[ViewVC] Diff of: cvs/Coro/Coro.pm

Comparing Coro/Coro.pm (file contents):
Revision 1.108 by root, Fri Jan 5 20:00:49 2007 UTC vs.
Revision 1.238 by root, Mon Nov 24 04:56:38 2008 UTC

…		…
1	=head1 NAME	1	=head1 NAME
2		2
3	Coro - coroutine process abstraction	3	Coro - the only real threads in perl
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use Coro;	7	use Coro;
8		8
9	async {	9	async {
10	# some asynchronous thread of execution	10	# some asynchronous thread of execution
		11	print "2\n";
		12	cede; # yield back to main
		13	print "4\n";
11	};	14	};
12		15	print "1\n";
13	# alternatively create an async coroutine like this:	16	cede; # yield to coroutine
14		17	print "3\n";
15	sub some_func : Coro {	18	cede; # and again
16	# some more async code	19
17	}	20	# use locking
18		21	use Coro::Semaphore;
19	cede;	22	my $lock = new Coro::Semaphore;
		23	my $locked;
		24
		25	$lock->down;
		26	$locked = 1;
		27	$lock->up;
20		28
21	=head1 DESCRIPTION	29	=head1 DESCRIPTION
22		30
23	This module collection manages coroutines. Coroutines are similar	31	For a tutorial-style introduction, please read the L<Coro::Intro>
24	to threads but don't run in parallel at the same time even on SMP	32	manpage. This manpage mainly contains reference information.
25	machines. The specific flavor of coroutine use din this module also
26	guarentees you that it will not switch between coroutines unless
27	necessary, at easily-identified points in your program, so locking and
28	parallel access are rarely an issue, making coroutine programming much
29	safer than threads programming.
30		33
31	(Perl, however, does not natively support real threads but instead does a	34	This module collection manages continuations in general, most often
32	very slow and memory-intensive emulation of processes using threads. This	35	in the form of cooperative threads (also called coroutines in the
33	is a performance win on Windows machines, and a loss everywhere else).	36	documentation). They are similar to kernel threads but don't (in general)
		37	run in parallel at the same time even on SMP machines. The specific flavor
		38	of thread offered by this module also guarantees you that it will not
		39	switch between threads unless necessary, at easily-identified points in
		40	your program, so locking and parallel access are rarely an issue, making
		41	thread programming much safer and easier than using other thread models.
34		42
		43	Unlike the so-called "Perl threads" (which are not actually real threads
		44	but only the windows process emulation ported to unix), Coro provides a
		45	full shared address space, which makes communication between threads
		46	very easy. And threads are fast, too: disabling the Windows process
		47	emulation code in your perl and using Coro can easily result in a two to
		48	four times speed increase for your programs.
		49
		50	Coro achieves that by supporting multiple running interpreters that share
		51	data, which is especially useful to code pseudo-parallel processes and
		52	for event-based programming, such as multiple HTTP-GET requests running
		53	concurrently. See L<Coro::AnyEvent> to learn more on how to integrate Coro
		54	into an event-based environment.
		55
35	In this module, coroutines are defined as "callchain + lexical variables +	56	In this module, a thread is defined as "callchain + lexical variables +
36	@_ + $_ + $@ + $/ + C stack), that is, a coroutine has its own callchain,	57	@_ + $_ + $@ + $/ + C stack), that is, a thread has its own callchain,
37	its own set of lexicals and its own set of perls most important global	58	its own set of lexicals and its own set of perls most important global
38	variables.	59	variables (see L<Coro::State> for more configuration and background info).
		60
		61	See also the C<SEE ALSO> section at the end of this document - the Coro
		62	module family is quite large.
39		63
40	=cut	64	=cut
41		65
42	package Coro;	66	package Coro;
43		67
44	use strict;	68	use strict qw(vars subs);
45	no warnings "uninitialized";	69	no warnings "uninitialized";
46		70
47	use Coro::State;	71	use Coro::State;
48		72
49	use base qw(Coro::State Exporter);	73	use base qw(Coro::State Exporter);
50		74
51	our $idle; # idle handler	75	our $idle; # idle handler
52	our $main; # main coroutine	76	our $main; # main coroutine
53	our $current; # current coroutine	77	our $current; # current coroutine
54		78
55	our $VERSION = '3.3';	79	our $VERSION = "5.0";
56		80
57	our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);	81	our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);
58	our %EXPORT_TAGS = (	82	our %EXPORT_TAGS = (
59	prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],	83	prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],
60	);	84	);
61	our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));	85	our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));
62		86
63	{	87	=head1 GLOBAL VARIABLES
64	my @async;
65	my $init;
66
67	# this way of handling attributes simply is NOT scalable ;()
68	sub import {
69	no strict 'refs';
70
71	Coro->export_to_level (1, @_);
72
73	my $old = *{(caller)[0]."::MODIFY_CODE_ATTRIBUTES"}{CODE};
74	*{(caller)[0]."::MODIFY_CODE_ATTRIBUTES"} = sub {
75	my ($package, $ref) = (shift, shift);
76	my @attrs;
77	for (@_) {
78	if ($_ eq "Coro") {
79	push @async, $ref;
80	unless ($init++) {
81	eval q{
82	sub INIT {
83	&async(pop @async) while @async;
84	}
85	};
86	}
87	} else {
88	push @attrs, $_;
89	}
90	}
91	return $old ? $old->($package, $ref, @attrs) : @attrs;
92	};
93	}
94
95	}
96		88
97	=over 4	89	=over 4
98		90
99	=item $main	91	=item $Coro::main
100		92
101	This coroutine represents the main program.	93	This variable stores the coroutine object that represents the main
		94	program. While you cna C<ready> it and do most other things you can do to
		95	coroutines, it is mainly useful to compare again C<$Coro::current>, to see
		96	whether you are running in the main program or not.
