[ViewVC] Diff of: cvs/libev/ev.pod

Comparing libev/ev.pod (file contents):
Revision 1.99 by root, Sat Dec 22 06:16:36 2007 UTC vs.
Revision 1.137 by root, Sun Mar 16 16:42:56 2008 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	#include <ev.h>	7	#include <ev.h>
8		8
9	=head1 EXAMPLE PROGRAM	9	=head2 EXAMPLE PROGRAM
10		10
		11	// a single header file is required
11	#include <ev.h>	12	#include <ev.h>
12		13
		14	// every watcher type has its own typedef'd struct
		15	// with the name ev_<type>
13	ev_io stdin_watcher;	16	ev_io stdin_watcher;
14	ev_timer timeout_watcher;	17	ev_timer timeout_watcher;
15		18
		19	// all watcher callbacks have a similar signature
16	/* called when data readable on stdin */	20	// this callback is called when data is readable on stdin
17	static void	21	static void
18	stdin_cb (EV_P_ struct ev_io *w, int revents)	22	stdin_cb (EV_P_ struct ev_io *w, int revents)
19	{	23	{
20	/* puts ("stdin ready"); */	24	puts ("stdin ready");
21	ev_io_stop (EV_A_ w); /* just a syntax example */	25	// for one-shot events, one must manually stop the watcher
22	ev_unloop (EV_A_ EVUNLOOP_ALL); /* leave all loop calls */	26	// with its corresponding stop function.
		27	ev_io_stop (EV_A_ w);
		28
		29	// this causes all nested ev_loop's to stop iterating
		30	ev_unloop (EV_A_ EVUNLOOP_ALL);
23	}	31	}
24		32
		33	// another callback, this time for a time-out
25	static void	34	static void
26	timeout_cb (EV_P_ struct ev_timer *w, int revents)	35	timeout_cb (EV_P_ struct ev_timer *w, int revents)
27	{	36	{
28	/* puts ("timeout"); */	37	puts ("timeout");
29	ev_unloop (EV_A_ EVUNLOOP_ONE); /* leave one loop call */	38	// this causes the innermost ev_loop to stop iterating
		39	ev_unloop (EV_A_ EVUNLOOP_ONE);
30	}	40	}
31		41
32	int	42	int
33	main (void)	43	main (void)
34	{	44	{
		45	// use the default event loop unless you have special needs
35	struct ev_loop *loop = ev_default_loop (0);	46	struct ev_loop *loop = ev_default_loop (0);
36		47
37	/* initialise an io watcher, then start it */	48	// initialise an io watcher, then start it
		49	// this one will watch for stdin to become readable
38	ev_io_init (&stdin_watcher, stdin_cb, /STDIN_FILENO/ 0, EV_READ);	50	ev_io_init (&stdin_watcher, stdin_cb, /STDIN_FILENO/ 0, EV_READ);
39	ev_io_start (loop, &stdin_watcher);	51	ev_io_start (loop, &stdin_watcher);
40		52
		53	// initialise a timer watcher, then start it
41	/* simple non-repeating 5.5 second timeout */	54	// simple non-repeating 5.5 second timeout
42	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);	55	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
43	ev_timer_start (loop, &timeout_watcher);	56	ev_timer_start (loop, &timeout_watcher);
44		57
45	/* loop till timeout or data ready */	58	// now wait for events to arrive
46	ev_loop (loop, 0);	59	ev_loop (loop, 0);
47		60
		61	// unloop was called, so exit
48	return 0;	62	return 0;
49	}	63	}
50		64
51	=head1 DESCRIPTION	65	=head1 DESCRIPTION
52		66
53	The newest version of this document is also available as a html-formatted	67	The newest version of this document is also available as an html-formatted
54	web page you might find easier to navigate when reading it for the first	68	web page you might find easier to navigate when reading it for the first
55	time: L<http://cvs.schmorp.de/libev/ev.html>.	69	time: L<http://cvs.schmorp.de/libev/ev.html>.
56		70
57	Libev is an event loop: you register interest in certain events (such as a	71	Libev is an event loop: you register interest in certain events (such as a
58	file descriptor being readable or a timeout occurring), and it will manage	72	file descriptor being readable or a timeout occurring), and it will manage
…		…
65	You register interest in certain events by registering so-called I<event	79	You register interest in certain events by registering so-called I<event
66	watchers>, which are relatively small C structures you initialise with the	80	watchers>, which are relatively small C structures you initialise with the
67	details of the event, and then hand it over to libev by I<starting> the	81	details of the event, and then hand it over to libev by I<starting> the
68	watcher.	82	watcher.
69		83
70	=head1 FEATURES	84	=head2 FEATURES
71		85
72	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the	86	Libev supports C<select>, C<poll>, the Linux-specific C<epoll>, the
73	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms	87	BSD-specific C<kqueue> and the Solaris-specific event port mechanisms
74	for file descriptor events (C<ev_io>), the Linux C<inotify> interface	88	for file descriptor events (C<ev_io>), the Linux C<inotify> interface
75	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers	89	(for C<ev_stat>), relative timers (C<ev_timer>), absolute timers
…		…
82		96
83	It also is quite fast (see this	97	It also is quite fast (see this
84	L<benchmark\|http://libev.schmorp.de/bench.html> comparing it to libevent	98	L<benchmark\|http://libev.schmorp.de/bench.html> comparing it to libevent
85	for example).	99	for example).
86		100
87	=head1 CONVENTIONS	101	=head2 CONVENTIONS
88		102
89	Libev is very configurable. In this manual the default configuration will	103	Libev is very configurable. In this manual the default (and most common)
90	be described, which supports multiple event loops. For more info about	104	configuration will be described, which supports multiple event loops. For
91	various configuration options please have a look at B<EMBED> section in	105	more info about various configuration options please have a look at
92	this manual. If libev was configured without support for multiple event	106	B<EMBED> section in this manual. If libev was configured without support
93	loops, then all functions taking an initial argument of name C<loop>	107	for multiple event loops, then all functions taking an initial argument of
94	(which is always of type C<struct ev_loop *>) will not have this argument.	108	name C<loop> (which is always of type C<struct ev_loop *>) will not have
		109	this argument.
95		110
96	=head1 TIME REPRESENTATION	111	=head2 TIME REPRESENTATION
97		112
98	Libev represents time as a single floating point number, representing the	113	Libev represents time as a single floating point number, representing the
99	(fractional) number of seconds since the (POSIX) epoch (somewhere near	114	(fractional) number of seconds since the (POSIX) epoch (somewhere near
100	the beginning of 1970, details are complicated, don't ask). This type is	115	the beginning of 1970, details are complicated, don't ask). This type is
101	called C<ev_tstamp>, which is what you should use too. It usually aliases	116	called C<ev_tstamp>, which is what you should use too. It usually aliases
…		…
260	flags. If that is troubling you, check C<ev_backend ()> afterwards).	275	flags. If that is troubling you, check C<ev_backend ()> afterwards).
261		276
262	If you don't know what event loop to use, use the one returned from this	277	If you don't know what event loop to use, use the one returned from this
263	function.	278	function.
264		279
		280	The default loop is the only loop that can handle C<ev_signal> and
		281	C<ev_child> watchers, and to do this, it always registers a handler
		282	for C<SIGCHLD>. If this is a problem for your app you can either
		283	create a dynamic loop with C<ev_loop_new> that doesn't do that, or you
		284	can simply overwrite the C<SIGCHLD> signal handler I<after> calling
		285	C<ev_default_init>.
		286
265	The flags argument can be used to specify special behaviour or specific	287	The flags argument can be used to specify special behaviour or specific
266	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).	288	backends to use, and is usually specified as C<0> (or C<EVFLAG_AUTO>).
267		289
268	The following flags are supported:	290	The following flags are supported:
269		291
…		…
290	enabling this flag.	312	enabling this flag.
291		313
292	This works by calling C<getpid ()> on every iteration of the loop,	314	This works by calling C<getpid ()> on every iteration of the loop,
293	and thus this might slow down your event loop if you do a lot of loop	315	and thus this might slow down your event loop if you do a lot of loop
294	iterations and little real work, but is usually not noticeable (on my	316	iterations and little real work, but is usually not noticeable (on my
295	Linux system for example, C<getpid> is actually a simple 5-insn sequence	317	GNU/Linux system for example, C<getpid> is actually a simple 5-insn sequence
296	without a syscall and thus I<very> fast, but my Linux system also has	318	without a syscall and thus I<very> fast, but my GNU/Linux system also has
297	C<pthread_atfork> which is even faster).	319	C<pthread_atfork> which is even faster).
298		320
299	The big advantage of this flag is that you can forget about fork (and	321	The big advantage of this flag is that you can forget about fork (and
300	forget about forgetting to tell libev about forking) when you use this	322	forget about forgetting to tell libev about forking) when you use this
301	flag.	323	flag.
…		…
306	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	328	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
307		329
308	This is your standard select(2) backend. Not I<completely> standard, as	330	This is your standard select(2) backend. Not I<completely> standard, as
309	libev tries to roll its own fd_set with no limits on the number of fds,	331	libev tries to roll its own fd_set with no limits on the number of fds,
310	but if that fails, expect a fairly low limit on the number of fds when	332	but if that fails, expect a fairly low limit on the number of fds when
311	using this backend. It doesn't scale too well (O(highest_fd)), but its usually	333	using this backend. It doesn't scale too well (O(highest_fd)), but its
312	the fastest backend for a low number of fds.	334	usually the fastest backend for a low number of (low-numbered :) fds.
