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

Comparing Coro/Coro.pm (file contents):
Revision 1.148 by root, Fri Oct 5 20:11:25 2007 UTC vs.
Revision 1.257 by root, Tue Jun 23 23:40:06 2009 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 coro
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 used in this module also
26	guarantees 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 in
32	very slow and memory-intensive emulation of processes using threads. This	35	the form of cooperative threads (also called coros, or simply "coro"
33	is a performance win on Windows machines, and a loss everywhere else).	36	in the documentation). They are similar to kernel threads but don't (in
		37	general) run in parallel at the same time even on SMP machines. The
		38	specific flavor of thread offered by this module also guarantees you that
		39	it will not switch between threads unless necessary, at easily-identified
		40	points in your program, so locking and parallel access are rarely an
		41	issue, making thread programming much safer and easier than using other
		42	thread models.
34		43
		44	Unlike the so-called "Perl threads" (which are not actually real threads
		45	but only the windows process emulation ported to unix, and as such act
		46	as processes), Coro provides a full shared address space, which makes
		47	communication between threads very easy. And Coro's threads are fast,
		48	too: disabling the Windows process emulation code in your perl and using
		49	Coro can easily result in a two to four times speed increase for your
		50	programs. A parallel matrix multiplication benchmark runs over 300 times
		51	faster on a single core than perl's pseudo-threads on a quad core using
		52	all four cores.
		53
		54	Coro achieves that by supporting multiple running interpreters that share
		55	data, which is especially useful to code pseudo-parallel processes and
		56	for event-based programming, such as multiple HTTP-GET requests running
		57	concurrently. See L<Coro::AnyEvent> to learn more on how to integrate Coro
		58	into an event-based environment.
		59
35	In this module, coroutines are defined as "callchain + lexical variables +	60	In this module, a thread is defined as "callchain + lexical variables +
36	@_ + $_ + $@ + $/ + C stack), that is, a coroutine has its own callchain,	61	some package variables + 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	62	its own set of lexicals and its own set of perls most important global
38	variables.	63	variables (see L<Coro::State> for more configuration and background info).
		64
		65	See also the C<SEE ALSO> section at the end of this document - the Coro
		66	module family is quite large.
39		67
40	=cut	68	=cut
41		69
42	package Coro;	70	package Coro;
43		71
44	use strict;	72	use strict qw(vars subs);
45	no warnings "uninitialized";	73	no warnings "uninitialized";
46		74
		75	use Guard ();
		76
47	use Coro::State;	77	use Coro::State;
48		78
49	use base qw(Coro::State Exporter);	79	use base qw(Coro::State Exporter);
50		80
51	our $idle; # idle handler	81	our $idle; # idle handler
52	our $main; # main coroutine	82	our $main; # main coro
53	our $current; # current coroutine	83	our $current; # current coro
54		84
55	our $VERSION = '4.01';	85	our $VERSION = 5.14;
56		86
57	our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);	87	our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);
58	our %EXPORT_TAGS = (	88	our %EXPORT_TAGS = (
59	prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],	89	prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],
60	);	90	);
61	our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));	91	our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));
62		92
63	{	93	=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		94
97	=over 4	95	=over 4
98		96
99	=item $main	97	=item $Coro::main
100		98
101	This coroutine represents the main program.	99	This variable stores the Coro object that represents the main
		100	program. While you cna C<ready> it and do most other things you can do to
		101	coro, it is mainly useful to compare again C<$Coro::current>, to see
		102	whether you are running in the main program or not.
102		103
103	=cut	104	=cut
104		105
105	$main = new Coro;	106	# $main is now being initialised by Coro::State
106		107
107	=item $current (or as function: current)	108	=item $Coro::current
108		109
109	The current coroutine (the last coroutine switched to). The initial value	110	The Coro object representing the current coro (the last
		111	coro that the Coro scheduler switched to). The initial value is
110	is C<$main> (of course).	112	C<$Coro::main> (of course).
111		113
112	This variable is B<strictly> I<read-only>. It is provided for performance	114	This variable is B<strictly> I<read-only>. You can take copies of the
113	reasons. If performance is not essential you are encouraged to use the	115	value stored in it and use it as any other Coro object, but you must
114	C<Coro::current> function instead.	116	not otherwise modify the variable itself.
115		117
116	=cut	118	=cut
117		119
118	$main->{desc} = "[main::]";
119
120	# maybe some other module used Coro::Specific before...
121	$main->{_specific} = $current->{_specific}
122	if $current;
123
124	_set_current $main;
125
126	sub current() { $current }	120	sub current() { $current } # [DEPRECATED]
127		121
128	=item $idle	122	=item $Coro::idle
129		123
		124	This variable is mainly useful to integrate Coro into event loops. It is
		125	usually better to rely on L<Coro::AnyEvent> or L<Coro::EV>, as this is
		126	pretty low-level functionality.
		127
		128	This variable stores either a Coro object or a callback.
		129
130	A callback that is called whenever the scheduler finds no ready coroutines	130	If it is a callback, the it is called whenever the scheduler finds no
131	to run. The default implementation prints "FATAL: deadlock detected" and	131	ready coros to run. The default implementation prints "FATAL:
132	exits, because the program has no other way to continue.	132	deadlock detected" and exits, because the program has no other way to
		133	continue.
133		134
		135	If it is a coro object, then this object will be readied (without
		136	invoking any ready hooks, however) when the scheduler finds no other ready
		137	coros to run.
		138
134	This hook is overwritten by modules such as C<Coro::Timer> and	139	This hook is overwritten by modules such as C<Coro::EV> and
135	C<Coro::Event> to wait on an external event that hopefully wake up a	140	C<Coro::AnyEvent> to wait on an external event that hopefully wake up a
136	coroutine so the scheduler can run it.	141	coro so the scheduler can run it.