102		97
103	=cut	98	=cut
104		99
105	$main = new Coro;	100	# $main is now being initialised by Coro::State
106		101
107	=item $current (or as function: current)	102	=item $Coro::current
108		103
109	The current coroutine (the last coroutine switched to). The initial value	104	The coroutine object representing the current coroutine (the last
		105	coroutine that the Coro scheduler switched to). The initial value is
110	is C<$main> (of course).	106	C<$Coro::main> (of course).
111		107
112	This variable is B<strictly> I<read-only>. It is provided for performance	108	This variable is B<strictly> I<read-only>. You can take copies of the
113	reasons. If performance is not essentiel you are encouraged to use the	109	value stored in it and use it as any other coroutine object, but you must
114	C<Coro::current> function instead.	110	not otherwise modify the variable itself.
115		111
116	=cut	112	=cut
117		113
118	# maybe some other module used Coro::Specific before...
119	$main->{specific} = $current->{specific}
120	if $current;
121
122	_set_current $main;
123
124	sub current() { $current }	114	sub current() { $current } # [DEPRECATED]
125		115
126	=item $idle	116	=item $Coro::idle
127		117
		118	This variable is mainly useful to integrate Coro into event loops. It is
		119	usually better to rely on L<Coro::AnyEvent> or L<Coro::EV>, as this is
		120	pretty low-level functionality.
		121
		122	This variable stores either a coroutine or a callback.
		123
128	A callback that is called whenever the scheduler finds no ready coroutines	124	If it is a callback, the it is called whenever the scheduler finds no
129	to run. The default implementation prints "FATAL: deadlock detected" and	125	ready coroutines to run. The default implementation prints "FATAL:
130	exits, because the program has no other way to continue.	126	deadlock detected" and exits, because the program has no other way to
		127	continue.
131		128
		129	If it is a coroutine object, then this object will be readied (without
		130	invoking any ready hooks, however) when the scheduler finds no other ready
		131	coroutines to run.
		132
132	This hook is overwritten by modules such as C<Coro::Timer> and	133	This hook is overwritten by modules such as C<Coro::EV> and
133	C<Coro::Event> to wait on an external event that hopefully wake up a	134	C<Coro::AnyEvent> to wait on an external event that hopefully wake up a
134	coroutine so the scheduler can run it.	135	coroutine so the scheduler can run it.
135		136
		137	Note that the callback I<must not>, under any circumstances, block
		138	the current coroutine. Normally, this is achieved by having an "idle
		139	coroutine" that calls the event loop and then blocks again, and then
		140	readying that coroutine in the idle handler, or by simply placing the idle
		141	coroutine in this variable.
		142
		143	See L<Coro::Event> or L<Coro::AnyEvent> for examples of using this
		144	technique.
		145
136	Please note that if your callback recursively invokes perl (e.g. for event	146	Please note that if your callback recursively invokes perl (e.g. for event
137	handlers), then it must be prepared to be called recursively.	147	handlers), then it must be prepared to be called recursively itself.
138		148
139	=cut	149	=cut
140		150
141	$idle = sub {	151	$idle = sub {
142	require Carp;	152	require Carp;
143	Carp::croak ("FATAL: deadlock detected");	153	Carp::croak ("FATAL: deadlock detected");
144	};	154	};
145		155
146	sub _cancel {
147	my ($self) = @_;
148
149	# free coroutine data and mark as destructed
150	$self->_destroy
151	or return;
152
153	# call all destruction callbacks
154	$_->(@{$self->{status}})
155	for @{(delete $self->{destroy_cb}) \|\| []};
156	}
157
158	# this coroutine is necessary because a coroutine	156	# this coroutine is necessary because a coroutine
159	# cannot destroy itself.	157	# cannot destroy itself.
160	my @destroy;	158	our @destroy;
161	my $manager;	159	our $manager;
162		160
163	$manager = new Coro sub {	161	$manager = new Coro sub {
164	while () {	162	while () {
165	(shift @destroy)->_cancel	163	Coro::_cancel shift @destroy
166	while @destroy;	164	while @destroy;
167		165
168	&schedule;	166	&schedule;
169	}	167	}
170	};	168	};
171		169	$manager->{desc} = "[coro manager]";
172	$manager->prio (PRIO_MAX);	170	$manager->prio (PRIO_MAX);
173		171
174	# static methods. not really.
