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

Comparing libev/ev.pod (file contents):
Revision 1.185 by root, Tue Sep 23 09:13:59 2008 UTC vs.
Revision 1.195 by root, Mon Oct 20 17:50:48 2008 UTC

…		…
214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for	214	C<ev_embeddable_backends () & ev_supported_backends ()>, likewise for
215	recommended ones.	215	recommended ones.
216		216
217	See the description of C<ev_embed> watchers for more info.	217	See the description of C<ev_embed> watchers for more info.
218		218
219	=item ev_set_allocator (void (cb)(void *ptr, long size))	219	=item ev_set_allocator (void (cb)(void *ptr, long size)) [NOT REENTRANT]
220		220
221	Sets the allocation function to use (the prototype is similar - the	221	Sets the allocation function to use (the prototype is similar - the
222	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is	222	semantics are identical to the C<realloc> C89/SuS/POSIX function). It is
223	used to allocate and free memory (no surprises here). If it returns zero	223	used to allocate and free memory (no surprises here). If it returns zero
224	when memory needs to be allocated (C<size != 0>), the library might abort	224	when memory needs to be allocated (C<size != 0>), the library might abort
…		…
250	}	250	}
251		251
252	...	252	...
253	ev_set_allocator (persistent_realloc);	253	ev_set_allocator (persistent_realloc);
254		254
255	=item ev_set_syserr_cb (void (cb)(const char msg));	255	=item ev_set_syserr_cb (void (cb)(const char msg)); [NOT REENTRANT]
256		256
257	Set the callback function to call on a retryable system call error (such	257	Set the callback function to call on a retryable system call error (such
258	as failed select, poll, epoll_wait). The message is a printable string	258	as failed select, poll, epoll_wait). The message is a printable string
259	indicating the system call or subsystem causing the problem. If this	259	indicating the system call or subsystem causing the problem. If this
260	callback is set, then libev will expect it to remedy the situation, no	260	callback is set, then libev will expect it to remedy the situation, no
…		…
685	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or	685	C<EVUNLOOP_ONE>, which will make the innermost C<ev_loop> call return, or
686	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.	686	C<EVUNLOOP_ALL>, which will make all nested C<ev_loop> calls return.
687		687
688	This "unloop state" will be cleared when entering C<ev_loop> again.	688	This "unloop state" will be cleared when entering C<ev_loop> again.
689		689
		690	It is safe to call C<ev_unloop> from otuside any C<ev_loop> calls.
		691
690	=item ev_ref (loop)	692	=item ev_ref (loop)
691		693
692	=item ev_unref (loop)	694	=item ev_unref (loop)
693		695
694	Ref/unref can be used to add or remove a reference count on the event	696	Ref/unref can be used to add or remove a reference count on the event
…		…
967		969
968	ev_io_start (EV_DEFAULT_UC, &w);	970	ev_io_start (EV_DEFAULT_UC, &w);
969		971
970	=item C<ev_TYPE_stop> (loop , ev_TYPE watcher)	972	=item C<ev_TYPE_stop> (loop , ev_TYPE watcher)
971		973
972	Stops the given watcher again (if active) and clears the pending	974	Stops the given watcher if active, and clears the pending status (whether
		975	the watcher was active or not).
		976
973	status. It is possible that stopped watchers are pending (for example,	977	It is possible that stopped watchers are pending - for example,
974	non-repeating timers are being stopped when they become pending), but	978	non-repeating timers are being stopped when they become pending - but
975	C<ev_TYPE_stop> ensures that the watcher is neither active nor pending. If	979	calling C<ev_TYPE_stop> ensures that the watcher is neither active nor
976	you want to free or reuse the memory used by the watcher it is therefore a	980	pending. If you want to free or reuse the memory used by the watcher it is
977	good idea to always call its C<ev_TYPE_stop> function.	981	therefore a good idea to always call its C<ev_TYPE_stop> function.
