPRI
		3624
		3625	=item EV_MAXPRI
		3626
		3627	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		3628	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		3629	provide for more priorities by overriding those symbols (usually defined
		3630	to be C<-2> and C<2>, respectively).
		3631
		3632	When doing priority-based operations, libev usually has to linearly search
		3633	all the priorities, so having many of them (hundreds) uses a lot of space
		3634	and time, so using the defaults of five priorities (-2 .. +2) is usually
		3635	fine.
		3636
		3637	If your embedding application does not need any priorities, defining these
		3638	both to C<0> will save some memory and CPU.
		3639
		3640	=item EV_PERIODIC_ENABLE
		3641
		3642	If undefined or defined to be C<1>, then periodic timers are supported. If
		3643	defined to be C<0>, then they are not. Disabling them saves a few kB of
		3644	code.
		3645
		3646	=item EV_IDLE_ENABLE
		3647
		3648	If undefined or defined to be C<1>, then idle watchers are supported. If
		3649	defined to be C<0>, then they are not. Disabling them saves a few kB of
		3650	code.
		3651
		3652	=item EV_EMBED_ENABLE
		3653
		3654	If undefined or defined to be C<1>, then embed watchers are supported. If
		3655	defined to be C<0>, then they are not. Embed watchers rely on most other
		3656	watcher types, which therefore must not be disabled.
		3657
		3658	=item EV_STAT_ENABLE
		3659
		3660	If undefined or defined to be C<1>, then stat watchers are supported. If
		3661	defined to be C<0>, then they are not.
		3662
		3663	=item EV_FORK_ENABLE
		3664
		3665	If undefined or defined to be C<1>, then fork watchers are supported. If
		3666	defined to be C<0>, then they are not.
		3667
		3668	=item EV_ASYNC_ENABLE
		3669
		3670	If undefined or defined to be C<1>, then async watchers are supported. If
		3671	defined to be C<0>, then they are not.
		3672
		3673	=item EV_MINIMAL
		3674
		3675	If you need to shave off some kilobytes of code at the expense of some
		3676	speed, define this symbol to C<1>. Currently this is used to override some
		3677	inlining decisions, saves roughly 30% code size on amd64. It also selects a
		3678	much smaller 2-heap for timer management over the default 4-heap.
		3679
		3680	=item EV_PID_HASHSIZE
		3681
		3682	C<ev_child> watchers use a small hash table to distribute workload by
		3683	pid. The default size is C<16> (or C<1> with C<EV_MINIMAL>), usually more
		3684	than enough. If you need to manage thousands of children you might want to
		3685	increase this value (I<must> be a power of two).
		3686
		3687	=item EV_INOTIFY_HASHSIZE
		3688
		3689	C<ev_stat> watchers use a small hash table to distribute workload by
		3690	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
		3691	usually more than enough. If you need to manage thousands of C<ev_stat>
		3692	watchers you might want to increase this value (I<must> be a power of
		3693	two).
		3694
		3695	=item EV_USE_4HEAP
		3696
		3697	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3698	timer and periodics heaps, libev uses a 4-heap when this symbol is defined
		3699	to C<1>. The 4-heap uses more complicated (longer) code but has noticeably
		3700	faster performance with many (thousands) of watchers.
		3701
		3702	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3703	(disabled).
		3704
		3705	=item EV_HEAP_CACHE_AT
		3706
		3707	Heaps are not very cache-efficient. To improve the cache-efficiency of the
		3708	timer and periodics heaps, libev can cache the timestamp (I<at>) within
		3709	the heap structure (selected by defining C<EV_HEAP_CACHE_AT> to C<1>),
		3710	which uses 8-12 bytes more per watcher and a few hundred bytes more code,
		3711	but avoids random read accesses on heap changes. This improves performance
		3712	noticeably with many (hundreds) of watchers.
		3713
		3714	The default is C<1> unless C<EV_MINIMAL> is set in which case it is C<0>
		3715	(disabled).
		3716
		3717	=item EV_VERIFY
		3718
		3719	Controls how much internal verification (see C<ev_loop_verify ()>) will
		3720	be done: If set to C<0>, no internal verification code will be compiled
		3721	in. If set to C<1>, then verification code will be compiled in, but not
		3722	called. If set to C<2>, then the internal verification code will be
		3723	called once per loop, which can slow down libev. If set to C<3>, then the
		3724	verification code will be called very frequently, which will slow down
		3725	libev considerably.