		335
		336	To get good performance out of this backend you need a high amount of
		337	parallelity (most of the file descriptors should be busy). If you are
		338	writing a server, you should C<accept ()> in a loop to accept as many
		339	connections as possible during one iteration. You might also want to have
		340	a look at C<ev_set_io_collect_interval ()> to increase the amount of
		341	readyness notifications you get per iteration.
313		342
314	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)	343	=item C<EVBACKEND_POLL> (value 2, poll backend, available everywhere except on windows)
315		344
316	And this is your standard poll(2) backend. It's more complicated than	345	And this is your standard poll(2) backend. It's more complicated
317	select, but handles sparse fds better and has no artificial limit on the	346	than select, but handles sparse fds better and has no artificial
318	number of fds you can use (except it will slow down considerably with a	347	limit on the number of fds you can use (except it will slow down
319	lot of inactive fds). It scales similarly to select, i.e. O(total_fds).	348	considerably with a lot of inactive fds). It scales similarly to select,
		349	i.e. O(total_fds). See the entry for C<EVBACKEND_SELECT>, above, for
		350	performance tips.
320		351
321	=item C<EVBACKEND_EPOLL> (value 4, Linux)	352	=item C<EVBACKEND_EPOLL> (value 4, Linux)
322		353
323	For few fds, this backend is a bit little slower than poll and select,	354	For few fds, this backend is a bit little slower than poll and select,
324	but it scales phenomenally better. While poll and select usually scale	355	but it scales phenomenally better. While poll and select usually scale
325	like O(total_fds) where n is the total number of fds (or the highest fd),	356	like O(total_fds) where n is the total number of fds (or the highest fd),
326	epoll scales either O(1) or O(active_fds). The epoll design has a number	357	epoll scales either O(1) or O(active_fds). The epoll design has a number
327	of shortcomings, such as silently dropping events in some hard-to-detect	358	of shortcomings, such as silently dropping events in some hard-to-detect
328	cases and rewiring a syscall per fd change, no fork support and bad	359	cases and rewiring a syscall per fd change, no fork support and bad
329	support for dup:	360	support for dup.
330		361
331	While stopping, setting and starting an I/O watcher in the same iteration	362	While stopping, setting and starting an I/O watcher in the same iteration
332	will result in some caching, there is still a syscall per such incident	363	will result in some caching, there is still a syscall per such incident
333	(because the fd could point to a different file description now), so its	364	(because the fd could point to a different file description now), so its
334	best to avoid that. Also, C<dup ()>'ed file descriptors might not work	365	best to avoid that. Also, C<dup ()>'ed file descriptors might not work
…		…
336		367
337	Please note that epoll sometimes generates spurious notifications, so you	368	Please note that epoll sometimes generates spurious notifications, so you
338	need to use non-blocking I/O or other means to avoid blocking when no data	369	need to use non-blocking I/O or other means to avoid blocking when no data
339	(or space) is available.	370	(or space) is available.
340		371
		372	Best performance from this backend is achieved by not unregistering all
		373	watchers for a file descriptor until it has been closed, if possible, i.e.
		374	keep at least one watcher active per fd at all times.
		375
		376	While nominally embeddeble in other event loops, this feature is broken in
		377	all kernel versions tested so far.
		378
341	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)	379	=item C<EVBACKEND_KQUEUE> (value 8, most BSD clones)
342		380
343	Kqueue deserves special mention, as at the time of this writing, it	381	Kqueue deserves special mention, as at the time of this writing, it
344	was broken on I<all> BSDs (usually it doesn't work with anything but	382	was broken on all BSDs except NetBSD (usually it doesn't work reliably
345	sockets and pipes, except on Darwin, where of course it's completely	383	with anything but sockets and pipes, except on Darwin, where of course
346	useless. On NetBSD, it seems to work for all the FD types I tested, so it
347	is used by default there). For this reason it's not being "autodetected"	384	it's completely useless). For this reason it's not being "autodetected"
348	unless you explicitly specify it explicitly in the flags (i.e. using	385	unless you explicitly specify it explicitly in the flags (i.e. using
349	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)	386	C<EVBACKEND_KQUEUE>) or libev was compiled on a known-to-be-good (-enough)
350	system like NetBSD.	387	system like NetBSD.
351		388
		389	You still can embed kqueue into a normal poll or select backend and use it
		390	only for sockets (after having made sure that sockets work with kqueue on
		391	the target platform). See C<ev_embed> watchers for more info.
		392
352	It scales in the same way as the epoll backend, but the interface to the	393	It scales in the same way as the epoll backend, but the interface to the
353	kernel is more efficient (which says nothing about its actual speed,	394	kernel is more efficient (which says nothing about its actual speed, of
354	of course). While stopping, setting and starting an I/O watcher does	395	course). While stopping, setting and starting an I/O watcher does never
355	never cause an extra syscall as with epoll, it still adds up to two event	396	cause an extra syscall as with C<EVBACKEND_EPOLL>, it still adds up to
356	changes per incident, support for C<fork ()> is very bad and it drops fds	397	two event changes per incident, support for C<fork ()> is very bad and it
357	silently in similarly hard-to-detetc cases.	398	drops fds silently in similarly hard-to-detect cases.
		399
		400	This backend usually performs well under most conditions.
		401
		402	While nominally embeddable in other event loops, this doesn't work
		403	everywhere, so you might need to test for this. And since it is broken
		404	almost everywhere, you should only use it when you have a lot of sockets
		405	(for which it usually works), by embedding it into another event loop
		406	(e.g. C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>) and using it only for
		407	sockets.
358		408
359	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)	409	=item C<EVBACKEND_DEVPOLL> (value 16, Solaris 8)
360		410
361	This is not implemented yet (and might never be).	411	This is not implemented yet (and might never be, unless you send me an
		412	implementation). According to reports, C</dev/poll> only supports sockets
		413	and is not embeddable, which would limit the usefulness of this backend
		414	immensely.
362		415
363	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	416	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
364		417
365	This uses the Solaris 10 event port mechanism. As with everything on Solaris,	418	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
366	it's really slow, but it still scales very well (O(active_fds)).	419	it's really slow, but it still scales very well (O(active_fds)).
367		420
368	Please note that solaris event ports can deliver a lot of spurious	421	Please note that solaris event ports can deliver a lot of spurious
369	notifications, so you need to use non-blocking I/O or other means to avoid	422	notifications, so you need to use non-blocking I/O or other means to avoid
370	blocking when no data (or space) is available.	423	blocking when no data (or space) is available.
371		424
		425	While this backend scales well, it requires one system call per active
		426	file descriptor per loop iteration. For small and medium numbers of file
		427	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
		428	might perform better.
		429
		430	On the positive side, ignoring the spurious readyness notifications, this
		431	backend actually performed to specification in all tests and is fully
		432	embeddable, which is a rare feat among the OS-specific backends.
		433
372	=item C<EVBACKEND_ALL>	434	=item C<EVBACKEND_ALL>
373		435
374	Try all backends (even potentially broken ones that wouldn't be tried	436	Try all backends (even potentially broken ones that wouldn't be tried
375	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	437	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
376	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	438	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
377		439
		440	It is definitely not recommended to use this flag.
		441
378	=back	442	=back
379		443
380	If one or more of these are ored into the flags value, then only these	444	If one or more of these are ored into the flags value, then only these
381	backends will be tried (in the reverse order as given here). If none are	445	backends will be tried (in the reverse order as listed here). If none are
382	specified, most compiled-in backend will be tried, usually in reverse	446	specified, all backends in C<ev_recommended_backends ()> will be tried.
383	order of their flag values :)
384		447
385	The most typical usage is like this:	448	The most typical usage is like this:
386		449
387	if (!ev_default_loop (0))	450	if (!ev_default_loop (0))
388	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");	451	fatal ("could not initialise libev, bad $LIBEV_FLAGS in environment?");
…		…
435	Like C<ev_default_destroy>, but destroys an event loop created by an	498	Like C<ev_default_destroy>, but destroys an event loop created by an
436	earlier call to C<ev_loop_new>.	499	earlier call to C<ev_loop_new>.
437		500
438	=item ev_default_fork ()	501	=item ev_default_fork ()
439		502
		503	This function sets a flag that causes subsequent C<ev_loop> iterations
440	This function reinitialises the kernel state for backends that have	504	to reinitialise the kernel state for backends that have one. Despite the
441	one. Despite the name, you can call it anytime, but it makes most sense	505	name, you can call it anytime, but it makes most sense after forking, in
442	after forking, in either the parent or child process (or both, but that	506	the child process (or both child and parent, but that again makes little
443	again makes little sense).	507	sense). You I<must> call it in the child before using any of the libev
		508	functions, and it will only take effect at the next C<ev_loop> iteration.
444		509
445	You I<must> call this function in the child process after forking if and	510	On the other hand, you only need to call this function in the child
446	only if you want to use the event library in both processes. If you just	511	process if and only if you want to use the event library in the child. If
447	fork+exec, you don't have to call it.	512	you just fork+exec, you don't have to call it at all.
448		513
449	The function itself is quite fast and it's usually not a problem to call	514	The function itself is quite fast and it's usually not a problem to call
450	it just in case after a fork. To make this easy, the function will fit in	515	it just in case after a fork. To make this easy, the function will fit in
451	quite nicely into a call to C<pthread_atfork>:	516	quite nicely into a call to C<pthread_atfork>:
452		517
453	pthread_atfork (0, 0, ev_default_fork);	518	pthread_atfork (0, 0, ev_default_fork);
454		519
455	At the moment, C<EVBACKEND_SELECT> and C<EVBACKEND_POLL> are safe to use
456	without calling this function, so if you force one of those backends you
457	do not need to care.