978		982
979	=item bool ev_is_active (ev_TYPE *watcher)	983	=item bool ev_is_active (ev_TYPE *watcher)
980		984
981	Returns a true value iff the watcher is active (i.e. it has been started	985	Returns a true value iff the watcher is active (i.e. it has been started
982	and not yet been stopped). As long as a watcher is active you must not modify	986	and not yet been stopped). As long as a watcher is active you must not modify
…		…
1625		1629
1626	=back	1630	=back
1627		1631
1628	=head3 Examples	1632	=head3 Examples
1629		1633
1630	Example: Try to exit cleanly on SIGINT and SIGTERM.	1634	Example: Try to exit cleanly on SIGINT.
1631		1635
1632	static void	1636	static void
1633	sigint_cb (struct ev_loop loop, struct ev_signal w, int revents)	1637	sigint_cb (struct ev_loop loop, struct ev_signal w, int revents)
1634	{	1638	{
1635	ev_unloop (loop, EVUNLOOP_ALL);	1639	ev_unloop (loop, EVUNLOOP_ALL);
1636	}	1640	}
1637		1641
1638	struct ev_signal signal_watcher;	1642	struct ev_signal signal_watcher;
1639	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);	1643	ev_signal_init (&signal_watcher, sigint_cb, SIGINT);
1640	ev_signal_start (loop, &sigint_cb);	1644	ev_signal_start (loop, &signal_watcher);
1641		1645
1642		1646
1643	=head2 C<ev_child> - watch out for process status changes	1647	=head2 C<ev_child> - watch out for process status changes
1644		1648
1645	Child watchers trigger when your process receives a SIGCHLD in response to	1649	Child watchers trigger when your process receives a SIGCHLD in response to
…		…
2242	So when you want to use this feature you will always have to be prepared	2246	So when you want to use this feature you will always have to be prepared
2243	that you cannot get an embeddable loop. The recommended way to get around	2247	that you cannot get an embeddable loop. The recommended way to get around
2244	this is to have a separate variables for your embeddable loop, try to	2248	this is to have a separate variables for your embeddable loop, try to
2245	create it, and if that fails, use the normal loop for everything.	2249	create it, and if that fails, use the normal loop for everything.
2246		2250
		2251	=head3 C<ev_embed> and fork
		2252
		2253	While the C<ev_embed> watcher is running, forks in the embedding loop will
		2254	automatically be applied to the embedded loop as well, so no special
		2255	fork handling is required in that case. When the watcher is not running,
		2256	however, it is still the task of the libev user to call C<ev_loop_fork ()>
		2257	as applicable.
		2258
2247	=head3 Watcher-Specific Functions and Data Members	2259	=head3 Watcher-Specific Functions and Data Members
2248		2260
2249	=over 4	2261	=over 4
2250		2262
2251	=item ev_embed_init (ev_embed , callback, struct ev_loop embedded_loop)	2263	=item ev_embed_init (ev_embed , callback, struct ev_loop embedded_loop)
…		…
2376	=over 4	2388	=over 4
2377		2389
2378	=item queueing from a signal handler context	2390	=item queueing from a signal handler context
2379		2391
2380	To implement race-free queueing, you simply add to the queue in the signal	2392	To implement race-free queueing, you simply add to the queue in the signal
2381	handler but you block the signal handler in the watcher callback. Here is an example that does that for	2393	handler but you block the signal handler in the watcher callback. Here is
2382	some fictitious SIGUSR1 handler:	2394	an example that does that for some fictitious SIGUSR1 handler:
2383		2395
2384	static ev_async mysig;	2396	static ev_async mysig;
2385		2397
2386	static void	2398	static void
2387	sigusr1_handler (void)	2399	sigusr1_handler (void)
…		…
2494	=over 4	2506	=over 4
2495		2507
2496	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)	2508	=item ev_once (loop, int fd, int events, ev_tstamp timeout, callback)
2497		2509
2498	This function combines a simple timer and an I/O watcher, calls your	2510	This function combines a simple timer and an I/O watcher, calls your
2499	callback on whichever event happens first and automatically stop both	2511	callback on whichever event happens first and automatically stops both
2500	watchers. This is useful if you want to wait for a single event on an fd	2512	watchers. This is useful if you want to wait for a single event on an fd
2501	or timeout without having to allocate/configure/start/stop/free one or	2513	or timeout without having to allocate/configure/start/stop/free one or
2502	more watchers yourself.	2514	more watchers yourself.