		3726
		3727	The default is C<1>, unless C<EV_MINIMAL> is set, in which case it will be
		3728	C<0>.
		3729
		3730	=item EV_COMMON
		3731
		3732	By default, all watchers have a C<void *data> member. By redefining
		3733	this macro to a something else you can include more and other types of
		3734	members. You have to define it each time you include one of the files,
		3735	though, and it must be identical each time.
		3736
		3737	For example, the perl EV module uses something like this:
		3738
		3739	#define EV_COMMON \
		3740	SV *self; /* contains this struct */ \
		3741	SV *cb_sv, *fh /* note no trailing ";" */
		3742
		3743	=item EV_CB_DECLARE (type)
		3744
		3745	=item EV_CB_INVOKE (watcher, revents)
		3746
		3747	=item ev_set_cb (ev, cb)
		3748
		3749	Can be used to change the callback member declaration in each watcher,
		3750	and the way callbacks are invoked and set. Must expand to a struct member
		3751	definition and a statement, respectively. See the F<ev.h> header file for
		3752	their default definitions. One possible use for overriding these is to
		3753	avoid the C<struct ev_loop *> as first argument in all cases, or to use
		3754	method calls instead of plain function calls in C++.
		3755
		3756	=back
		3757
		3758	=head2 EXPORTED API SYMBOLS
		3759
		3760	If you need to re-export the API (e.g. via a DLL) and you need a list of
		3761	exported symbols, you can use the provided F<Symbol.*> files which list
		3762	all public symbols, one per line:
		3763
		3764	Symbols.ev for libev proper
		3765	Symbols.event for the libevent emulation
		3766
		3767	This can also be used to rename all public symbols to avoid clashes with
		3768	multiple versions of libev linked together (which is obviously bad in
		3769	itself, but sometimes it is inconvenient to avoid this).
		3770
		3771	A sed command like this will create wrapper C<#define>'s that you need to
		3772	include before including F<ev.h>:
		3773
		3774	<Symbols.ev sed -e "s/.*/#define & myprefix_&/" >wrap.h
		3775
		3776	This would create a file F<wrap.h> which essentially looks like this:
		3777
		3778	#define ev_backend myprefix_ev_backend
		3779	#define ev_check_start myprefix_ev_check_start
		3780	#define ev_check_stop myprefix_ev_check_stop
		3781	...
		3782
		3783	=head2 EXAMPLES
		3784
		3785	For a real-world example of a program the includes libev
		3786	verbatim, you can have a look at the EV perl module
		3787	(L<http://software.schmorp.de/pkg/EV.html>). It has the libev files in
		3788	the F<libev/> subdirectory and includes them in the F<EV/EVAPI.h> (public
		3789	interface) and F<EV.xs> (implementation) files. Only the F<EV.xs> file
		3790	will be compiled. It is pretty complex because it provides its own header
		3791	file.
		3792
		3793	The usage in rxvt-unicode is simpler. It has a F<ev_cpp.h> header file
		3794	that everybody includes and which overrides some configure choices:
		3795
		3796	#define EV_MINIMAL 1
		3797	#define EV_USE_POLL 0
		3798	#define EV_MULTIPLICITY 0
		3799	#define EV_PERIODIC_ENABLE 0
		3800	#define EV_STAT_ENABLE 0
		3801	#define EV_FORK_ENABLE 0
		3802	#define EV_CONFIG_H <config.h>
		3803	#define EV_MINPRI 0
		3804	#define EV_MAXPRI 0
		3805
		3806	#include "ev++.h"
		3807
		3808	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
		3809
		3810	#include "ev_cpp.h"
		3811	#include "ev.c"
		3812
		3813	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
		3814
		3815	=head2 THREADS AND COROUTINES
		3816
		3817	=head3 THREADS
		3818
		3819	All libev functions are reentrant and thread-safe unless explicitly
		3820	documented otherwise, but libev implements no locking itself. This means
		3821	that you can use as many loops as you want in parallel, as long as there
		3822	are no concurrent calls into any libev function with the same loop
		3823	parameter (C<ev_default_*> calls have an implicit default loop parameter,
		3824	of course): libev guarantees that different event loops share no data
		3825	structures that need any locking.
		3826
		3827	Or to put it differently: calls with different loop parameters can be done
		3828	concurrently from multiple threads, calls with the same loop parameter
		3829	must be done serially (but can be done from different threads, as long as
		3830	only one thread ever is inside a call at any point in time, e.g. by using
		3831	a mutex per loop).