		142
		143	Note that the callback I<must not>, under any circumstances, block
		144	the current coro. Normally, this is achieved by having an "idle
		145	coro" that calls the event loop and then blocks again, and then
		146	readying that coro in the idle handler, or by simply placing the idle
		147	coro in this variable.
		148
		149	See L<Coro::Event> or L<Coro::AnyEvent> for examples of using this
		150	technique.
137		151
138	Please note that if your callback recursively invokes perl (e.g. for event	152	Please note that if your callback recursively invokes perl (e.g. for event
139	handlers), then it must be prepared to be called recursively.	153	handlers), then it must be prepared to be called recursively itself.
140		154
141	=cut	155	=cut
142		156
143	$idle = sub {	157	$idle = sub {
144	require Carp;	158	require Carp;
145	Carp::croak ("FATAL: deadlock detected");	159	Carp::croak ("FATAL: deadlock detected");
146	};	160	};
147		161
148	sub _cancel {
149	my ($self) = @_;
150
151	# free coroutine data and mark as destructed
152	$self->_destroy
153	or return;
154
155	# call all destruction callbacks
156	$_->(@{$self->{_status}})
157	for @{(delete $self->{_on_destroy}) \|\| []};
158	}
159
160	# this coroutine is necessary because a coroutine	162	# this coro is necessary because a coro
161	# cannot destroy itself.	163	# cannot destroy itself.
162	my @destroy;	164	our @destroy;
163	my $manager;	165	our $manager;
164		166
165	$manager = new Coro sub {	167	$manager = new Coro sub {
166	while () {	168	while () {
167	(shift @destroy)->_cancel	169	Coro::State::cancel shift @destroy
168	while @destroy;	170	while @destroy;
169		171
170	&schedule;	172	&schedule;
171	}	173	}
172	};	174	};
173	$manager->desc ("[coro manager]");	175	$manager->{desc} = "[coro manager]";
174	$manager->prio (PRIO_MAX);	176	$manager->prio (PRIO_MAX);
175		177
176	# static methods. not really.
177
178	=back	178	=back
179		179
180	=head2 STATIC METHODS	180	=head1 SIMPLE CORO CREATION
181
182	Static methods are actually functions that operate on the current coroutine only.
183		181
184	=over 4	182	=over 4
185		183
186	=item async { ... } [@args...]	184	=item async { ... } [@args...]
187		185
188	Create a new asynchronous coroutine and return it's coroutine object	186	Create a new coro and return its Coro object (usually
189	(usually unused). When the sub returns the new coroutine is automatically	187	unused). The coro will be put into the ready queue, so
		188	it will start running automatically on the next scheduler run.
		189
		190	The first argument is a codeblock/closure that should be executed in the
		191	coro. When it returns argument returns the coro is automatically
190	terminated.	192	terminated.
191		193
		194	The remaining arguments are passed as arguments to the closure.
		195
192	See the C<Coro::State::new> constructor for info about the coroutine	196	See the C<Coro::State::new> constructor for info about the coro
193	environment.	197	environment in which coro are executed.
194		198
195	Calling C<exit> in a coroutine will do the same as calling exit outside	199	Calling C<exit> in a coro will do the same as calling exit outside
196	the coroutine. Likewise, when the coroutine dies, the program will exit,	200	the coro. Likewise, when the coro dies, the program will exit,
197	just as it would in the main program.	201	just as it would in the main program.
198		202
		203	If you do not want that, you can provide a default C<die> handler, or
		204	simply avoid dieing (by use of C<eval>).
		205
199	# create a new coroutine that just prints its arguments	206	Example: Create a new coro that just prints its arguments.
		207
200	async {	208	async {
201	print "@_\n";	209	print "@_\n";
202	} 1,2,3,4;	210	} 1,2,3,4;
203		211
204	=cut	212	=cut
209	$coro	217	$coro
210	}	218	}
211		219
212	=item async_pool { ... } [@args...]	220	=item async_pool { ... } [@args...]
213		221
214	Similar to C<async>, but uses a coroutine pool, so you should not call	222	Similar to C<async>, but uses a coro pool, so you should not call
215	terminate or join (although you are allowed to), and you get a coroutine	223	terminate or join on it (although you are allowed to), and you get a
216	that might have executed other code already (which can be good or bad :).	224	coro that might have executed other code already (which can be good
		225	or bad :).
217		226
		227	On the plus side, this function is about twice as fast as creating (and
		228	destroying) a completely new coro, so if you need a lot of generic
		229	coros in quick successsion, use C<async_pool>, not C<async>.
		230
218	Also, the block is executed in an C<eval> context and a warning will be	231	The code block is executed in an C<eval> context and a warning will be
219	issued in case of an exception instead of terminating the program, as	232	issued in case of an exception instead of terminating the program, as
220	C<async> does. As the coroutine is being reused, stuff like C<on_destroy>	233	C<async> does. As the coro is being reused, stuff like C<on_destroy>
221	will not work in the expected way, unless you call terminate or cancel,	234	will not work in the expected way, unless you call terminate or cancel,
222	which somehow defeats the purpose of pooling.	235	which somehow defeats the purpose of pooling (but is fine in the
		236	exceptional case).