2503		2515
2504	If C<fd> is less than 0, then no I/O watcher will be started and events	2516	If C<fd> is less than 0, then no I/O watcher will be started and the
2505	is being ignored. Otherwise, an C<ev_io> watcher for the given C<fd> and	2517	C<events> argument is being ignored. Otherwise, an C<ev_io> watcher for
2506	C<events> set will be created and started.	2518	the given C<fd> and C<events> set will be created and started.
2507		2519
2508	If C<timeout> is less than 0, then no timeout watcher will be	2520	If C<timeout> is less than 0, then no timeout watcher will be
2509	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and	2521	started. Otherwise an C<ev_timer> watcher with after = C<timeout> (and
2510	repeat = 0) will be started. While C<0> is a valid timeout, it is of	2522	repeat = 0) will be started. C<0> is a valid timeout.
2511	dubious value.
2512		2523
2513	The callback has the type C<void (cb)(int revents, void arg)> and gets	2524	The callback has the type C<void (cb)(int revents, void arg)> and gets
2514	passed an C<revents> set like normal event callbacks (a combination of	2525	passed an C<revents> set like normal event callbacks (a combination of
2515	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>	2526	C<EV_ERROR>, C<EV_READ>, C<EV_WRITE> or C<EV_TIMEOUT>) and the C<arg>
2516	value passed to C<ev_once>:	2527	value passed to C<ev_once>. Note that it is possible to receive I<both>
		2528	a timeout and an io event at the same time - you probably should give io
		2529	events precedence.
		2530
		2531	Example: wait up to ten seconds for data to appear on STDIN_FILENO.
2517		2532
2518	static void stdin_ready (int revents, void *arg)	2533	static void stdin_ready (int revents, void *arg)
2519	{	2534	{
		2535	if (revents & EV_READ)
		2536	/* stdin might have data for us, joy! */;
2520	if (revents & EV_TIMEOUT)	2537	else if (revents & EV_TIMEOUT)
2521	/* doh, nothing entered */;	2538	/* doh, nothing entered */;
2522	else if (revents & EV_READ)
2523	/* stdin might have data for us, joy! */;
2524	}	2539	}
2525		2540
2526	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);	2541	ev_once (STDIN_FILENO, EV_READ, 10., stdin_ready, 0);
2527		2542
2528	=item ev_feed_event (ev_loop , watcher , int revents)	2543	=item ev_feed_event (ev_loop , watcher , int revents)
…		…
3298	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:	3313	And a F<ev_cpp.C> implementation file that contains libev proper and is compiled:
3299		3314
3300	#include "ev_cpp.h"	3315	#include "ev_cpp.h"
3301	#include "ev.c"	3316	#include "ev.c"
3302		3317
		3318	=head1 INTERACTION WITH OTHER PROGRAMS OR LIBRARIES
3303		3319
3304	=head1 THREADS AND COROUTINES	3320	=head2 THREADS AND COROUTINES
3305		3321
3306	=head2 THREADS	3322	=head3 THREADS
3307		3323
3308	Libev itself is thread-safe (unless the opposite is specifically	3324	All libev functions are reentrant and thread-safe unless explicitly
3309	documented for a function), but it uses no locking itself. This means that	3325	documented otherwise, but libev implements no locking itself. This means
3310	you can use as many loops as you want in parallel, as long as only one	3326	that you can use as many loops as you want in parallel, as long as there
3311	thread ever calls into one libev function with the same loop parameter:	3327	are no concurrent calls into any libev function with the same loop
		3328	parameter (C<ev_default_*> calls have an implicit default loop parameter,
3312	libev guarantees that different event loops share no data structures that	3329	of course): libev guarantees that different event loops share no data
3313	need locking.	3330	structures that need any locking.
3314		3331
3315	Or to put it differently: calls with different loop parameters can be done	3332	Or to put it differently: calls with different loop parameters can be done
3316	concurrently from multiple threads, calls with the same loop parameter	3333	concurrently from multiple threads, calls with the same loop parameter
3317	must be done serially (but can be done from different threads, as long as	3334	must be done serially (but can be done from different threads, as long as
3318	only one thread ever is inside a call at any point in time, e.g. by using	3335	only one thread ever is inside a call at any point in time, e.g. by using
3319	a mutex per loop).	3336	a mutex per loop).
