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

Comparing Coro/Coro.pm (file contents):
Revision 1.224 by root, Wed Nov 19 05:52:42 2008 UTC vs.
Revision 1.268 by root, Thu Oct 1 23:16:27 2009 UTC

1=head1 NAME 1=head1 NAME
2 2
3Coro - coroutine process abstraction 3Coro - the only real threads in perl
4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use Coro; 7 use Coro;
8 8
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 coro
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 use Coro::Semaphore;
26 $locked = 1; 26 $locked = 1;
27 $lock->up; 27 $lock->up;
28 28
29=head1 DESCRIPTION 29=head1 DESCRIPTION
30 30
31This module collection manages coroutines. Coroutines are similar to 31For a tutorial-style introduction, please read the L<Coro::Intro>
32threads but don't (in general) run in parallel at the same time even 32manpage. This manpage mainly contains reference information.
33on SMP machines. The specific flavor of coroutine used in this module
34also guarantees you that it will not switch between coroutines unless
35necessary, at easily-identified points in your program, so locking and
36parallel access are rarely an issue, making coroutine programming much
37safer and easier than threads programming.
38 33
39Unlike a normal perl program, however, coroutines allow you to have 34This module collection manages continuations in general, most often in
40multiple running interpreters that share data, which is especially useful 35the form of cooperative threads (also called coros, or simply "coro"
41to code pseudo-parallel processes and for event-based programming, such as 36in the documentation). They are similar to kernel threads but don't (in
42multiple HTTP-GET requests running concurrently. See L<Coro::AnyEvent> to 37general) run in parallel at the same time even on SMP machines. The
43learn more. 38specific flavor of thread offered by this module also guarantees you that
39it will not switch between threads unless necessary, at easily-identified
40points in your program, so locking and parallel access are rarely an
41issue, making thread programming much safer and easier than using other
42thread models.
44 43
45Coroutines are also useful because Perl has no support for threads (the so 44Unlike the so-called "Perl threads" (which are not actually real threads
46called "threads" that perl offers are nothing more than the (bad) process 45but only the windows process emulation (see section of same name for more
47emulation coming from the Windows platform: On standard operating systems 46details) ported to unix, and as such act as processes), Coro provides
48they serve no purpose whatsoever, except by making your programs slow and 47a full shared address space, which makes communication between threads
49making them use a lot of memory. Best disable them when building perl, or 48very easy. And Coro's threads are fast, too: disabling the Windows
50aks your software vendor/distributor to do it for you). 49process emulation code in your perl and using Coro can easily result in
50a two to four times speed increase for your programs. A parallel matrix
51multiplication benchmark runs over 300 times faster on a single core than
52perl's pseudo-threads on a quad core using all four cores.
51 53
54Coro achieves that by supporting multiple running interpreters that share
55data, which is especially useful to code pseudo-parallel processes and
56for event-based programming, such as multiple HTTP-GET requests running
57concurrently. See L<Coro::AnyEvent> to learn more on how to integrate Coro
58into an event-based environment.
59
52In this module, coroutines are defined as "callchain + lexical variables + 60In this module, a thread is defined as "callchain + lexical variables +
53@_ + $_ + $@ + $/ + C stack), that is, a coroutine has its own callchain, 61some package variables + C stack), that is, a thread has its own callchain,
54its own set of lexicals and its own set of perls most important global 62its own set of lexicals and its own set of perls most important global
55variables (see L<Coro::State> for more configuration). 63variables (see L<Coro::State> for more configuration and background info).
64
65See also the C<SEE ALSO> section at the end of this document - the Coro
66module family is quite large.
