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

Comparing libev/ev.pod (file contents):
Revision 1.345 by root, Wed Nov 10 14:36:42 2010 UTC vs.
Revision 1.370 by root, Thu Jun 2 23:42:40 2011 UTC

…		…
58	ev_timer_start (loop, &timeout_watcher);	58	ev_timer_start (loop, &timeout_watcher);
59		59
60	// now wait for events to arrive	60	// now wait for events to arrive
61	ev_run (loop, 0);	61	ev_run (loop, 0);
62		62
63	// unloop was called, so exit	63	// break was called, so exit
64	return 0;	64	return 0;
65	}	65	}
66		66
67	=head1 ABOUT THIS DOCUMENT	67	=head1 ABOUT THIS DOCUMENT
68		68
…		…
299	}	299	}
300		300
301	...	301	...
302	ev_set_syserr_cb (fatal_error);	302	ev_set_syserr_cb (fatal_error);
303		303
		304	=item ev_feed_signal (int signum)
		305
		306	This function can be used to "simulate" a signal receive. It is completely
		307	safe to call this function at any time, from any context, including signal
		308	handlers or random threads.
		309
		310	Its main use is to customise signal handling in your process, especially
		311	in the presence of threads. For example, you could block signals
		312	by default in all threads (and specifying C<EVFLAG_NOSIGMASK> when
		313	creating any loops), and in one thread, use C<sigwait> or any other
		314	mechanism to wait for signals, then "deliver" them to libev by calling
		315	C<ev_feed_signal>.
		316
304	=back	317	=back
305		318
306	=head1 FUNCTIONS CONTROLLING EVENT LOOPS	319	=head1 FUNCTIONS CONTROLLING EVENT LOOPS
307		320
308	An event loop is described by a C<struct ev_loop *> (the C<struct> is	321	An event loop is described by a C<struct ev_loop *> (the C<struct> is
…		…
419		432
420	Signalfd will not be used by default as this changes your signal mask, and	433	Signalfd will not be used by default as this changes your signal mask, and
421	there are a lot of shoddy libraries and programs (glib's threadpool for	434	there are a lot of shoddy libraries and programs (glib's threadpool for
422	example) that can't properly initialise their signal masks.	435	example) that can't properly initialise their signal masks.
423		436
		437	=item C<EVFLAG_NOSIGMASK>
		438
		439	When this flag is specified, then libev will avoid to modify the signal
		440	mask. Specifically, this means you ahve to make sure signals are unblocked
		441	when you want to receive them.
		442
		443	This behaviour is useful when you want to do your own signal handling, or
		444	want to handle signals only in specific threads and want to avoid libev
		445	unblocking the signals.
		446
		447	It's also required by POSIX in a threaded program, as libev calls
		448	C<sigprocmask>, whose behaviour is officially unspecified.
		449
		450	This flag's behaviour will become the default in future versions of libev.
		451
424	=item C<EVBACKEND_SELECT> (value 1, portable select backend)	452	=item C<EVBACKEND_SELECT> (value 1, portable select backend)
425		453
426	This is your standard select(2) backend. Not I<completely> standard, as	454	This is your standard select(2) backend. Not I<completely> standard, as
427	libev tries to roll its own fd_set with no limits on the number of fds,	455	libev tries to roll its own fd_set with no limits on the number of fds,
428	but if that fails, expect a fairly low limit on the number of fds when	456	but if that fails, expect a fairly low limit on the number of fds when
…		…
455	=item C<EVBACKEND_EPOLL> (value 4, Linux)	483	=item C<EVBACKEND_EPOLL> (value 4, Linux)
456		484
457	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9	485	Use the linux-specific epoll(7) interface (for both pre- and post-2.6.9
458	kernels).	486	kernels).
459		487
460	For few fds, this backend is a bit little slower than poll and select,	488	For few fds, this backend is a bit little slower than poll and select, but
461	but it scales phenomenally better. While poll and select usually scale	489	it scales phenomenally better. While poll and select usually scale like
462	like O(total_fds) where n is the total number of fds (or the highest fd),	490	O(total_fds) where total_fds is the total number of fds (or the highest
463	epoll scales either O(1) or O(active_fds).	491	fd), epoll scales either O(1) or O(active_fds).
464		492
465	The epoll mechanism deserves honorable mention as the most misdesigned	493	The epoll mechanism deserves honorable mention as the most misdesigned
466	of the more advanced event mechanisms: mere annoyances include silently	494	of the more advanced event mechanisms: mere annoyances include silently
467	dropping file descriptors, requiring a system call per change per file	495	dropping file descriptors, requiring a system call per change per file
468	descriptor (and unnecessary guessing of parameters), problems with dup,	496	descriptor (and unnecessary guessing of parameters), problems with dup,
…		…
471	0.1ms) and so on. The biggest issue is fork races, however - if a program	499	0.1ms) and so on. The biggest issue is fork races, however - if a program
472	forks then I<both> parent and child process have to recreate the epoll	500	forks then I<both> parent and child process have to recreate the epoll
473	set, which can take considerable time (one syscall per file descriptor)	501	set, which can take considerable time (one syscall per file descriptor)
474	and is of course hard to detect.	502	and is of course hard to detect.
475		503
476	Epoll is also notoriously buggy - embedding epoll fds I<should> work, but	504	Epoll is also notoriously buggy - embedding epoll fds I<should> work,
477	of course I<doesn't>, and epoll just loves to report events for totally	505	but of course I<doesn't>, and epoll just loves to report events for
478	I<different> file descriptors (even already closed ones, so one cannot	506	totally I<different> file descriptors (even already closed ones, so
479	even remove them from the set) than registered in the set (especially	507	one cannot even remove them from the set) than registered in the set
480	on SMP systems). Libev tries to counter these spurious notifications by	508	(especially on SMP systems). Libev tries to counter these spurious
481	employing an additional generation counter and comparing that against the	509	notifications by employing an additional generation counter and comparing
482	events to filter out spurious ones, recreating the set when required. Last	510	that against the events to filter out spurious ones, recreating the set
		511	when required. Epoll also errornously rounds down timeouts, but gives you
		512	no way to know when and by how much, so sometimes you have to busy-wait
		513	because epoll returns immediately despite a nonzero timeout. And last
483	not least, it also refuses to work with some file descriptors which work	514	not least, it also refuses to work with some file descriptors which work
484	perfectly fine with C<select> (files, many character devices...).	515	perfectly fine with C<select> (files, many character devices...).
485		516
486	Epoll is truly the train wreck analog among event poll mechanisms.	517	Epoll is truly the train wreck among event poll mechanisms, a frankenpoll,
		518	cobbled together in a hurry, no thought to design or interaction with
		519	others. Oh, the pain, will it ever stop...