175
176	=back	172	=back
177		173
178	=head2 STATIC METHODS	174	=head1 SIMPLE COROUTINE CREATION
179
180	Static methods are actually functions that operate on the current coroutine only.
181		175
182	=over 4	176	=over 4
183		177
184	=item async { ... } [@args...]	178	=item async { ... } [@args...]
185		179
186	Create a new asynchronous coroutine and return it's coroutine object	180	Create a new coroutine and return it's coroutine object (usually
187	(usually unused). When the sub returns the new coroutine is automatically	181	unused). The coroutine will be put into the ready queue, so
		182	it will start running automatically on the next scheduler run.
		183
		184	The first argument is a codeblock/closure that should be executed in the
		185	coroutine. When it returns argument returns the coroutine is automatically
188	terminated.	186	terminated.
189		187
190	Calling C<exit> in a coroutine will not work correctly, so do not do that.	188	The remaining arguments are passed as arguments to the closure.
191		189
192	When the coroutine dies, the program will exit, just as in the main	190	See the C<Coro::State::new> constructor for info about the coroutine
193	program.	191	environment in which coroutines are executed.
194		192
		193	Calling C<exit> in a coroutine will do the same as calling exit outside
		194	the coroutine. Likewise, when the coroutine dies, the program will exit,
		195	just as it would in the main program.
		196
		197	If you do not want that, you can provide a default C<die> handler, or
		198	simply avoid dieing (by use of C<eval>).
		199
195	# create a new coroutine that just prints its arguments	200	Example: Create a new coroutine that just prints its arguments.
		201
196	async {	202	async {
197	print "@_\n";	203	print "@_\n";
198	} 1,2,3,4;	204	} 1,2,3,4;
199		205
200	=cut	206	=cut
206	}	212	}
207		213
208	=item async_pool { ... } [@args...]	214	=item async_pool { ... } [@args...]
209		215
210	Similar to C<async>, but uses a coroutine pool, so you should not call	216	Similar to C<async>, but uses a coroutine pool, so you should not call
211	terminate or join (although you are allowed to), and you get a coroutine	217	terminate or join on it (although you are allowed to), and you get a
212	that might have executed other code already (which can be good or bad :).	218	coroutine that might have executed other code already (which can be good
		219	or bad :).
213		220
		221	On the plus side, this function is about twice as fast as creating (and
		222	destroying) a completely new coroutine, so if you need a lot of generic
		223	coroutines in quick successsion, use C<async_pool>, not C<async>.
		224
214	Also, the block is executed in an C<eval> context and a warning will be	225	The code block is executed in an C<eval> context and a warning will be
215	issued in case of an exception instead of terminating the program, as	226	issued in case of an exception instead of terminating the program, as
216	C<async> does. As the coroutine is being reused, stuff like C<on_destroy>	227	C<async> does. As the coroutine is being reused, stuff like C<on_destroy>
217	will not work in the expected way, unless you call terminate or cancel,	228	will not work in the expected way, unless you call terminate or cancel,
218	which somehow defeats the purpose of pooling.	229	which somehow defeats the purpose of pooling (but is fine in the
		230	exceptional case).
219		231
220	The priority will be reset to C<0> after each job, otherwise the coroutine	232	The priority will be reset to C<0> after each run, tracing will be
221	will be re-used "as-is".	233	disabled, the description will be reset and the default output filehandle
		234	gets restored, so you can change all these. Otherwise the coroutine will
		235	be re-used "as-is": most notably if you change other per-coroutine global
		236	stuff such as C<$/> you I<must needs> revert that change, which is most
		237	simply done by using local as in: C<< local $/ >>.
222		238
223	The pool size is limited to 8 idle coroutines (this can be adjusted by	239	The idle pool size is limited to C<8> idle coroutines (this can be
224	changing $Coro::POOL_SIZE), and there can be as many non-idle coros as	240	adjusted by changing $Coro::POOL_SIZE), but there can be as many non-idle
225	required.	241	coros as required.
226		242
227	If you are concerned about pooled coroutines growing a lot because a	243	If you are concerned about pooled coroutines growing a lot because a
228	single C<async_pool> used a lot of stackspace you can e.g. C<async_pool {	244	single C<async_pool> used a lot of stackspace you can e.g. C<async_pool
229	terminate }> once per second or so to slowly replenish the pool.	245	{ terminate }> once per second or so to slowly replenish the pool. In
		246	addition to that, when the stacks used by a handler grows larger than 32kb
		247	(adjustable via $Coro::POOL_RSS) it will also be destroyed.
230		248
231	=cut	249	=cut
232		250
233	our $POOL_SIZE = 8;	251	our $POOL_SIZE = 8;
		252	our $POOL_RSS = 32 * 1024;
234	our @pool;	253	our @async_pool;
235		254
236	sub pool_handler {	255	sub pool_handler {
237	while () {	256	while () {
238	my ($cb, @arg) = @{ delete $current->{_invoke} };
239
240	eval {	257	eval {
241	$cb->(@arg);	258	&{&_pool_handler} while 1;
242	};	259	};
		260
243	warn $@ if $@;	261	warn $@ if $@;
244
245	last if @pool >= $POOL_SIZE;
246	push @pool, $current;
247
248	$current->prio (0);
249	schedule;
250	}	262	}
251	}	263	}
252		264
253	sub async_pool(&@) {	265	=back
254	# this is also inlined into the unlock_scheduler
255	my $coro = (pop @pool or new Coro \&pool_handler);
256		266
257	$coro->{_invoke} = [@_];	267	=head1 STATIC METHODS
258	$coro->ready;
259		268
260	$coro	269	Static methods are actually functions that implicitly operate on the
261	}	270	current coroutine.
