ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/Coro/Coro.pm
(Generate patch)

Comparing Coro/Coro.pm (file contents):
Revision 1.175 by root, Sun Apr 6 19:23:50 2008 UTC vs.
Revision 1.224 by root, Wed Nov 19 05:52:42 2008 UTC

… …
2 2
3Coro - coroutine process abstraction 3Coro - coroutine process abstraction
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"; 11 print "2\n";
12 cede; # yield back to main 12 cede; # yield back to main
13 print "4\n"; 13 print "4\n";
14 }; 14 };
15 print "1\n"; 15 print "1\n";
16 cede; # yield to coroutine 16 cede; # yield to coroutine
17 print "3\n"; 17 print "3\n";
18 cede; # and again 18 cede; # and again
19 19
20 # use locking 20 # use locking
21 use Coro::Semaphore;
21 my $lock = new Coro::Semaphore; 22 my $lock = new Coro::Semaphore;
22 my $locked; 23 my $locked;
23 24
24 $lock->down; 25 $lock->down;
25 $locked = 1; 26 $locked = 1;
26 $lock->up; 27 $lock->up;
27 28
28=head1 DESCRIPTION 29=head1 DESCRIPTION
29 30
30This module collection manages coroutines. Coroutines are similar 31This module collection manages coroutines. Coroutines are similar to
31to threads but don't run in parallel at the same time even on SMP 32threads but don't (in general) run in parallel at the same time even
32machines. The specific flavor of coroutine used in this module also 33on SMP machines. The specific flavor of coroutine used in this module
33guarantees you that it will not switch between coroutines unless 34also guarantees you that it will not switch between coroutines unless
34necessary, at easily-identified points in your program, so locking and 35necessary, at easily-identified points in your program, so locking and
35parallel access are rarely an issue, making coroutine programming much 36parallel access are rarely an issue, making coroutine programming much
36safer than threads programming. 37safer and easier than threads programming.
37 38
38(Perl, however, does not natively support real threads but instead does a 39Unlike a normal perl program, however, coroutines allow you to have
39very slow and memory-intensive emulation of processes using threads. This 40multiple running interpreters that share data, which is especially useful
40is a performance win on Windows machines, and a loss everywhere else). 41to code pseudo-parallel processes and for event-based programming, such as
42multiple HTTP-GET requests running concurrently. See L<Coro::AnyEvent> to
43learn more.
44
45Coroutines are also useful because Perl has no support for threads (the so
46called "threads" that perl offers are nothing more than the (bad) process
47emulation coming from the Windows platform: On standard operating systems
48they serve no purpose whatsoever, except by making your programs slow and
49making them use a lot of memory. Best disable them when building perl, or
50aks your software vendor/distributor to do it for you).
41 51
42In this module, coroutines are defined as "callchain + lexical variables + 52In this module, coroutines are defined as "callchain + lexical variables +
43@_ + $_ + $@ + $/ + C stack), that is, a coroutine has its own callchain, 53@_ + $_ + $@ + $/ + C stack), that is, a coroutine has its own callchain,
44its own set of lexicals and its own set of perls most important global 54its own set of lexicals and its own set of perls most important global
45variables (see L<Coro::State> for more configuration). 55variables (see L<Coro::State> for more configuration).
46 56
47=cut 57=cut
48 58
49package Coro; 59package Coro;
50 60
51use strict; 61use strict qw(vars subs);
52no warnings "uninitialized"; 62no warnings "uninitialized";
53 63
54use Coro::State; 64use Coro::State;
55 65
56use base qw(Coro::State Exporter); 66use base qw(Coro::State Exporter);
57 67
58our $idle; # idle handler 68our $idle; # idle handler
59our $main; # main coroutine 69our $main; # main coroutine
60our $current; # current coroutine 70our $current; # current coroutine
61 71
62our $VERSION = '4.49'; 72our $VERSION = 5.0;
63 73
64our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub); 74our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);
65our %EXPORT_TAGS = ( 75our %EXPORT_TAGS = (
66 prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)], 76 prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],
67); 77);
68our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready)); 78our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));
69 79
70{
71 my @async;
72 my $init;
73
74 # this way of handling attributes simply is NOT scalable ;()
75 sub import {
76 no strict 'refs';
77
78 Coro->export_to_level (1, @_);
79
80 my $old = *{(caller)[0]."::MODIFY_CODE_ATTRIBUTES"}{CODE};
81 *{(caller)[0]."::MODIFY_CODE_ATTRIBUTES"} = sub {
82 my ($package, $ref) = (shift, shift);
83 my @attrs;
84 for (@_) {
85 if ($_ eq "Coro") {
86 push @async, $ref;
87 unless ($init++) {
88 eval q{
89 sub INIT {
90 &async(pop @async) while @async;
91 }
92 };
93 }
94 } else {
95 push @attrs, $_;
96 }
97 }
98 return $old ? $old->($package, $ref, @attrs) : @attrs;
99 };
100 }
101
102}
103
104=over 4 80=over 4
105 81
106=item $main 82=item $Coro::main
107 83
108This coroutine represents the main program. 84This variable stores the coroutine object that represents the main
85program. While you cna C<ready> it and do most other things you can do to
86coroutines, it is mainly useful to compare again C<$Coro::current>, to see
87whether you are running in the main program or not.