56 67
57=cut 68=cut
58 69
59package Coro; 70package Coro;
60 71
61use strict qw(vars subs); 72use common::sense;
62no warnings "uninitialized"; 73
74use Carp ();
75
76use Guard ();
63 77
64use Coro::State; 78use Coro::State;
65 79
66use base qw(Coro::State Exporter); 80use base qw(Coro::State Exporter);
67 81
68our $idle; # idle handler 82our $idle; # idle handler
69our $main; # main coroutine 83our $main; # main coro
70our $current; # current coroutine 84our $current; # current coro
71 85
72our $VERSION = 5.0; 86our $VERSION = 5.17;
73 87
74our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub); 88our @EXPORT = qw(async async_pool cede schedule terminate current unblock_sub);
75our %EXPORT_TAGS = ( 89our %EXPORT_TAGS = (
76 prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)], 90 prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],
77); 91);
78our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready)); 92our @EXPORT_OK = (@{$EXPORT_TAGS{prio}}, qw(nready));
79 93
94=head1 GLOBAL VARIABLES
95
80=over 4 96=over 4
81 97
82=item $Coro::main 98=item $Coro::main
83 99
84This variable stores the coroutine object that represents the main 100This variable stores the Coro object that represents the main
85program. While you cna C<ready> it and do most other things you can do to 101program. While you cna C<ready> it and do most other things you can do to
86coroutines, it is mainly useful to compare again C<$Coro::current>, to see 102coro, it is mainly useful to compare again C<$Coro::current>, to see
87whether you are running in the main program or not. 103whether you are running in the main program or not.
88 104
89=cut 105=cut
90 106
91# $main is now being initialised by Coro::State 107# $main is now being initialised by Coro::State
92 108
93=item $Coro::current 109=item $Coro::current
94 110
95The coroutine object representing the current coroutine (the last 111The Coro object representing the current coro (the last
96coroutine that the Coro scheduler switched to). The initial value is 112coro that the Coro scheduler switched to). The initial value is
97C<$Coro::main> (of course). 113C<$Coro::main> (of course).
98 114
99This variable is B<strictly> I<read-only>. You can take copies of the 115This variable is B<strictly> I<read-only>. You can take copies of the
100value stored in it and use it as any other coroutine object, but you must 116value stored in it and use it as any other Coro object, but you must
101not otherwise modify the variable itself. 117not otherwise modify the variable itself.
102 118
103=cut 119=cut
104 120
105sub current() { $current } # [DEPRECATED] 121sub current() { $current } # [DEPRECATED]
106 122
107=item $Coro::idle 123=item $Coro::idle
108 124
109This variable is mainly useful to integrate Coro into event loops. It is 125This variable is mainly useful to integrate Coro into event loops. It is
110usually better to rely on L<Coro::AnyEvent> or LC<Coro::EV>, as this is 126usually better to rely on L<Coro::AnyEvent> or L<Coro::EV>, as this is
111pretty low-level functionality. 127pretty low-level functionality.
112 128
113This variable stores a callback that is called whenever the scheduler 129This variable stores either a Coro object or a callback.
130
131If it is a callback, the it is called whenever the scheduler finds no
114finds no ready coroutines to run. The default implementation prints 132ready coros to run. The default implementation prints "FATAL:
115"FATAL: deadlock detected" and exits, because the program has no other way 133deadlock detected" and exits, because the program has no other way to
116to continue. 134continue.
117 135
136If it is a coro object, then this object will be readied (without
137invoking any ready hooks, however) when the scheduler finds no other ready
138coros to run.
139
118This hook is overwritten by modules such as C<Coro::Timer> and 140This hook is overwritten by modules such as C<Coro::EV> and
119C<Coro::AnyEvent> to wait on an external event that hopefully wake up a 141C<Coro::AnyEvent> to wait on an external event that hopefully wake up a
120coroutine so the scheduler can run it. 142coro so the scheduler can run it.
121 143
122Note that the callback I<must not>, under any circumstances, block 144Note that the callback I<must not>, under any circumstances, block
123the current coroutine. Normally, this is achieved by having an "idle 145the current coro. Normally, this is achieved by having an "idle
124coroutine" that calls the event loop and then blocks again, and then 146coro" that calls the event loop and then blocks again, and then
125readying that coroutine in the idle handler. 147readying that coro in the idle handler, or by simply placing the idle
148coro in this variable.