		271
		272	=over 4
262		273
263	=item schedule	274	=item schedule
264		275
265	Calls the scheduler. Please note that the current coroutine will not be put	276	Calls the scheduler. The scheduler will find the next coroutine that is
		277	to be run from the ready queue and switches to it. The next coroutine
		278	to be run is simply the one with the highest priority that is longest
		279	in its ready queue. If there is no coroutine ready, it will clal the
		280	C<$Coro::idle> hook.
		281
		282	Please note that the current coroutine will I<not> be put into the ready
266	into the ready queue, so calling this function usually means you will	283	queue, so calling this function usually means you will never be called
267	never be called again unless something else (e.g. an event handler) calls	284	again unless something else (e.g. an event handler) calls C<< ->ready >>,
268	ready.	285	thus waking you up.
269		286
270	The canonical way to wait on external events is this:	287	This makes C<schedule> I<the> generic method to use to block the current
		288	coroutine and wait for events: first you remember the current coroutine in
		289	a variable, then arrange for some callback of yours to call C<< ->ready
		290	>> on that once some event happens, and last you call C<schedule> to put
		291	yourself to sleep. Note that a lot of things can wake your coroutine up,
		292	so you need to check whether the event indeed happened, e.g. by storing the
		293	status in a variable.
271		294
		295	See B<HOW TO WAIT FOR A CALLBACK>, below, for some ways to wait for callbacks.
		296
		297	=item cede
		298
		299	"Cede" to other coroutines. This function puts the current coroutine into
		300	the ready queue and calls C<schedule>, which has the effect of giving
		301	up the current "timeslice" to other coroutines of the same or higher
		302	priority. Once your coroutine gets its turn again it will automatically be
		303	resumed.
		304
		305	This function is often called C<yield> in other languages.
		306
		307	=item Coro::cede_notself
		308
		309	Works like cede, but is not exported by default and will cede to I<any>
		310	coroutine, regardless of priority. This is useful sometimes to ensure
		311	progress is made.
		312
		313	=item terminate [arg...]
		314
		315	Terminates the current coroutine with the given status values (see L<cancel>).
		316
		317	=item killall
		318
		319	Kills/terminates/cancels all coroutines except the currently running
		320	one. This is useful after a fork, either in the child or the parent, as
		321	usually only one of them should inherit the running coroutines.
		322
		323	Note that while this will try to free some of the main programs resources,
		324	you cannot free all of them, so if a coroutine that is not the main
		325	program calls this function, there will be some one-time resource leak.
		326
		327	=cut
		328
		329	sub killall {
		330	for (Coro::State::list) {
		331	$_->cancel
		332	if $_ != $current && UNIVERSAL::isa $_, "Coro";
272	{	333	}
273	# remember current coroutine	334	}
		335
		336	=back
		337
		338	=head1 COROUTINE OBJECT METHODS
		339
		340	These are the methods you can call on coroutine objects (or to create
		341	them).
		342
		343	=over 4
		344
		345	=item new Coro \&sub [, @args...]
		346
		347	Create a new coroutine and return it. When the sub returns, the coroutine
		348	automatically terminates as if C<terminate> with the returned values were
		349	called. To make the coroutine run you must first put it into the ready
		350	queue by calling the ready method.
		351
		352	See C<async> and C<Coro::State::new> for additional info about the
		353	coroutine environment.
		354
		355	=cut
		356
		357	sub _terminate {
		358	terminate &{+shift};
		359	}
		360
		361	=item $success = $coroutine->ready
		362
		363	Put the given coroutine into the end of its ready queue (there is one
		364	queue for each priority) and return true. If the coroutine is already in
		365	the ready queue, do nothing and return false.
		366
		367	This ensures that the scheduler will resume this coroutine automatically
		368	once all the coroutines of higher priority and all coroutines of the same
		369	priority that were put into the ready queue earlier have been resumed.
		370
		371	=item $is_ready = $coroutine->is_ready
		372
		373	Return whether the coroutine is currently the ready queue or not,
		374
		375	=item $coroutine->cancel (arg...)
		376
		377	Terminates the given coroutine and makes it return the given arguments as
		378	status (default: the empty list). Never returns if the coroutine is the
		379	current coroutine.
		380
		381	=cut
		382
		383	sub cancel {
		384	my $self = shift;
		385
		386	if ($current == $self) {
		387	terminate @_;
		388	} else {
		389	$self->{_status} = [@_];
		390	$self->_cancel;
		391	}
		392	}
		393
		394	=item $coroutine->schedule_to
		395
		396	Puts the current coroutine to sleep (like C<Coro::schedule>), but instead
		397	of continuing with the next coro from the ready queue, always switch to
		398	the given coroutine object (regardless of priority etc.). The readyness
		399	state of that coroutine isn't changed.