223		237
224	The priority will be reset to C<0> after each job, tracing will be	238	The priority will be reset to C<0> after each run, tracing will be
225	disabled, the description will be reset and the default output filehandle	239	disabled, the description will be reset and the default output filehandle
226	gets restored, so you can change alkl these. Otherwise the coroutine will	240	gets restored, so you can change all these. Otherwise the coro will
227	be re-used "as-is": most notably if you change other per-coroutine global	241	be re-used "as-is": most notably if you change other per-coro global
228	stuff such as C<$/> you need to revert that change, which is most simply	242	stuff such as C<$/> you I<must needs> revert that change, which is most
229	done by using local as in C< local $/ >.	243	simply done by using local as in: C<< local $/ >>.
230		244
231	The pool size is limited to 8 idle coroutines (this can be adjusted by	245	The idle pool size is limited to C<8> idle coros (this can be
232	changing $Coro::POOL_SIZE), and there can be as many non-idle coros as	246	adjusted by changing $Coro::POOL_SIZE), but there can be as many non-idle
233	required.	247	coros as required.
234		248
235	If you are concerned about pooled coroutines growing a lot because a	249	If you are concerned about pooled coros growing a lot because a
236	single C<async_pool> used a lot of stackspace you can e.g. C<async_pool	250	single C<async_pool> used a lot of stackspace you can e.g. C<async_pool
237	{ terminate }> once per second or so to slowly replenish the pool. In	251	{ terminate }> once per second or so to slowly replenish the pool. In
238	addition to that, when the stacks used by a handler grows larger than 16kb	252	addition to that, when the stacks used by a handler grows larger than 32kb
239	(adjustable with $Coro::POOL_RSS) it will also exit.	253	(adjustable via $Coro::POOL_RSS) it will also be destroyed.
240		254
241	=cut	255	=cut
242		256
243	our $POOL_SIZE = 8;	257	our $POOL_SIZE = 8;
244	our $POOL_RSS = 16 * 1024;	258	our $POOL_RSS = 32 * 1024;
245	our @async_pool;	259	our @async_pool;
246		260
247	sub pool_handler {	261	sub pool_handler {
248	my $cb;
249
250	while () {	262	while () {
251	eval {	263	eval {
252	while () {	264	&{&_pool_handler} while 1;
253	_pool_1 $cb;
254	&$cb;
255	_pool_2 $cb;
256	&schedule;
257	}
258	};	265	};
259		266
260	last if $@ eq "\3terminate\2\n";
261	warn $@ if $@;	267	warn $@ if $@;
262	}	268	}
263	}	269	}
264		270
265	sub async_pool(&@) {	271	=back
266	# this is also inlined into the unlock_scheduler
267	my $coro = (pop @async_pool) \|\| new Coro \&pool_handler;
268		272
269	$coro->{_invoke} = [@_];	273	=head1 STATIC METHODS
270	$coro->ready;
271		274
272	$coro	275	Static methods are actually functions that implicitly operate on the
273	}	276	current coro.
		277
		278	=over 4
274		279
275	=item schedule	280	=item schedule
276		281
277	Calls the scheduler. Please note that the current coroutine will not be put	282	Calls the scheduler. The scheduler will find the next coro that is
		283	to be run from the ready queue and switches to it. The next coro
		284	to be run is simply the one with the highest priority that is longest
		285	in its ready queue. If there is no coro ready, it will clal the
		286	C<$Coro::idle> hook.
		287
		288	Please note that the current coro will I<not> be put into the ready
278	into the ready queue, so calling this function usually means you will	289	queue, so calling this function usually means you will never be called
279	never be called again unless something else (e.g. an event handler) calls	290	again unless something else (e.g. an event handler) calls C<< ->ready >>,
280	ready.	291	thus waking you up.
281		292
282	The canonical way to wait on external events is this:	293	This makes C<schedule> I<the> generic method to use to block the current
		294	coro and wait for events: first you remember the current coro in
		295	a variable, then arrange for some callback of yours to call C<< ->ready
		296	>> on that once some event happens, and last you call C<schedule> to put
		297	yourself to sleep. Note that a lot of things can wake your coro up,
		298	so you need to check whether the event indeed happened, e.g. by storing the
		299	status in a variable.
283		300
284	{	301	See B<HOW TO WAIT FOR A CALLBACK>, below, for some ways to wait for callbacks.
285	# remember current coroutine
286	my $current = $Coro::current;
287		302
288	# register a hypothetical event handler	303	=item cede
289	on_event_invoke sub {	304
290	# wake up sleeping coroutine	305	"Cede" to other coros. This function puts the current coro into
291	$current->ready;	306	the ready queue and calls C<schedule>, which has the effect of giving
292	undef $current;	307	up the current "timeslice" to other coros of the same or higher
		308	priority. Once your coro gets its turn again it will automatically be
		309	resumed.
		310
		311	This function is often called C<yield> in other languages.
		312
		313	=item Coro::cede_notself
		314
		315	Works like cede, but is not exported by default and will cede to I<any>
		316	coro, regardless of priority. This is useful sometimes to ensure
		317	progress is made.
		318
		319	=item terminate [arg...]
		320
		321	Terminates the current coro with the given status values (see L<cancel>).
		322
		323	=item Coro::on_enter BLOCK, Coro::on_leave BLOCK
		324
		325	These function install enter and leave winders in the current scope. The
		326	enter block will be executed when on_enter is called and whenever the
		327	current coro is re-entered by the scheduler, while the leave block is
		328	executed whenever the current coro is blocked by the scheduler, and
		329	also when the containing scope is exited (by whatever means, be it exit,
		330	die, last etc.).
		331
		332	I<Neither invoking the scheduler, nor exceptions, are allowed within those
		333	BLOCKs>. That means: do not even think about calling C<die> without an
		334	eval, and do not even think of entering the scheduler in any way.
		335
		336	Since both BLOCKs are tied to the current scope, they will automatically
		337	be removed when the current scope exits.