458
459	=item ev_loop_fork (loop)	520	=item ev_loop_fork (loop)
460		521
461	Like C<ev_default_fork>, but acts on an event loop created by	522	Like C<ev_default_fork>, but acts on an event loop created by
462	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	523	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
463	after fork, and how you do this is entirely your own problem.	524	after fork, and how you do this is entirely your own problem.
		525
		526	=item int ev_is_default_loop (loop)
		527
		528	Returns true when the given loop actually is the default loop, false otherwise.
464		529
465	=item unsigned int ev_loop_count (loop)	530	=item unsigned int ev_loop_count (loop)
466		531
467	Returns the count of loop iterations for the loop, which is identical to	532	Returns the count of loop iterations for the loop, which is identical to
468	the number of times libev did poll for new events. It starts at C<0> and	533	the number of times libev did poll for new events. It starts at C<0> and
…		…
513	usually a better approach for this kind of thing.	578	usually a better approach for this kind of thing.
514		579
515	Here are the gory details of what C<ev_loop> does:	580	Here are the gory details of what C<ev_loop> does:
516		581
517	- Before the first iteration, call any pending watchers.	582	- Before the first iteration, call any pending watchers.
518	* If there are no active watchers (reference count is zero), return.	583	* If EVFLAG_FORKCHECK was used, check for a fork.
519	- Queue all prepare watchers and then call all outstanding watchers.	584	- If a fork was detected, queue and call all fork watchers.
		585	- Queue and call all prepare watchers.
520	- If we have been forked, recreate the kernel state.	586	- If we have been forked, recreate the kernel state.
521	- Update the kernel state with all outstanding changes.	587	- Update the kernel state with all outstanding changes.
522	- Update the "event loop time".	588	- Update the "event loop time".
523	- Calculate for how long to block.	589	- Calculate for how long to sleep or block, if at all
		590	(active idle watchers, EVLOOP_NONBLOCK or not having
		591	any active watchers at all will result in not sleeping).
		592	- Sleep if the I/O and timer collect interval say so.
524	- Block the process, waiting for any events.	593	- Block the process, waiting for any events.
525	- Queue all outstanding I/O (fd) events.	594	- Queue all outstanding I/O (fd) events.
526	- Update the "event loop time" and do time jump handling.	595	- Update the "event loop time" and do time jump handling.
527	- Queue all outstanding timers.	596	- Queue all outstanding timers.
528	- Queue all outstanding periodics.	597	- Queue all outstanding periodics.
529	- If no events are pending now, queue all idle watchers.	598	- If no events are pending now, queue all idle watchers.
530	- Queue all check watchers.	599	- Queue all check watchers.
531	- Call all queued watchers in reverse order (i.e. check watchers first).	600	- Call all queued watchers in reverse order (i.e. check watchers first).
532	Signals and child watchers are implemented as I/O watchers, and will	601	Signals and child watchers are implemented as I/O watchers, and will
533	be handled here by queueing them when their watcher gets executed.	602	be handled here by queueing them when their watcher gets executed.
534	- If ev_unloop has been called or EVLOOP_ONESHOT or EVLOOP_NONBLOCK	603	- If ev_unloop has been called, or EVLOOP_ONESHOT or EVLOOP_NONBLOCK
535	were used, return, otherwise continue with step *.	604	were used, or there are no active watchers, return, otherwise
		605	continue with step *.
536		606
537	Example: Queue some jobs and then loop until no events are outsanding	607	Example: Queue some jobs and then loop until no events are outstanding
538	anymore.	608	anymore.
539		609
540	... queue jobs here, make sure they register event watchers as long	610	... queue jobs here, make sure they register event watchers as long
541	... as they still have work to do (even an idle watcher will do..)	611	... as they still have work to do (even an idle watcher will do..)
542	ev_loop (my_loop, 0);	612	ev_loop (my_loop, 0);
…		…
546		616
547	Can be used to make a call to C<ev_loop> return early (but only after it	617	Can be used to make a call to C<ev_loop> return early (but only after it
548	has processed all outstanding events). The C<how> argument must be either	618	has processed all outstanding events). The C<how> argument must be either
549	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	619	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
550	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	620	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
		621
		622	This "unloop state" will be cleared when entering C<ev_loop> again.
551		623
552	=item ev_ref (loop)	624	=item ev_ref (loop)
553		625
554	=item ev_unref (loop)	626	=item ev_unref (loop)
555		627
…		…
560	returning, ev_unref() after starting, and ev_ref() before stopping it. For	632	returning, ev_unref() after starting, and ev_ref() before stopping it. For
561	example, libev itself uses this for its internal signal pipe: It is not	633	example, libev itself uses this for its internal signal pipe: It is not
562	visible to the libev user and should not keep C<ev_loop> from exiting if	634	visible to the libev user and should not keep C<ev_loop> from exiting if
563	no event watchers registered by it are active. It is also an excellent	635	no event watchers registered by it are active. It is also an excellent
564	way to do this for generic recurring timers or from within third-party	636	way to do this for generic recurring timers or from within third-party
565	libraries. Just remember to I<unref after start> and I<ref before stop>.	637	libraries. Just remember to I<unref after start> and I<ref before stop>
		638	(but only if the watcher wasn't active before, or was active before,
		639	respectively).
566		640
567	Example: Create a signal watcher, but keep it from keeping C<ev_loop>	641	Example: Create a signal watcher, but keep it from keeping C<ev_loop>
568	running when nothing else is active.	642	running when nothing else is active.
569		643
570	struct ev_signal exitsig;	644	struct ev_signal exitsig;
…		…
596	overhead for the actual polling but can deliver many events at once.	670	overhead for the actual polling but can deliver many events at once.
597		671
598	By setting a higher I<io collect interval> you allow libev to spend more	672	By setting a higher I<io collect interval> you allow libev to spend more
599	time collecting I/O events, so you can handle more events per iteration,	673	time collecting I/O events, so you can handle more events per iteration,
600	at the cost of increasing latency. Timeouts (both C<ev_periodic> and	674	at the cost of increasing latency. Timeouts (both C<ev_periodic> and
601	C<ev_timer>) will be not affected. Setting this to a non-null bvalue will	675	C<ev_timer>) will be not affected. Setting this to a non-null value will
602	introduce an additional C<ev_sleep ()> call into most loop iterations.	676	introduce an additional C<ev_sleep ()> call into most loop iterations.
603		677
604	Likewise, by setting a higher I<timeout collect interval> you allow libev	678	Likewise, by setting a higher I<timeout collect interval> you allow libev
605	to spend more time collecting timeouts, at the expense of increased	679	to spend more time collecting timeouts, at the expense of increased
606	latency (the watcher callback will be called later). C<ev_io> watchers	680	latency (the watcher callback will be called later). C<ev_io> watchers
…		…
718		792
719	=item C<EV_FORK>	793	=item C<EV_FORK>
720		794
721	The event loop has been resumed in the child process after fork (see	795	The event loop has been resumed in the child process after fork (see
722	C<ev_fork>).	796	C<ev_fork>).
		797
		798	=item C<EV_ASYNC>
		799
		800	The given async watcher has been asynchronously notified (see C<ev_async>).
723		801
724	=item C<EV_ERROR>	802	=item C<EV_ERROR>
725		803
726	An unspecified error has occured, the watcher has been stopped. This might	804	An unspecified error has occured, the watcher has been stopped. This might
727	happen because the watcher could not be properly started because libev	805	happen because the watcher could not be properly started because libev
…		…
945	In general you can register as many read and/or write event watchers per	1023	In general you can register as many read and/or write event watchers per
946	fd as you want (as long as you don't confuse yourself). Setting all file	1024	fd as you want (as long as you don't confuse yourself). Setting all file
947	descriptors to non-blocking mode is also usually a good idea (but not	1025	descriptors to non-blocking mode is also usually a good idea (but not
948	required if you know what you are doing).	1026	required if you know what you are doing).
949		1027
950	You have to be careful with dup'ed file descriptors, though. Some backends
951	(the linux epoll backend is a notable example) cannot handle dup'ed file
952	descriptors correctly if you register interest in two or more fds pointing
953	to the same underlying file/socket/etc. description (that is, they share
954	the same underlying "file open").
955
956	If you must do this, then force the use of a known-to-be-good backend	1028	If you must do this, then force the use of a known-to-be-good backend
957	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and	1029	(at the time of this writing, this includes only C<EVBACKEND_SELECT> and
958	C<EVBACKEND_POLL>).	1030	C<EVBACKEND_POLL>).
959		1031
960	Another thing you have to watch out for is that it is quite easy to	1032	Another thing you have to watch out for is that it is quite easy to
…		…
994	optimisations to libev.	1066	optimisations to libev.
995		1067
996	=head3 The special problem of dup'ed file descriptors	1068	=head3 The special problem of dup'ed file descriptors
997		1069
998	Some backends (e.g. epoll), cannot register events for file descriptors,	1070	Some backends (e.g. epoll), cannot register events for file descriptors,
999	but only events for the underlying file descriptions. That menas when you	1071	but only events for the underlying file descriptions. That means when you
1000	have C<dup ()>'ed file descriptors and register events for them, only one	1072	have C<dup ()>'ed file descriptors or weirder constellations, and register
1001	file descriptor might actually receive events.	1073	events for them, only one file descriptor might actually receive events.