487		520
488	While stopping, setting and starting an I/O watcher in the same iteration	521	While stopping, setting and starting an I/O watcher in the same iteration
489	will result in some caching, there is still a system call per such	522	will result in some caching, there is still a system call per such
490	incident (because the same I<file descriptor> could point to a different	523	incident (because the same I<file descriptor> could point to a different
491	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed	524	I<file description> now), so its best to avoid that. Also, C<dup ()>'ed
…		…
557	=item C<EVBACKEND_PORT> (value 32, Solaris 10)	590	=item C<EVBACKEND_PORT> (value 32, Solaris 10)
558		591
559	This uses the Solaris 10 event port mechanism. As with everything on Solaris,	592	This uses the Solaris 10 event port mechanism. As with everything on Solaris,
560	it's really slow, but it still scales very well (O(active_fds)).	593	it's really slow, but it still scales very well (O(active_fds)).
561		594
562	Please note that Solaris event ports can deliver a lot of spurious
563	notifications, so you need to use non-blocking I/O or other means to avoid
564	blocking when no data (or space) is available.
565
566	While this backend scales well, it requires one system call per active	595	While this backend scales well, it requires one system call per active
567	file descriptor per loop iteration. For small and medium numbers of file	596	file descriptor per loop iteration. For small and medium numbers of file
568	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend	597	descriptors a "slow" C<EVBACKEND_SELECT> or C<EVBACKEND_POLL> backend
569	might perform better.	598	might perform better.
570		599
571	On the positive side, with the exception of the spurious readiness	600	On the positive side, this backend actually performed fully to
572	notifications, this backend actually performed fully to specification
573	in all tests and is fully embeddable, which is a rare feat among the	601	specification in all tests and is fully embeddable, which is a rare feat
574	OS-specific backends (I vastly prefer correctness over speed hacks).	602	among the OS-specific backends (I vastly prefer correctness over speed
		603	hacks).
		604
		605	On the negative side, the interface is I<bizarre> - so bizarre that
		606	even sun itself gets it wrong in their code examples: The event polling
		607	function sometimes returning events to the caller even though an error
		608	occurred, but with no indication whether it has done so or not (yes, it's
		609	even documented that way) - deadly for edge-triggered interfaces where
		610	you absolutely have to know whether an event occurred or not because you
		611	have to re-arm the watcher.
		612
		613	Fortunately libev seems to be able to work around these idiocies.
575		614
576	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as	615	This backend maps C<EV_READ> and C<EV_WRITE> in the same way as
577	C<EVBACKEND_POLL>.	616	C<EVBACKEND_POLL>.
578		617
579	=item C<EVBACKEND_ALL>	618	=item C<EVBACKEND_ALL>
580		619
581	Try all backends (even potentially broken ones that wouldn't be tried	620	Try all backends (even potentially broken ones that wouldn't be tried
582	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as	621	with C<EVFLAG_AUTO>). Since this is a mask, you can do stuff such as
583	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.	622	C<EVBACKEND_ALL & ~EVBACKEND_KQUEUE>.
584		623
585	It is definitely not recommended to use this flag.	624	It is definitely not recommended to use this flag, use whatever
		625	C<ev_recommended_backends ()> returns, or simply do not specify a backend
		626	at all.
		627
		628	=item C<EVBACKEND_MASK>
		629
		630	Not a backend at all, but a mask to select all backend bits from a
		631	C<flags> value, in case you want to mask out any backends from a flags
		632	value (e.g. when modifying the C<LIBEV_FLAGS> environment variable).
586		633
587	=back	634	=back
588		635
589	If one or more of the backend flags are or'ed into the flags value,	636	If one or more of the backend flags are or'ed into the flags value,
590	then only these backends will be tried (in the reverse order as listed	637	then only these backends will be tried (in the reverse order as listed
…		…
781	This is useful if you are waiting for some external event in conjunction	828	This is useful if you are waiting for some external event in conjunction
782	with something not expressible using other libev watchers (i.e. "roll your	829	with something not expressible using other libev watchers (i.e. "roll your
783	own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is	830	own C<ev_run>"). However, a pair of C<ev_prepare>/C<ev_check> watchers is
784	usually a better approach for this kind of thing.	831	usually a better approach for this kind of thing.
785		832
786	Here are the gory details of what C<ev_run> does:	833	Here are the gory details of what C<ev_run> does (this is for your
		834	understanding, not a guarantee that things will work exactly like this in
		835	future versions):
787		836
788	- Increment loop depth.	837	- Increment loop depth.
789	- Reset the ev_break status.	838	- Reset the ev_break status.
790	- Before the first iteration, call any pending watchers.	839	- Before the first iteration, call any pending watchers.
791	LOOP:	840	LOOP:
…		…
824	anymore.	873	anymore.
825		874
826	... queue jobs here, make sure they register event watchers as long	875	... queue jobs here, make sure they register event watchers as long
827	... as they still have work to do (even an idle watcher will do..)	876	... as they still have work to do (even an idle watcher will do..)
828	ev_run (my_loop, 0);	877	ev_run (my_loop, 0);
829	... jobs done or somebody called unloop. yeah!	878	... jobs done or somebody called break. yeah!
830		879
831	=item ev_break (loop, how)	880	=item ev_break (loop, how)
832		881
833	Can be used to make a call to C<ev_run> return early (but only after it	882	Can be used to make a call to C<ev_run> return early (but only after it
834	has processed all outstanding events). The C<how> argument must be either	883	has processed all outstanding events). The C<how> argument must be either
…		…
867	running when nothing else is active.	916	running when nothing else is active.
868		917
869	ev_signal exitsig;	918	ev_signal exitsig;
870	ev_signal_init (&exitsig, sig_cb, SIGINT);	919	ev_signal_init (&exitsig, sig_cb, SIGINT);
871	ev_signal_start (loop, &exitsig);	920	ev_signal_start (loop, &exitsig);
872	evf_unref (loop);	921	ev_unref (loop);
873		922
874	Example: For some weird reason, unregister the above signal handler again.	923	Example: For some weird reason, unregister the above signal handler again.
875		924
876	ev_ref (loop);	925	ev_ref (loop);
877	ev_signal_stop (loop, &exitsig);	926	ev_signal_stop (loop, &exitsig);
…		…
1316	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related	1365	See also C<ev_feed_fd_event> and C<ev_feed_signal_event> for related
1317	functions that do not need a watcher.	1366	functions that do not need a watcher.
1318		1367
1319	=back	1368	=back
1320		1369
1321	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	1370	See also the L<ASSOCIATING CUSTOM DATA WITH A WATCHER> and L<BUILDING YOUR
1322		1371	OWN COMPOSITE WATCHERS> idioms.
1323	Each watcher has, by default, a member C<void *data> that you can change
1324	and read at any time: libev will completely ignore it. This can be used
1325	to associate arbitrary data with your watcher. If you need more data and
1326	don't want to allocate memory and store a pointer to it in that data
1327	member, you can also "subclass" the watcher type and provide your own
1328	data:
1329
1330	struct my_io
1331	{
1332	ev_io io;
1333	int otherfd;
1334	void *somedata;
1335	struct whatever *mostinteresting;
1336	};
1337
1338	...