109 88
110=cut 89=cut
111 90
112$main = new Coro; 91# $main is now being initialised by Coro::State
113 92
114=item $current (or as function: current) 93=item $Coro::current
115 94
116The current coroutine (the last coroutine switched to). The initial value 95The coroutine object representing the current coroutine (the last
96coroutine that the Coro scheduler switched to). The initial value is
117is C<$main> (of course). 97C<$Coro::main> (of course).
118 98
119This variable is B<strictly> I<read-only>. It is provided for performance 99This variable is B<strictly> I<read-only>. You can take copies of the
120reasons. If performance is not essential you are encouraged to use the 100value stored in it and use it as any other coroutine object, but you must
121C<Coro::current> function instead. 101not otherwise modify the variable itself.
122 102
123=cut 103=cut
124 104
125$main->{desc} = "[main::]";
126
127# maybe some other module used Coro::Specific before...
128$main->{_specific} = $current->{_specific}
129 if $current;
130
131_set_current $main;
132
133sub current() { $current } 105sub current() { $current } # [DEPRECATED]
134 106
135=item $idle 107=item $Coro::idle
136 108
137A callback that is called whenever the scheduler finds no ready coroutines 109This variable is mainly useful to integrate Coro into event loops. It is
138to run. The default implementation prints "FATAL: deadlock detected" and 110usually better to rely on L<Coro::AnyEvent> or LC<Coro::EV>, as this is
139exits, because the program has no other way to continue. 111pretty low-level functionality.
112
113This variable stores a callback that is called whenever the scheduler
114finds no ready coroutines to run. The default implementation prints
115"FATAL: deadlock detected" and exits, because the program has no other way
116to continue.
140 117
141This hook is overwritten by modules such as C<Coro::Timer> and 118This hook is overwritten by modules such as C<Coro::Timer> and
142C<Coro::Event> to wait on an external event that hopefully wake up a 119C<Coro::AnyEvent> to wait on an external event that hopefully wake up a
143coroutine so the scheduler can run it. 120coroutine so the scheduler can run it.
121
122Note that the callback I<must not>, under any circumstances, block
123the current coroutine. Normally, this is achieved by having an "idle
124coroutine" that calls the event loop and then blocks again, and then
125readying that coroutine in the idle handler.
126
127See L<Coro::Event> or L<Coro::AnyEvent> for examples of using this
128technique.
144 129
145Please note that if your callback recursively invokes perl (e.g. for event 130Please note that if your callback recursively invokes perl (e.g. for event
146handlers), then it must be prepared to be called recursively itself. 131handlers), then it must be prepared to be called recursively itself.
147 132
148=cut 133=cut
… …
159 $self->_destroy 144 $self->_destroy
160 or return; 145 or return;
161 146
162 # call all destruction callbacks 147 # call all destruction callbacks
163 $_->(@{$self->{_status}}) 148 $_->(@{$self->{_status}})
164 for @{(delete $self->{_on_destroy}) || []}; 149 for @{ delete $self->{_on_destroy} || [] };
165} 150}
166 151
167# this coroutine is necessary because a coroutine 152# this coroutine is necessary because a coroutine
168# cannot destroy itself. 153# cannot destroy itself.
