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

Comparing libev/ev.pod (file contents):
Revision 1.63 by root, Thu Nov 29 20:05:59 2007 UTC vs.
Revision 1.74 by root, Sat Dec 8 14:12:08 2007 UTC

…		…
47		47
48	return 0;	48	return 0;
49	}	49	}
50		50
51	=head1 DESCRIPTION	51	=head1 DESCRIPTION
		52
		53	The newest version of this document is also available as a html-formatted
		54	web page you might find easier to navigate when reading it for the first
		55	time: L<http://cvs.schmorp.de/libev/ev.html>.
52		56
53	Libev is an event loop: you register interest in certain events (such as a	57	Libev is an event loop: you register interest in certain events (such as a
54	file descriptor being readable or a timeout occuring), and it will manage	58	file descriptor being readable or a timeout occuring), and it will manage
55	these event sources and provide your program with events.	59	these event sources and provide your program with events.
56		60
…		…
274	a fork, you can also make libev check for a fork in each iteration by	278	a fork, you can also make libev check for a fork in each iteration by
275	enabling this flag.	279	enabling this flag.
276		280
277	This works by calling C<getpid ()> on every iteration of the loop,	281	This works by calling C<getpid ()> on every iteration of the loop,
278	and thus this might slow down your event loop if you do a lot of loop	282	and thus this might slow down your event loop if you do a lot of loop
279	iterations and little real work, but is usually not noticable (on my	283	iterations and little real work, but is usually not noticeable (on my
280	Linux system for example, C<getpid> is actually a simple 5-insn sequence	284	Linux system for example, C<getpid> is actually a simple 5-insn sequence
281	without a syscall and thus I<very> fast, but my Linux system also has	285	without a syscall and thus I<very> fast, but my Linux system also has
282	C<pthread_atfork> which is even faster).	286	C<pthread_atfork> which is even faster).
283		287
284	The big advantage of this flag is that you can forget about fork (and	288	The big advantage of this flag is that you can forget about fork (and
…		…
429	=item ev_loop_fork (loop)	433	=item ev_loop_fork (loop)
430		434
431	Like C<ev_default_fork>, but acts on an event loop created by	435	Like C<ev_default_fork>, but acts on an event loop created by
432	C<ev_loop_new>. Yes, you have to call this on every allocated event loop	436	C<ev_loop_new>. Yes, you have to call this on every allocated event loop
433	after fork, and how you do this is entirely your own problem.	437	after fork, and how you do this is entirely your own problem.
		438
		439	=item unsigned int ev_loop_count (loop)
		440
		441	Returns the count of loop iterations for the loop, which is identical to
		442	the number of times libev did poll for new events. It starts at C<0> and
		443	happily wraps around with enough iterations.
		444
		445	This value can sometimes be useful as a generation counter of sorts (it
		446	"ticks" the number of loop iterations), as it roughly corresponds with
		447	C<ev_prepare> and C<ev_check> calls.
434		448
435	=item unsigned int ev_backend (loop)	449	=item unsigned int ev_backend (loop)
436		450
437	Returns one of the C<EVBACKEND_*> flags indicating the event backend in	451	Returns one of the C<EVBACKEND_*> flags indicating the event backend in
438	use.	452	use.
…		…
722	=item bool ev_is_pending (ev_TYPE *watcher)	736	=item bool ev_is_pending (ev_TYPE *watcher)
723		737
724	Returns a true value iff the watcher is pending, (i.e. it has outstanding	738	Returns a true value iff the watcher is pending, (i.e. it has outstanding
725	events but its callback has not yet been invoked). As long as a watcher	739	events but its callback has not yet been invoked). As long as a watcher
726	is pending (but not active) you must not call an init function on it (but	740	is pending (but not active) you must not call an init function on it (but
727	C<ev_TYPE_set> is safe) and you must make sure the watcher is available to	741	C<ev_TYPE_set> is safe), you must not change its priority, and you must
728	libev (e.g. you cnanot C<free ()> it).	742	make sure the watcher is available to libev (e.g. you cannot C<free ()>
		743	it).
729		744
730	=item callback ev_cb (ev_TYPE *watcher)	745	=item callback ev_cb (ev_TYPE *watcher)
731		746
732	Returns the callback currently set on the watcher.	747	Returns the callback currently set on the watcher.
733		748
734	=item ev_cb_set (ev_TYPE *watcher, callback)	749	=item ev_cb_set (ev_TYPE *watcher, callback)
735		750
736	Change the callback. You can change the callback at virtually any time	751	Change the callback. You can change the callback at virtually any time
737	(modulo threads).	752	(modulo threads).