		3832
		3833	Specifically to support threads (and signal handlers), libev implements
		3834	so-called C<ev_async> watchers, which allow some limited form of
		3835	concurrency on the same event loop, namely waking it up "from the
		3836	outside".
		3837
		3838	If you want to know which design (one loop, locking, or multiple loops
		3839	without or something else still) is best for your problem, then I cannot
		3840	help you, but here is some generic advice:
		3841
		3842	=over 4
		3843
		3844	=item * most applications have a main thread: use the default libev loop
		3845	in that thread, or create a separate thread running only the default loop.
		3846
		3847	This helps integrating other libraries or software modules that use libev
		3848	themselves and don't care/know about threading.
		3849
		3850	=item * one loop per thread is usually a good model.
		3851
		3852	Doing this is almost never wrong, sometimes a better-performance model
		3853	exists, but it is always a good start.
		3854
		3855	=item * other models exist, such as the leader/follower pattern, where one
		3856	loop is handed through multiple threads in a kind of round-robin fashion.
		3857
		3858	Choosing a model is hard - look around, learn, know that usually you can do
		3859	better than you currently do :-)
		3860
		3861	=item * often you need to talk to some other thread which blocks in the
		3862	event loop.
		3863
		3864	C<ev_async> watchers can be used to wake them up from other threads safely
		3865	(or from signal contexts...).
		3866
		3867	An example use would be to communicate signals or other events that only
		3868	work in the default loop by registering the signal watcher with the
		3869	default loop and triggering an C<ev_async> watcher from the default loop
		3870	watcher callback into the event loop interested in the signal.
		3871
		3872	=back
		3873
		3874	=head3 COROUTINES
		3875
		3876	Libev is very accommodating to coroutines ("cooperative threads"):
		3877	libev fully supports nesting calls to its functions from different
		3878	coroutines (e.g. you can call C<ev_loop> on the same loop from two
		3879	different coroutines, and switch freely between both coroutines running the
		3880	loop, as long as you don't confuse yourself). The only exception is that
		3881	you must not do this from C<ev_periodic> reschedule callbacks.
		3882
		3883	Care has been taken to ensure that libev does not keep local state inside
		3884	C<ev_loop>, and other calls do not usually allow for coroutine switches as
		3885	they do not call any callbacks.
		3886
		3887	=head2 COMPILER WARNINGS
		3888
		3889	Depending on your compiler and compiler settings, you might get no or a
		3890	lot of warnings when compiling libev code. Some people are apparently
		3891	scared by this.
		3892
		3893	However, these are unavoidable for many reasons. For one, each compiler
		3894	has different warnings, and each user has different tastes regarding
		3895	warning options. "Warn-free" code therefore cannot be a goal except when
		3896	targeting a specific compiler and compiler-version.
		3897
		3898	Another reason is that some compiler warnings require elaborate
		3899	workarounds, or other changes to the code that make it less clear and less
		3900	maintainable.
		3901
		3902	And of course, some compiler warnings are just plain stupid, or simply
		3903	wrong (because they don't actually warn about the condition their message
		3904	seems to warn about). For example, certain older gcc versions had some
		3905	warnings that resulted an extreme number of false positives. These have
		3906	been fixed, but some people still insist on making code warn-free with
		3907	such buggy versions.
		3908
		3909	While libev is written to generate as few warnings as possible,
		3910	"warn-free" code is not a goal, and it is recommended not to build libev
		3911	with any compiler warnings enabled unless you are prepared to cope with
		3912	them (e.g. by ignoring them). Remember that warnings are just that:
		3913	warnings, not errors, or proof of bugs.
		3914
		3915
		3916	=head2 VALGRIND
		3917
		3918	Valgrind has a special section here because it is a popular tool that is
		3919	highly useful. Unfortunately, valgrind reports are very hard to interpret.
		3920
		3921	If you think you found a bug (memory leak, uninitialised data access etc.)
		3922	in libev, then check twice: If valgrind reports something like:
		3923
		3924	==2274== definitely lost: 0 bytes in 0 blocks.
		3925	==2274== possibly lost: 0 bytes in 0 blocks.
		3926	==2274== still reachable: 256 bytes in 1 blocks.
		3927
		3928	Then there is no memory leak, just as memory accounted to global variables
		3929	is not a memleak - the memory is still being referenced, and didn't leak.