169my @destroy; 154my @destroy;
… …
175 while @destroy; 160 while @destroy;
176 161
177 &schedule; 162 &schedule;
178 } 163 }
179}; 164};
180$manager->desc ("[coro manager]"); 165$manager->{desc} = "[coro manager]";
181$manager->prio (PRIO_MAX); 166$manager->prio (PRIO_MAX);
182 167
183# static methods. not really.
184
185=back 168=back
186 169
187=head2 STATIC METHODS 170=head2 SIMPLE COROUTINE CREATION
188
189Static methods are actually functions that operate on the current coroutine only.
190 171
191=over 4 172=over 4
192 173
193=item async { ... } [@args...] 174=item async { ... } [@args...]
194 175
195Create a new asynchronous coroutine and return it's coroutine object 176Create a new coroutine and return it's coroutine object (usually
196(usually unused). When the sub returns the new coroutine is automatically 177unused). The coroutine will be put into the ready queue, so
178it will start running automatically on the next scheduler run.
179
180The first argument is a codeblock/closure that should be executed in the
181coroutine. When it returns argument returns the coroutine is automatically
197terminated. 182terminated.
198 183
184The remaining arguments are passed as arguments to the closure.
185
199See the C<Coro::State::new> constructor for info about the coroutine 186See the C<Coro::State::new> constructor for info about the coroutine
200environment in which coroutines run. 187environment in which coroutines are executed.
201 188
202Calling C<exit> in a coroutine will do the same as calling exit outside 189Calling C<exit> in a coroutine will do the same as calling exit outside
203the coroutine. Likewise, when the coroutine dies, the program will exit, 190the coroutine. Likewise, when the coroutine dies, the program will exit,
204just as it would in the main program. 191just as it would in the main program.
205 192
193If you do not want that, you can provide a default C<die> handler, or
194simply avoid dieing (by use of C<eval>).
195
206 # create a new coroutine that just prints its arguments 196Example: Create a new coroutine that just prints its arguments.
197
207 async { 198 async {
208 print "@_\n"; 199 print "@_\n";
209 } 1,2,3,4; 200 } 1,2,3,4;
210 201
211=cut 202=cut
… …
217} 208}
218 209
219=item async_pool { ... } [@args...] 210=item async_pool { ... } [@args...]
220 211
221Similar to C<async>, but uses a coroutine pool, so you should not call 212Similar to C<async>, but uses a coroutine pool, so you should not call
222terminate or join (although you are allowed to), and you get a coroutine 213terminate or join on it (although you are allowed to), and you get a
223that might have executed other code already (which can be good or bad :). 214coroutine that might have executed other code already (which can be good
215or bad :).
224 216
217On the plus side, this function is faster than creating (and destroying)
218a completly new coroutine, so if you need a lot of generic coroutines in
219quick successsion, use C<async_pool>, not C<async>.
220
225Also, the block is executed in an C<eval> context and a warning will be 221The code block is executed in an C<eval> context and a warning will be
226issued in case of an exception instead of terminating the program, as 222issued in case of an exception instead of terminating the program, as
227C<async> does. As the coroutine is being reused, stuff like C<on_destroy> 223C<async> does. As the coroutine is being reused, stuff like C<on_destroy>
228will not work in the expected way, unless you call terminate or cancel, 224will not work in the expected way, unless you call terminate or cancel,
229which somehow defeats the purpose of pooling. 225which somehow defeats the purpose of pooling (but is fine in the
226exceptional case).
230 227
231The priority will be reset to C<0> after each job, tracing will be 228The priority will be reset to C<0> after each run, tracing will be
232disabled, the description will be reset and the default output filehandle 229disabled, the description will be reset and the default output filehandle
233gets restored, so you can change alkl these. Otherwise the coroutine will 230gets restored, so you can change all these. Otherwise the coroutine will
234be re-used "as-is": most notably if you change other per-coroutine global 231be re-used "as-is": most notably if you change other per-coroutine global
235stuff such as C<$/> you need to revert that change, which is most simply 232stuff such as C<$/> you I<must needs> revert that change, which is most
236done by using local as in C< local $/ >. 233simply done by using local as in: C<< local $/ >>.
237 234
238The pool size is limited to 8 idle coroutines (this can be adjusted by 235The idle pool size is limited to C<8> idle coroutines (this can be
239changing $Coro::POOL_SIZE), and there can be as many non-idle coros as 236adjusted by changing $Coro::POOL_SIZE), but there can be as many non-idle
240required. 237coros as required.