		338
		339	These functions implement the same concept as C<dynamic-wind> in scheme
		340	does, and are useful when you want to localise some resource to a specific
		341	coro.
		342
		343	They slow down thread switching considerably for coros that use them
		344	(about 40% for a BLOCK with a single assignment, so thread switching is
		345	still reasonably fast if the handlers are fast).
		346
		347	These functions are best understood by an example: The following function
		348	will change the current timezone to "Antarctica/South_Pole", which
		349	requires a call to C<tzset>, but by using C<on_enter> and C<on_leave>,
		350	which remember/change the current timezone and restore the previous
		351	value, respectively, the timezone is only changed for the coro that
		352	installed those handlers.
		353
		354	use POSIX qw(tzset);
		355
		356	async {
		357	my $old_tz; # store outside TZ value here
		358
		359	Coro::on_enter {
		360	$old_tz = $ENV{TZ}; # remember the old value
		361
		362	$ENV{TZ} = "Antarctica/South_Pole";
		363	tzset; # enable new value
293	};	364	};
294		365
295	# call schedule until event occurred.	366	Coro::on_leave {
296	# in case we are woken up for other reasons	367	$ENV{TZ} = $old_tz;
297	# (current still defined), loop.	368	tzset; # restore old value
298	Coro::schedule while $current;	369	};
		370
		371	# at this place, the timezone is Antarctica/South_Pole,
		372	# without disturbing the TZ of any other coro.
299	}	373	};
300		374
301	=item cede	375	This can be used to localise about any resource (locale, uid, current
		376	working directory etc.) to a block, despite the existance of other
		377	coros.
302		378
303	"Cede" to other coroutines. This function puts the current coroutine into the	379	Another interesting example implements time-sliced multitasking using
304	ready queue and calls C<schedule>, which has the effect of giving up the	380	interval timers (this could obviously be optimised, but does the job):
305	current "timeslice" to other coroutines of the same or higher priority.
306		381
307	Returns true if at least one coroutine switch has happened.	382	# "timeslice" the given block
		383	sub timeslice(&) {
		384	use Time::HiRes ();
308		385
309	=item Coro::cede_notself	386	Coro::on_enter {
		387	# on entering the thread, we set an VTALRM handler to cede
		388	$SIG{VTALRM} = sub { cede };
		389	# and then start the interval timer
		390	Time::HiRes::setitimer &Time::HiRes::ITIMER_VIRTUAL, 0.01, 0.01;
		391	};
		392	Coro::on_leave {
		393	# on leaving the thread, we stop the interval timer again
		394	Time::HiRes::setitimer &Time::HiRes::ITIMER_VIRTUAL, 0, 0;
		395	};
310		396
311	Works like cede, but is not exported by default and will cede to any	397	&{+shift};
312	coroutine, regardless of priority, once.	398	}
313		399
314	Returns true if at least one coroutine switch has happened.	400	# use like this:
		401	timeslice {
		402	# The following is an endless loop that would normally
		403	# monopolise the process. Since it runs in a timesliced
		404	# environment, it will regularly cede to other threads.
		405	while () { }
		406	};
315		407
316	=item terminate [arg...]
317
318	Terminates the current coroutine with the given status values (see L<cancel>).
319		408
320	=item killall	409	=item killall
321		410
322	Kills/terminates/cancels all coroutines except the currently running	411	Kills/terminates/cancels all coros except the currently running one.
323	one. This is useful after a fork, either in the child or the parent, as
324	usually only one of them should inherit the running coroutines.
325		412
326	=cut	413	Note that while this will try to free some of the main interpreter
		414	resources if the calling coro isn't the main coro, but one
		415	cannot free all of them, so if a coro that is not the main coro
		416	calls this function, there will be some one-time resource leak.
327		417
328	sub terminate {	418	=cut
329	$current->cancel (@_);
330	}
331		419
332	sub killall {	420	sub killall {
333	for (Coro::State::list) {	421	for (Coro::State::list) {
334	$_->cancel	422	$_->cancel
335	if $_ != $current && UNIVERSAL::isa $_, "Coro";	423	if $_ != $current && UNIVERSAL::isa $_, "Coro";
336	}	424	}
337	}	425	}
338		426
339	=back	427	=back
340		428
341	# dynamic methods
342
343	=head2 COROUTINE METHODS	429	=head1 CORO OBJECT METHODS
344		430
345	These are the methods you can call on coroutine objects.	431	These are the methods you can call on coro objects (or to create
		432	them).
346		433
347	=over 4	434	=over 4
348		435
349	=item new Coro \&sub [, @args...]	436	=item new Coro \&sub [, @args...]
350		437
351	Create a new coroutine and return it. When the sub returns the coroutine	438	Create a new coro and return it. When the sub returns, the coro
352	automatically terminates as if C<terminate> with the returned values were	439	automatically terminates as if C<terminate> with the returned values were
353	called. To make the coroutine run you must first put it into the ready queue	440	called. To make the coro run you must first put it into the ready
354	by calling the ready method.	441	queue by calling the ready method.
355		442
356	See C<async> and C<Coro::State::new> for additional info about the	443	See C<async> and C<Coro::State::new> for additional info about the
357	coroutine environment.	444	coro environment.
358		445
359	=cut	446	=cut
360		447
361	sub _run_coro {	448	sub _coro_run {
362	terminate &{+shift};	449	terminate &{+shift};
363	}	450	}
364		451
365	sub new {
366	my $class = shift;
367
368	$class->SUPER::new (\&_run_coro, @_)
369	}
370
371	=item $success = $coroutine->ready	452	=item $success = $coro->ready
372		453
373	Put the given coroutine into the ready queue (according to it's priority)	454	Put the given coro into the end of its ready queue (there is one
374	and return true. If the coroutine is already in the ready queue, do nothing	455	queue for each priority) and return true. If the coro is already in
375	and return false.	456	the ready queue, do nothing and return false.