		3930
		3931	Similarly, under some circumstances, valgrind might report kernel bugs
		3932	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		3933	although an acceptable workaround has been found here), or it might be
		3934	confused.
		3935
		3936	Keep in mind that valgrind is a very good tool, but only a tool. Don't
		3937	make it into some kind of religion.
		3938
		3939	If you are unsure about something, feel free to contact the mailing list
		3940	with the full valgrind report and an explanation on why you think this
		3941	is a bug in libev (best check the archives, too :). However, don't be
		3942	annoyed when you get a brisk "this is no bug" answer and take the chance
		3943	of learning how to interpret valgrind properly.
		3944
		3945	If you need, for some reason, empty reports from valgrind for your project
		3946	I suggest using suppression lists.
		3947
		3948
		3949	=head1 PORTABILITY NOTES
		3950
		3951	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
		3952
		3953	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		3954	requires, and its I/O model is fundamentally incompatible with the POSIX
		3955	model. Libev still offers limited functionality on this platform in
		3956	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		3957	descriptors. This only applies when using Win32 natively, not when using
		3958	e.g. cygwin.
		3959
		3960	Lifting these limitations would basically require the full
		3961	re-implementation of the I/O system. If you are into these kinds of
		3962	things, then note that glib does exactly that for you in a very portable
		3963	way (note also that glib is the slowest event library known to man).
		3964
		3965	There is no supported compilation method available on windows except
		3966	embedding it into other applications.
		3967
		3968	Sensible signal handling is officially unsupported by Microsoft - libev
		3969	tries its best, but under most conditions, signals will simply not work.
		3970
		3971	Not a libev limitation but worth mentioning: windows apparently doesn't
		3972	accept large writes: instead of resulting in a partial write, windows will
		3973	either accept everything or return C<ENOBUFS> if the buffer is too large,
		3974	so make sure you only write small amounts into your sockets (less than a
		3975	megabyte seems safe, but this apparently depends on the amount of memory
		3976	available).
		3977
		3978	Due to the many, low, and arbitrary limits on the win32 platform and
		3979	the abysmal performance of winsockets, using a large number of sockets
		3980	is not recommended (and not reasonable). If your program needs to use
		3981	more than a hundred or so sockets, then likely it needs to use a totally
		3982	different implementation for windows, as libev offers the POSIX readiness
		3983	notification model, which cannot be implemented efficiently on windows
		3984	(due to Microsoft monopoly games).
		3985
		3986	A typical way to use libev under windows is to embed it (see the embedding
		3987	section for details) and use the following F<evwrap.h> header file instead
		3988	of F<ev.h>:
		3989
		3990	#define EV_STANDALONE /* keeps ev from requiring config.h */
		3991	#define EV_SELECT_IS_WINSOCKET 1 /* configure libev for windows select */
		3992
		3993	#include "ev.h"
		3994
		3995	And compile the following F<evwrap.c> file into your project (make sure
		3996	you do I<not> compile the F<ev.c> or any other embedded source files!):
		3997
		3998	#include "evwrap.h"
		3999	#include "ev.c"
		4000
		4001	=over 4
		4002
		4003	=item The winsocket select function
		4004
		4005	The winsocket C<select> function doesn't follow POSIX in that it
		4006	requires socket I<handles> and not socket I<file descriptors> (it is
		4007	also extremely buggy). This makes select very inefficient, and also
		4008	requires a mapping from file descriptors to socket handles (the Microsoft
		4009	C runtime provides the function C<_open_osfhandle> for this). See the
		4010	discussion of the C<EV_SELECT_USE_FD_SET>, C<EV_SELECT_IS_WINSOCKET> and
		4011	C<EV_FD_TO_WIN32_HANDLE> preprocessor symbols for more info.
		4012
		4013	The configuration for a "naked" win32 using the Microsoft runtime
		4014	libraries and raw winsocket select is:
		4015
		4016	#define EV_USE_SELECT 1
		4017	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		4018
		4019	Note that winsockets handling of fd sets is O(n), so you can easily get a
		4020	complexity in the O(n²) range when using win32.
		4021
		4022	=item Limited number of file descriptors
		4023
		4024	Windows has numerous arbitrary (and low) limits on things.
		4025
		4026	Early versions of winsocket's select only supported waiting for a maximum
		4027	of C<64> handles (probably owning to the fact that all windows kernels
		4028	can only wait for C<64> things at the same time internally; Microsoft
		4029	recommends spawning a chain of threads and wait for 63 handles and the
		4030	previous thread in each. Sounds great!).