241 238
242If you are concerned about pooled coroutines growing a lot because a 239If you are concerned about pooled coroutines growing a lot because a
243single C<async_pool> used a lot of stackspace you can e.g. C<async_pool 240single C<async_pool> used a lot of stackspace you can e.g. C<async_pool
244{ terminate }> once per second or so to slowly replenish the pool. In 241{ terminate }> once per second or so to slowly replenish the pool. In
245addition to that, when the stacks used by a handler grows larger than 16kb 242addition to that, when the stacks used by a handler grows larger than 16kb
246(adjustable with $Coro::POOL_RSS) it will also exit. 243(adjustable via $Coro::POOL_RSS) it will also be destroyed.
247 244
248=cut 245=cut
249 246
250our $POOL_SIZE = 8; 247our $POOL_SIZE = 8;
251our $POOL_RSS = 16 * 1024; 248our $POOL_RSS = 16 * 1024;
… …
262 _pool_2 $cb; 259 _pool_2 $cb;
263 &schedule; 260 &schedule;
264 } 261 }
265 }; 262 };
266 263
264 if ($@) {
267 last if $@ eq "\3async_pool terminate\2\n"; 265 last if $@ eq "\3async_pool terminate\2\n";
268 warn $@ if $@; 266 warn $@;
267 }
269 } 268 }
270} 269}
271 270
272sub async_pool(&@) { 271sub async_pool(&@) {
273 # this is also inlined into the unlock_scheduler 272 # this is also inlined into the unblock_scheduler
274 my $coro = (pop @async_pool) || new Coro \&pool_handler; 273 my $coro = (pop @async_pool) || new Coro \&pool_handler;
275 274
276 $coro->{_invoke} = [@_]; 275 $coro->{_invoke} = [@_];
277 $coro->ready; 276 $coro->ready;
278 277
279 $coro 278 $coro
280} 279}
281 280
281=back
282
283=head2 STATIC METHODS
284
285Static methods are actually functions that operate on the current coroutine.
286
287=over 4
288
282=item schedule 289=item schedule
283 290
284Calls the scheduler. Please note that the current coroutine will not be put 291Calls the scheduler. The scheduler will find the next coroutine that is
292to be run from the ready queue and switches to it. The next coroutine
293to be run is simply the one with the highest priority that is longest
294in its ready queue. If there is no coroutine ready, it will clal the
295C<$Coro::idle> hook.
296
297Please note that the current coroutine will I<not> be put into the ready
285into the ready queue, so calling this function usually means you will 298queue, so calling this function usually means you will never be called
286never be called again unless something else (e.g. an event handler) calls 299again unless something else (e.g. an event handler) calls C<< ->ready >>,
287ready. 300thus waking you up.
288 301
289The canonical way to wait on external events is this: 302This makes C<schedule> I<the> generic method to use to block the current
303coroutine and wait for events: first you remember the current coroutine in
304a variable, then arrange for some callback of yours to call C<< ->ready
305>> on that once some event happens, and last you call C<schedule> to put
306yourself to sleep. Note that a lot of things can wake your coroutine up,
307so you need to check whether the event indeed happened, e.g. by storing the
308status in a variable.
290 309
291 { 310See B<HOW TO WAIT FOR A CALLBACK>, below, for some ways to wait for callbacks.
292 # remember current coroutine
293 my $current = $Coro::current;
294
295 # register a hypothetical event handler
296 on_event_invoke sub {
297 # wake up sleeping coroutine
298 $current->ready;
299 undef $current;
300 };
301
302 # call schedule until event occurred.
303 # in case we are woken up for other reasons
304 # (current still defined), loop.
305 Coro::schedule while $current;
306 }
307 311
308=item cede 312=item cede
309 313
310"Cede" to other coroutines. This function puts the current coroutine into the 314"Cede" to other coroutines. This function puts the current coroutine into
311ready queue and calls C<schedule>, which has the effect of giving up the 315the ready queue and calls C<schedule>, which has the effect of giving
312current "timeslice" to other coroutines of the same or higher priority. 316up the current "timeslice" to other coroutines of the same or higher
317priority. Once your coroutine gets its turn again it will automatically be
318resumed.
319
320This function is often called C<yield> in other languages.