		753
		754	=item ev_set_priority (ev_TYPE *watcher, priority)
		755
		756	=item int ev_priority (ev_TYPE *watcher)
		757
		758	Set and query the priority of the watcher. The priority is a small
		759	integer between C<EV_MAXPRI> (default: C<2>) and C<EV_MINPRI>
		760	(default: C<-2>). Pending watchers with higher priority will be invoked
		761	before watchers with lower priority, but priority will not keep watchers
		762	from being executed (except for C<ev_idle> watchers).
		763
		764	This means that priorities are I<only> used for ordering callback
		765	invocation after new events have been received. This is useful, for
		766	example, to reduce latency after idling, or more often, to bind two
		767	watchers on the same event and make sure one is called first.
		768
		769	If you need to suppress invocation when higher priority events are pending
		770	you need to look at C<ev_idle> watchers, which provide this functionality.
		771
		772	You I<must not> change the priority of a watcher as long as it is active or
		773	pending.
		774
		775	The default priority used by watchers when no priority has been set is
		776	always C<0>, which is supposed to not be too high and not be too low :).
		777
		778	Setting a priority outside the range of C<EV_MINPRI> to C<EV_MAXPRI> is
		779	fine, as long as you do not mind that the priority value you query might
		780	or might not have been adjusted to be within valid range.
		781
		782	=item ev_invoke (loop, ev_TYPE *watcher, int revents)
		783
		784	Invoke the C<watcher> with the given C<loop> and C<revents>. Neither
		785	C<loop> nor C<revents> need to be valid as long as the watcher callback
		786	can deal with that fact.
		787
		788	=item int ev_clear_pending (loop, ev_TYPE *watcher)
		789
		790	If the watcher is pending, this function returns clears its pending status
		791	and returns its C<revents> bitset (as if its callback was invoked). If the
		792	watcher isn't pending it does nothing and returns C<0>.
738		793
739	=back	794	=back
740		795
741		796
742	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER	797	=head2 ASSOCIATING CUSTOM DATA WITH A WATCHER
…		…
848	it is best to always use non-blocking I/O: An extra C<read>(2) returning	903	it is best to always use non-blocking I/O: An extra C<read>(2) returning
849	C<EAGAIN> is far preferable to a program hanging until some data arrives.	904	C<EAGAIN> is far preferable to a program hanging until some data arrives.
850		905
851	If you cannot run the fd in non-blocking mode (for example you should not	906	If you cannot run the fd in non-blocking mode (for example you should not
852	play around with an Xlib connection), then you have to seperately re-test	907	play around with an Xlib connection), then you have to seperately re-test
853	wether a file descriptor is really ready with a known-to-be good interface	908	whether a file descriptor is really ready with a known-to-be good interface
854	such as poll (fortunately in our Xlib example, Xlib already does this on	909	such as poll (fortunately in our Xlib example, Xlib already does this on
855	its own, so its quite safe to use).	910	its own, so its quite safe to use).
856		911
857	=over 4	912	=over 4
858		913
…		…
1341	ev_stat_start (loop, &passwd);	1396	ev_stat_start (loop, &passwd);
1342		1397
1343		1398
1344	=head2 C<ev_idle> - when you've got nothing better to do...	1399	=head2 C<ev_idle> - when you've got nothing better to do...
1345		1400
1346	Idle watchers trigger events when there are no other events are pending	1401	Idle watchers trigger events when no other events of the same or higher
1347	(prepare, check and other idle watchers do not count). That is, as long	1402	priority are pending (prepare, check and other idle watchers do not
1348	as your process is busy handling sockets or timeouts (or even signals,	1403	count).
1349	imagine) it will not be triggered. But when your process is idle all idle	1404
1350	watchers are being called again and again, once per event loop iteration -	1405	That is, as long as your process is busy handling sockets or timeouts
		1406	(or even signals, imagine) of the same or higher priority it will not be
		1407	triggered. But when your process is idle (or only lower-priority watchers
		1408	are pending), the idle watchers are being called once per event loop
1351	until stopped, that is, or your process receives more events and becomes	1409	iteration - until stopped, that is, or your process receives more events
1352	busy.	1410	and becomes busy again with higher priority stuff.
1353		1411
1354	The most noteworthy effect is that as long as any idle watchers are	1412	The most noteworthy effect is that as long as any idle watchers are
1355	active, the process will not block when waiting for new events.	1413	active, the process will not block when waiting for new events.