3320		3337
3321	Specifically to support threads (and signal handlers), libev implements	3338	Specifically to support threads (and signal handlers), libev implements
3322	so-called C<ev_async> watchers, which allow some limited form of	3339	so-called C<ev_async> watchers, which allow some limited form of
3323	concurrency on the same event loop.	3340	concurrency on the same event loop, namely waking it up "from the
		3341	outside".
3324		3342
3325	If you want to know which design (one loop, locking, or multiple loops	3343	If you want to know which design (one loop, locking, or multiple loops
3326	without or something else still) is best for your problem, then I cannot	3344	without or something else still) is best for your problem, then I cannot
3327	help you. I can give some generic advice however:	3345	help you, but here is some generic advice:
3328		3346
3329	=over 4	3347	=over 4
3330		3348
3331	=item * most applications have a main thread: use the default libev loop	3349	=item * most applications have a main thread: use the default libev loop
3332	in that thread, or create a separate thread running only the default loop.	3350	in that thread, or create a separate thread running only the default loop.
…		…
3356	default loop and triggering an C<ev_async> watcher from the default loop	3374	default loop and triggering an C<ev_async> watcher from the default loop
3357	watcher callback into the event loop interested in the signal.	3375	watcher callback into the event loop interested in the signal.
3358		3376
3359	=back	3377	=back
3360		3378
3361	=head2 COROUTINES	3379	=head3 COROUTINES
3362		3380
3363	Libev is much more accommodating to coroutines ("cooperative threads"):	3381	Libev is very accommodating to coroutines ("cooperative threads"):
3364	libev fully supports nesting calls to it's functions from different	3382	libev fully supports nesting calls to its functions from different
3365	coroutines (e.g. you can call C<ev_loop> on the same loop from two	3383	coroutines (e.g. you can call C<ev_loop> on the same loop from two
3366	different coroutines and switch freely between both coroutines running the	3384	different coroutines, and switch freely between both coroutines running the
3367	loop, as long as you don't confuse yourself). The only exception is that	3385	loop, as long as you don't confuse yourself). The only exception is that
3368	you must not do this from C<ev_periodic> reschedule callbacks.	3386	you must not do this from C<ev_periodic> reschedule callbacks.
3369		3387
3370	Care has been taken to ensure that libev does not keep local state inside	3388	Care has been taken to ensure that libev does not keep local state inside
3371	C<ev_loop>, and other calls do not usually allow coroutine switches.	3389	C<ev_loop>, and other calls do not usually allow for coroutine switches as
		3390	they do not clal any callbacks.
3372		3391
		3392	=head2 COMPILER WARNINGS
3373		3393
3374	=head1 COMPLEXITIES	3394	Depending on your compiler and compiler settings, you might get no or a
		3395	lot of warnings when compiling libev code. Some people are apparently
		3396	scared by this.
3375		3397
3376	In this section the complexities of (many of) the algorithms used inside	3398	However, these are unavoidable for many reasons. For one, each compiler
3377	libev will be explained. For complexity discussions about backends see the	3399	has different warnings, and each user has different tastes regarding
3378	documentation for C<ev_default_init>.	3400	warning options. "Warn-free" code therefore cannot be a goal except when
		3401	targeting a specific compiler and compiler-version.
3379		3402
3380	All of the following are about amortised time: If an array needs to be	3403	Another reason is that some compiler warnings require elaborate
3381	extended, libev needs to realloc and move the whole array, but this	3404	workarounds, or other changes to the code that make it less clear and less
3382	happens asymptotically never with higher number of elements, so O(1) might	3405	maintainable.
3383	mean it might do a lengthy realloc operation in rare cases, but on average
3384	it is much faster and asymptotically approaches constant time.
3385		3406
3386	=over 4	3407	And of course, some compiler warnings are just plain stupid, or simply
		3408	wrong (because they don't actually warn about the condition their message
		3409	seems to warn about). For example, certain older gcc versions had some
		3410	warnings that resulted an extreme number of false positives. These have
		3411	been fixed, but some people still insist on making code warn-free with
		3412	such buggy versions.
3387		3413
3388	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)	3414	While libev is written to generate as few warnings as possible,
		3415	"warn-free" code is not a goal, and it is recommended not to build libev
		3416	with any compiler warnings enabled unless you are prepared to cope with
		3417	them (e.g. by ignoring them). Remember that warnings are just that:
		3418	warnings, not errors, or proof of bugs.