1002		1074
1003	There is no workaorund possible except not registering events	1075	There is no workaround possible except not registering events
1004	for potentially C<dup ()>'ed file descriptors or to resort to	1076	for potentially C<dup ()>'ed file descriptors, or to resort to
1005	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.	1077	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
1006		1078
1007	=head3 The special problem of fork	1079	=head3 The special problem of fork
1008		1080
1009	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit	1081	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
…		…
1035	=item int events [read-only]	1107	=item int events [read-only]
1036		1108
1037	The events being watched.	1109	The events being watched.
1038		1110
1039	=back	1111	=back
		1112
		1113	=head3 Examples
1040		1114
1041	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well	1115	Example: Call C<stdin_readable_cb> when STDIN_FILENO has become, well
1042	readable, but only once. Since it is likely line-buffered, you could	1116	readable, but only once. Since it is likely line-buffered, you could
1043	attempt to read a whole line in the callback.	1117	attempt to read a whole line in the callback.
1044		1118
…		…
1097	configure a timer to trigger every 10 seconds, then it will trigger at	1171	configure a timer to trigger every 10 seconds, then it will trigger at
1098	exactly 10 second intervals. If, however, your program cannot keep up with	1172	exactly 10 second intervals. If, however, your program cannot keep up with
1099	the timer (because it takes longer than those 10 seconds to do stuff) the	1173	the timer (because it takes longer than those 10 seconds to do stuff) the
1100	timer will not fire more than once per event loop iteration.	1174	timer will not fire more than once per event loop iteration.
1101		1175
1102	=item ev_timer_again (loop)	1176	=item ev_timer_again (loop, ev_timer *)
1103		1177
1104	This will act as if the timer timed out and restart it again if it is	1178	This will act as if the timer timed out and restart it again if it is
1105	repeating. The exact semantics are:	1179	repeating. The exact semantics are:
1106		1180
1107	If the timer is pending, its pending status is cleared.	1181	If the timer is pending, its pending status is cleared.
…		…
1142	or C<ev_timer_again> is called and determines the next timeout (if any),	1216	or C<ev_timer_again> is called and determines the next timeout (if any),
1143	which is also when any modifications are taken into account.	1217	which is also when any modifications are taken into account.
1144		1218
1145	=back	1219	=back
1146		1220
		1221	=head3 Examples
		1222
1147	Example: Create a timer that fires after 60 seconds.	1223	Example: Create a timer that fires after 60 seconds.
1148		1224
1149	static void	1225	static void
1150	one_minute_cb (struct ev_loop loop, struct ev_timer w, int revents)	1226	one_minute_cb (struct ev_loop loop, struct ev_timer w, int revents)
1151	{	1227	{
…		…
1214	In this configuration the watcher triggers an event at the wallclock time	1290	In this configuration the watcher triggers an event at the wallclock time
1215	C<at> and doesn't repeat. It will not adjust when a time jump occurs,	1291	C<at> and doesn't repeat. It will not adjust when a time jump occurs,
1216	that is, if it is to be run at January 1st 2011 then it will run when the	1292	that is, if it is to be run at January 1st 2011 then it will run when the
1217	system time reaches or surpasses this time.	1293	system time reaches or surpasses this time.
1218		1294
1219	=item * non-repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)	1295	=item * repeating interval timer (at = offset, interval > 0, reschedule_cb = 0)
1220		1296
1221	In this mode the watcher will always be scheduled to time out at the next	1297	In this mode the watcher will always be scheduled to time out at the next
1222	C<at + N * interval> time (for some integer N, which can also be negative)	1298	C<at + N * interval> time (for some integer N, which can also be negative)
1223	and then repeat, regardless of any time jumps.	1299	and then repeat, regardless of any time jumps.
1224		1300
…		…
1307		1383
1308	When active, contains the absolute time that the watcher is supposed to	1384	When active, contains the absolute time that the watcher is supposed to
1309	trigger next.	1385	trigger next.
1310		1386
1311	=back	1387	=back
		1388
		1389	=head3 Examples
1312		1390
1313	Example: Call a callback every hour, or, more precisely, whenever the	1391	Example: Call a callback every hour, or, more precisely, whenever the
1314	system clock is divisible by 3600. The callback invocation times have	1392	system clock is divisible by 3600. The callback invocation times have
1315	potentially a lot of jittering, but good long-term stability.	1393	potentially a lot of jittering, but good long-term stability.
1316		1394
…		…
1356	with the kernel (thus it coexists with your own signal handlers as long	1434	with the kernel (thus it coexists with your own signal handlers as long
1357	as you don't register any with libev). Similarly, when the last signal	1435	as you don't register any with libev). Similarly, when the last signal
1358	watcher for a signal is stopped libev will reset the signal handler to	1436	watcher for a signal is stopped libev will reset the signal handler to
1359	SIG_DFL (regardless of what it was set to before).	1437	SIG_DFL (regardless of what it was set to before).
1360		1438
		1439	If possible and supported, libev will install its handlers with
		1440	C<SA_RESTART> behaviour enabled, so syscalls should not be unduly
		1441	interrupted. If you have a problem with syscalls getting interrupted by
		1442	signals you can block all signals in an C<ev_check> watcher and unblock
		1443	them in an C<ev_prepare> watcher.
		1444
1361	=head3 Watcher-Specific Functions and Data Members	1445	=head3 Watcher-Specific Functions and Data Members
1362		1446
1363	=over 4	1447	=over 4
1364		1448
1365	=item ev_signal_init (ev_signal *, callback, int signum)	1449	=item ev_signal_init (ev_signal *, callback, int signum)
…		…
1373		1457
1374	The signal the watcher watches out for.	1458	The signal the watcher watches out for.
1375		1459
1376	=back	1460	=back
1377		1461
		1462	=head3 Examples
		1463
		1464	Example: Try to exit cleanly on SIGINT and SIGTERM.
		1465
		1466	static void
		1467	sigint_cb (struct ev_loop loop, struct ev_signal w, int revents)
		1468	{
		1469	ev_unloop (loop, EVUNLOOP_ALL);
		1470	}
		1471
		1472	struct ev_signal signal_watcher;
		1473	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
		1474	ev_signal_start (loop, &sigint_cb);
		1475
1378		1476
1379	=head2 C<ev_child> - watch out for process status changes	1477	=head2 C<ev_child> - watch out for process status changes
1380		1478
1381	Child watchers trigger when your process receives a SIGCHLD in response to	1479	Child watchers trigger when your process receives a SIGCHLD in response to
1382	some child status changes (most typically when a child of yours dies).	1480	some child status changes (most typically when a child of yours dies). It
		1481	is permissible to install a child watcher I<after> the child has been
		1482	forked (which implies it might have already exited), as long as the event
		1483	loop isn't entered (or is continued from a watcher).
		1484
		1485	Only the default event loop is capable of handling signals, and therefore
		1486	you can only rgeister child watchers in the default event loop.
		1487
		1488	=head3 Process Interaction
		1489
		1490	Libev grabs C<SIGCHLD> as soon as the default event loop is
		1491	initialised. This is necessary to guarantee proper behaviour even if
		1492	the first child watcher is started after the child exits. The occurance
		1493	of C<SIGCHLD> is recorded asynchronously, but child reaping is done
		1494	synchronously as part of the event loop processing. Libev always reaps all
		1495	children, even ones not watched.
		1496
		1497	=head3 Overriding the Built-In Processing
		1498
		1499	Libev offers no special support for overriding the built-in child
		1500	processing, but if your application collides with libev's default child
		1501	handler, you can override it easily by installing your own handler for
		1502	C<SIGCHLD> after initialising the default loop, and making sure the
		1503	default loop never gets destroyed. You are encouraged, however, to use an
		1504	event-based approach to child reaping and thus use libev's support for
		1505	that, so other libev users can use C<ev_child> watchers freely.
1383		1506
1384	=head3 Watcher-Specific Functions and Data Members	1507	=head3 Watcher-Specific Functions and Data Members
1385		1508
1386	=over 4	1509	=over 4
1387		1510
1388	=item ev_child_init (ev_child *, callback, int pid)	1511	=item ev_child_init (ev_child *, callback, int pid, int trace)
1389		1512
1390	=item ev_child_set (ev_child *, int pid)	1513	=item ev_child_set (ev_child *, int pid, int trace)
1391		1514
1392	Configures the watcher to wait for status changes of process C<pid> (or	1515	Configures the watcher to wait for status changes of process C<pid> (or
1393	I<any> process if C<pid> is specified as C<0>). The callback can look	1516	I<any> process if C<pid> is specified as C<0>). The callback can look
1394	at the C<rstatus> member of the C<ev_child> watcher structure to see	1517	at the C<rstatus> member of the C<ev_child> watcher structure to see
1395	the status word (use the macros from C<sys/wait.h> and see your systems	1518	the status word (use the macros from C<sys/wait.h> and see your systems
1396	C<waitpid> documentation). The C<rpid> member contains the pid of the	1519	C<waitpid> documentation). The C<rpid> member contains the pid of the
1397	process causing the status change.	1520	process causing the status change. C<trace> must be either C<0> (only
		1521	activate the watcher when the process terminates) or C<1> (additionally
		1522	activate the watcher when the process is stopped or continued).
1398		1523
1399	=item int pid [read-only]	1524	=item int pid [read-only]
1400		1525
1401	The process id this watcher watches out for, or C<0>, meaning any process id.	1526	The process id this watcher watches out for, or C<0>, meaning any process id.
1402		1527
…		…
1409	The process exit/trace status caused by C<rpid> (see your systems	1534	The process exit/trace status caused by C<rpid> (see your systems
1410	C<waitpid> and C<sys/wait.h> documentation for details).	1535	C<waitpid> and C<sys/wait.h> documentation for details).
1411		1536
1412	=back	1537	=back
1413		1538
1414	Example: Try to exit cleanly on SIGINT and SIGTERM.	1539	=head3 Examples
		1540
		1541	Example: C<fork()> a new process and install a child handler to wait for
		1542	its completion.