1356		1414
1357	Apart from keeping your process non-blocking (which is a useful	1415	Apart from keeping your process non-blocking (which is a useful
…		…
1457		1515
1458	// create io watchers for each fd and a timer before blocking	1516	// create io watchers for each fd and a timer before blocking
1459	static void	1517	static void
1460	adns_prepare_cb (ev_loop loop, ev_prepare w, int revents)	1518	adns_prepare_cb (ev_loop loop, ev_prepare w, int revents)
1461	{	1519	{
1462	int timeout = 3600000;truct pollfd fds [nfd];	1520	int timeout = 3600000;
		1521	struct pollfd fds [nfd];
1463	// actual code will need to loop here and realloc etc.	1522	// actual code will need to loop here and realloc etc.
1464	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));	1523	adns_beforepoll (ads, fds, &nfd, &timeout, timeval_from (ev_time ()));
1465		1524
1466	/* the callback is illegal, but won't be called as we stop during check */	1525	/* the callback is illegal, but won't be called as we stop during check */
1467	ev_timer_init (&tw, 0, timeout * 1e-3);	1526	ev_timer_init (&tw, 0, timeout * 1e-3);
…		…
1701		1760
1702	To use it,	1761	To use it,
1703		1762
1704	#include <ev++.h>	1763	#include <ev++.h>
1705		1764
1706	(it is not installed by default). This automatically includes F<ev.h>	1765	This automatically includes F<ev.h> and puts all of its definitions (many
1707	and puts all of its definitions (many of them macros) into the global	1766	of them macros) into the global namespace. All C++ specific things are
1708	namespace. All C++ specific things are put into the C<ev> namespace.	1767	put into the C<ev> namespace. It should support all the same embedding
		1768	options as F<ev.h>, most notably C<EV_MULTIPLICITY>.
1709		1769
1710	It should support all the same embedding options as F<ev.h>, most notably	1770	Care has been taken to keep the overhead low. The only data member the C++
1711	C<EV_MULTIPLICITY>.	1771	classes add (compared to plain C-style watchers) is the event loop pointer
		1772	that the watcher is associated with (or no additional members at all if
		1773	you disable C<EV_MULTIPLICITY> when embedding libev).
		1774
		1775	Currently, functions, and static and non-static member functions can be
		1776	used as callbacks. Other types should be easy to add as long as they only
		1777	need one additional pointer for context. If you need support for other
		1778	types of functors please contact the author (preferably after implementing
		1779	it).
1712		1780
1713	Here is a list of things available in the C<ev> namespace:	1781	Here is a list of things available in the C<ev> namespace:
1714		1782
1715	=over 4	1783	=over 4
1716		1784
…		…
1732		1800
1733	All of those classes have these methods:	1801	All of those classes have these methods:
1734		1802
1735	=over 4	1803	=over 4
1736		1804
1737	=item ev::TYPE::TYPE (object , object::method )	1805	=item ev::TYPE::TYPE ()
1738		1806
1739	=item ev::TYPE::TYPE (object , object::method , struct ev_loop *)	1807	=item ev::TYPE::TYPE (struct ev_loop *)
1740		1808
1741	=item ev::TYPE::~TYPE	1809	=item ev::TYPE::~TYPE
1742		1810
1743	The constructor takes a pointer to an object and a method pointer to	1811	The constructor (optionally) takes an event loop to associate the watcher
1744	the event handler callback to call in this class. The constructor calls	1812	with. If it is omitted, it will use C<EV_DEFAULT>.
1745	C<ev_init> for you, which means you have to call the C<set> method	1813
1746	before starting it. If you do not specify a loop then the constructor	1814	The constructor calls C<ev_init> for you, which means you have to call the
1747	automatically associates the default loop with this watcher.	1815	C<set> method before starting it.
		1816
		1817	It will not set a callback, however: You have to call the templated C<set>
		1818	method to set a callback before you can start the watcher.
		1819
		1820	(The reason why you have to use a method is a limitation in C++ which does
		1821	not allow explicit template arguments for constructors).
1748		1822
1749	The destructor automatically stops the watcher if it is active.	1823	The destructor automatically stops the watcher if it is active.
		1824
		1825	=item w->set<class, &class::method> (object *)
		1826
		1827	This method sets the callback method to call. The method has to have a
		1828	signature of C<void (*)(ev_TYPE &, int)>, it receives the watcher as
		1829	first argument and the C<revents> as second. The object must be given as
		1830	parameter and is stored in the C<data> member of the watcher.
		1831
		1832	This method synthesizes efficient thunking code to call your method from
		1833	the C callback that libev requires. If your compiler can inline your
		1834	callback (i.e. it is visible to it at the place of the C<set> call and
		1835	your compiler is good :), then the method will be fully inlined into the
		1836	thunking function, making it as fast as a direct C callback.