		4031
		4032	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		4033	to some high number (e.g. C<2048>) before compiling the winsocket select
		4034	call (which might be in libev or elsewhere, for example, perl and many
		4035	other interpreters do their own select emulation on windows).
		4036
		4037	Another limit is the number of file descriptors in the Microsoft runtime
		4038	libraries, which by default is C<64> (there must be a hidden I<64>
		4039	fetish or something like this inside Microsoft). You can increase this
		4040	by calling C<_setmaxstdio>, which can increase this limit to C<2048>
		4041	(another arbitrary limit), but is broken in many versions of the Microsoft
		4042	runtime libraries. This might get you to about C<512> or C<2048> sockets
		4043	(depending on windows version and/or the phase of the moon). To get more,
		4044	you need to wrap all I/O functions and provide your own fd management, but
		4045	the cost of calling select (O(n²)) will likely make this unworkable.
		4046
		4047	=back
		4048
		4049	=head2 PORTABILITY REQUIREMENTS
		4050
		4051	In addition to a working ISO-C implementation and of course the
		4052	backend-specific APIs, libev relies on a few additional extensions:
		4053
		4054	=over 4
		4055
		4056	=item C<void (*)(ev_watcher_type *, int revents)> must have compatible
		4057	calling conventions regardless of C<ev_watcher_type *>.
		4058
		4059	Libev assumes not only that all watcher pointers have the same internal
		4060	structure (guaranteed by POSIX but not by ISO C for example), but it also
		4061	assumes that the same (machine) code can be used to call any watcher
		4062	callback: The watcher callbacks have different type signatures, but libev
		4063	calls them using an C<ev_watcher *> internally.
		4064
		4065	=item C<sig_atomic_t volatile> must be thread-atomic as well
		4066
		4067	The type C<sig_atomic_t volatile> (or whatever is defined as
		4068	C<EV_ATOMIC_T>) must be atomic with respect to accesses from different
		4069	threads. This is not part of the specification for C<sig_atomic_t>, but is
		4070	believed to be sufficiently portable.
		4071
		4072	=item C<sigprocmask> must work in a threaded environment
		4073
		4074	Libev uses C<sigprocmask> to temporarily block signals. This is not
		4075	allowed in a threaded program (C<pthread_sigmask> has to be used). Typical
		4076	pthread implementations will either allow C<sigprocmask> in the "main
		4077	thread" or will block signals process-wide, both behaviours would
		4078	be compatible with libev. Interaction between C<sigprocmask> and
		4079	C<pthread_sigmask> could complicate things, however.
		4080
		4081	The most portable way to handle signals is to block signals in all threads
		4082	except the initial one, and run the default loop in the initial thread as
		4083	well.
		4084
		4085	=item C<long> must be large enough for common memory allocation sizes
		4086
		4087	To improve portability and simplify its API, libev uses C<long> internally
		4088	instead of C<size_t> when allocating its data structures. On non-POSIX
		4089	systems (Microsoft...) this might be unexpectedly low, but is still at
		4090	least 31 bits everywhere, which is enough for hundreds of millions of
		4091	watchers.
		4092
		4093	=item C<double> must hold a time value in seconds with enough accuracy
		4094
		4095	The type C<double> is used to represent timestamps. It is required to
		4096	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
		4097	enough for at least into the year 4000. This requirement is fulfilled by
		4098	implementations implementing IEEE 754, which is basically all existing
		4099	ones. With IEEE 754 doubles, you get microsecond accuracy until at least
		4100	2200.
		4101
		4102	=back
		4103
		4104	If you know of other additional requirements drop me a note.
		4105
		4106
		4107	=head1 ALGORITHMIC COMPLEXITIES
		4108
		4109	In this section the complexities of (many of) the algorithms used inside
		4110	libev will be documented. For complexity discussions about backends see
		4111	the documentation for C<ev_default_init>.
		4112
		4113	All of the following are about amortised time: If an array needs to be
		4114	extended, libev needs to realloc and move the whole array, but this
		4115	happens asymptotically rarer with higher number of elements, so O(1) might
		4116	mean that libev does a lengthy realloc operation in rare cases, but on
		4117	average it is much faster and asymptotically approaches constant time.
		4118
		4119	=over 4
		4120
		4121	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
		4122
		4123	This means that, when you have a watcher that triggers in one hour and
		4124	there are 100 watchers that would trigger before that, then inserting will
		4125	have to skip roughly seven (C<ld 100>) of these watchers.