		1543
		1544	ev_child cw;
1415		1545
1416	static void	1546	static void
1417	sigint_cb (struct ev_loop loop, struct ev_signal w, int revents)	1547	child_cb (EV_P_ struct ev_child *w, int revents)
1418	{	1548	{
1419	ev_unloop (loop, EVUNLOOP_ALL);	1549	ev_child_stop (EV_A_ w);
		1550	printf ("process %d exited with status %x\n", w->rpid, w->rstatus);
1420	}	1551	}
1421		1552
1422	struct ev_signal signal_watcher;	1553	pid_t pid = fork ();
1423	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1554
1424	ev_signal_start (loop, &sigint_cb);	1555	if (pid < 0)
		1556	// error
		1557	else if (pid == 0)
		1558	{
		1559	// the forked child executes here
		1560	exit (1);
		1561	}
		1562	else
		1563	{
		1564	ev_child_init (&cw, child_cb, pid, 0);
		1565	ev_child_start (EV_DEFAULT_ &cw);
		1566	}
1425		1567
1426		1568
1427	=head2 C<ev_stat> - did the file attributes just change?	1569	=head2 C<ev_stat> - did the file attributes just change?
1428		1570
1429	This watches a filesystem path for attribute changes. That is, it calls	1571	This watches a filesystem path for attribute changes. That is, it calls
…		…
1458	semantics of C<ev_stat> watchers, which means that libev sometimes needs	1600	semantics of C<ev_stat> watchers, which means that libev sometimes needs
1459	to fall back to regular polling again even with inotify, but changes are	1601	to fall back to regular polling again even with inotify, but changes are
1460	usually detected immediately, and if the file exists there will be no	1602	usually detected immediately, and if the file exists there will be no
1461	polling.	1603	polling.
1462		1604
		1605	=head3 ABI Issues (Largefile Support)
		1606
		1607	Libev by default (unless the user overrides this) uses the default
		1608	compilation environment, which means that on systems with optionally
		1609	disabled large file support, you get the 32 bit version of the stat
		1610	structure. When using the library from programs that change the ABI to
		1611	use 64 bit file offsets the programs will fail. In that case you have to
		1612	compile libev with the same flags to get binary compatibility. This is
		1613	obviously the case with any flags that change the ABI, but the problem is
		1614	most noticably with ev_stat and largefile support.
		1615
		1616	=head3 Inotify
		1617
		1618	When C<inotify (7)> support has been compiled into libev (generally only
		1619	available on Linux) and present at runtime, it will be used to speed up
		1620	change detection where possible. The inotify descriptor will be created lazily
		1621	when the first C<ev_stat> watcher is being started.
		1622
		1623	Inotify presense does not change the semantics of C<ev_stat> watchers
		1624	except that changes might be detected earlier, and in some cases, to avoid
		1625	making regular C<stat> calls. Even in the presense of inotify support
		1626	there are many cases where libev has to resort to regular C<stat> polling.
		1627
		1628	(There is no support for kqueue, as apparently it cannot be used to
		1629	implement this functionality, due to the requirement of having a file
		1630	descriptor open on the object at all times).
		1631
		1632	=head3 The special problem of stat time resolution
		1633
		1634	The C<stat ()> syscall only supports full-second resolution portably, and
		1635	even on systems where the resolution is higher, many filesystems still
		1636	only support whole seconds.
		1637
		1638	That means that, if the time is the only thing that changes, you might
		1639	miss updates: on the first update, C<ev_stat> detects a change and calls
		1640	your callback, which does something. When there is another update within
		1641	the same second, C<ev_stat> will be unable to detect it.
		1642
		1643	The solution to this is to delay acting on a change for a second (or till
		1644	the next second boundary), using a roughly one-second delay C<ev_timer>
		1645	(C<ev_timer_set (w, 0., 1.01); ev_timer_again (loop, w)>). The C<.01>
		1646	is added to work around small timing inconsistencies of some operating
		1647	systems.
		1648
1463	=head3 Watcher-Specific Functions and Data Members	1649	=head3 Watcher-Specific Functions and Data Members
1464		1650
1465	=over 4	1651	=over 4
1466		1652
1467	=item ev_stat_init (ev_stat , callback, const char path, ev_tstamp interval)	1653	=item ev_stat_init (ev_stat , callback, const char path, ev_tstamp interval)
…		…
1476		1662
1477	The callback will be receive C<EV_STAT> when a change was detected,	1663	The callback will be receive C<EV_STAT> when a change was detected,
1478	relative to the attributes at the time the watcher was started (or the	1664	relative to the attributes at the time the watcher was started (or the
1479	last change was detected).	1665	last change was detected).
1480		1666
1481	=item ev_stat_stat (ev_stat *)	1667	=item ev_stat_stat (loop, ev_stat *)
1482		1668
1483	Updates the stat buffer immediately with new values. If you change the	1669	Updates the stat buffer immediately with new values. If you change the
1484	watched path in your callback, you could call this fucntion to avoid	1670	watched path in your callback, you could call this fucntion to avoid
1485	detecting this change (while introducing a race condition). Can also be	1671	detecting this change (while introducing a race condition). Can also be
1486	useful simply to find out the new values.	1672	useful simply to find out the new values.
…		…
1504	=item const char *path [read-only]	1690	=item const char *path [read-only]
1505		1691
1506	The filesystem path that is being watched.	1692	The filesystem path that is being watched.
1507		1693
1508	=back	1694	=back
		1695
		1696	=head3 Examples
1509		1697
1510	Example: Watch C</etc/passwd> for attribute changes.	1698	Example: Watch C</etc/passwd> for attribute changes.
1511		1699
1512	static void	1700	static void
1513	passwd_cb (struct ev_loop loop, ev_stat w, int revents)	1701	passwd_cb (struct ev_loop loop, ev_stat w, int revents)
…		…
1526	}	1714	}
1527		1715
1528	...	1716	...
1529	ev_stat passwd;	1717	ev_stat passwd;
1530		1718
1531	ev_stat_init (&passwd, passwd_cb, "/etc/passwd");	1719	ev_stat_init (&passwd, passwd_cb, "/etc/passwd", 0.);
1532	ev_stat_start (loop, &passwd);	1720	ev_stat_start (loop, &passwd);
		1721
		1722	Example: Like above, but additionally use a one-second delay so we do not
		1723	miss updates (however, frequent updates will delay processing, too, so
		1724	one might do the work both on C<ev_stat> callback invocation I<and> on
		1725	C<ev_timer> callback invocation).
		1726
		1727	static ev_stat passwd;
		1728	static ev_timer timer;
		1729
		1730	static void
		1731	timer_cb (EV_P_ ev_timer *w, int revents)
		1732	{
		1733	ev_timer_stop (EV_A_ w);
		1734
		1735	/* now it's one second after the most recent passwd change */
		1736	}
		1737
		1738	static void
		1739	stat_cb (EV_P_ ev_stat *w, int revents)
		1740	{
		1741	/* reset the one-second timer */
		1742	ev_timer_again (EV_A_ &timer);
		1743	}
		1744
		1745	...
		1746	ev_stat_init (&passwd, stat_cb, "/etc/passwd", 0.);
		1747	ev_stat_start (loop, &passwd);
		1748	ev_timer_init (&timer, timer_cb, 0., 1.01);
1533		1749
1534		1750
1535	=head2 C<ev_idle> - when you've got nothing better to do...	1751	=head2 C<ev_idle> - when you've got nothing better to do...
1536		1752
1537	Idle watchers trigger events when no other events of the same or higher	1753	Idle watchers trigger events when no other events of the same or higher
…		…
1563	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,	1779	kind. There is a C<ev_idle_set> macro, but using it is utterly pointless,
1564	believe me.	1780	believe me.
1565		1781
1566	=back	1782	=back
1567		1783
		1784	=head3 Examples
		1785
1568	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the	1786	Example: Dynamically allocate an C<ev_idle> watcher, start it, and in the
1569	callback, free it. Also, use no error checking, as usual.	1787	callback, free it. Also, use no error checking, as usual.
1570		1788
1571	static void	1789	static void
1572	idle_cb (struct ev_loop loop, struct ev_idle w, int revents)	1790	idle_cb (struct ev_loop loop, struct ev_idle w, int revents)
1573	{	1791	{
1574	free (w);	1792	free (w);
1575	// now do something you wanted to do when the program has	1793	// now do something you wanted to do when the program has
1576	// no longer asnything immediate to do.	1794	// no longer anything immediate to do.
1577	}	1795	}
1578		1796
1579	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));	1797	struct ev_idle *idle_watcher = malloc (sizeof (struct ev_idle));
1580	ev_idle_init (idle_watcher, idle_cb);	1798	ev_idle_init (idle_watcher, idle_cb);
1581	ev_idle_start (loop, idle_cb);	1799	ev_idle_start (loop, idle_cb);
…		…
1623		1841
1624	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)	1842	It is recommended to give C<ev_check> watchers highest (C<EV_MAXPRI>)
1625	priority, to ensure that they are being run before any other watchers	1843	priority, to ensure that they are being run before any other watchers
1626	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,	1844	after the poll. Also, C<ev_check> watchers (and C<ev_prepare> watchers,
1627	too) should not activate ("feed") events into libev. While libev fully	1845	too) should not activate ("feed") events into libev. While libev fully
1628	supports this, they will be called before other C<ev_check> watchers did	1846	supports this, they will be called before other C<ev_check> watchers
1629	their job. As C<ev_check> watchers are often used to embed other event	1847	did their job. As C<ev_check> watchers are often used to embed other
1630	loops those other event loops might be in an unusable state until their	1848	(non-libev) event loops those other event loops might be in an unusable
1631	C<ev_check> watcher ran (always remind yourself to coexist peacefully with	1849	state until their C<ev_check> watcher ran (always remind yourself to
1632	others).	1850	coexist peacefully with others).