376		457
		458	This ensures that the scheduler will resume this coro automatically
		459	once all the coro of higher priority and all coro of the same
		460	priority that were put into the ready queue earlier have been resumed.
		461
		462	=item $coro->suspend
		463
		464	Suspends the specified coro. A suspended coro works just like any other
		465	coro, except that the scheduler will not select a suspended coro for
		466	execution.
		467
		468	Suspending a coro can be useful when you want to keep the coro from
		469	running, but you don't want to destroy it, or when you want to temporarily
		470	freeze a coro (e.g. for debugging) to resume it later.
		471
		472	A scenario for the former would be to suspend all (other) coros after a
		473	fork and keep them alive, so their destructors aren't called, but new
		474	coros can be created.
		475
		476	=item $coro->resume
		477
		478	If the specified coro was suspended, it will be resumed. Note that when
		479	the coro was in the ready queue when it was suspended, it might have been
		480	unreadied by the scheduler, so an activation might have been lost.
		481
		482	To avoid this, it is best to put a suspended coro into the ready queue
		483	unconditionally, as every synchronisation mechanism must protect itself
		484	against spurious wakeups, and the one in the Coro family certainly do
		485	that.
		486
377	=item $is_ready = $coroutine->is_ready	487	=item $is_ready = $coro->is_ready
378		488
379	Return wether the coroutine is currently the ready queue or not,	489	Returns true iff the Coro object is in the ready queue. Unless the Coro
		490	object gets destroyed, it will eventually be scheduled by the scheduler.
380		491
		492	=item $is_running = $coro->is_running
		493
		494	Returns true iff the Coro object is currently running. Only one Coro object
		495	can ever be in the running state (but it currently is possible to have
		496	multiple running Coro::States).
		497
		498	=item $is_suspended = $coro->is_suspended
		499
		500	Returns true iff this Coro object has been suspended. Suspended Coros will
		501	not ever be scheduled.
		502
381	=item $coroutine->cancel (arg...)	503	=item $coro->cancel (arg...)
382		504
383	Terminates the given coroutine and makes it return the given arguments as	505	Terminates the given Coro and makes it return the given arguments as
384	status (default: the empty list). Never returns if the coroutine is the	506	status (default: the empty list). Never returns if the Coro is the
385	current coroutine.	507	current Coro.
386		508
387	=cut	509	=cut
388		510
389	sub cancel {	511	sub cancel {
390	my $self = shift;	512	my $self = shift;
391	$self->{_status} = [@_];
392		513
393	if ($current == $self) {	514	if ($current == $self) {
394	push @destroy, $self;	515	terminate @_;
395	$manager->ready;
396	&schedule while 1;
397	} else {	516	} else {
398	$self->_cancel;	517	$self->{_status} = [@_];
		518	Coro::State::cancel $self;
399	}	519	}
400	}	520	}
401		521
		522	=item $coro->schedule_to
		523
		524	Puts the current coro to sleep (like C<Coro::schedule>), but instead
		525	of continuing with the next coro from the ready queue, always switch to
		526	the given coro object (regardless of priority etc.). The readyness
		527	state of that coro isn't changed.
		528
		529	This is an advanced method for special cases - I'd love to hear about any
		530	uses for this one.
		531
		532	=item $coro->cede_to
		533
		534	Like C<schedule_to>, but puts the current coro into the ready
		535	queue. This has the effect of temporarily switching to the given
		536	coro, and continuing some time later.
		537
		538	This is an advanced method for special cases - I'd love to hear about any
		539	uses for this one.
		540
		541	=item $coro->throw ([$scalar])
		542
		543	If C<$throw> is specified and defined, it will be thrown as an exception
		544	inside the coro at the next convenient point in time. Otherwise
		545	clears the exception object.
		546
		547	Coro will check for the exception each time a schedule-like-function
		548	returns, i.e. after each C<schedule>, C<cede>, C<< Coro::Semaphore->down
		549	>>, C<< Coro::Handle->readable >> and so on. Most of these functions
		550	detect this case and return early in case an exception is pending.
		551
		552	The exception object will be thrown "as is" with the specified scalar in
		553	C<$@>, i.e. if it is a string, no line number or newline will be appended
		554	(unlike with C<die>).
		555
		556	This can be used as a softer means than C<cancel> to ask a coro to
		557	end itself, although there is no guarantee that the exception will lead to
		558	termination, and if the exception isn't caught it might well end the whole
		559	program.
		560
		561	You might also think of C<throw> as being the moral equivalent of
		562	C<kill>ing a coro with a signal (in this case, a scalar).
		563
402	=item $coroutine->join	564	=item $coro->join
403		565
404	Wait until the coroutine terminates and return any values given to the	566	Wait until the coro terminates and return any values given to the
405	C<terminate> or C<cancel> functions. C<join> can be called concurrently	567	C<terminate> or C<cancel> functions. C<join> can be called concurrently
406	from multiple coroutines.	568	from multiple coro, and all will be resumed and given the status
		569	return once the C<$coro> terminates.