		400
		401	This is an advanced method for special cases - I'd love to hear about any
		402	uses for this one.
		403
		404	=item $coroutine->cede_to
		405
		406	Like C<schedule_to>, but puts the current coroutine into the ready
		407	queue. This has the effect of temporarily switching to the given
		408	coroutine, and continuing some time later.
		409
		410	This is an advanced method for special cases - I'd love to hear about any
		411	uses for this one.
		412
		413	=item $coroutine->throw ([$scalar])
		414
		415	If C<$throw> is specified and defined, it will be thrown as an exception
		416	inside the coroutine at the next convenient point in time. Otherwise
		417	clears the exception object.
		418
		419	Coro will check for the exception each time a schedule-like-function
		420	returns, i.e. after each C<schedule>, C<cede>, C<< Coro::Semaphore->down
		421	>>, C<< Coro::Handle->readable >> and so on. Most of these functions
		422	detect this case and return early in case an exception is pending.
		423
		424	The exception object will be thrown "as is" with the specified scalar in
		425	C<$@>, i.e. if it is a string, no line number or newline will be appended
		426	(unlike with C<die>).
		427
		428	This can be used as a softer means than C<cancel> to ask a coroutine to
		429	end itself, although there is no guarantee that the exception will lead to
		430	termination, and if the exception isn't caught it might well end the whole
		431	program.
		432
		433	You might also think of C<throw> as being the moral equivalent of
		434	C<kill>ing a coroutine with a signal (in this case, a scalar).
		435
		436	=item $coroutine->join
		437
		438	Wait until the coroutine terminates and return any values given to the
		439	C<terminate> or C<cancel> functions. C<join> can be called concurrently
		440	from multiple coroutines, and all will be resumed and given the status
		441	return once the C<$coroutine> terminates.
		442
		443	=cut
		444
		445	sub join {
		446	my $self = shift;
		447
		448	unless ($self->{_status}) {
274	my $current = $Coro::current;	449	my $current = $current;
275		450
276	# register a hypothetical event handler	451	push @{$self->{_on_destroy}}, sub {
277	on_event_invoke sub {
278	# wake up sleeping coroutine
279	$current->ready;	452	$current->ready;
280	undef $current;	453	undef $current;
281	};	454	};
282		455
283	# call schedule until event occured.
284	# in case we are woken up for other reasons
285	# (current still defined), loop.
286	Coro::schedule while $current;
287	}
288
289	=item cede
290
291	"Cede" to other coroutines. This function puts the current coroutine into the
292	ready queue and calls C<schedule>, which has the effect of giving up the
293	current "timeslice" to other coroutines of the same or higher priority.
294
295	Returns true if at least one coroutine switch has happened.
296
297	=item Coro::cede_notself
298
299	Works like cede, but is not exported by default and will cede to any
300	coroutine, regardless of priority, once.
301
302	Returns true if at least one coroutine switch has happened.
303
304	=item terminate [arg...]
305
306	Terminates the current coroutine with the given status values (see L<cancel>).
307
308	=cut
309
310	sub terminate {
311	$current->cancel (@_);
312	}
313
314	=back
315
316	# dynamic methods
317
318	=head2 COROUTINE METHODS
319
320	These are the methods you can call on coroutine objects.
321
322	=over 4
323
324	=item new Coro \&sub [, @args...]
325
326	Create a new coroutine and return it. When the sub returns the coroutine
327	automatically terminates as if C<terminate> with the returned values were
328	called. To make the coroutine run you must first put it into the ready queue
329	by calling the ready method.
330
331	Calling C<exit> in a coroutine will not work correctly, so do not do that.
332
333	=cut
334
335	sub _run_coro {
336	terminate &{+shift};
337	}
338
339	sub new {
340	my $class = shift;
341
342	$class->SUPER::new (\&_run_coro, @_)
343	}
344
345	=item $success = $coroutine->ready
346
347	Put the given coroutine into the ready queue (according to it's priority)
348	and return true. If the coroutine is already in the ready queue, do nothing
349	and return false.
350
351	=item $is_ready = $coroutine->is_ready
352
353	Return wether the coroutine is currently the ready queue or not,
354
355	=item $coroutine->cancel (arg...)
356
357	Terminates the given coroutine and makes it return the given arguments as
358	status (default: the empty list). Never returns if the coroutine is the
359	current coroutine.
360
361	=cut
362
363	sub cancel {
364	my $self = shift;
365	$self->{status} = [@_];
366
367	if ($current == $self) {
368	push @destroy, $self;
369	$manager->ready;
370	&schedule while 1;
371	} else {
372	$self->_cancel;
373	}
374	}
375
376	=item $coroutine->join
377
378	Wait until the coroutine terminates and return any values given to the
379	C<terminate> or C<cancel> functions. C<join> can be called multiple times
380	from multiple coroutine.