		1837
		1838	Example: simple class declaration and watcher initialisation
		1839
		1840	struct myclass
		1841	{
		1842	void io_cb (ev::io &w, int revents) { }
		1843	}
		1844
		1845	myclass obj;
		1846	ev::io iow;
		1847	iow.set <myclass, &myclass::io_cb> (&obj);
		1848
		1849	=item w->set (void (function)(watcher &w, int), void data = 0)
		1850
		1851	Also sets a callback, but uses a static method or plain function as
		1852	callback. The optional C<data> argument will be stored in the watcher's
		1853	C<data> member and is free for you to use.
		1854
		1855	See the method-C<set> above for more details.
1750		1856
1751	=item w->set (struct ev_loop *)	1857	=item w->set (struct ev_loop *)
1752		1858
1753	Associates a different C<struct ev_loop> with this watcher. You can only	1859	Associates a different C<struct ev_loop> with this watcher. You can only
1754	do this when the watcher is inactive (and not pending either).	1860	do this when the watcher is inactive (and not pending either).
1755		1861
1756	=item w->set ([args])	1862	=item w->set ([args])
1757		1863
1758	Basically the same as C<ev_TYPE_set>, with the same args. Must be	1864	Basically the same as C<ev_TYPE_set>, with the same args. Must be
1759	called at least once. Unlike the C counterpart, an active watcher gets	1865	called at least once. Unlike the C counterpart, an active watcher gets
1760	automatically stopped and restarted.	1866	automatically stopped and restarted when reconfiguring it with this
		1867	method.
1761		1868
1762	=item w->start ()	1869	=item w->start ()
1763		1870
1764	Starts the watcher. Note that there is no C<loop> argument as the	1871	Starts the watcher. Note that there is no C<loop> argument, as the
1765	constructor already takes the loop.	1872	constructor already stores the event loop.
1766		1873
1767	=item w->stop ()	1874	=item w->stop ()
1768		1875
1769	Stops the watcher if it is active. Again, no C<loop> argument.	1876	Stops the watcher if it is active. Again, no C<loop> argument.
1770		1877
…		…
1795		1902
1796	myclass ();	1903	myclass ();
1797	}	1904	}
1798		1905
1799	myclass::myclass (int fd)	1906	myclass::myclass (int fd)
1800	: io (this, &myclass::io_cb),
1801	idle (this, &myclass::idle_cb)
1802	{	1907	{
		1908	io .set <myclass, &myclass::io_cb > (this);
		1909	idle.set <myclass, &myclass::idle_cb> (this);
		1910
1803	io.start (fd, ev::READ);	1911	io.start (fd, ev::READ);
1804	}	1912	}
1805		1913
1806		1914
1807	=head1 MACRO MAGIC	1915	=head1 MACRO MAGIC
1808		1916
1809	Libev can be compiled with a variety of options, the most fundemantal is	1917	Libev can be compiled with a variety of options, the most fundemantal is
1810	C<EV_MULTIPLICITY>. This option determines wether (most) functions and	1918	C<EV_MULTIPLICITY>. This option determines whether (most) functions and
1811	callbacks have an initial C<struct ev_loop *> argument.	1919	callbacks have an initial C<struct ev_loop *> argument.
1812		1920
1813	To make it easier to write programs that cope with either variant, the	1921	To make it easier to write programs that cope with either variant, the
1814	following macros are defined:	1922	following macros are defined:
1815		1923
…		…
1849	loop, if multiple loops are supported ("ev loop default").	1957	loop, if multiple loops are supported ("ev loop default").
1850		1958
1851	=back	1959	=back
1852		1960
1853	Example: Declare and initialise a check watcher, utilising the above	1961	Example: Declare and initialise a check watcher, utilising the above
1854	macros so it will work regardless of wether multiple loops are supported	1962	macros so it will work regardless of whether multiple loops are supported
1855	or not.	1963	or not.
1856		1964
1857	static void	1965	static void
1858	check_cb (EV_P_ ev_timer *w, int revents)	1966	check_cb (EV_P_ ev_timer *w, int revents)
1859	{	1967	{
…		…
2084	will have the C<struct ev_loop *> as first argument, and you can create	2192	will have the C<struct ev_loop *> as first argument, and you can create
2085	additional independent event loops. Otherwise there will be no support	2193	additional independent event loops. Otherwise there will be no support
2086	for multiple event loops and there is no first event loop pointer	2194	for multiple event loops and there is no first event loop pointer
2087	argument. Instead, all functions act on the single default loop.	2195	argument. Instead, all functions act on the single default loop.