407		570
408	=cut	571	=cut
409		572
410	sub join {	573	sub join {
411	my $self = shift;	574	my $self = shift;
…		…
422	}	585	}
423		586
424	wantarray ? @{$self->{_status}} : $self->{_status}[0];	587	wantarray ? @{$self->{_status}} : $self->{_status}[0];
425	}	588	}
426		589
427	=item $coroutine->on_destroy (\&cb)	590	=item $coro->on_destroy (\&cb)
428		591
429	Registers a callback that is called when this coroutine gets destroyed,	592	Registers a callback that is called when this coro gets destroyed,
430	but before it is joined. The callback gets passed the terminate arguments,	593	but before it is joined. The callback gets passed the terminate arguments,
431	if any.	594	if any, and I<must not> die, under any circumstances.
432		595
433	=cut	596	=cut
434		597
435	sub on_destroy {	598	sub on_destroy {
436	my ($self, $cb) = @_;	599	my ($self, $cb) = @_;
437		600
438	push @{ $self->{_on_destroy} }, $cb;	601	push @{ $self->{_on_destroy} }, $cb;
439	}	602	}
440		603
441	=item $oldprio = $coroutine->prio ($newprio)	604	=item $oldprio = $coro->prio ($newprio)
442		605
443	Sets (or gets, if the argument is missing) the priority of the	606	Sets (or gets, if the argument is missing) the priority of the
444	coroutine. Higher priority coroutines get run before lower priority	607	coro. Higher priority coro get run before lower priority
445	coroutines. Priorities are small signed integers (currently -4 .. +3),	608	coro. Priorities are small signed integers (currently -4 .. +3),
446	that you can refer to using PRIO_xxx constants (use the import tag :prio	609	that you can refer to using PRIO_xxx constants (use the import tag :prio
447	to get then):	610	to get then):
448		611
449	PRIO_MAX > PRIO_HIGH > PRIO_NORMAL > PRIO_LOW > PRIO_IDLE > PRIO_MIN	612	PRIO_MAX > PRIO_HIGH > PRIO_NORMAL > PRIO_LOW > PRIO_IDLE > PRIO_MIN
450	3 > 1 > 0 > -1 > -3 > -4	613	3 > 1 > 0 > -1 > -3 > -4
451		614
452	# set priority to HIGH	615	# set priority to HIGH
453	current->prio(PRIO_HIGH);	616	current->prio (PRIO_HIGH);
454		617
455	The idle coroutine ($Coro::idle) always has a lower priority than any	618	The idle coro ($Coro::idle) always has a lower priority than any
456	existing coroutine.	619	existing coro.
457		620
458	Changing the priority of the current coroutine will take effect immediately,	621	Changing the priority of the current coro will take effect immediately,
459	but changing the priority of coroutines in the ready queue (but not	622	but changing the priority of coro in the ready queue (but not
460	running) will only take effect after the next schedule (of that	623	running) will only take effect after the next schedule (of that
461	coroutine). This is a bug that will be fixed in some future version.	624	coro). This is a bug that will be fixed in some future version.
462		625
463	=item $newprio = $coroutine->nice ($change)	626	=item $newprio = $coro->nice ($change)
464		627
465	Similar to C<prio>, but subtract the given value from the priority (i.e.	628	Similar to C<prio>, but subtract the given value from the priority (i.e.
466	higher values mean lower priority, just as in unix).	629	higher values mean lower priority, just as in unix).
467		630
468	=item $olddesc = $coroutine->desc ($newdesc)	631	=item $olddesc = $coro->desc ($newdesc)
469		632
470	Sets (or gets in case the argument is missing) the description for this	633	Sets (or gets in case the argument is missing) the description for this
471	coroutine. This is just a free-form string you can associate with a coroutine.	634	coro. This is just a free-form string you can associate with a
		635	coro.
472		636
473	This method simply sets the C<< $coroutine->{desc} >> member to the given string. You	637	This method simply sets the C<< $coro->{desc} >> member to the given
474	can modify this member directly if you wish.	638	string. You can modify this member directly if you wish.
475		639
476	=cut	640	=cut
477		641
478	sub desc {	642	sub desc {
479	my $old = $_[0]{desc};	643	my $old = $_[0]{desc};
480	$_[0]{desc} = $_[1] if @_ > 1;	644	$_[0]{desc} = $_[1] if @_ > 1;
481	$old;	645	$old;
482	}	646	}
483		647
		648	sub transfer {
		649	require Carp;
		650	Carp::croak ("You must not call ->transfer on Coro objects. Use Coro::State objects or the ->schedule_to method. Caught");
		651	}
		652
484	=back	653	=back
485		654
486	=head2 GLOBAL FUNCTIONS	655	=head1 GLOBAL FUNCTIONS
487		656
488	=over 4	657	=over 4
489		658
490	=item Coro::nready	659	=item Coro::nready
491		660
492	Returns the number of coroutines that are currently in the ready state,	661	Returns the number of coro that are currently in the ready state,
493	i.e. that can be switched to. The value C<0> means that the only runnable	662	i.e. that can be switched to by calling C<schedule> directory or
		663	indirectly. The value C<0> means that the only runnable coro is the
494	coroutine is the currently running one, so C<cede> would have no effect,	664	currently running one, so C<cede> would have no effect, and C<schedule>
495	and C<schedule> would cause a deadlock unless there is an idle handler	665	would cause a deadlock unless there is an idle handler that wakes up some
496	that wakes up some coroutines.	666	coro.
497		667
498	=item my $guard = Coro::guard { ... }	668	=item my $guard = Coro::guard { ... }
499		669
500	This creates and returns a guard object. Nothing happens until the object	670	This function still exists, but is deprecated. Please use the
501	gets destroyed, in which case the codeblock given as argument will be	671	C<Guard::guard> function instead.