381
382	=cut
383
384	sub join {
385	my $self = shift;
386
387	unless ($self->{status}) {
388	my $current = $current;
389
390	push @{$self->{destroy_cb}}, sub {
391	$current->ready;
392	undef $current;
393	};
394
395	&schedule while $current;	456	&schedule while $current;
396	}	457	}
397		458
398	wantarray ? @{$self->{status}} : $self->{status}[0];	459	wantarray ? @{$self->{_status}} : $self->{_status}[0];
399	}	460	}
400		461
401	=item $coroutine->on_destroy (\&cb)	462	=item $coroutine->on_destroy (\&cb)
402		463
403	Registers a callback that is called when this coroutine gets destroyed,	464	Registers a callback that is called when this coroutine gets destroyed,
404	but before it is joined. The callback gets passed the terminate arguments,	465	but before it is joined. The callback gets passed the terminate arguments,
405	if any.	466	if any, and I<must not> die, under any circumstances.
406		467
407	=cut	468	=cut
408		469
409	sub on_destroy {	470	sub on_destroy {
410	my ($self, $cb) = @_;	471	my ($self, $cb) = @_;
411		472
412	push @{ $self->{destroy_cb} }, $cb;	473	push @{ $self->{_on_destroy} }, $cb;
413	}	474	}
414		475
415	=item $oldprio = $coroutine->prio ($newprio)	476	=item $oldprio = $coroutine->prio ($newprio)
416		477
417	Sets (or gets, if the argument is missing) the priority of the	478	Sets (or gets, if the argument is missing) the priority of the
…		…
440	higher values mean lower priority, just as in unix).	501	higher values mean lower priority, just as in unix).
441		502
442	=item $olddesc = $coroutine->desc ($newdesc)	503	=item $olddesc = $coroutine->desc ($newdesc)
443		504
444	Sets (or gets in case the argument is missing) the description for this	505	Sets (or gets in case the argument is missing) the description for this
445	coroutine. This is just a free-form string you can associate with a coroutine.	506	coroutine. This is just a free-form string you can associate with a
		507	coroutine.
		508
		509	This method simply sets the C<< $coroutine->{desc} >> member to the given
		510	string. You can modify this member directly if you wish.
446		511
447	=cut	512	=cut
448		513
449	sub desc {	514	sub desc {
450	my $old = $_[0]{desc};	515	my $old = $_[0]{desc};
451	$_[0]{desc} = $_[1] if @_ > 1;	516	$_[0]{desc} = $_[1] if @_ > 1;
452	$old;	517	$old;
453	}	518	}
454		519
		520	sub transfer {
		521	require Carp;
		522	Carp::croak ("You must not call ->transfer on Coro objects. Use Coro::State objects or the ->schedule_to method. Caught");
		523	}
		524
455	=back	525	=back
456		526
457	=head2 GLOBAL FUNCTIONS	527	=head1 GLOBAL FUNCTIONS
458		528
459	=over 4	529	=over 4
460		530
461	=item Coro::nready	531	=item Coro::nready
462		532
463	Returns the number of coroutines that are currently in the ready state,	533	Returns the number of coroutines that are currently in the ready state,
464	i.e. that can be swicthed to. The value C<0> means that the only runnable	534	i.e. that can be switched to by calling C<schedule> directory or
		535	indirectly. The value C<0> means that the only runnable coroutine is the
465	coroutine is the currently running one, so C<cede> would have no effect,	536	currently running one, so C<cede> would have no effect, and C<schedule>
466	and C<schedule> would cause a deadlock unless there is an idle handler	537	would cause a deadlock unless there is an idle handler that wakes up some
467	that wakes up some coroutines.	538	coroutines.
468		539
469	=item my $guard = Coro::guard { ... }	540	=item my $guard = Coro::guard { ... }
470		541
471	This creates and returns a guard object. Nothing happens until the objetc	542	This creates and returns a guard object. Nothing happens until the object
472	gets destroyed, in which case the codeblock given as argument will be	543	gets destroyed, in which case the codeblock given as argument will be
473	executed. This is useful to free locks or other resources in case of a	544	executed. This is useful to free locks or other resources in case of a
474	runtime error or when the coroutine gets canceled, as in both cases the	545	runtime error or when the coroutine gets canceled, as in both cases the
475	guard block will be executed. The guard object supports only one method,	546	guard block will be executed. The guard object supports only one method,
476	C<< ->cancel >>, which will keep the codeblock from being executed.	547	C<< ->cancel >>, which will keep the codeblock from being executed.
…		…
501		572
502		573
503	=item unblock_sub { ... }	574	=item unblock_sub { ... }
504		575
505	This utility function takes a BLOCK or code reference and "unblocks" it,	576	This utility function takes a BLOCK or code reference and "unblocks" it,
506	returning the new coderef. This means that the new coderef will return	577	returning a new coderef. Unblocking means that calling the new coderef
507	immediately without blocking, returning nothing, while the original code	578	will return immediately without blocking, returning nothing, while the
508	ref will be called (with parameters) from within its own coroutine.	579	original code ref will be called (with parameters) from within another
		580	coroutine.