313 321
314=item Coro::cede_notself 322=item Coro::cede_notself
315 323
316Works like cede, but is not exported by default and will cede to any 324Works like cede, but is not exported by default and will cede to I<any>
317coroutine, regardless of priority, once. 325coroutine, regardless of priority. This is useful sometimes to ensure
326progress is made.
318 327
319=item terminate [arg...] 328=item terminate [arg...]
320 329
321Terminates the current coroutine with the given status values (see L<cancel>). 330Terminates the current coroutine with the given status values (see L<cancel>).
322 331
323=item killall 332=item killall
324 333
325Kills/terminates/cancels all coroutines except the currently running 334Kills/terminates/cancels all coroutines except the currently running
326one. This is useful after a fork, either in the child or the parent, as 335one. This is useful after a fork, either in the child or the parent, as
327usually only one of them should inherit the running coroutines. 336usually only one of them should inherit the running coroutines.
337
338Note that while this will try to free some of the main programs resources,
339you cannot free all of them, so if a coroutine that is not the main
340program calls this function, there will be some one-time resource leak.
328 341
329=cut 342=cut
330 343
331sub terminate { 344sub terminate {
332 $current->cancel (@_); 345 $current->cancel (@_);
… …
339 } 352 }
340} 353}
341 354
342=back 355=back
343 356
344# dynamic methods
345
346=head2 COROUTINE METHODS 357=head2 COROUTINE METHODS
347 358
348These are the methods you can call on coroutine objects. 359These are the methods you can call on coroutine objects (or to create
360them).
349 361
350=over 4 362=over 4
351 363
352=item new Coro \&sub [, @args...] 364=item new Coro \&sub [, @args...]
353 365
354Create a new coroutine and return it. When the sub returns the coroutine 366Create a new coroutine and return it. When the sub returns, the coroutine
355automatically terminates as if C<terminate> with the returned values were 367automatically terminates as if C<terminate> with the returned values were
356called. To make the coroutine run you must first put it into the ready queue 368called. To make the coroutine run you must first put it into the ready
357by calling the ready method. 369queue by calling the ready method.
358 370
359See C<async> and C<Coro::State::new> for additional info about the 371See C<async> and C<Coro::State::new> for additional info about the
360coroutine environment. 372coroutine environment.
361 373
362=cut 374=cut
… …
371 $class->SUPER::new (\&_run_coro, @_) 383 $class->SUPER::new (\&_run_coro, @_)
372} 384}
373 385
374=item $success = $coroutine->ready 386=item $success = $coroutine->ready
375 387
376Put the given coroutine into the ready queue (according to it's priority) 388Put the given coroutine into the end of its ready queue (there is one
377and return true. If the coroutine is already in the ready queue, do nothing 389queue for each priority) and return true. If the coroutine is already in
378and return false. 390the ready queue, do nothing and return false.
391
392This ensures that the scheduler will resume this coroutine automatically
393once all the coroutines of higher priority and all coroutines of the same
394priority that were put into the ready queue earlier have been resumed.
379 395
380=item $is_ready = $coroutine->is_ready 396=item $is_ready = $coroutine->is_ready
381 397
382Return wether the coroutine is currently the ready queue or not, 398Return whether the coroutine is currently the ready queue or not,
383 399
384=item $coroutine->cancel (arg...) 400=item $coroutine->cancel (arg...)
385 401
386Terminates the given coroutine and makes it return the given arguments as 402Terminates the given coroutine and makes it return the given arguments as
387status (default: the empty list). Never returns if the coroutine is the 403status (default: the empty list). Never returns if the coroutine is the
… …
400 } else { 416 } else {
401 $self->_cancel; 417 $self->_cancel;
402 } 418 }
403} 419}
404 420
421=item $coroutine->throw ([$scalar])
422
423If C<$throw> is specified and defined, it will be thrown as an exception
424inside the coroutine at the next convenient point in time. Otherwise
425clears the exception object.
426
427Coro will check for the exception each time a schedule-like-function
428returns, i.e. after each C<schedule>, C<cede>, C<< Coro::Semaphore->down
429>>, C<< Coro::Handle->readable >> and so on. Most of these functions
430detect this case and return early in case an exception is pending.
431
432The exception object will be thrown "as is" with the specified scalar in
433C<$@>, i.e. if it is a string, no line number or newline will be appended
434(unlike with C<die>).