502	executed. This is useful to free locks or other resources in case of a
503	runtime error or when the coroutine gets canceled, as in both cases the
504	guard block will be executed. The guard object supports only one method,
505	C<< ->cancel >>, which will keep the codeblock from being executed.
506		672
507	Example: set some flag and clear it again when the coroutine gets canceled
508	or the function returns:
509
510	sub do_something {
511	my $guard = Coro::guard { $busy = 0 };
512	$busy = 1;
513
514	# do something that requires $busy to be true
515	}
516
517	=cut	673	=cut
518		674
519	sub guard(&) {	675	BEGIN { *guard = \&Guard::guard }
520	bless \(my $cb = $_[0]), "Coro::guard"
521	}
522
523	sub Coro::guard::cancel {
524	${$_[0]} = sub { };
525	}
526
527	sub Coro::guard::DESTROY {
528	${$_[0]}->();
529	}
530
531		676
532	=item unblock_sub { ... }	677	=item unblock_sub { ... }
533		678
534	This utility function takes a BLOCK or code reference and "unblocks" it,	679	This utility function takes a BLOCK or code reference and "unblocks" it,
535	returning the new coderef. This means that the new coderef will return	680	returning a new coderef. Unblocking means that calling the new coderef
536	immediately without blocking, returning nothing, while the original code	681	will return immediately without blocking, returning nothing, while the
537	ref will be called (with parameters) from within its own coroutine.	682	original code ref will be called (with parameters) from within another
		683	coro.
538		684
539	The reason this function exists is that many event libraries (such as the	685	The reason this function exists is that many event libraries (such as the
540	venerable L<Event\|Event> module) are not coroutine-safe (a weaker form	686	venerable L<Event\|Event> module) are not thread-safe (a weaker form
541	of thread-safety). This means you must not block within event callbacks,	687	of reentrancy). This means you must not block within event callbacks,
542	otherwise you might suffer from crashes or worse.	688	otherwise you might suffer from crashes or worse. The only event library
		689	currently known that is safe to use without C<unblock_sub> is L<EV>.
543		690
544	This function allows your callbacks to block by executing them in another	691	This function allows your callbacks to block by executing them in another
545	coroutine where it is safe to block. One example where blocking is handy	692	coro where it is safe to block. One example where blocking is handy
546	is when you use the L<Coro::AIO\|Coro::AIO> functions to save results to	693	is when you use the L<Coro::AIO\|Coro::AIO> functions to save results to
547	disk.	694	disk, for example.
548		695
549	In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when	696	In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when
550	creating event callbacks that want to block.	697	creating event callbacks that want to block.
		698
		699	If your handler does not plan to block (e.g. simply sends a message to
		700	another coro, or puts some other coro into the ready queue), there is
		701	no reason to use C<unblock_sub>.
		702
		703	Note that you also need to use C<unblock_sub> for any other callbacks that
		704	are indirectly executed by any C-based event loop. For example, when you
		705	use a module that uses L<AnyEvent> (and you use L<Coro::AnyEvent>) and it
		706	provides callbacks that are the result of some event callback, then you
		707	must not block either, or use C<unblock_sub>.
551		708
552	=cut	709	=cut
553		710
554	our @unblock_queue;	711	our @unblock_queue;
555		712
…		…
558	# return immediately and can be reused) and because we cannot cede	715	# return immediately and can be reused) and because we cannot cede
559	# inside an event callback.	716	# inside an event callback.
560	our $unblock_scheduler = new Coro sub {	717	our $unblock_scheduler = new Coro sub {
561	while () {	718	while () {
562	while (my $cb = pop @unblock_queue) {	719	while (my $cb = pop @unblock_queue) {
563	# this is an inlined copy of async_pool	720	&async_pool (@$cb);
564	my $coro = (pop @async_pool) \|\| new Coro \&pool_handler;
565		721
566	$coro->{_invoke} = $cb;
567	$coro->ready;
568	cede; # for short-lived callbacks, this reduces pressure on the coro pool	722	# for short-lived callbacks, this reduces pressure on the coro pool
		723	# as the chance is very high that the async_poll coro will be back
		724	# in the idle state when cede returns
		725	cede;
569	}	726	}
570	schedule; # sleep well	727	schedule; # sleep well
571	}	728	}
572	};	729	};
573	$unblock_scheduler->desc ("[unblock_sub scheduler]");	730	$unblock_scheduler->{desc} = "[unblock_sub scheduler]";
574		731
575	sub unblock_sub(&) {	732	sub unblock_sub(&) {
576	my $cb = shift;	733	my $cb = shift;
577		734
578	sub {	735	sub {
579	unshift @unblock_queue, [$cb, @_];	736	unshift @unblock_queue, [$cb, @_];
580	$unblock_scheduler->ready;	737	$unblock_scheduler->ready;
581	}	738	}
582	}	739	}
583		740
		741	=item $cb = Coro::rouse_cb
		742
		743	Create and return a "rouse callback". That's a code reference that,
		744	when called, will remember a copy of its arguments and notify the owner
		745	coro of the callback.
		746
		747	See the next function.
		748
		749	=item @args = Coro::rouse_wait [$cb]
		750
		751	Wait for the specified rouse callback (or the last one that was created in
		752	this coro).
		753
		754	As soon as the callback is invoked (or when the callback was invoked
		755	before C<rouse_wait>), it will return the arguments originally passed to
		756	the rouse callback.
		757
		758	See the section B<HOW TO WAIT FOR A CALLBACK> for an actual usage example.
		759
584	=back	760	=back
585		761
586	=cut	762	=cut
587		763
588	1;	764	1;
589		765
		766	=head1 HOW TO WAIT FOR A CALLBACK
		767
		768	It is very common for a coro to wait for some callback to be
		769	called. This occurs naturally when you use coro in an otherwise
		770	event-based program, or when you use event-based libraries.