1339	struct my_io w;
1340	ev_io_init (&w.io, my_cb, fd, EV_READ);
1341
1342	And since your callback will be called with a pointer to the watcher, you
1343	can cast it back to your own type:
1344
1345	static void my_cb (struct ev_loop loop, ev_io w_, int revents)
1346	{
1347	struct my_io w = (struct my_io )w_;
1348	...
1349	}
1350
1351	More interesting and less C-conformant ways of casting your callback type
1352	instead have been omitted.
1353
1354	Another common scenario is to use some data structure with multiple
1355	embedded watchers:
1356
1357	struct my_biggy
1358	{
1359	int some_data;
1360	ev_timer t1;
1361	ev_timer t2;
1362	}
1363
1364	In this case getting the pointer to C<my_biggy> is a bit more
1365	complicated: Either you store the address of your C<my_biggy> struct
1366	in the C<data> member of the watcher (for woozies), or you need to use
1367	some pointer arithmetic using C<offsetof> inside your watchers (for real
1368	programmers):
1369
1370	#include <stddef.h>
1371
1372	static void
1373	t1_cb (EV_P_ ev_timer *w, int revents)
1374	{
1375	struct my_biggy big = (struct my_biggy *)
1376	(((char *)w) - offsetof (struct my_biggy, t1));
1377	}
1378
1379	static void
1380	t2_cb (EV_P_ ev_timer *w, int revents)
1381	{
1382	struct my_biggy big = (struct my_biggy *)
1383	(((char *)w) - offsetof (struct my_biggy, t2));
1384	}
1385		1372
1386	=head2 WATCHER STATES	1373	=head2 WATCHER STATES
1387		1374
1388	There are various watcher states mentioned throughout this manual -	1375	There are various watcher states mentioned throughout this manual -
1389	active, pending and so on. In this section these states and the rules to	1376	active, pending and so on. In this section these states and the rules to
…		…
1396		1383
1397	Before a watcher can be registered with the event looop it has to be	1384	Before a watcher can be registered with the event looop it has to be
1398	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to	1385	initialised. This can be done with a call to C<ev_TYPE_init>, or calls to
1399	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.	1386	C<ev_init> followed by the watcher-specific C<ev_TYPE_set> function.
1400		1387
1401	In this state it is simply some block of memory that is suitable for use	1388	In this state it is simply some block of memory that is suitable for
1402	in an event loop. It can be moved around, freed, reused etc. at will.	1389	use in an event loop. It can be moved around, freed, reused etc. at
		1390	will - as long as you either keep the memory contents intact, or call
		1391	C<ev_TYPE_init> again.
1403		1392
1404	=item started/running/active	1393	=item started/running/active
1405		1394
1406	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes	1395	Once a watcher has been started with a call to C<ev_TYPE_start> it becomes
1407	property of the event loop, and is actively waiting for events. While in	1396	property of the event loop, and is actively waiting for events. While in
…		…
1435	latter will clear any pending state the watcher might be in, regardless	1424	latter will clear any pending state the watcher might be in, regardless
1436	of whether it was active or not, so stopping a watcher explicitly before	1425	of whether it was active or not, so stopping a watcher explicitly before
1437	freeing it is often a good idea.	1426	freeing it is often a good idea.
1438		1427
1439	While stopped (and not pending) the watcher is essentially in the	1428	While stopped (and not pending) the watcher is essentially in the
1440	initialised state, that is it can be reused, moved, modified in any way	1429	initialised state, that is, it can be reused, moved, modified in any way
1441	you wish.	1430	you wish (but when you trash the memory block, you need to C<ev_TYPE_init>
		1431	it again).
1442		1432
1443	=back	1433	=back
1444		1434
1445	=head2 WATCHER PRIORITY MODELS	1435	=head2 WATCHER PRIORITY MODELS
1446		1436
…		…
1575	In general you can register as many read and/or write event watchers per	1565	In general you can register as many read and/or write event watchers per
1576	fd as you want (as long as you don't confuse yourself). Setting all file	1566	fd as you want (as long as you don't confuse yourself). Setting all file
1577	descriptors to non-blocking mode is also usually a good idea (but not	1567	descriptors to non-blocking mode is also usually a good idea (but not
1578	required if you know what you are doing).	1568	required if you know what you are doing).
1579		1569
1580	If you cannot use non-blocking mode, then force the use of a
1581	known-to-be-good backend (at the time of this writing, this includes only
1582	C<EVBACKEND_SELECT> and C<EVBACKEND_POLL>). The same applies to file
1583	descriptors for which non-blocking operation makes no sense (such as
1584	files) - libev doesn't guarantee any specific behaviour in that case.
1585
1586	Another thing you have to watch out for is that it is quite easy to	1570	Another thing you have to watch out for is that it is quite easy to
1587	receive "spurious" readiness notifications, that is your callback might	1571	receive "spurious" readiness notifications, that is, your callback might
1588	be called with C<EV_READ> but a subsequent C<read>(2) will actually block	1572	be called with C<EV_READ> but a subsequent C<read>(2) will actually block
1589	because there is no data. Not only are some backends known to create a	1573	because there is no data. It is very easy to get into this situation even
1590	lot of those (for example Solaris ports), it is very easy to get into	1574	with a relatively standard program structure. Thus it is best to always
1591	this situation even with a relatively standard program structure. Thus	1575	use non-blocking I/O: An extra C<read>(2) returning C<EAGAIN> is far
1592	it is best to always use non-blocking I/O: An extra C<read>(2) returning
1593	C<EAGAIN> is far preferable to a program hanging until some data arrives.	1576	preferable to a program hanging until some data arrives.
1594		1577
1595	If you cannot run the fd in non-blocking mode (for example you should	1578	If you cannot run the fd in non-blocking mode (for example you should
1596	not play around with an Xlib connection), then you have to separately	1579	not play around with an Xlib connection), then you have to separately
1597	re-test whether a file descriptor is really ready with a known-to-be good	1580	re-test whether a file descriptor is really ready with a known-to-be good
1598	interface such as poll (fortunately in our Xlib example, Xlib already	1581	interface such as poll (fortunately in the case of Xlib, it already does
1599	does this on its own, so its quite safe to use). Some people additionally	1582	this on its own, so its quite safe to use). Some people additionally
1600	use C<SIGALRM> and an interval timer, just to be sure you won't block	1583	use C<SIGALRM> and an interval timer, just to be sure you won't block
1601	indefinitely.	1584	indefinitely.