126 149
127See L<Coro::Event> or L<Coro::AnyEvent> for examples of using this 150See L<Coro::Event> or L<Coro::AnyEvent> for examples of using this
128technique. 151technique.
129 152
130Please note that if your callback recursively invokes perl (e.g. for event 153Please note that if your callback recursively invokes perl (e.g. for event
131handlers), then it must be prepared to be called recursively itself. 154handlers), then it must be prepared to be called recursively itself.
132 155
133=cut 156=cut
134 157
135$idle = sub { 158$idle = sub {
136 require Carp; 159 warn "oi\n";#d#
137 Carp::croak ("FATAL: deadlock detected"); 160 Carp::confess ("FATAL: deadlock detected");
138}; 161};
139 162
140sub _cancel {
141 my ($self) = @_;
142
143 # free coroutine data and mark as destructed
144 $self->_destroy
145 or return;
146
147 # call all destruction callbacks
148 $_->(@{$self->{_status}})
149 for @{ delete $self->{_on_destroy} || [] };
150}
151
152# this coroutine is necessary because a coroutine 163# this coro is necessary because a coro
153# cannot destroy itself. 164# cannot destroy itself.
154my @destroy; 165our @destroy;
155my $manager; 166our $manager;
156 167
157$manager = new Coro sub { 168$manager = new Coro sub {
158 while () { 169 while () {
159 (shift @destroy)->_cancel 170 Coro::State::cancel shift @destroy
160 while @destroy; 171 while @destroy;
161 172
162 &schedule; 173 &schedule;
163 } 174 }
164}; 175};
165$manager->{desc} = "[coro manager]"; 176$manager->{desc} = "[coro manager]";
166$manager->prio (PRIO_MAX); 177$manager->prio (PRIO_MAX);
167 178
168=back 179=back
169 180
170=head2 SIMPLE COROUTINE CREATION 181=head1 SIMPLE CORO CREATION
171 182
172=over 4 183=over 4
173 184
174=item async { ... } [@args...] 185=item async { ... } [@args...]
175 186
176Create a new coroutine and return it's coroutine object (usually 187Create a new coro and return its Coro object (usually
177unused). The coroutine will be put into the ready queue, so 188unused). The coro will be put into the ready queue, so
178it will start running automatically on the next scheduler run. 189it will start running automatically on the next scheduler run.
179 190
180The first argument is a codeblock/closure that should be executed in the 191The first argument is a codeblock/closure that should be executed in the
181coroutine. When it returns argument returns the coroutine is automatically 192coro. When it returns argument returns the coro is automatically
182terminated. 193terminated.
183 194
184The remaining arguments are passed as arguments to the closure. 195The remaining arguments are passed as arguments to the closure.
185 196
186See the C<Coro::State::new> constructor for info about the coroutine 197See the C<Coro::State::new> constructor for info about the coro
187environment in which coroutines are executed. 198environment in which coro are executed.
188 199
189Calling C<exit> in a coroutine will do the same as calling exit outside 200Calling C<exit> in a coro will do the same as calling exit outside
190the coroutine. Likewise, when the coroutine dies, the program will exit, 201the coro. Likewise, when the coro dies, the program will exit,
191just as it would in the main program. 202just as it would in the main program.
192 203
193If you do not want that, you can provide a default C<die> handler, or 204If you do not want that, you can provide a default C<die> handler, or
194simply avoid dieing (by use of C<eval>). 205simply avoid dieing (by use of C<eval>).
195 206
196Example: Create a new coroutine that just prints its arguments. 207Example: Create a new coro that just prints its arguments.
197 208
198 async { 209 async {
199 print "@_\n"; 210 print "@_\n";
200 } 1,2,3,4; 211 } 1,2,3,4;
201 212
202=cut
203
204sub async(&@) {
205 my $coro = new Coro @_;
206 $coro->ready;
207 $coro
208}
209
210=item async_pool { ... } [@args...] 213=item async_pool { ... } [@args...]