		771
		772	These typically register a callback for some event, and call that callback
		773	when the event occured. In a coro, however, you typically want to
		774	just wait for the event, simplyifying things.
		775
		776	For example C<< AnyEvent->child >> registers a callback to be called when
		777	a specific child has exited:
		778
		779	my $child_watcher = AnyEvent->child (pid => $pid, cb => sub { ... });
		780
		781	But from within a coro, you often just want to write this:
		782
		783	my $status = wait_for_child $pid;
		784
		785	Coro offers two functions specifically designed to make this easy,
		786	C<Coro::rouse_cb> and C<Coro::rouse_wait>.
		787
		788	The first function, C<rouse_cb>, generates and returns a callback that,
		789	when invoked, will save its arguments and notify the coro that
		790	created the callback.
		791
		792	The second function, C<rouse_wait>, waits for the callback to be called
		793	(by calling C<schedule> to go to sleep) and returns the arguments
		794	originally passed to the callback.
		795
		796	Using these functions, it becomes easy to write the C<wait_for_child>
		797	function mentioned above:
		798
		799	sub wait_for_child($) {
		800	my ($pid) = @_;
		801
		802	my $watcher = AnyEvent->child (pid => $pid, cb => Coro::rouse_cb);
		803
		804	my ($rpid, $rstatus) = Coro::rouse_wait;
		805	$rstatus
		806	}
		807
		808	In the case where C<rouse_cb> and C<rouse_wait> are not flexible enough,
		809	you can roll your own, using C<schedule>:
		810
		811	sub wait_for_child($) {
		812	my ($pid) = @_;
		813
		814	# store the current coro in $current,
		815	# and provide result variables for the closure passed to ->child
		816	my $current = $Coro::current;
		817	my ($done, $rstatus);
		818
		819	# pass a closure to ->child
		820	my $watcher = AnyEvent->child (pid => $pid, cb => sub {
		821	$rstatus = $_[1]; # remember rstatus
		822	$done = 1; # mark $rstatus as valud
		823	});
		824
		825	# wait until the closure has been called
		826	schedule while !$done;
		827
		828	$rstatus
		829	}
		830
		831
590	=head1 BUGS/LIMITATIONS	832	=head1 BUGS/LIMITATIONS
591		833
592	- you must make very sure that no coro is still active on global	834	=over 4
593	destruction. very bad things might happen otherwise (usually segfaults).
594		835
		836	=item fork with pthread backend
		837
		838	When Coro is compiled using the pthread backend (which isn't recommended
		839	but required on many BSDs as their libcs are completely broken), then
		840	coro will not survive a fork. There is no known workaround except to
		841	fix your libc and use a saner backend.
		842
		843	=item perl process emulation ("threads")
		844
595	- this module is not thread-safe. You should only ever use this module	845	This module is not perl-pseudo-thread-safe. You should only ever use this
596	from the same thread (this requirement might be loosened in the future	846	module from the first thread (this requirement might be removed in the
597	to allow per-thread schedulers, but Coro::State does not yet allow	847	future to allow per-thread schedulers, but Coro::State does not yet allow
598	this).	848	this). I recommend disabling thread support and using processes, as having
		849	the windows process emulation enabled under unix roughly halves perl
		850	performance, even when not used.
		851
		852	=item coro switching is not signal safe
		853
		854	You must not switch to another coro from within a signal handler
		855	(only relevant with %SIG - most event libraries provide safe signals).
		856
		857	That means you I<MUST NOT> call any function that might "block" the
		858	current coro - C<cede>, C<schedule> C<< Coro::Semaphore->down >> or
		859	anything that calls those. Everything else, including calling C<ready>,
		860	works.
		861
		862	=back
		863
599		864
600	=head1 SEE ALSO	865	=head1 SEE ALSO
601		866
		867	Event-Loop integration: L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
		868
		869	Debugging: L<Coro::Debug>.
		870
602	Support/Utility: L<Coro::Specific>, L<Coro::State>, L<Coro::Util>.	871	Support/Utility: L<Coro::Specific>, L<Coro::Util>.
603		872
604	Locking/IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>, L<Coro::SemaphoreSet>, L<Coro::RWLock>.	873	Locking and IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>,
		874	L<Coro::SemaphoreSet>, L<Coro::RWLock>.
605		875
606	Event/IO: L<Coro::Timer>, L<Coro::Event>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::Select>.	876	I/O and Timers: L<Coro::Timer>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::AIO>.
607		877
608	Embedding: L<Coro:MakeMaker>	878	Compatibility with other modules: L<Coro::LWP> (but see also L<AnyEvent::HTTP> for
		879	a better-working alternative), L<Coro::BDB>, L<Coro::Storable>,
		880	L<Coro::Select>.
		881
		882	XS API: L<Coro::MakeMaker>.
		883
		884	Low level Configuration, Thread Environment, Continuations: L<Coro::State>.
609		885
610	=head1 AUTHOR	886	=head1 AUTHOR
611		887
612	Marc Lehmann <schmorp@schmorp.de>	888	Marc Lehmann <schmorp@schmorp.de>
613	http://home.schmorp.de/	889	http://home.schmorp.de/

Diff Legend

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

Comparing Coro/Coro.pm (file contents): Revision 1.148 by root, Fri Oct 5 20:11:25 2007 UTC vs. Revision 1.257 by root, Tue Jun 23 23:40:06 2009 UTC

Diff Legend

Comparing Coro/Coro.pm (file contents):
Revision 1.148 by root, Fri Oct 5 20:11:25 2007 UTC vs.
Revision 1.257 by root, Tue Jun 23 23:40:06 2009 UTC