509		581
510	The reason this fucntion exists is that many event libraries (such as the	582	The reason this function exists is that many event libraries (such as the
511	venerable L<Event\|Event> module) are not coroutine-safe (a weaker form	583	venerable L<Event\|Event> module) are not coroutine-safe (a weaker form
512	of thread-safety). This means you must not block within event callbacks,	584	of reentrancy). This means you must not block within event callbacks,
513	otherwise you might suffer from crashes or worse.	585	otherwise you might suffer from crashes or worse. The only event library
		586	currently known that is safe to use without C<unblock_sub> is L<EV>.
514		587
515	This function allows your callbacks to block by executing them in another	588	This function allows your callbacks to block by executing them in another
516	coroutine where it is safe to block. One example where blocking is handy	589	coroutine where it is safe to block. One example where blocking is handy
517	is when you use the L<Coro::AIO\|Coro::AIO> functions to save results to	590	is when you use the L<Coro::AIO\|Coro::AIO> functions to save results to
518	disk.	591	disk, for example.
519		592
520	In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when	593	In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when
521	creating event callbacks that want to block.	594	creating event callbacks that want to block.
		595
		596	If your handler does not plan to block (e.g. simply sends a message to
		597	another coroutine, or puts some other coroutine into the ready queue),
		598	there is no reason to use C<unblock_sub>.
		599
		600	Note that you also need to use C<unblock_sub> for any other callbacks that
		601	are indirectly executed by any C-based event loop. For example, when you
		602	use a module that uses L<AnyEvent> (and you use L<Coro::AnyEvent>) and it
		603	provides callbacks that are the result of some event callback, then you
		604	must not block either, or use C<unblock_sub>.
522		605
523	=cut	606	=cut
524		607
525	our @unblock_queue;	608	our @unblock_queue;
526		609
527	# we create a special coro because we want to cede,	610	# we create a special coro because we want to cede,
528	# to reduce pressure on the coro pool (because most callbacks	611	# to reduce pressure on the coro pool (because most callbacks
529	# return immediately and can be reused) and because we cannot cede	612	# return immediately and can be reused) and because we cannot cede
530	# inside an event callback.	613	# inside an event callback.
531	our $unblock_scheduler = async {	614	our $unblock_scheduler = new Coro sub {
532	while () {	615	while () {
533	while (my $cb = pop @unblock_queue) {	616	while (my $cb = pop @unblock_queue) {
534	# this is an inlined copy of async_pool	617	&async_pool (@$cb);
535	my $coro = (pop @pool or new Coro \&pool_handler);
536		618
537	$coro->{_invoke} = $cb;
538	$coro->ready;
539	cede; # for short-lived callbacks, this reduces pressure on the coro pool	619	# for short-lived callbacks, this reduces pressure on the coro pool
		620	# as the chance is very high that the async_poll coro will be back
		621	# in the idle state when cede returns
		622	cede;
540	}	623	}
541	schedule; # sleep well	624	schedule; # sleep well
542	}	625	}
543	};	626	};
		627	$unblock_scheduler->{desc} = "[unblock_sub scheduler]";
544		628
545	sub unblock_sub(&) {	629	sub unblock_sub(&) {
546	my $cb = shift;	630	my $cb = shift;
547		631
548	sub {	632	sub {
549	unshift @unblock_queue, [$cb, @_];	633	unshift @unblock_queue, [$cb, @_];
550	$unblock_scheduler->ready;	634	$unblock_scheduler->ready;
551	}	635	}
552	}	636	}
553		637
		638	=item $cb = Coro::rouse_cb
		639
		640	Create and return a "rouse callback". That's a code reference that,
		641	when called, will remember a copy of its arguments and notify the owner
		642	coroutine of the callback.
		643
		644	See the next function.
		645
		646	=item @args = Coro::rouse_wait [$cb]
		647
		648	Wait for the specified rouse callback (or the last one that was created in
		649	this coroutine).
		650
		651	As soon as the callback is invoked (or when the callback was invoked
		652	before C<rouse_wait>), it will return the arguments originally passed to
		653	the rouse callback.
		654
		655	See the section B<HOW TO WAIT FOR A CALLBACK> for an actual usage example.
		656
554	=back	657	=back
555		658
556	=cut	659	=cut
557		660
558	1;	661	1;
559		662
		663	=head1 HOW TO WAIT FOR A CALLBACK
		664
		665	It is very common for a coroutine to wait for some callback to be
		666	called. This occurs naturally when you use coroutines in an otherwise
		667	event-based program, or when you use event-based libraries.
		668
		669	These typically register a callback for some event, and call that callback
		670	when the event occured. In a coroutine, however, you typically want to
		671	just wait for the event, simplyifying things.
		672
		673	For example C<< AnyEvent->child >> registers a callback to be called when
		674	a specific child has exited:
		675
		676	my $child_watcher = AnyEvent->child (pid => $pid, cb => sub { ... });
		677
		678	But from withina coroutine, you often just want to write this:
		679
		680	my $status = wait_for_child $pid;
		681
		682	Coro offers two functions specifically designed to make this easy,
		683	C<Coro::rouse_cb> and C<Coro::rouse_wait>.
		684
		685	The first function, C<rouse_cb>, generates and returns a callback that,
		686	when invoked, will save it's arguments and notify the coroutine that
		687	created the callback.