1633		1851
1634	=head3 Watcher-Specific Functions and Data Members	1852	=head3 Watcher-Specific Functions and Data Members
1635		1853
1636	=over 4	1854	=over 4
1637		1855
…		…
1642	Initialises and configures the prepare or check watcher - they have no	1860	Initialises and configures the prepare or check watcher - they have no
1643	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>	1861	parameters of any kind. There are C<ev_prepare_set> and C<ev_check_set>
1644	macros, but using them is utterly, utterly and completely pointless.	1862	macros, but using them is utterly, utterly and completely pointless.
1645		1863
1646	=back	1864	=back
		1865
		1866	=head3 Examples
1647		1867
1648	There are a number of principal ways to embed other event loops or modules	1868	There are a number of principal ways to embed other event loops or modules
1649	into libev. Here are some ideas on how to include libadns into libev	1869	into libev. Here are some ideas on how to include libadns into libev
1650	(there is a Perl module named C<EV::ADNS> that does this, which you could	1870	(there is a Perl module named C<EV::ADNS> that does this, which you could
1651	use for an actually working example. Another Perl module named C<EV::Glib>	1871	use for an actually working example. Another Perl module named C<EV::Glib>
…		…
1776	=head2 C<ev_embed> - when one backend isn't enough...	1996	=head2 C<ev_embed> - when one backend isn't enough...
1777		1997
1778	This is a rather advanced watcher type that lets you embed one event loop	1998	This is a rather advanced watcher type that lets you embed one event loop
1779	into another (currently only C<ev_io> events are supported in the embedded	1999	into another (currently only C<ev_io> events are supported in the embedded
1780	loop, other types of watchers might be handled in a delayed or incorrect	2000	loop, other types of watchers might be handled in a delayed or incorrect
1781	fashion and must not be used). (See portability notes, below).	2001	fashion and must not be used).
1782		2002
1783	There are primarily two reasons you would want that: work around bugs and	2003	There are primarily two reasons you would want that: work around bugs and
1784	prioritise I/O.	2004	prioritise I/O.
1785		2005
1786	As an example for a bug workaround, the kqueue backend might only support	2006	As an example for a bug workaround, the kqueue backend might only support
…		…
1820	portable one.	2040	portable one.
1821		2041
1822	So when you want to use this feature you will always have to be prepared	2042	So when you want to use this feature you will always have to be prepared
1823	that you cannot get an embeddable loop. The recommended way to get around	2043	that you cannot get an embeddable loop. The recommended way to get around
1824	this is to have a separate variables for your embeddable loop, try to	2044	this is to have a separate variables for your embeddable loop, try to
1825	create it, and if that fails, use the normal loop for everything:	2045	create it, and if that fails, use the normal loop for everything.
		2046
		2047	=head3 Watcher-Specific Functions and Data Members
		2048
		2049	=over 4
		2050
		2051	=item ev_embed_init (ev_embed , callback, struct ev_loop embedded_loop)
		2052
		2053	=item ev_embed_set (ev_embed , callback, struct ev_loop embedded_loop)
		2054
		2055	Configures the watcher to embed the given loop, which must be
		2056	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
		2057	invoked automatically, otherwise it is the responsibility of the callback
		2058	to invoke it (it will continue to be called until the sweep has been done,
		2059	if you do not want thta, you need to temporarily stop the embed watcher).
		2060
		2061	=item ev_embed_sweep (loop, ev_embed *)
		2062
		2063	Make a single, non-blocking sweep over the embedded loop. This works
		2064	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
		2065	apropriate way for embedded loops.
		2066
		2067	=item struct ev_loop *other [read-only]
		2068
		2069	The embedded event loop.
		2070
		2071	=back
		2072
		2073	=head3 Examples
		2074
		2075	Example: Try to get an embeddable event loop and embed it into the default
		2076	event loop. If that is not possible, use the default loop. The default
		2077	loop is stored in C<loop_hi>, while the mebeddable loop is stored in
		2078	C<loop_lo> (which is C<loop_hi> in the acse no embeddable loop can be
		2079	used).
1826		2080
1827	struct ev_loop *loop_hi = ev_default_init (0);	2081	struct ev_loop *loop_hi = ev_default_init (0);
1828	struct ev_loop *loop_lo = 0;	2082	struct ev_loop *loop_lo = 0;
1829	struct ev_embed embed;	2083	struct ev_embed embed;
1830		2084
…		…
1841	ev_embed_start (loop_hi, &embed);	2095	ev_embed_start (loop_hi, &embed);
1842	}	2096	}
1843	else	2097	else
1844	loop_lo = loop_hi;	2098	loop_lo = loop_hi;
1845		2099
1846	=head2 Portability notes	2100	Example: Check if kqueue is available but not recommended and create
		2101	a kqueue backend for use with sockets (which usually work with any
		2102	kqueue implementation). Store the kqueue/socket-only event loop in
		2103	C<loop_socket>. (One might optionally use C<EVFLAG_NOENV>, too).
1847		2104
1848	Kqueue is nominally embeddable, but this is broken on all BSDs that I	2105	struct ev_loop *loop = ev_default_init (0);
1849	tried, in various ways. Usually the embedded event loop will simply never	2106	struct ev_loop *loop_socket = 0;
1850	receive events, sometimes it will only trigger a few times, sometimes in a	2107	struct ev_embed embed;
1851	loop. Epoll is also nominally embeddable, but many Linux kernel versions	2108
1852	will always eport the epoll fd as ready, even when no events are pending.	2109	if (ev_supported_backends () & ~ev_recommended_backends () & EVBACKEND_KQUEUE)
		2110	if ((loop_socket = ev_loop_new (EVBACKEND_KQUEUE))
		2111	{
		2112	ev_embed_init (&embed, 0, loop_socket);
		2113	ev_embed_start (loop, &embed);
		2114	}
1853		2115
1854	While libev allows embedding these backends (they are contained in	2116	if (!loop_socket)
1855	C<ev_embeddable_backends ()>), take extreme care that it will actually	2117	loop_socket = loop;
1856	work.
1857		2118
1858	When in doubt, create a dynamic event loop forced to use sockets (this	2119	// now use loop_socket for all sockets, and loop for everything else
1859	usually works) and possibly another thread and a pipe or so to report to
1860	your main event loop.
1861
1862	=head3 Watcher-Specific Functions and Data Members
1863
1864	=over 4
1865
1866	=item ev_embed_init (ev_embed , callback, struct ev_loop embedded_loop)
1867
1868	=item ev_embed_set (ev_embed , callback, struct ev_loop embedded_loop)
1869
1870	Configures the watcher to embed the given loop, which must be
1871	embeddable. If the callback is C<0>, then C<ev_embed_sweep> will be
1872	invoked automatically, otherwise it is the responsibility of the callback
1873	to invoke it (it will continue to be called until the sweep has been done,
1874	if you do not want thta, you need to temporarily stop the embed watcher).
1875
1876	=item ev_embed_sweep (loop, ev_embed *)
1877
1878	Make a single, non-blocking sweep over the embedded loop. This works
1879	similarly to C<ev_loop (embedded_loop, EVLOOP_NONBLOCK)>, but in the most
1880	apropriate way for embedded loops.
1881
1882	=item struct ev_loop *other [read-only]
1883
1884	The embedded event loop.
1885
1886	=back
1887		2120
1888		2121
1889	=head2 C<ev_fork> - the audacity to resume the event loop after a fork	2122	=head2 C<ev_fork> - the audacity to resume the event loop after a fork
1890		2123
1891	Fork watchers are called when a C<fork ()> was detected (usually because	2124	Fork watchers are called when a C<fork ()> was detected (usually because
…		…
1907	believe me.	2140	believe me.
1908		2141
1909	=back	2142	=back
1910		2143
1911		2144
		2145	=head2 C<ev_async> - how to wake up another event loop
		2146
		2147	In general, you cannot use an C<ev_loop> from multiple threads or other
		2148	asynchronous sources such as signal handlers (as opposed to multiple event
		2149	loops - those are of course safe to use in different threads).
		2150
		2151	Sometimes, however, you need to wake up another event loop you do not
		2152	control, for example because it belongs to another thread. This is what
		2153	C<ev_async> watchers do: as long as the C<ev_async> watcher is active, you
		2154	can signal it by calling C<ev_async_send>, which is thread- and signal
		2155	safe.
		2156
		2157	This functionality is very similar to C<ev_signal> watchers, as signals,
		2158	too, are asynchronous in nature, and signals, too, will be compressed
		2159	(i.e. the number of callback invocations may be less than the number of
		2160	C<ev_async_sent> calls).
		2161
		2162	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
		2163	just the default loop.
		2164
		2165	=head3 Queueing
		2166
		2167	C<ev_async> does not support queueing of data in any way. The reason
		2168	is that the author does not know of a simple (or any) algorithm for a
		2169	multiple-writer-single-reader queue that works in all cases and doesn't
		2170	need elaborate support such as pthreads.
		2171
		2172	That means that if you want to queue data, you have to provide your own
		2173	queue. But at least I can tell you would implement locking around your
		2174	queue:
		2175
		2176	=over 4
		2177
		2178	=item queueing from a signal handler context
		2179
		2180	To implement race-free queueing, you simply add to the queue in the signal
		2181	handler but you block the signal handler in the watcher callback. Here is an example that does that for
		2182	some fictitiuous SIGUSR1 handler:
		2183
		2184	static ev_async mysig;
		2185
		2186	static void
		2187	sigusr1_handler (void)
		2188	{
		2189	sometype data;
		2190
		2191	// no locking etc.