211 214
212Similar to C<async>, but uses a coroutine pool, so you should not call 215Similar to C<async>, but uses a coro pool, so you should not call
213terminate or join on it (although you are allowed to), and you get a 216terminate or join on it (although you are allowed to), and you get a
214coroutine that might have executed other code already (which can be good 217coro that might have executed other code already (which can be good
215or bad :). 218or bad :).
216 219
217On the plus side, this function is faster than creating (and destroying) 220On the plus side, this function is about twice as fast as creating (and
218a completly new coroutine, so if you need a lot of generic coroutines in 221destroying) a completely new coro, so if you need a lot of generic
219quick successsion, use C<async_pool>, not C<async>. 222coros in quick successsion, use C<async_pool>, not C<async>.
220 223
221The code block is executed in an C<eval> context and a warning will be 224The code block is executed in an C<eval> context and a warning will be
222issued in case of an exception instead of terminating the program, as 225issued in case of an exception instead of terminating the program, as
223C<async> does. As the coroutine is being reused, stuff like C<on_destroy> 226C<async> does. As the coro is being reused, stuff like C<on_destroy>
224will not work in the expected way, unless you call terminate or cancel, 227will not work in the expected way, unless you call terminate or cancel,
225which somehow defeats the purpose of pooling (but is fine in the 228which somehow defeats the purpose of pooling (but is fine in the
226exceptional case). 229exceptional case).
227 230
228The priority will be reset to C<0> after each run, tracing will be 231The priority will be reset to C<0> after each run, tracing will be
229disabled, the description will be reset and the default output filehandle 232disabled, the description will be reset and the default output filehandle
230gets restored, so you can change all these. Otherwise the coroutine will 233gets restored, so you can change all these. Otherwise the coro will
231be re-used "as-is": most notably if you change other per-coroutine global 234be re-used "as-is": most notably if you change other per-coro global
232stuff such as C<$/> you I<must needs> revert that change, which is most 235stuff such as C<$/> you I<must needs> revert that change, which is most
233simply done by using local as in: C<< local $/ >>. 236simply done by using local as in: C<< local $/ >>.
234 237
235The idle pool size is limited to C<8> idle coroutines (this can be 238The idle pool size is limited to C<8> idle coros (this can be
236adjusted by changing $Coro::POOL_SIZE), but there can be as many non-idle 239adjusted by changing $Coro::POOL_SIZE), but there can be as many non-idle
237coros as required. 240coros as required.
238 241
239If you are concerned about pooled coroutines growing a lot because a 242If you are concerned about pooled coros growing a lot because a
240single C<async_pool> used a lot of stackspace you can e.g. C<async_pool 243single C<async_pool> used a lot of stackspace you can e.g. C<async_pool
241{ terminate }> once per second or so to slowly replenish the pool. In 244{ terminate }> once per second or so to slowly replenish the pool. In
242addition to that, when the stacks used by a handler grows larger than 16kb 245addition to that, when the stacks used by a handler grows larger than 32kb
243(adjustable via $Coro::POOL_RSS) it will also be destroyed. 246(adjustable via $Coro::POOL_RSS) it will also be destroyed.
244 247
245=cut 248=cut
246 249
247our $POOL_SIZE = 8; 250our $POOL_SIZE = 8;
248our $POOL_RSS = 16 * 1024; 251our $POOL_RSS = 32 * 1024;
249our @async_pool; 252our @async_pool;
250 253
251sub pool_handler { 254sub pool_handler {
252 my $cb;
253
254 while () { 255 while () {
255 eval { 256 eval {
256 while () { 257 &{&_pool_handler} while 1;
257 _pool_1 $cb;
258 &$cb;
259 _pool_2 $cb;
260 &schedule;
261 }
262 }; 258 };
263 259
264 if ($@) {
265 last if $@ eq "\3async_pool terminate\2\n";
266 warn $@; 260 warn $@ if $@;
267 }
268 } 261 }
269} 262}
270 263
271sub async_pool(&@) {
272 # this is also inlined into the unblock_scheduler
273 my $coro = (pop @async_pool) || new Coro \&pool_handler;
274
275 $coro->{_invoke} = [@_];
276 $coro->ready;
277
278 $coro
279}
280
281=back 264=back
282 265
283=head2 STATIC METHODS 266=head1 STATIC METHODS
284 267
285Static methods are actually functions that operate on the current coroutine. 268Static methods are actually functions that implicitly operate on the
269current coro.