3389		3419
3390	This means that, when you have a watcher that triggers in one hour and
3391	there are 100 watchers that would trigger before that then inserting will
3392	have to skip roughly seven (C<ld 100>) of these watchers.
3393		3420
3394	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)	3421	=head2 VALGRIND
3395		3422
3396	That means that changing a timer costs less than removing/adding them	3423	Valgrind has a special section here because it is a popular tool that is
3397	as only the relative motion in the event queue has to be paid for.	3424	highly useful. Unfortunately, valgrind reports are very hard to interpret.
3398		3425
3399	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)	3426	If you think you found a bug (memory leak, uninitialised data access etc.)
		3427	in libev, then check twice: If valgrind reports something like:
3400		3428
3401	These just add the watcher into an array or at the head of a list.	3429	==2274== definitely lost: 0 bytes in 0 blocks.
		3430	==2274== possibly lost: 0 bytes in 0 blocks.
		3431	==2274== still reachable: 256 bytes in 1 blocks.
3402		3432
3403	=item Stopping check/prepare/idle/fork/async watchers: O(1)	3433	Then there is no memory leak, just as memory accounted to global variables
		3434	is not a memleak - the memory is still being refernced, and didn't leak.
3404		3435
3405	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))	3436	Similarly, under some circumstances, valgrind might report kernel bugs
		3437	as if it were a bug in libev (e.g. in realloc or in the poll backend,
		3438	although an acceptable workaround has been found here), or it might be
		3439	confused.
3406		3440
3407	These watchers are stored in lists then need to be walked to find the	3441	Keep in mind that valgrind is a very good tool, but only a tool. Don't
3408	correct watcher to remove. The lists are usually short (you don't usually	3442	make it into some kind of religion.
3409	have many watchers waiting for the same fd or signal).
3410		3443
3411	=item Finding the next timer in each loop iteration: O(1)	3444	If you are unsure about something, feel free to contact the mailing list
		3445	with the full valgrind report and an explanation on why you think this
		3446	is a bug in libev (best check the archives, too :). However, don't be
		3447	annoyed when you get a brisk "this is no bug" answer and take the chance
		3448	of learning how to interpret valgrind properly.
3412		3449
3413	By virtue of using a binary or 4-heap, the next timer is always found at a	3450	If you need, for some reason, empty reports from valgrind for your project
3414	fixed position in the storage array.	3451	I suggest using suppression lists.
3415		3452
3416	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
3417		3453
3418	A change means an I/O watcher gets started or stopped, which requires	3454	=head1 PORTABILITY NOTES
3419	libev to recalculate its status (and possibly tell the kernel, depending
3420	on backend and whether C<ev_io_set> was used).
3421		3455
3422	=item Activating one watcher (putting it into the pending state): O(1)
3423
3424	=item Priority handling: O(number_of_priorities)
3425
3426	Priorities are implemented by allocating some space for each
3427	priority. When doing priority-based operations, libev usually has to
3428	linearly search all the priorities, but starting/stopping and activating
3429	watchers becomes O(1) with respect to priority handling.
3430
3431	=item Sending an ev_async: O(1)
3432
3433	=item Processing ev_async_send: O(number_of_async_watchers)
3434
3435	=item Processing signals: O(max_signal_number)
3436
3437	Sending involves a system call I<iff> there were no other C<ev_async_send>
3438	calls in the current loop iteration. Checking for async and signal events
3439	involves iterating over all running async watchers or all signal numbers.
3440
3441	=back
3442
3443
3444	=head1 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS	3456	=head2 WIN32 PLATFORM LIMITATIONS AND WORKAROUNDS
3445		3457
3446	Win32 doesn't support any of the standards (e.g. POSIX) that libev	3458	Win32 doesn't support any of the standards (e.g. POSIX) that libev
3447	requires, and its I/O model is fundamentally incompatible with the POSIX	3459	requires, and its I/O model is fundamentally incompatible with the POSIX
3448	model. Libev still offers limited functionality on this platform in	3460	model. Libev still offers limited functionality on this platform in
3449	the form of the C<EVBACKEND_SELECT> backend, and only supports socket	3461	the form of the C<EVBACKEND_SELECT> backend, and only supports socket
…		…
3536	wrap all I/O functions and provide your own fd management, but the cost of	3548	wrap all I/O functions and provide your own fd management, but the cost of
3537	calling select (O(n²)) will likely make this unworkable.	3549	calling select (O(n²)) will likely make this unworkable.