		2192	queue_put (data);
		2193	ev_async_send (EV_DEFAULT_ &mysig);
		2194	}
		2195
		2196	static void
		2197	mysig_cb (EV_P_ ev_async *w, int revents)
		2198	{
		2199	sometype data;
		2200	sigset_t block, prev;
		2201
		2202	sigemptyset (&block);
		2203	sigaddset (&block, SIGUSR1);
		2204	sigprocmask (SIG_BLOCK, &block, &prev);
		2205
		2206	while (queue_get (&data))
		2207	process (data);
		2208
		2209	if (sigismember (&prev, SIGUSR1)
		2210	sigprocmask (SIG_UNBLOCK, &block, 0);
		2211	}
		2212
		2213	(Note: pthreads in theory requires you to use C<pthread_setmask>
		2214	instead of C<sigprocmask> when you use threads, but libev doesn't do it
		2215	either...).
		2216
		2217	=item queueing from a thread context
		2218
		2219	The strategy for threads is different, as you cannot (easily) block
		2220	threads but you can easily preempt them, so to queue safely you need to
		2221	employ a traditional mutex lock, such as in this pthread example:
		2222
		2223	static ev_async mysig;
		2224	static pthread_mutex_t mymutex = PTHREAD_MUTEX_INITIALIZER;
		2225
		2226	static void
		2227	otherthread (void)
		2228	{
		2229	// only need to lock the actual queueing operation
		2230	pthread_mutex_lock (&mymutex);
		2231	queue_put (data);
		2232	pthread_mutex_unlock (&mymutex);
		2233
		2234	ev_async_send (EV_DEFAULT_ &mysig);
		2235	}
		2236
		2237	static void
		2238	mysig_cb (EV_P_ ev_async *w, int revents)
		2239	{
		2240	pthread_mutex_lock (&mymutex);
		2241
		2242	while (queue_get (&data))
		2243	process (data);
		2244
		2245	pthread_mutex_unlock (&mymutex);
		2246	}
		2247
		2248	=back
		2249
		2250
		2251	=head3 Watcher-Specific Functions and Data Members
		2252
		2253	=over 4
		2254
		2255	=item ev_async_init (ev_async *, callback)
		2256
		2257	Initialises and configures the async watcher - it has no parameters of any
		2258	kind. There is a C<ev_asynd_set> macro, but using it is utterly pointless,
		2259	believe me.
		2260
		2261	=item ev_async_send (loop, ev_async *)
		2262
		2263	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
		2264	an C<EV_ASYNC> event on the watcher into the event loop. Unlike
		2265	C<ev_feed_event>, this call is safe to do in other threads, signal or
		2266	similar contexts (see the dicusssion of C<EV_ATOMIC_T> in the embedding
		2267	section below on what exactly this means).
		2268
		2269	This call incurs the overhead of a syscall only once per loop iteration,
		2270	so while the overhead might be noticable, it doesn't apply to repeated
		2271	calls to C<ev_async_send>.
		2272
		2273	=back
		2274
		2275
1912	=head1 OTHER FUNCTIONS	2276	=head1 OTHER FUNCTIONS
1913		2277
1914	There are some other functions of possible interest. Described. Here. Now.	2278	There are some other functions of possible interest. Described. Here. Now.
1915		2279
1916	=over 4	2280	=over 4
…		…
2143	Example: Define a class with an IO and idle watcher, start one of them in	2507	Example: Define a class with an IO and idle watcher, start one of them in
2144	the constructor.	2508	the constructor.
2145		2509
2146	class myclass	2510	class myclass
2147	{	2511	{
2148	ev_io io; void io_cb (ev::io &w, int revents);	2512	ev::io io; void io_cb (ev::io &w, int revents);
2149	ev_idle idle void idle_cb (ev::idle &w, int revents);	2513	ev:idle idle void idle_cb (ev::idle &w, int revents);
2150		2514
2151	myclass ();	2515	myclass (int fd)
2152	}
2153
2154	myclass::myclass (int fd)
2155	{	2516	{
2156	io .set <myclass, &myclass::io_cb > (this);	2517	io .set <myclass, &myclass::io_cb > (this);
2157	idle.set <myclass, &myclass::idle_cb> (this);	2518	idle.set <myclass, &myclass::idle_cb> (this);
2158		2519
2159	io.start (fd, ev::READ);	2520	io.start (fd, ev::READ);
		2521	}
2160	}	2522	};
		2523
		2524
		2525	=head1 OTHER LANGUAGE BINDINGS
		2526
		2527	Libev does not offer other language bindings itself, but bindings for a
		2528	numbe rof languages exist in the form of third-party packages. If you know
		2529	any interesting language binding in addition to the ones listed here, drop
		2530	me a note.
		2531
		2532	=over 4
		2533
		2534	=item Perl
		2535
		2536	The EV module implements the full libev API and is actually used to test
		2537	libev. EV is developed together with libev. Apart from the EV core module,
		2538	there are additional modules that implement libev-compatible interfaces
		2539	to C<libadns> (C<EV::ADNS>), C<Net::SNMP> (C<Net::SNMP::EV>) and the
		2540	C<libglib> event core (C<Glib::EV> and C<EV::Glib>).
		2541
		2542	It can be found and installed via CPAN, its homepage is found at
		2543	L<http://software.schmorp.de/pkg/EV>.
		2544
		2545	=item Ruby
		2546
		2547	Tony Arcieri has written a ruby extension that offers access to a subset
		2548	of the libev API and adds filehandle abstractions, asynchronous DNS and
		2549	more on top of it. It can be found via gem servers. Its homepage is at
		2550	L<http://rev.rubyforge.org/>.
		2551
		2552	=item D
		2553
		2554	Leandro Lucarella has written a D language binding (F<ev.d>) for libev, to
		2555	be found at L<http://git.llucax.com.ar/?p=software/ev.d.git;a=summary>.
		2556
		2557	=back
2161		2558
2162		2559
2163	=head1 MACRO MAGIC	2560	=head1 MACRO MAGIC
2164		2561
2165	Libev can be compiled with a variety of options, the most fundamantal	2562	Libev can be compiled with a variety of options, the most fundamantal
…		…
2370	wants osf handles on win32 (this is the case when the select to	2767	wants osf handles on win32 (this is the case when the select to
2371	be used is the winsock select). This means that it will call	2768	be used is the winsock select). This means that it will call
2372	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,	2769	C<_get_osfhandle> on the fd to convert it to an OS handle. Otherwise,
2373	it is assumed that all these functions actually work on fds, even	2770	it is assumed that all these functions actually work on fds, even
2374	on win32. Should not be defined on non-win32 platforms.	2771	on win32. Should not be defined on non-win32 platforms.
		2772
		2773	=item EV_FD_TO_WIN32_HANDLE
		2774
		2775	If C<EV_SELECT_IS_WINSOCKET> is enabled, then libev needs a way to map
		2776	file descriptors to socket handles. When not defining this symbol (the
		2777	default), then libev will call C<_get_osfhandle>, which is usually
		2778	correct. In some cases, programs use their own file descriptor management,
		2779	in which case they can provide this function to map fds to socket handles.
2375		2780
2376	=item EV_USE_POLL	2781	=item EV_USE_POLL
2377		2782
2378	If defined to be C<1>, libev will compile in support for the C<poll>(2)	2783	If defined to be C<1>, libev will compile in support for the C<poll>(2)
2379	backend. Otherwise it will be enabled on non-win32 platforms. It	2784	backend. Otherwise it will be enabled on non-win32 platforms. It
…		…
2413		2818
2414	If defined to be C<1>, libev will compile in support for the Linux inotify	2819	If defined to be C<1>, libev will compile in support for the Linux inotify
2415	interface to speed up C<ev_stat> watchers. Its actual availability will	2820	interface to speed up C<ev_stat> watchers. Its actual availability will
2416	be detected at runtime.	2821	be detected at runtime.
2417		2822
		2823	=item EV_ATOMIC_T
		2824
		2825	Libev requires an integer type (suitable for storing C<0> or C<1>) whose
		2826	access is atomic with respect to other threads or signal contexts. No such
		2827	type is easily found in the C language, so you can provide your own type
		2828	that you know is safe for your purposes. It is used both for signal handler "locking"
		2829	as well as for signal and thread safety in C<ev_async> watchers.
		2830
		2831	In the absense of this define, libev will use C<sig_atomic_t volatile>
		2832	(from F<signal.h>), which is usually good enough on most platforms.
		2833
2418	=item EV_H	2834	=item EV_H
2419		2835
2420	The name of the F<ev.h> header file used to include it. The default if	2836	The name of the F<ev.h> header file used to include it. The default if
2421	undefined is C<< <ev.h> >> in F<event.h> and C<"ev.h"> in F<ev.c>. This	2837	undefined is C<"ev.h"> in F<event.h>, F<ev.c> and F<ev++.h>. This can be
2422	can be used to virtually rename the F<ev.h> header file in case of conflicts.	2838	used to virtually rename the F<ev.h> header file in case of conflicts.
2423		2839
2424	=item EV_CONFIG_H	2840	=item EV_CONFIG_H
2425		2841
2426	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override	2842	If C<EV_STANDALONE> isn't C<1>, this variable can be used to override
2427	F<ev.c>'s idea of where to find the F<config.h> file, similarly to	2843	F<ev.c>'s idea of where to find the F<config.h> file, similarly to
2428	C<EV_H>, above.	2844	C<EV_H>, above.
2429		2845
2430	=item EV_EVENT_H	2846	=item EV_EVENT_H
2431		2847
2432	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea	2848	Similarly to C<EV_H>, this macro can be used to override F<event.c>'s idea
2433	of how the F<event.h> header can be found.	2849	of how the F<event.h> header can be found, the default is C<"event.h">.