2088		2196
		2197	=item EV_MINPRI
		2198
		2199	=item EV_MAXPRI
		2200
		2201	The range of allowed priorities. C<EV_MINPRI> must be smaller or equal to
		2202	C<EV_MAXPRI>, but otherwise there are no non-obvious limitations. You can
		2203	provide for more priorities by overriding those symbols (usually defined
		2204	to be C<-2> and C<2>, respectively).
		2205
		2206	When doing priority-based operations, libev usually has to linearly search
		2207	all the priorities, so having many of them (hundreds) uses a lot of space
		2208	and time, so using the defaults of five priorities (-2 .. +2) is usually
		2209	fine.
		2210
		2211	If your embedding app does not need any priorities, defining these both to
		2212	C<0> will save some memory and cpu.
		2213
2089	=item EV_PERIODIC_ENABLE	2214	=item EV_PERIODIC_ENABLE
2090		2215
2091	If undefined or defined to be C<1>, then periodic timers are supported. If	2216	If undefined or defined to be C<1>, then periodic timers are supported. If
		2217	defined to be C<0>, then they are not. Disabling them saves a few kB of
		2218	code.
		2219
		2220	=item EV_IDLE_ENABLE
		2221
		2222	If undefined or defined to be C<1>, then idle watchers are supported. If
2092	defined to be C<0>, then they are not. Disabling them saves a few kB of	2223	defined to be C<0>, then they are not. Disabling them saves a few kB of
2093	code.	2224	code.
2094		2225
2095	=item EV_EMBED_ENABLE	2226	=item EV_EMBED_ENABLE
2096		2227
…		…
2189		2320
2190	In this section the complexities of (many of) the algorithms used inside	2321	In this section the complexities of (many of) the algorithms used inside
2191	libev will be explained. For complexity discussions about backends see the	2322	libev will be explained. For complexity discussions about backends see the
2192	documentation for C<ev_default_init>.	2323	documentation for C<ev_default_init>.
2193		2324
		2325	All of the following are about amortised time: If an array needs to be
		2326	extended, libev needs to realloc and move the whole array, but this
		2327	happens asymptotically never with higher number of elements, so O(1) might
		2328	mean it might do a lengthy realloc operation in rare cases, but on average
		2329	it is much faster and asymptotically approaches constant time.
		2330
2194	=over 4	2331	=over 4
2195		2332
2196	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	2333	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
2197		2334
		2335	This means that, when you have a watcher that triggers in one hour and
		2336	there are 100 watchers that would trigger before that then inserting will
		2337	have to skip those 100 watchers.
		2338
2198	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)	2339	=item Changing timer/periodic watchers (by autorepeat, again): O(log skipped_other_timers)
2199		2340
		2341	That means that for changing a timer costs less than removing/adding them
		2342	as only the relative motion in the event queue has to be paid for.
		2343
2200	=item Starting io/check/prepare/idle/signal/child watchers: O(1)	2344	=item Starting io/check/prepare/idle/signal/child watchers: O(1)
2201		2345
		2346	These just add the watcher into an array or at the head of a list.
2202	=item Stopping check/prepare/idle watchers: O(1)	2347	=item Stopping check/prepare/idle watchers: O(1)
2203		2348
2204	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	2349	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
2205		2350
		2351	These watchers are stored in lists then need to be walked to find the
		2352	correct watcher to remove. The lists are usually short (you don't usually
		2353	have many watchers waiting for the same fd or signal).
		2354
2206	=item Finding the next timer per loop iteration: O(1)	2355	=item Finding the next timer per loop iteration: O(1)
2207		2356
2208	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)	2357	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
2209		2358
		2359	A change means an I/O watcher gets started or stopped, which requires
		2360	libev to recalculate its status (and possibly tell the kernel).
		2361
2210	=item Activating one watcher: O(1)	2362	=item Activating one watcher: O(1)
2211		2363
		2364	=item Priority handling: O(number_of_priorities)
		2365
		2366	Priorities are implemented by allocating some space for each
		2367	priority. When doing priority-based operations, libev usually has to
		2368	linearly search all the priorities.
		2369
2212	=back	2370	=back
2213		2371
2214		2372
2215	=head1 AUTHOR	2373	=head1 AUTHOR
2216		2374

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.63 by root, Thu Nov 29 20:05:59 2007 UTC vs. Revision 1.74 by root, Sat Dec 8 14:12:08 2007 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.63 by root, Thu Nov 29 20:05:59 2007 UTC vs.
Revision 1.74 by root, Sat Dec 8 14:12:08 2007 UTC