3538		3550
3539	=back	3551	=back
3540		3552
3541
3542	=head1 PORTABILITY REQUIREMENTS	3553	=head2 PORTABILITY REQUIREMENTS
3543		3554
3544	In addition to a working ISO-C implementation, libev relies on a few	3555	In addition to a working ISO-C implementation and of course the
3545	additional extensions:	3556	backend-specific APIs, libev relies on a few additional extensions:
3546		3557
3547	=over 4	3558	=over 4
3548		3559
3549	=item C<void ()(ev_watcher_type , int revents)> must have compatible	3560	=item C<void ()(ev_watcher_type , int revents)> must have compatible
3550	calling conventions regardless of C<ev_watcher_type *>.	3561	calling conventions regardless of C<ev_watcher_type *>.
…		…
3575	except the initial one, and run the default loop in the initial thread as	3586	except the initial one, and run the default loop in the initial thread as
3576	well.	3587	well.
3577		3588
3578	=item C<long> must be large enough for common memory allocation sizes	3589	=item C<long> must be large enough for common memory allocation sizes
3579		3590
3580	To improve portability and simplify using libev, libev uses C<long>	3591	To improve portability and simplify its API, libev uses C<long> internally
3581	internally instead of C<size_t> when allocating its data structures. On	3592	instead of C<size_t> when allocating its data structures. On non-POSIX
3582	non-POSIX systems (Microsoft...) this might be unexpectedly low, but	3593	systems (Microsoft...) this might be unexpectedly low, but is still at
3583	is still at least 31 bits everywhere, which is enough for hundreds of	3594	least 31 bits everywhere, which is enough for hundreds of millions of
3584	millions of watchers.	3595	watchers.
3585		3596
3586	=item C<double> must hold a time value in seconds with enough accuracy	3597	=item C<double> must hold a time value in seconds with enough accuracy
3587		3598
3588	The type C<double> is used to represent timestamps. It is required to	3599	The type C<double> is used to represent timestamps. It is required to
3589	have at least 51 bits of mantissa (and 9 bits of exponent), which is good	3600	have at least 51 bits of mantissa (and 9 bits of exponent), which is good
…		…
3593	=back	3604	=back
3594		3605
3595	If you know of other additional requirements drop me a note.	3606	If you know of other additional requirements drop me a note.
3596		3607
3597		3608
3598	=head1 COMPILER WARNINGS	3609	=head1 ALGORITHMIC COMPLEXITIES
3599		3610
3600	Depending on your compiler and compiler settings, you might get no or a	3611	In this section the complexities of (many of) the algorithms used inside
3601	lot of warnings when compiling libev code. Some people are apparently	3612	libev will be documented. For complexity discussions about backends see
3602	scared by this.	3613	the documentation for C<ev_default_init>.
3603		3614
3604	However, these are unavoidable for many reasons. For one, each compiler	3615	All of the following are about amortised time: If an array needs to be
3605	has different warnings, and each user has different tastes regarding	3616	extended, libev needs to realloc and move the whole array, but this
3606	warning options. "Warn-free" code therefore cannot be a goal except when	3617	happens asymptotically rarer with higher number of elements, so O(1) might
3607	targeting a specific compiler and compiler-version.	3618	mean that libev does a lengthy realloc operation in rare cases, but on
		3619	average it is much faster and asymptotically approaches constant time.
3608		3620
3609	Another reason is that some compiler warnings require elaborate	3621	=over 4
3610	workarounds, or other changes to the code that make it less clear and less
3611	maintainable.
3612		3622
3613	And of course, some compiler warnings are just plain stupid, or simply	3623	=item Starting and stopping timer/periodic watchers: O(log skipped_other_timers)
3614	wrong (because they don't actually warn about the condition their message
3615	seems to warn about).
3616		3624
3617	While libev is written to generate as few warnings as possible,	3625	This means that, when you have a watcher that triggers in one hour and
3618	"warn-free" code is not a goal, and it is recommended not to build libev	3626	there are 100 watchers that would trigger before that, then inserting will
3619	with any compiler warnings enabled unless you are prepared to cope with	3627	have to skip roughly seven (C<ld 100>) of these watchers.