435
436This can be used as a softer means than C<cancel> to ask a coroutine to
437end itself, although there is no guarantee that the exception will lead to
438termination, and if the exception isn't caught it might well end the whole
439program.
440
441You might also think of C<throw> as being the moral equivalent of
442C<kill>ing a coroutine with a signal (in this case, a scalar).
443
405=item $coroutine->join 444=item $coroutine->join
406 445
407Wait until the coroutine terminates and return any values given to the 446Wait until the coroutine terminates and return any values given to the
408C<terminate> or C<cancel> functions. C<join> can be called concurrently 447C<terminate> or C<cancel> functions. C<join> can be called concurrently
409from multiple coroutines. 448from multiple coroutines, and all will be resumed and given the status
449return once the C<$coroutine> terminates.
410 450
411=cut 451=cut
412 452
413sub join { 453sub join {
414 my $self = shift; 454 my $self = shift;
… …
429 469
430=item $coroutine->on_destroy (\&cb) 470=item $coroutine->on_destroy (\&cb)
431 471
432Registers a callback that is called when this coroutine gets destroyed, 472Registers a callback that is called when this coroutine gets destroyed,
433but before it is joined. The callback gets passed the terminate arguments, 473but before it is joined. The callback gets passed the terminate arguments,
434if any. 474if any, and I<must not> die, under any circumstances.
435 475
436=cut 476=cut
437 477
438sub on_destroy { 478sub on_destroy {
439 my ($self, $cb) = @_; 479 my ($self, $cb) = @_;
… …
469higher values mean lower priority, just as in unix). 509higher values mean lower priority, just as in unix).
470 510
471=item $olddesc = $coroutine->desc ($newdesc) 511=item $olddesc = $coroutine->desc ($newdesc)
472 512
473Sets (or gets in case the argument is missing) the description for this 513Sets (or gets in case the argument is missing) the description for this
474coroutine. This is just a free-form string you can associate with a coroutine. 514coroutine. This is just a free-form string you can associate with a
515coroutine.
475 516
476This method simply sets the C<< $coroutine->{desc} >> member to the given string. You 517This method simply sets the C<< $coroutine->{desc} >> member to the given
477can modify this member directly if you wish. 518string. You can modify this member directly if you wish.
478
479=item $coroutine->throw ([$scalar])
480
481If C<$throw> is specified and defined, it will be thrown as an exception
482inside the coroutine at the next convinient point in time (usually after
483it gains control at the next schedule/transfer/cede). Otherwise clears the
484exception object.
485
486The exception object will be thrown "as is" with the specified scalar in
487C<$@>, i.e. if it is a string, no line number or newline will be appended
488(unlike with C<die>).
489
490This can be used as a softer means than C<cancel> to ask a coroutine to
491end itself, although there is no guarentee that the exception will lead to
492termination, and if the exception isn't caught it might well end the whole
493program.
494 519
495=cut 520=cut
496 521
497sub desc { 522sub desc {
498 my $old = $_[0]{desc}; 523 my $old = $_[0]{desc};
… …
507=over 4 532=over 4
508 533
509=item Coro::nready 534=item Coro::nready
510 535
511Returns the number of coroutines that are currently in the ready state, 536Returns the number of coroutines that are currently in the ready state,
512i.e. that can be switched to. The value C<0> means that the only runnable 537i.e. that can be switched to by calling C<schedule> directory or
538indirectly. The value C<0> means that the only runnable coroutine is the
513coroutine is the currently running one, so C<cede> would have no effect, 539currently running one, so C<cede> would have no effect, and C<schedule>
514and C<schedule> would cause a deadlock unless there is an idle handler 540would cause a deadlock unless there is an idle handler that wakes up some
515that wakes up some coroutines. 541coroutines.
516 542
517=item my $guard = Coro::guard { ... } 543=item my $guard = Coro::guard { ... }
518 544
519This creates and returns a guard object. Nothing happens until the object 545This creates and returns a guard object. Nothing happens until the object
520gets destroyed, in which case the codeblock given as argument will be 546gets destroyed, in which case the codeblock given as argument will be
… …
549 575
550 576
551=item unblock_sub { ... } 577=item unblock_sub { ... }
552 578
553This utility function takes a BLOCK or code reference and "unblocks" it, 579This utility function takes a BLOCK or code reference and "unblocks" it,
554returning the new coderef. This means that the new coderef will return 580returning a new coderef. Unblocking means that calling the new coderef
555immediately without blocking, returning nothing, while the original code 581will return immediately without blocking, returning nothing, while the
556ref will be called (with parameters) from within its own coroutine. 582original code ref will be called (with parameters) from within another
583coroutine.