		688
		689	The second function, C<rouse_wait>, waits for the callback to be called
		690	(by calling C<schedule> to go to sleep) and returns the arguments
		691	originally passed to the callback.
		692
		693	Using these functions, it becomes easy to write the C<wait_for_child>
		694	function mentioned above:
		695
		696	sub wait_for_child($) {
		697	my ($pid) = @_;
		698
		699	my $watcher = AnyEvent->child (pid => $pid, cb => Coro::rouse_cb);
		700
		701	my ($rpid, $rstatus) = Coro::rouse_wait;
		702	$rstatus
		703	}
		704
		705	In the case where C<rouse_cb> and C<rouse_wait> are not flexible enough,
		706	you can roll your own, using C<schedule>:
		707
		708	sub wait_for_child($) {
		709	my ($pid) = @_;
		710
		711	# store the current coroutine in $current,
		712	# and provide result variables for the closure passed to ->child
		713	my $current = $Coro::current;
		714	my ($done, $rstatus);
		715
		716	# pass a closure to ->child
		717	my $watcher = AnyEvent->child (pid => $pid, cb => sub {
		718	$rstatus = $_[1]; # remember rstatus
		719	$done = 1; # mark $rstatus as valud
		720	});
		721
		722	# wait until the closure has been called
		723	schedule while !$done;
		724
		725	$rstatus
		726	}
		727
		728
560	=head1 BUGS/LIMITATIONS	729	=head1 BUGS/LIMITATIONS
561		730
562	- you must make very sure that no coro is still active on global	731	=over 4
563	destruction. very bad things might happen otherwise (usually segfaults).
564		732
		733	=item fork with pthread backend
		734
		735	When Coro is compiled using the pthread backend (which isn't recommended
		736	but required on many BSDs as their libcs are completely broken), then
		737	coroutines will not survive a fork. There is no known workaround except to
		738	fix your libc and use a saner backend.
		739
		740	=item perl process emulation ("threads")
		741
565	- this module is not thread-safe. You should only ever use this module	742	This module is not perl-pseudo-thread-safe. You should only ever use this
566	from the same thread (this requirement might be losened in the future	743	module from the first thread (this requirement might be removed in the
567	to allow per-thread schedulers, but Coro::State does not yet allow	744	future to allow per-thread schedulers, but Coro::State does not yet allow
568	this).	745	this). I recommend disabling thread support and using processes, as having
		746	the windows process emulation enabled under unix roughly halves perl
		747	performance, even when not used.
		748
		749	=item coroutine switching not signal safe
		750
		751	You must not switch to another coroutine from within a signal handler
		752	(only relevant with %SIG - most event libraries provide safe signals).
		753
		754	That means you I<MUST NOT> call any function that might "block" the
		755	current coroutine - C<cede>, C<schedule> C<< Coro::Semaphore->down >> or
		756	anything that calls those. Everything else, including calling C<ready>,
		757	works.
		758
		759	=back
		760
569		761
570	=head1 SEE ALSO	762	=head1 SEE ALSO
571		763
		764	Event-Loop integration: L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
		765
		766	Debugging: L<Coro::Debug>.
		767
572	Support/Utility: L<Coro::Cont>, L<Coro::Specific>, L<Coro::State>, L<Coro::Util>.	768	Support/Utility: L<Coro::Specific>, L<Coro::Util>.
573		769
574	Locking/IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>, L<Coro::SemaphoreSet>, L<Coro::RWLock>.	770	Locking and IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>,
		771	L<Coro::SemaphoreSet>, L<Coro::RWLock>.
575		772
576	Event/IO: L<Coro::Timer>, L<Coro::Event>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::Select>.	773	I/O and Timers: L<Coro::Timer>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::AIO>.
577		774
578	Embedding: L<Coro:MakeMaker>	775	Compatibility with other modules: L<Coro::LWP> (but see also L<AnyEvent::HTTP> for
		776	a better-working alternative), L<Coro::BDB>, L<Coro::Storable>,
		777	L<Coro::Select>.
		778
		779	XS API: L<Coro::MakeMaker>.
		780
		781	Low level Configuration, Thread Environment, Continuations: L<Coro::State>.
579		782
580	=head1 AUTHOR	783	=head1 AUTHOR
581		784
582	Marc Lehmann <schmorp@schmorp.de>	785	Marc Lehmann <schmorp@schmorp.de>
583	http://home.schmorp.de/	786	http://home.schmorp.de/

Diff Legend

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

Comparing Coro/Coro.pm (file contents): Revision 1.108 by root, Fri Jan 5 20:00:49 2007 UTC vs. Revision 1.238 by root, Mon Nov 24 04:56:38 2008 UTC

Diff Legend

Comparing Coro/Coro.pm (file contents):
Revision 1.108 by root, Fri Jan 5 20:00:49 2007 UTC vs.
Revision 1.238 by root, Mon Nov 24 04:56:38 2008 UTC