1602		1585
1603	But really, best use non-blocking mode.	1586	But really, best use non-blocking mode.
1604		1587
…		…
1632		1615
1633	There is no workaround possible except not registering events	1616	There is no workaround possible except not registering events
1634	for potentially C<dup ()>'ed file descriptors, or to resort to	1617	for potentially C<dup ()>'ed file descriptors, or to resort to
1635	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.	1618	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
1636		1619
		1620	=head3 The special problem of files
		1621
		1622	Many people try to use C<select> (or libev) on file descriptors
		1623	representing files, and expect it to become ready when their program
		1624	doesn't block on disk accesses (which can take a long time on their own).
		1625
		1626	However, this cannot ever work in the "expected" way - you get a readiness
		1627	notification as soon as the kernel knows whether and how much data is
		1628	there, and in the case of open files, that's always the case, so you
		1629	always get a readiness notification instantly, and your read (or possibly
		1630	write) will still block on the disk I/O.
		1631
		1632	Another way to view it is that in the case of sockets, pipes, character
		1633	devices and so on, there is another party (the sender) that delivers data
		1634	on its own, but in the case of files, there is no such thing: the disk
		1635	will not send data on its own, simply because it doesn't know what you
		1636	wish to read - you would first have to request some data.
		1637
		1638	Since files are typically not-so-well supported by advanced notification
		1639	mechanism, libev tries hard to emulate POSIX behaviour with respect
		1640	to files, even though you should not use it. The reason for this is
		1641	convenience: sometimes you want to watch STDIN or STDOUT, which is
		1642	usually a tty, often a pipe, but also sometimes files or special devices
		1643	(for example, C<epoll> on Linux works with F</dev/random> but not with
		1644	F</dev/urandom>), and even though the file might better be served with
		1645	asynchronous I/O instead of with non-blocking I/O, it is still useful when
		1646	it "just works" instead of freezing.
		1647
		1648	So avoid file descriptors pointing to files when you know it (e.g. use
		1649	libeio), but use them when it is convenient, e.g. for STDIN/STDOUT, or
		1650	when you rarely read from a file instead of from a socket, and want to
		1651	reuse the same code path.
		1652
1637	=head3 The special problem of fork	1653	=head3 The special problem of fork
1638		1654
1639	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit	1655	Some backends (epoll, kqueue) do not support C<fork ()> at all or exhibit
1640	useless behaviour. Libev fully supports fork, but needs to be told about	1656	useless behaviour. Libev fully supports fork, but needs to be told about
1641	it in the child.	1657	it in the child if you want to continue to use it in the child.
1642		1658
1643	To support fork in your programs, you either have to call	1659	To support fork in your child processes, you have to call C<ev_loop_fork
1644	C<ev_default_fork ()> or C<ev_loop_fork ()> after a fork in the child,	1660	()> after a fork in the child, enable C<EVFLAG_FORKCHECK>, or resort to
1645	enable C<EVFLAG_FORKCHECK>, or resort to C<EVBACKEND_SELECT> or	1661	C<EVBACKEND_SELECT> or C<EVBACKEND_POLL>.
1646	C<EVBACKEND_POLL>.
1647		1662
1648	=head3 The special problem of SIGPIPE	1663	=head3 The special problem of SIGPIPE
1649		1664
1650	While not really specific to libev, it is easy to forget about C<SIGPIPE>:	1665	While not really specific to libev, it is easy to forget about C<SIGPIPE>:
1651	when writing to a pipe whose other end has been closed, your program gets	1666	when writing to a pipe whose other end has been closed, your program gets
…		…
2141		2156
2142	Another way to think about it (for the mathematically inclined) is that	2157	Another way to think about it (for the mathematically inclined) is that
2143	C<ev_periodic> will try to run the callback in this mode at the next possible	2158	C<ev_periodic> will try to run the callback in this mode at the next possible
2144	time where C<time = offset (mod interval)>, regardless of any time jumps.	2159	time where C<time = offset (mod interval)>, regardless of any time jumps.
2145		2160
2146	For numerical stability it is preferable that the C<offset> value is near	2161	The C<interval> I<MUST> be positive, and for numerical stability, the
2147	C<ev_now ()> (the current time), but there is no range requirement for	2162	interval value should be higher than C<1/8192> (which is around 100
2148	this value, and in fact is often specified as zero.	2163	microseconds) and C<offset> should be higher than C<0> and should have
		2164	at most a similar magnitude as the current time (say, within a factor of
		2165	ten). Typical values for offset are, in fact, C<0> or something between
		2166	C<0> and C<interval>, which is also the recommended range.
2149		2167
2150	Note also that there is an upper limit to how often a timer can fire (CPU	2168	Note also that there is an upper limit to how often a timer can fire (CPU
2151	speed for example), so if C<interval> is very small then timing stability	2169	speed for example), so if C<interval> is very small then timing stability
2152	will of course deteriorate. Libev itself tries to be exact to be about one	2170	will of course deteriorate. Libev itself tries to be exact to be about one
2153	millisecond (if the OS supports it and the machine is fast enough).	2171	millisecond (if the OS supports it and the machine is fast enough).
…		…
2296	=head3 The special problem of inheritance over fork/execve/pthread_create	2314	=head3 The special problem of inheritance over fork/execve/pthread_create
2297		2315
2298	Both the signal mask (C<sigprocmask>) and the signal disposition	2316	Both the signal mask (C<sigprocmask>) and the signal disposition
2299	(C<sigaction>) are unspecified after starting a signal watcher (and after	2317	(C<sigaction>) are unspecified after starting a signal watcher (and after
2300	stopping it again), that is, libev might or might not block the signal,	2318	stopping it again), that is, libev might or might not block the signal,
2301	and might or might not set or restore the installed signal handler.	2319	and might or might not set or restore the installed signal handler (but
		2320	see C<EVFLAG_NOSIGMASK>).
2302		2321
2303	While this does not matter for the signal disposition (libev never	2322	While this does not matter for the signal disposition (libev never
2304	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on	2323	sets signals to C<SIG_IGN>, so handlers will be reset to C<SIG_DFL> on
2305	C<execve>), this matters for the signal mask: many programs do not expect	2324	C<execve>), this matters for the signal mask: many programs do not expect
2306	certain signals to be blocked.	2325	certain signals to be blocked.
…		…
2319	I<has> to modify the signal mask, at least temporarily.	2338	I<has> to modify the signal mask, at least temporarily.
2320		2339
2321	So I can't stress this enough: I<If you do not reset your signal mask when	2340	So I can't stress this enough: I<If you do not reset your signal mask when
2322	you expect it to be empty, you have a race condition in your code>. This	2341	you expect it to be empty, you have a race condition in your code>. This
2323	is not a libev-specific thing, this is true for most event libraries.	2342	is not a libev-specific thing, this is true for most event libraries.
		2343
		2344	=head3 The special problem of threads signal handling
		2345
		2346	POSIX threads has problematic signal handling semantics, specifically,
		2347	a lot of functionality (sigfd, sigwait etc.) only really works if all
		2348	threads in a process block signals, which is hard to achieve.