557 584
558The reason this function exists is that many event libraries (such as the 585The reason this function exists is that many event libraries (such as the
559venerable L<Event|Event> module) are not coroutine-safe (a weaker form 586venerable L<Event|Event> module) are not coroutine-safe (a weaker form
560of thread-safety). This means you must not block within event callbacks, 587of thread-safety). This means you must not block within event callbacks,
561otherwise you might suffer from crashes or worse. 588otherwise you might suffer from crashes or worse. The only event library
589currently known that is safe to use without C<unblock_sub> is L<EV>.
562 590
563This function allows your callbacks to block by executing them in another 591This function allows your callbacks to block by executing them in another
564coroutine where it is safe to block. One example where blocking is handy 592coroutine where it is safe to block. One example where blocking is handy
565is when you use the L<Coro::AIO|Coro::AIO> functions to save results to 593is when you use the L<Coro::AIO|Coro::AIO> functions to save results to
566disk. 594disk, for example.
567 595
568In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when 596In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when
569creating event callbacks that want to block. 597creating event callbacks that want to block.
598
599If your handler does not plan to block (e.g. simply sends a message to
600another coroutine, or puts some other coroutine into the ready queue),
601there is no reason to use C<unblock_sub>.
602
603Note that you also need to use C<unblock_sub> for any other callbacks that
604are indirectly executed by any C-based event loop. For example, when you
605use a module that uses L<AnyEvent> (and you use L<Coro::AnyEvent>) and it
606provides callbacks that are the result of some event callback, then you
607must not block either, or use C<unblock_sub>.
570 608
571=cut 609=cut
572 610
573our @unblock_queue; 611our @unblock_queue;
574 612
… …
587 cede; # for short-lived callbacks, this reduces pressure on the coro pool 625 cede; # for short-lived callbacks, this reduces pressure on the coro pool
588 } 626 }
589 schedule; # sleep well 627 schedule; # sleep well
590 } 628 }
591}; 629};
592$unblock_scheduler->desc ("[unblock_sub scheduler]"); 630$unblock_scheduler->{desc} = "[unblock_sub scheduler]";
593 631
594sub unblock_sub(&) { 632sub unblock_sub(&) {
595 my $cb = shift; 633 my $cb = shift;
596 634
597 sub { 635 sub {
598 unshift @unblock_queue, [$cb, @_]; 636 unshift @unblock_queue, [$cb, @_];
599 $unblock_scheduler->ready; 637 $unblock_scheduler->ready;
600 } 638 }
601} 639}
602 640
641=item $cb = Coro::rouse_cb
642
643Create and return a "rouse callback". That's a code reference that, when
644called, will save its arguments and notify the owner coroutine of the
645callback.
646
647See the next function.
648
649=item @args = Coro::rouse_wait [$cb]
650
651Wait for the specified rouse callback (or the last one tht was created in
652this coroutine).
653
654As soon as the callback is invoked (or when the calback was invoked before
655C<rouse_wait>), it will return a copy of the arguments originally passed
656to the rouse callback.
657
658See the section B<HOW TO WAIT FOR A CALLBACK> for an actual usage example.
659
603=back 660=back
604 661
605=cut 662=cut
606 663
6071; 6641;
608 665
666=head1 HOW TO WAIT FOR A CALLBACK
667
668It is very common for a coroutine to wait for some callback to be
669called. This occurs naturally when you use coroutines in an otherwise
670event-based program, or when you use event-based libraries.
671
672These typically register a callback for some event, and call that callback
673when the event occured. In a coroutine, however, you typically want to
674just wait for the event, simplyifying things.
675
676For example C<< AnyEvent->child >> registers a callback to be called when
677a specific child has exited:
678
679 my $child_watcher = AnyEvent->child (pid => $pid, cb => sub { ... });
680
681But from withina coroutine, you often just want to write this:
682
683 my $status = wait_for_child $pid;
684
685Coro offers two functions specifically designed to make this easy,
686C<Coro::rouse_cb> and C<Coro::rouse_wait>.
687
688The first function, C<rouse_cb>, generates and returns a callback that,
689when invoked, will save it's arguments and notify the coroutine that
690created the callback.