		4126
		4127	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
		4128
		4129	That means that changing a timer costs less than removing/adding them,
		4130	as only the relative motion in the event queue has to be paid for.
		4131
		4132	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
		4133
		4134	These just add the watcher into an array or at the head of a list.
		4135
		4136	=item Stopping check/prepare/idle/fork/async watchers: O(1)
		4137
		4138	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
		4139
		4140	These watchers are stored in lists, so they need to be walked to find the
		4141	correct watcher to remove. The lists are usually short (you don't usually
		4142	have many watchers waiting for the same fd or signal: one is typical, two
		4143	is rare).
		4144
		4145	=item Finding the next timer in each loop iteration: O(1)
		4146
		4147	By virtue of using a binary or 4-heap, the next timer is always found at a
		4148	fixed position in the storage array.
		4149
		4150	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		4151
		4152	A change means an I/O watcher gets started or stopped, which requires
		4153	libev to recalculate its status (and possibly tell the kernel, depending
		4154	on backend and whether C<ev_io_set> was used).
		4155
		4156	=item Activating one watcher (putting it into the pending state): O(1)
		4157
		4158	=item Priority handling: O(number_of_priorities)
		4159
		4160	Priorities are implemented by allocating some space for each
		4161	priority. When doing priority-based operations, libev usually has to
		4162	linearly search all the priorities, but starting/stopping and activating
		4163	watchers becomes O(1) with respect to priority handling.
		4164
		4165	=item Sending an ev_async: O(1)
		4166
		4167	=item Processing ev_async_send: O(number_of_async_watchers)
		4168
		4169	=item Processing signals: O(max_signal_number)
		4170
		4171	Sending involves a system call I<iff> there were no other C<ev_async_send>
		4172	calls in the current loop iteration. Checking for async and signal events
		4173	involves iterating over all running async watchers or all signal numbers.
		4174
		4175	=back
		4176
		4177
		4178	=head1 GLOSSARY
		4179
		4180	=over 4
		4181
		4182	=item active
		4183
		4184	A watcher is active as long as it has been started (has been attached to
		4185	an event loop) but not yet stopped (disassociated from the event loop).
		4186
		4187	=item application
		4188
		4189	In this document, an application is whatever is using libev.
		4190
		4191	=item callback
		4192
		4193	The address of a function that is called when some event has been
		4194	detected. Callbacks are being passed the event loop, the watcher that
		4195	received the event, and the actual event bitset.
		4196
		4197	=item callback invocation
		4198
		4199	The act of calling the callback associated with a watcher.
		4200
		4201	=item event
		4202
		4203	A change of state of some external event, such as data now being available
		4204	for reading on a file descriptor, time having passed or simply not having
		4205	any other events happening anymore.
		4206
		4207	In libev, events are represented as single bits (such as C<EV_READ> or
		4208	C<EV_TIMEOUT>).
		4209
		4210	=item event library
		4211
		4212	A software package implementing an event model and loop.
		4213
		4214	=item event loop
		4215
		4216	An entity that handles and processes external events and converts them
		4217	into callback invocations.
		4218
		4219	=item event model
		4220
		4221	The model used to describe how an event loop handles and processes
		4222	watchers and events.
		4223
		4224	=item pending
		4225
		4226	A watcher is pending as soon as the corresponding event has been detected,
		4227	and stops being pending as soon as the watcher will be invoked or its
		4228	pending status is explicitly cleared by the application.
		4229
		4230	A watcher can be pending, but not active. Stopping a watcher also clears
		4231	its pending status.
		4232
		4233	=item real time
		4234
		4235	The physical time that is observed. It is apparently strictly monotonic :)
		4236
		4237	=item wall-clock time
		4238
		4239	The time and date as shown on clocks. Unlike real time, it can actually
		4240	be wrong and jump forwards and backwards, e.g. when the you adjust your
		4241	clock.
		4242
		4243	=item watcher
		4244
		4245	A data structure that describes interest in certain events. Watchers need
		4246	to be started (attached to an event loop) before they can receive events.
		4247
		4248	=item watcher invocation
		4249
		4250	The act of calling the callback associated with a watcher.
		4251
		4252	=back
878		4253
879	=head1 AUTHOR	4254	=head1 AUTHOR
880		4255
881	Marc Lehmann <libev@schmorp.de>.	4256	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael Magnusson.
882		4257

Diff Legend

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