		2349
		2350	When you want to use sigwait (or mix libev signal handling with your own
		2351	for the same signals), you can tackle this problem by globally blocking
		2352	all signals before creating any threads (or creating them with a fully set
		2353	sigprocmask) and also specifying the C<EVFLAG_NOSIGMASK> when creating
		2354	loops. Then designate one thread as "signal receiver thread" which handles
		2355	these signals. You can pass on any signals that libev might be interested
		2356	in by calling C<ev_feed_signal>.
2324		2357
2325	=head3 Watcher-Specific Functions and Data Members	2358	=head3 Watcher-Specific Functions and Data Members
2326		2359
2327	=over 4	2360	=over 4
2328		2361
…		…
3163	atexit (program_exits);	3196	atexit (program_exits);
3164		3197
3165		3198
3166	=head2 C<ev_async> - how to wake up an event loop	3199	=head2 C<ev_async> - how to wake up an event loop
3167		3200
3168	In general, you cannot use an C<ev_run> from multiple threads or other	3201	In general, you cannot use an C<ev_loop> from multiple threads or other
3169	asynchronous sources such as signal handlers (as opposed to multiple event	3202	asynchronous sources such as signal handlers (as opposed to multiple event
3170	loops - those are of course safe to use in different threads).	3203	loops - those are of course safe to use in different threads).
3171		3204
3172	Sometimes, however, you need to wake up an event loop you do not control,	3205	Sometimes, however, you need to wake up an event loop you do not control,
3173	for example because it belongs to another thread. This is what C<ev_async>	3206	for example because it belongs to another thread. This is what C<ev_async>
…		…
3175	it by calling C<ev_async_send>, which is thread- and signal safe.	3208	it by calling C<ev_async_send>, which is thread- and signal safe.
3176		3209
3177	This functionality is very similar to C<ev_signal> watchers, as signals,	3210	This functionality is very similar to C<ev_signal> watchers, as signals,
3178	too, are asynchronous in nature, and signals, too, will be compressed	3211	too, are asynchronous in nature, and signals, too, will be compressed
3179	(i.e. the number of callback invocations may be less than the number of	3212	(i.e. the number of callback invocations may be less than the number of
3180	C<ev_async_sent> calls).	3213	C<ev_async_sent> calls). In fact, you could use signal watchers as a kind
		3214	of "global async watchers" by using a watcher on an otherwise unused
		3215	signal, and C<ev_feed_signal> to signal this watcher from another thread,
		3216	even without knowing which loop owns the signal.
3181		3217
3182	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not	3218	Unlike C<ev_signal> watchers, C<ev_async> works with any event loop, not
3183	just the default loop.	3219	just the default loop.
3184		3220
3185	=head3 Queueing	3221	=head3 Queueing
…		…
3280	trust me.	3316	trust me.
3281		3317
3282	=item ev_async_send (loop, ev_async *)	3318	=item ev_async_send (loop, ev_async *)
3283		3319
3284	Sends/signals/activates the given C<ev_async> watcher, that is, feeds	3320	Sends/signals/activates the given C<ev_async> watcher, that is, feeds
3285	an C<EV_ASYNC> event on the watcher into the event loop. Unlike	3321	an C<EV_ASYNC> event on the watcher into the event loop, and instantly
		3322	returns.
		3323
3286	C<ev_feed_event>, this call is safe to do from other threads, signal or	3324	Unlike C<ev_feed_event>, this call is safe to do from other threads,
3287	similar contexts (see the discussion of C<EV_ATOMIC_T> in the embedding	3325	signal or similar contexts (see the discussion of C<EV_ATOMIC_T> in the
3288	section below on what exactly this means).	3326	embedding section below on what exactly this means).
3289		3327
3290	Note that, as with other watchers in libev, multiple events might get	3328	Note that, as with other watchers in libev, multiple events might get
3291	compressed into a single callback invocation (another way to look at this	3329	compressed into a single callback invocation (another way to look at this
3292	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,	3330	is that C<ev_async> watchers are level-triggered, set on C<ev_async_send>,
3293	reset when the event loop detects that).	3331	reset when the event loop detects that).
…		…
3361	Feed an event on the given fd, as if a file descriptor backend detected	3399	Feed an event on the given fd, as if a file descriptor backend detected
3362	the given events it.	3400	the given events it.
3363		3401
3364	=item ev_feed_signal_event (loop, int signum)	3402	=item ev_feed_signal_event (loop, int signum)
3365		3403
3366	Feed an event as if the given signal occurred (C<loop> must be the default	3404	Feed an event as if the given signal occurred. See also C<ev_feed_signal>,
3367	loop!).	3405	which is async-safe.
3368		3406
3369	=back	3407	=back
3370		3408
3371		3409
3372	=head1 COMMON OR USEFUL IDIOMS (OR BOTH)	3410	=head1 COMMON OR USEFUL IDIOMS (OR BOTH)
3373		3411
3374	This section explains some common idioms that are not immediately	3412	This section explains some common idioms that are not immediately
3375	obvious. Note that examples are sprinkled over the whole manual, and this	3413	obvious. Note that examples are sprinkled over the whole manual, and this
3376	section only contains stuff that wouldn't fit anywhere else.	3414	section only contains stuff that wouldn't fit anywhere else.
3377		3415
3378	=over 4	3416	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
3379		3417
3380	=item Model/nested event loop invocations and exit conditions.	3418	Each watcher has, by default, a C<void *data> member that you can read
		3419	or modify at any time: libev will completely ignore it. This can be used
		3420	to associate arbitrary data with your watcher. If you need more data and
		3421	don't want to allocate memory separately and store a pointer to it in that
		3422	data member, you can also "subclass" the watcher type and provide your own
		3423	data:
		3424
		3425	struct my_io
		3426	{
		3427	ev_io io;
		3428	int otherfd;
		3429	void *somedata;
		3430	struct whatever *mostinteresting;
		3431	};
		3432
		3433	...
		3434	struct my_io w;
		3435	ev_io_init (&w.io, my_cb, fd, EV_READ);
		3436
		3437	And since your callback will be called with a pointer to the watcher, you
		3438	can cast it back to your own type:
		3439
		3440	static void my_cb (struct ev_loop loop, ev_io w_, int revents)
		3441	{
		3442	struct my_io w = (struct my_io )w_;
		3443	...
		3444	}
		3445
		3446	More interesting and less C-conformant ways of casting your callback
		3447	function type instead have been omitted.