3620	them (e.g. by ignoring them). Remember that warnings are just that:
3621	warnings, not errors, or proof of bugs.
3622		3628
		3629	=item Changing timer/periodic watchers (by autorepeat or calling again): O(log skipped_other_timers)
3623		3630
3624	=head1 VALGRIND	3631	That means that changing a timer costs less than removing/adding them,
		3632	as only the relative motion in the event queue has to be paid for.
3625		3633
3626	Valgrind has a special section here because it is a popular tool that is	3634	=item Starting io/check/prepare/idle/signal/child/fork/async watchers: O(1)
3627	highly useful, but valgrind reports are very hard to interpret.
3628		3635
3629	If you think you found a bug (memory leak, uninitialised data access etc.)	3636	These just add the watcher into an array or at the head of a list.
3630	in libev, then check twice: If valgrind reports something like:
3631		3637
3632	==2274== definitely lost: 0 bytes in 0 blocks.	3638	=item Stopping check/prepare/idle/fork/async watchers: O(1)
3633	==2274== possibly lost: 0 bytes in 0 blocks.
3634	==2274== still reachable: 256 bytes in 1 blocks.
3635		3639
3636	Then there is no memory leak. Similarly, under some circumstances,	3640	=item Stopping an io/signal/child watcher: O(number_of_watchers_for_this_(fd/signal/pid % EV_PID_HASHSIZE))
3637	valgrind might report kernel bugs as if it were a bug in libev, or it
3638	might be confused (it is a very good tool, but only a tool).
3639		3641
3640	If you are unsure about something, feel free to contact the mailing list	3642	These watchers are stored in lists, so they need to be walked to find the
3641	with the full valgrind report and an explanation on why you think this is	3643	correct watcher to remove. The lists are usually short (you don't usually
3642	a bug in libev. However, don't be annoyed when you get a brisk "this is	3644	have many watchers waiting for the same fd or signal: one is typical, two
3643	no bug" answer and take the chance of learning how to interpret valgrind	3645	is rare).
3644	properly.
3645		3646
3646	If you need, for some reason, empty reports from valgrind for your project	3647	=item Finding the next timer in each loop iteration: O(1)
3647	I suggest using suppression lists.	3648
		3649	By virtue of using a binary or 4-heap, the next timer is always found at a
		3650	fixed position in the storage array.
		3651
		3652	=item Each change on a file descriptor per loop iteration: O(number_of_watchers_for_this_fd)
		3653
		3654	A change means an I/O watcher gets started or stopped, which requires
		3655	libev to recalculate its status (and possibly tell the kernel, depending
		3656	on backend and whether C<ev_io_set> was used).
		3657
		3658	=item Activating one watcher (putting it into the pending state): O(1)
		3659
		3660	=item Priority handling: O(number_of_priorities)
		3661
		3662	Priorities are implemented by allocating some space for each
		3663	priority. When doing priority-based operations, libev usually has to
		3664	linearly search all the priorities, but starting/stopping and activating
		3665	watchers becomes O(1) with respect to priority handling.
		3666
		3667	=item Sending an ev_async: O(1)
		3668
		3669	=item Processing ev_async_send: O(number_of_async_watchers)
		3670
		3671	=item Processing signals: O(max_signal_number)
		3672
		3673	Sending involves a system call I<iff> there were no other C<ev_async_send>
		3674	calls in the current loop iteration. Checking for async and signal events
		3675	involves iterating over all running async watchers or all signal numbers.
		3676
		3677	=back
3648		3678
3649		3679
3650	=head1 AUTHOR	3680	=head1 AUTHOR
3651		3681
3652	Marc Lehmann <libev@schmorp.de>.	3682	Marc Lehmann <libev@schmorp.de>.

Diff Legend

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

Comparing libev/ev.pod (file contents): Revision 1.185 by root, Tue Sep 23 09:13:59 2008 UTC vs. Revision 1.195 by root, Mon Oct 20 17:50:48 2008 UTC

Diff Legend

Comparing libev/ev.pod (file contents):
Revision 1.185 by root, Tue Sep 23 09:13:59 2008 UTC vs.
Revision 1.195 by root, Mon Oct 20 17:50:48 2008 UTC