2434		2850
2435	=item EV_PROTOTYPES	2851	=item EV_PROTOTYPES
2436		2852
2437	If defined to be C<0>, then F<ev.h> will not define any function	2853	If defined to be C<0>, then F<ev.h> will not define any function
2438	prototypes, but still define all the structs and other symbols. This is	2854	prototypes, but still define all the structs and other symbols. This is
…		…
2489	=item EV_FORK_ENABLE	2905	=item EV_FORK_ENABLE
2490		2906
2491	If undefined or defined to be C<1>, then fork watchers are supported. If	2907	If undefined or defined to be C<1>, then fork watchers are supported. If
2492	defined to be C<0>, then they are not.	2908	defined to be C<0>, then they are not.
2493		2909
		2910	=item EV_ASYNC_ENABLE
		2911
		2912	If undefined or defined to be C<1>, then async watchers are supported. If
		2913	defined to be C<0>, then they are not.
		2914
2494	=item EV_MINIMAL	2915	=item EV_MINIMAL
2495		2916
2496	If you need to shave off some kilobytes of code at the expense of some	2917	If you need to shave off some kilobytes of code at the expense of some
2497	speed, define this symbol to C<1>. Currently only used for gcc to override	2918	speed, define this symbol to C<1>. Currently only used for gcc to override
2498	some inlining decisions, saves roughly 30% codesize of amd64.	2919	some inlining decisions, saves roughly 30% codesize of amd64.
…		…
2504	than enough. If you need to manage thousands of children you might want to	2925	than enough. If you need to manage thousands of children you might want to
2505	increase this value (I<must> be a power of two).	2926	increase this value (I<must> be a power of two).
2506		2927
2507	=item EV_INOTIFY_HASHSIZE	2928	=item EV_INOTIFY_HASHSIZE
2508		2929
2509	C<ev_staz> watchers use a small hash table to distribute workload by	2930	C<ev_stat> watchers use a small hash table to distribute workload by
2510	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),	2931	inotify watch id. The default size is C<16> (or C<1> with C<EV_MINIMAL>),
2511	usually more than enough. If you need to manage thousands of C<ev_stat>	2932	usually more than enough. If you need to manage thousands of C<ev_stat>
2512	watchers you might want to increase this value (I<must> be a power of	2933	watchers you might want to increase this value (I<must> be a power of
2513	two).	2934	two).
2514		2935
…		…
2610		3031
2611	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3032	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2612		3033
2613	This means that, when you have a watcher that triggers in one hour and	3034	This means that, when you have a watcher that triggers in one hour and
2614	there are 100 watchers that would trigger before that then inserting will	3035	there are 100 watchers that would trigger before that then inserting will
2615	have to skip those 100 watchers.	3036	have to skip roughly seven (C<ld 100>) of these watchers.
2616		3037
2617	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	3038	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
2618		3039
2619	That means that for changing a timer costs less than removing/adding them	3040	That means that changing a timer costs less than removing/adding them
2620	as only the relative motion in the event queue has to be paid for.	3041	as only the relative motion in the event queue has to be paid for.
2621		3042
2622	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	3043	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
2623		3044
2624	These just add the watcher into an array or at the head of a list.	3045	These just add the watcher into an array or at the head of a list.
		3046
2625	=item Stopping check/prepare/idle watchers: O(1)	3047	=item Stopping check/prepare/idle/fork/async watchers: O(1)
2626		3048
2627	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3049	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2628		3050
2629	These watchers are stored in lists then need to be walked to find the	3051	These watchers are stored in lists then need to be walked to find the
2630	correct watcher to remove. The lists are usually short (you don't usually	3052	correct watcher to remove. The lists are usually short (you don't usually
2631	have many watchers waiting for the same fd or signal).	3053	have many watchers waiting for the same fd or signal).
2632		3054
2633	=item Finding the next timer per loop iteration: O(1)	3055	=item Finding the next timer in each loop iteration: O(1)
		3056
		3057	By virtue of using a binary heap, the next timer is always found at the
		3058	beginning of the storage array.
2634		3059
2635	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	3060	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2636		3061
2637	A change means an I/O watcher gets started or stopped, which requires	3062	A change means an I/O watcher gets started or stopped, which requires
2638	libev to recalculate its status (and possibly tell the kernel).	3063	libev to recalculate its status (and possibly tell the kernel, depending
		3064	on backend and wether C<ev_io_set> was used).
2639		3065
2640	=item Activating one watcher: O(1)	3066	=item Activating one watcher (putting it into the pending state): O(1)
2641		3067
2642	=item Priority handling: O(number_of_priorities)	3068	=item Priority handling: O(number_of_priorities)
2643		3069
2644	Priorities are implemented by allocating some space for each	3070	Priorities are implemented by allocating some space for each
2645	priority. When doing priority-based operations, libev usually has to	3071	priority. When doing priority-based operations, libev usually has to
2646	linearly search all the priorities.	3072	linearly search all the priorities, but starting/stopping and activating
		3073	watchers becomes O(1) w.r.t. priority handling.
		3074
		3075	=item Sending an ev_async: O(1)
		3076
		3077	=item Processing ev_async_send: O(number_of_async_watchers)
		3078
		3079	=item Processing signals: O(max_signal_number)
		3080
		3081	Sending involves a syscall I<iff> there were no other C<ev_async_send>
		3082	calls in the current loop iteration. Checking for async and signal events
		3083	involves iterating over all running async watchers or all signal numbers.
2647		3084
2648	=back	3085	=back
2649		3086
2650		3087
		3088	=head1 Win32 platform limitations and workarounds
		3089
		3090	Win32 doesn't support any of the standards (e.g. POSIX) that libev
		3091	requires, and its I/O model is fundamentally incompatible with the POSIX
		3092	model. Libev still offers limited functionality on this platform in
		3093	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
		3094	descriptors. This only applies when using Win32 natively, not when using
		3095	e.g. cygwin.
		3096
		3097	There is no supported compilation method available on windows except
		3098	embedding it into other applications.
		3099
		3100	Due to the many, low, and arbitrary limits on the win32 platform and the
		3101	abysmal performance of winsockets, using a large number of sockets is not
		3102	recommended (and not reasonable). If your program needs to use more than
		3103	a hundred or so sockets, then likely it needs to use a totally different
		3104	implementation for windows, as libev offers the POSIX model, which cannot
		3105	be implemented efficiently on windows (microsoft monopoly games).
		3106
		3107	=over 4
		3108
		3109	=item The winsocket select function
		3110
		3111	The winsocket C<select> function doesn't follow POSIX in that it requires
		3112	socket I<handles> and not socket I<file descriptors>. This makes select
		3113	very inefficient, and also requires a mapping from file descriptors
		3114	to socket handles. See the discussion of the C<EV_SELECT_USE_FD_SET>,
		3115	C<EV_SELECT_IS_WINSOCKET> and C<EV_FD_TO_WIN32_HANDLE> preprocessor
		3116	symbols for more info.
		3117
		3118	The configuration for a "naked" win32 using the microsoft runtime
		3119	libraries and raw winsocket select is:
		3120
		3121	#define EV_USE_SELECT 1
		3122	#define EV_SELECT_IS_WINSOCKET 1 /* forces EV_SELECT_USE_FD_SET, too */
		3123
		3124	Note that winsockets handling of fd sets is O(n), so you can easily get a
		3125	complexity in the O(n²) range when using win32.
		3126
		3127	=item Limited number of file descriptors
		3128
		3129	Windows has numerous arbitrary (and low) limits on things. Early versions
		3130	of winsocket's select only supported waiting for a max. of C<64> handles
		3131	(probably owning to the fact that all windows kernels can only wait for
		3132	C<64> things at the same time internally; microsoft recommends spawning a
		3133	chain of threads and wait for 63 handles and the previous thread in each).
		3134
		3135	Newer versions support more handles, but you need to define C<FD_SETSIZE>
		3136	to some high number (e.g. C<2048>) before compiling the winsocket select
		3137	call (which might be in libev or elsewhere, for example, perl does its own
		3138	select emulation on windows).
		3139
		3140	Another limit is the number of file descriptors in the microsoft runtime
		3141	libraries, which by default is C<64> (there must be a hidden I<64> fetish
		3142	or something like this inside microsoft). You can increase this by calling
		3143	C<_setmaxstdio>, which can increase this limit to C<2048> (another
		3144	arbitrary limit), but is broken in many versions of the microsoft runtime
		3145	libraries.
		3146
		3147	This might get you to about C<512> or C<2048> sockets (depending on
		3148	windows version and/or the phase of the moon). To get more, you need to
		3149	wrap all I/O functions and provide your own fd management, but the cost of
		3150	calling select (O(n²)) will likely make this unworkable.
		3151
		3152	=back
		3153
		3154
2651	=head1 AUTHOR	3155	=head1 AUTHOR
2652		3156
2653	Marc Lehmann <libev@schmorp.de>.	3157	Marc Lehmann <libev@schmorp.de>.
2654		3158

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.99 by root, Sat Dec 22 06:16:36 2007 UTC vs. Revision 1.137 by root, Sun Mar 16 16:42:56 2008 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.99 by root, Sat Dec 22 06:16:36 2007 UTC vs.
Revision 1.137 by root, Sun Mar 16 16:42:56 2008 UTC