		3448
		3449	=head2 BUILDING YOUR OWN COMPOSITE WATCHERS
		3450
		3451	Another common scenario is to use some data structure with multiple
		3452	embedded watchers, in effect creating your own watcher that combines
		3453	multiple libev event sources into one "super-watcher":
		3454
		3455	struct my_biggy
		3456	{
		3457	int some_data;
		3458	ev_timer t1;
		3459	ev_timer t2;
		3460	}
		3461
		3462	In this case getting the pointer to C<my_biggy> is a bit more
		3463	complicated: Either you store the address of your C<my_biggy> struct in
		3464	the C<data> member of the watcher (for woozies or C++ coders), or you need
		3465	to use some pointer arithmetic using C<offsetof> inside your watchers (for
		3466	real programmers):
		3467
		3468	#include <stddef.h>
		3469
		3470	static void
		3471	t1_cb (EV_P_ ev_timer *w, int revents)
		3472	{
		3473	struct my_biggy big = (struct my_biggy *)
		3474	(((char *)w) - offsetof (struct my_biggy, t1));
		3475	}
		3476
		3477	static void
		3478	t2_cb (EV_P_ ev_timer *w, int revents)
		3479	{
		3480	struct my_biggy big = (struct my_biggy *)
		3481	(((char *)w) - offsetof (struct my_biggy, t2));
		3482	}
		3483
		3484	=head2 MODEL/NESTED EVENT LOOP INVOCATIONS AND EXIT CONDITIONS
3381		3485
3382	Often (especially in GUI toolkits) there are places where you have	3486	Often (especially in GUI toolkits) there are places where you have
3383	I<modal> interaction, which is most easily implemented by recursively	3487	I<modal> interaction, which is most easily implemented by recursively
3384	invoking C<ev_run>.	3488	invoking C<ev_run>.
3385		3489
…		…
3414	exit_main_loop = 1;	3518	exit_main_loop = 1;
3415		3519
3416	// exit both	3520	// exit both
3417	exit_main_loop = exit_nested_loop = 1;	3521	exit_main_loop = exit_nested_loop = 1;
3418		3522
3419	=back	3523	=head2 THREAD LOCKING EXAMPLE
		3524
		3525	Here is a fictitious example of how to run an event loop in a different
		3526	thread from where callbacks are being invoked and watchers are
		3527	created/added/removed.
		3528
		3529	For a real-world example, see the C<EV::Loop::Async> perl module,
		3530	which uses exactly this technique (which is suited for many high-level
		3531	languages).
		3532
		3533	The example uses a pthread mutex to protect the loop data, a condition
		3534	variable to wait for callback invocations, an async watcher to notify the
		3535	event loop thread and an unspecified mechanism to wake up the main thread.
		3536
		3537	First, you need to associate some data with the event loop:
		3538
		3539	typedef struct {
		3540	mutex_t lock; /* global loop lock */
		3541	ev_async async_w;
		3542	thread_t tid;
		3543	cond_t invoke_cv;
		3544	} userdata;
		3545
		3546	void prepare_loop (EV_P)
		3547	{
		3548	// for simplicity, we use a static userdata struct.
		3549	static userdata u;
		3550
		3551	ev_async_init (&u->async_w, async_cb);
		3552	ev_async_start (EV_A_ &u->async_w);
		3553
		3554	pthread_mutex_init (&u->lock, 0);
		3555	pthread_cond_init (&u->invoke_cv, 0);
		3556
		3557	// now associate this with the loop
		3558	ev_set_userdata (EV_A_ u);
		3559	ev_set_invoke_pending_cb (EV_A_ l_invoke);
		3560	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
		3561
		3562	// then create the thread running ev_run
		3563	pthread_create (&u->tid, 0, l_run, EV_A);
		3564	}
		3565
		3566	The callback for the C<ev_async> watcher does nothing: the watcher is used
		3567	solely to wake up the event loop so it takes notice of any new watchers
		3568	that might have been added:
		3569
		3570	static void
		3571	async_cb (EV_P_ ev_async *w, int revents)
		3572	{
		3573	// just used for the side effects
		3574	}
		3575
		3576	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
		3577	protecting the loop data, respectively.
		3578
		3579	static void
		3580	l_release (EV_P)
		3581	{
		3582	userdata *u = ev_userdata (EV_A);
		3583	pthread_mutex_unlock (&u->lock);
		3584	}
		3585
		3586	static void
		3587	l_acquire (EV_P)
		3588	{
		3589	userdata *u = ev_userdata (EV_A);
		3590	pthread_mutex_lock (&u->lock);
		3591	}
		3592
		3593	The event loop thread first acquires the mutex, and then jumps straight
		3594	into C<ev_run>:
		3595
		3596	void *
		3597	l_run (void *thr_arg)
		3598	{
		3599	struct ev_loop loop = (struct ev_loop )thr_arg;
		3600
		3601	l_acquire (EV_A);
		3602	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
		3603	ev_run (EV_A_ 0);
		3604	l_release (EV_A);
		3605
		3606	return 0;
		3607	}
		3608
		3609	Instead of invoking all pending watchers, the C<l_invoke> callback will
		3610	signal the main thread via some unspecified mechanism (signals? pipe
		3611	writes? C<Async::Interrupt>?) and then waits until all pending watchers
		3612	have been called (in a while loop because a) spurious wakeups are possible
		3613	and b) skipping inter-thread-communication when there are no pending
		3614	watchers is very beneficial):
		3615
		3616	static void
		3617	l_invoke (EV_P)
		3618	{
		3619	userdata *u = ev_userdata (EV_A);
		3620
		3621	while (ev_pending_count (EV_A))
		3622	{
		3623	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
		3624	pthread_cond_wait (&u->invoke_cv, &u->lock);
		3625	}
		3626	}
		3627
		3628	Now, whenever the main thread gets told to invoke pending watchers, it
		3629	will grab the lock, call C<ev_invoke_pending> and then signal the loop
		3630	thread to continue:
		3631
		3632	static void
		3633	real_invoke_pending (EV_P)
		3634	{
		3635	userdata *u = ev_userdata (EV_A);
		3636
		3637	pthread_mutex_lock (&u->lock);
		3638	ev_invoke_pending (EV_A);
		3639	pthread_cond_signal (&u->invoke_cv);
		3640	pthread_mutex_unlock (&u->lock);
		3641	}
		3642
		3643	Whenever you want to start/stop a watcher or do other modifications to an
		3644	event loop, you will now have to lock:
		3645
		3646	ev_timer timeout_watcher;
		3647	userdata *u = ev_userdata (EV_A);
		3648
		3649	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
		3650
		3651	pthread_mutex_lock (&u->lock);
		3652	ev_timer_start (EV_A_ &timeout_watcher);
		3653	ev_async_send (EV_A_ &u->async_w);
		3654	pthread_mutex_unlock (&u->lock);
		3655
		3656	Note that sending the C<ev_async> watcher is required because otherwise
		3657	an event loop currently blocking in the kernel will have no knowledge
		3658	about the newly added timer. By waking up the loop it will pick up any new
		3659	watchers in the next event loop iteration.
		3660
		3661	=head2 THREADS, COROUTINES, CONTINUATIONS, QUEUES... INSTEAD OF CALLBACKS
		3662
		3663	While the overhead of a callback that e.g. schedules a thread is small, it
		3664	is still an overhead. If you embed libev, and your main usage is with some
		3665	kind of threads or coroutines, you might want to customise libev so that
		3666	doesn't need callbacks anymore.