691
692The second function, C<rouse_wait>, waits for the callback to be called
693(by calling C<schedule> to go to sleep) and returns the arguments
694originally passed to the callback.
695
696Using these functions, it becomes easy to write the C<wait_for_child>
697function mentioned above:
698
699 sub wait_for_child($) {
700 my ($pid) = @_;
701
702 my $watcher = AnyEvent->child (pid => $pid, cb => Coro::rouse_cb);
703
704 my ($rpid, $rstatus) = Coro::rouse_wait;
705 $rstatus
706 }
707
708In the case where C<rouse_cb> and C<rouse_wait> are not flexible enough,
709you can roll your own, using C<schedule>:
710
711 sub wait_for_child($) {
712 my ($pid) = @_;
713
714 # store the current coroutine in $current,
715 # and provide result variables for the closure passed to ->child
716 my $current = $Coro::current;
717 my ($done, $rstatus);
718
719 # pass a closure to ->child
720 my $watcher = AnyEvent->child (pid => $pid, cb => sub {
721 $rstatus = $_[1]; # remember rstatus
722 $done = 1; # mark $rstatus as valud
723 });
724
725 # wait until the closure has been called
726 schedule while !$done;
727
728 $rstatus
729 }
730
731
609=head1 BUGS/LIMITATIONS 732=head1 BUGS/LIMITATIONS
610 733
611 - you must make very sure that no coro is still active on global 734=over 4
612 destruction. very bad things might happen otherwise (usually segfaults).
613 735
736=item fork with pthread backend
737
738When Coro is compiled using the pthread backend (which isn't recommended
739but required on many BSDs as their libcs are completely broken), then
740coroutines will not survive a fork. There is no known workaround except to
741fix your libc and use a saner backend.
742
743=item perl process emulation ("threads")
744
614 - this module is not thread-safe. You should only ever use this module 745This module is not perl-pseudo-thread-safe. You should only ever use this
615 from the same thread (this requirement might be loosened in the future 746module from the same thread (this requirement might be removed in the
616 to allow per-thread schedulers, but Coro::State does not yet allow 747future to allow per-thread schedulers, but Coro::State does not yet allow
617 this). 748this). I recommend disabling thread support and using processes, as having
749the windows process emulation enabled under unix roughly halves perl
750performance, even when not used.
751
752=item coroutine switching not signal safe
753
754You must not switch to another coroutine from within a signal handler
755(only relevant with %SIG - most event libraries provide safe signals).
756
757That means you I<MUST NOT> call any function that might "block" the
758current coroutine - C<cede>, C<schedule> C<< Coro::Semaphore->down >> or
759anything that calls those. Everything else, including calling C<ready>,
760works.
761
762=back
763
618 764
619=head1 SEE ALSO 765=head1 SEE ALSO
620 766
621Lower level Configuration, Coroutine Environment: L<Coro::State>. 767Event-Loop integration: L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
622 768
623Debugging: L<Coro::Debug>. 769Debugging: L<Coro::Debug>.
624 770
625Support/Utility: L<Coro::Specific>, L<Coro::Util>. 771Support/Utility: L<Coro::Specific>, L<Coro::Util>.
626 772
627Locking/IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>, L<Coro::SemaphoreSet>, L<Coro::RWLock>. 773Locking/IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>, L<Coro::SemaphoreSet>, L<Coro::RWLock>.
628 774
629Event/IO: L<Coro::Timer>, L<Coro::Event>, L<Coro::Handle>, L<Coro::Socket>. 775IO/Timers: L<Coro::Timer>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::AIO>.
630 776
631Compatibility: L<Coro::LWP>, L<Coro::Storable>, L<Coro::Select>. 777Compatibility: L<Coro::LWP>, L<Coro::BDB>, L<Coro::Storable>, L<Coro::Select>.
632 778
633Embedding: L<Coro::MakeMaker>. 779XS API: L<Coro::MakeMaker>.
780
781Low level Configuration, Coroutine Environment: L<Coro::State>.
634 782
635=head1 AUTHOR 783=head1 AUTHOR
636 784
637 Marc Lehmann <schmorp@schmorp.de> 785 Marc Lehmann <schmorp@schmorp.de>
638 http://home.schmorp.de/ 786 http://home.schmorp.de/

Diff Legend

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