		3667
		3668	Imagine you have coroutines that you can switch to using a function
		3669	C<switch_to (coro)>, that libev runs in a coroutine called C<libev_coro>
		3670	and that due to some magic, the currently active coroutine is stored in a
		3671	global called C<current_coro>. Then you can build your own "wait for libev
		3672	event" primitive by changing C<EV_CB_DECLARE> and C<EV_CB_INVOKE> (note
		3673	the differing C<;> conventions):
		3674
		3675	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3676	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb)
		3677
		3678	That means instead of having a C callback function, you store the
		3679	coroutine to switch to in each watcher, and instead of having libev call
		3680	your callback, you instead have it switch to that coroutine.
		3681
		3682	A coroutine might now wait for an event with a function called
		3683	C<wait_for_event>. (the watcher needs to be started, as always, but it doesn't
		3684	matter when, or whether the watcher is active or not when this function is
		3685	called):
		3686
		3687	void
		3688	wait_for_event (ev_watcher *w)
		3689	{
		3690	ev_cb_set (w) = current_coro;
		3691	switch_to (libev_coro);
		3692	}
		3693
		3694	That basically suspends the coroutine inside C<wait_for_event> and
		3695	continues the libev coroutine, which, when appropriate, switches back to
		3696	this or any other coroutine. I am sure if you sue this your own :)
		3697
		3698	You can do similar tricks if you have, say, threads with an event queue -
		3699	instead of storing a coroutine, you store the queue object and instead of
		3700	switching to a coroutine, you push the watcher onto the queue and notify
		3701	any waiters.
		3702
		3703	To embed libev, see L<EMBEDDING>, but in short, it's easiest to create two
		3704	files, F<my_ev.h> and F<my_ev.c> that include the respective libev files:
		3705
		3706	// my_ev.h
		3707	#define EV_CB_DECLARE(type) struct my_coro *cb;
		3708	#define EV_CB_INVOKE(watcher) switch_to ((watcher)->cb);
		3709	#include "../libev/ev.h"
		3710
		3711	// my_ev.c
		3712	#define EV_H "my_ev.h"
		3713	#include "../libev/ev.c"
		3714
		3715	And then use F<my_ev.h> when you would normally use F<ev.h>, and compile
		3716	F<my_ev.c> into your project. When properly specifying include paths, you
		3717	can even use F<ev.h> as header file name directly.
3420		3718
3421		3719
3422	=head1 LIBEVENT EMULATION	3720	=head1 LIBEVENT EMULATION
3423		3721
3424	Libev offers a compatibility emulation layer for libevent. It cannot	3722	Libev offers a compatibility emulation layer for libevent. It cannot
…		…
3427	=over 4	3725	=over 4
3428		3726
3429	=item * Only the libevent-1.4.1-beta API is being emulated.	3727	=item * Only the libevent-1.4.1-beta API is being emulated.
3430		3728
3431	This was the newest libevent version available when libev was implemented,	3729	This was the newest libevent version available when libev was implemented,
3432	and is still mostly uncanged in 2010.	3730	and is still mostly unchanged in 2010.
3433		3731
3434	=item * Use it by including <event.h>, as usual.	3732	=item * Use it by including <event.h>, as usual.
3435		3733
3436	=item * The following members are fully supported: ev_base, ev_callback,	3734	=item * The following members are fully supported: ev_base, ev_callback,
3437	ev_arg, ev_fd, ev_res, ev_events.	3735	ev_arg, ev_fd, ev_res, ev_events.
…		…
3472	Care has been taken to keep the overhead low. The only data member the C++	3770	Care has been taken to keep the overhead low. The only data member the C++
3473	classes add (compared to plain C-style watchers) is the event loop pointer	3771	classes add (compared to plain C-style watchers) is the event loop pointer
3474	that the watcher is associated with (or no additional members at all if	3772	that the watcher is associated with (or no additional members at all if
3475	you disable C<EV_MULTIPLICITY> when embedding libev).	3773	you disable C<EV_MULTIPLICITY> when embedding libev).
3476		3774
3477	Currently, functions, and static and non-static member functions can be	3775	Currently, functions, static and non-static member functions and classes
3478	used as callbacks. Other types should be easy to add as long as they only	3776	with C<operator ()> can be used as callbacks. Other types should be easy
3479	need one additional pointer for context. If you need support for other	3777	to add as long as they only need one additional pointer for context. If
3480	types of functors please contact the author (preferably after implementing	3778	you need support for other types of functors please contact the author
3481	it).	3779	(preferably after implementing it).
3482		3780
3483	Here is a list of things available in the C<ev> namespace:	3781	Here is a list of things available in the C<ev> namespace:
3484		3782
3485	=over 4	3783	=over 4
3486		3784
…		…
3914	F<event.h> that are not directly supported by the libev core alone.	4212	F<event.h> that are not directly supported by the libev core alone.
3915		4213
3916	In standalone mode, libev will still try to automatically deduce the	4214	In standalone mode, libev will still try to automatically deduce the
3917	configuration, but has to be more conservative.	4215	configuration, but has to be more conservative.
3918		4216
		4217	=item EV_USE_FLOOR
		4218
		4219	If defined to be C<1>, libev will use the C<floor ()> function for its
		4220	periodic reschedule calculations, otherwise libev will fall back on a
		4221	portable (slower) implementation. If you enable this, you usually have to
		4222	link against libm or something equivalent. Enabling this when the C<floor>
		4223	function is not available will fail, so the safe default is to not enable
		4224	this.
		4225
3919	=item EV_USE_MONOTONIC	4226	=item EV_USE_MONOTONIC
3920		4227
3921	If defined to be C<1>, libev will try to detect the availability of the	4228	If defined to be C<1>, libev will try to detect the availability of the
3922	monotonic clock option at both compile time and runtime. Otherwise no	4229	monotonic clock option at both compile time and runtime. Otherwise no
3923	use of the monotonic clock option will be attempted. If you enable this,	4230	use of the monotonic clock option will be attempted. If you enable this,
…		…
4354	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	4661	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
4355		4662
4356	#include "ev_cpp.h"	4663	#include "ev_cpp.h"
4357	#include "ev.c"	4664	#include "ev.c"
4358		4665
4359	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES	4666	=head1 INTERACTION WITH OTHER PROGRAMS, LIBRARIES OR THE ENVIRONMENT
4360		4667
4361	=head2 THREADS AND COROUTINES	4668	=head2 THREADS AND COROUTINES
4362		4669
4363	=head3 THREADS	4670	=head3 THREADS
4364		4671
…		…
4415	default loop and triggering an C<ev_async> watcher from the default loop	4722	default loop and triggering an C<ev_async> watcher from the default loop
4416	watcher callback into the event loop interested in the signal.	4723	watcher callback into the event loop interested in the signal.
4417		4724
4418	=back	4725	=back
4419		4726
4420	=head4 THREAD LOCKING EXAMPLE	4727	See also L<THREAD LOCKING EXAMPLE>.
4421
4422	Here is a fictitious example of how to run an event loop in a different
4423	thread than where callbacks are being invoked and watchers are
4424	created/added/removed.
4425
4426	For a real-world example, see the C<EV::Loop::Async> perl module,
4427	which uses exactly this technique (which is suited for many high-level
4428	languages).
4429
4430	The example uses a pthread mutex to protect the loop data, a condition
4431	variable to wait for callback invocations, an async watcher to notify the
4432	event loop thread and an unspecified mechanism to wake up the main thread.
4433
4434	First, you need to associate some data with the event loop:
4435
4436	typedef struct {
4437	mutex_t lock; /* global loop lock */
4438	ev_async async_w;
4439	thread_t tid;
4440	cond_t invoke_cv;
4441	} userdata;
4442
4443	void prepare_loop (EV_P)
4444	{
4445	// for simplicity, we use a static userdata struct.
4446	static userdata u;
4447
4448	ev_async_init (&u->async_w, async_cb);
4449	ev_async_start (EV_A_ &u->async_w);
4450
4451	pthread_mutex_init (&u->lock, 0);
4452	pthread_cond_init (&u->invoke_cv, 0);
4453
4454	// now associate this with the loop
4455	ev_set_userdata (EV_A_ u);
4456	ev_set_invoke_pending_cb (EV_A_ l_invoke);
4457	ev_set_loop_release_cb (EV_A_ l_release, l_acquire);
4458
4459	// then create the thread running ev_loop
4460	pthread_create (&u->tid, 0, l_run, EV_A);
4461	}
4462
4463	The callback for the C<ev_async> watcher does nothing: the watcher is used
4464	solely to wake up the event loop so it takes notice of any new watchers
4465	that might have been added:
4466
4467	static void
4468	async_cb (EV_P_ ev_async *w, int revents)
4469	{
4470	// just used for the side effects
4471	}
4472
4473	The C<l_release> and C<l_acquire> callbacks simply unlock/lock the mutex
4474	protecting the loop data, respectively.
4475
4476	static void
4477	l_release (EV_P)
4478	{
4479	userdata *u = ev_userdata (EV_A);
4480	pthread_mutex_unlock (&u->lock);
4481	}
4482
4483	static void
4484	l_acquire (EV_P)
4485	{
4486	userdata *u = ev_userdata (EV_A);
4487	pthread_mutex_lock (&u->lock);
4488	}
4489
4490	The event loop thread first acquires the mutex, and then jumps straight
4491	into C<ev_run>:
4492
4493	void *
4494	l_run (void *thr_arg)
4495	{
4496	struct ev_loop loop = (struct ev_loop )thr_arg;
4497
4498	l_acquire (EV_A);
4499	pthread_setcanceltype (PTHREAD_CANCEL_ASYNCHRONOUS, 0);
4500	ev_run (EV_A_ 0);
4501	l_release (EV_A);
4502
4503	return 0;
4504	}
4505
4506	Instead of invoking all pending watchers, the C<l_invoke> callback will
4507	signal the main thread via some unspecified mechanism (signals? pipe
4508	writes? C<Async::Interrupt>?) and then waits until all pending watchers
4509	have been called (in a while loop because a) spurious wakeups are possible
4510	and b) skipping inter-thread-communication when there are no pending
4511	watchers is very beneficial):
4512
4513	static void
4514	l_invoke (EV_P)
4515	{
4516	userdata *u = ev_userdata (EV_A);
4517
4518	while (ev_pending_count (EV_A))
4519	{
4520	wake_up_other_thread_in_some_magic_or_not_so_magic_way ();
4521	pthread_cond_wait (&u->invoke_cv, &u->lock);
4522	}
4523	}
4524
4525	Now, whenever the main thread gets told to invoke pending watchers, it
4526	will grab the lock, call C<ev_invoke_pending> and then signal the loop
4527	thread to continue:
4528
4529	static void
4530	real_invoke_pending (EV_P)
4531	{
4532	userdata *u = ev_userdata (EV_A);
4533
4534	pthread_mutex_lock (&u->lock);
4535	ev_invoke_pending (EV_A);
4536	pthread_cond_signal (&u->invoke_cv);
4537	pthread_mutex_unlock (&u->lock);
4538	}
4539
4540	Whenever you want to start/stop a watcher or do other modifications to an
4541	event loop, you will now have to lock:
4542
4543	ev_timer timeout_watcher;
4544	userdata *u = ev_userdata (EV_A);
4545
4546	ev_timer_init (&timeout_watcher, timeout_cb, 5.5, 0.);
4547
4548	pthread_mutex_lock (&u->lock);
4549	ev_timer_start (EV_A_ &timeout_watcher);
4550	ev_async_send (EV_A_ &u->async_w);
4551	pthread_mutex_unlock (&u->lock);
4552
4553	Note that sending the C<ev_async> watcher is required because otherwise
4554	an event loop currently blocking in the kernel will have no knowledge
4555	about the newly added timer. By waking up the loop it will pick up any new
4556	watchers in the next event loop iteration.
4557		4728
4558	=head3 COROUTINES	4729	=head3 COROUTINES
4559		4730
4560	Libev is very accommodating to coroutines ("cooperative threads"):	4731	Libev is very accommodating to coroutines ("cooperative threads"):
4561	libev fully supports nesting calls to its functions from different	4732	libev fully supports nesting calls to its functions from different
…		…
5070	The physical time that is observed. It is apparently strictly monotonic :)	5241	The physical time that is observed. It is apparently strictly monotonic :)
5071		5242
5072	=item wall-clock time	5243	=item wall-clock time
5073		5244
5074	The time and date as shown on clocks. Unlike real time, it can actually	5245	The time and date as shown on clocks. Unlike real time, it can actually
5075	be wrong and jump forwards and backwards, e.g. when the you adjust your	5246	be wrong and jump forwards and backwards, e.g. when you adjust your
5076	clock.	5247	clock.
5077		5248
5078	=item watcher	5249	=item watcher
5079		5250
5080	A data structure that describes interest in certain events. Watchers need	5251	A data structure that describes interest in certain events. Watchers need
…		…
5083	=back	5254	=back
5084		5255
5085	=head1 AUTHOR	5256	=head1 AUTHOR
5086		5257
5087	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael	5258	Marc Lehmann <libev@schmorp.de>, with repeated corrections by Mikael
5088	Magnusson and Emanuele Giaquinta.	5259	Magnusson and Emanuele Giaquinta, and minor corrections by many others.
5089		5260

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.345 by root, Wed Nov 10 14:36:42 2010 UTC vs. Revision 1.370 by root, Thu Jun 2 23:42:40 2011 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.345 by root, Wed Nov 10 14:36:42 2010 UTC vs.
Revision 1.370 by root, Thu Jun 2 23:42:40 2011 UTC