286 270
287=over 4 271=over 4
288 272
289=item schedule 273=item schedule
290 274
291Calls the scheduler. The scheduler will find the next coroutine that is 275Calls the scheduler. The scheduler will find the next coro that is
292to be run from the ready queue and switches to it. The next coroutine 276to be run from the ready queue and switches to it. The next coro
293to be run is simply the one with the highest priority that is longest 277to be run is simply the one with the highest priority that is longest
294in its ready queue. If there is no coroutine ready, it will clal the 278in its ready queue. If there is no coro ready, it will clal the
295C<$Coro::idle> hook. 279C<$Coro::idle> hook.
296 280
297Please note that the current coroutine will I<not> be put into the ready 281Please note that the current coro will I<not> be put into the ready
298queue, so calling this function usually means you will never be called 282queue, so calling this function usually means you will never be called
299again unless something else (e.g. an event handler) calls C<< ->ready >>, 283again unless something else (e.g. an event handler) calls C<< ->ready >>,
300thus waking you up. 284thus waking you up.
301 285
302This makes C<schedule> I<the> generic method to use to block the current 286This 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 287coro and wait for events: first you remember the current coro in
304a variable, then arrange for some callback of yours to call C<< ->ready 288a 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 289>> 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, 290yourself to sleep. Note that a lot of things can wake your coro up,
307so you need to check whether the event indeed happened, e.g. by storing the 291so you need to check whether the event indeed happened, e.g. by storing the
308status in a variable. 292status in a variable.
309 293
310See B<HOW TO WAIT FOR A CALLBACK>, below, for some ways to wait for callbacks. 294See B<HOW TO WAIT FOR A CALLBACK>, below, for some ways to wait for callbacks.
311 295
312=item cede 296=item cede
313 297
314"Cede" to other coroutines. This function puts the current coroutine into 298"Cede" to other coros. This function puts the current coro into
315the ready queue and calls C<schedule>, which has the effect of giving 299the ready queue and calls C<schedule>, which has the effect of giving
316up the current "timeslice" to other coroutines of the same or higher 300up the current "timeslice" to other coros of the same or higher
317priority. Once your coroutine gets its turn again it will automatically be 301priority. Once your coro gets its turn again it will automatically be
318resumed. 302resumed.
319 303
320This function is often called C<yield> in other languages. 304This function is often called C<yield> in other languages.
321 305
322=item Coro::cede_notself 306=item Coro::cede_notself
323 307
324Works like cede, but is not exported by default and will cede to I<any> 308Works like cede, but is not exported by default and will cede to I<any>
325coroutine, regardless of priority. This is useful sometimes to ensure 309coro, regardless of priority. This is useful sometimes to ensure
326progress is made. 310progress is made.
327 311
328=item terminate [arg...] 312=item terminate [arg...]
329 313
330Terminates the current coroutine with the given status values (see L<cancel>). 314Terminates the current coro with the given status values (see L<cancel>).
315
316=item Coro::on_enter BLOCK, Coro::on_leave BLOCK
317
318These function install enter and leave winders in the current scope. The
319enter block will be executed when on_enter is called and whenever the
320current coro is re-entered by the scheduler, while the leave block is
321executed whenever the current coro is blocked by the scheduler, and
322also when the containing scope is exited (by whatever means, be it exit,
323die, last etc.).
324
325I<Neither invoking the scheduler, nor exceptions, are allowed within those
326BLOCKs>. That means: do not even think about calling C<die> without an
327eval, and do not even think of entering the scheduler in any way.
328
329Since both BLOCKs are tied to the current scope, they will automatically
330be removed when the current scope exits.
331
332These functions implement the same concept as C<dynamic-wind> in scheme
333does, and are useful when you want to localise some resource to a specific
334coro.
335
336They slow down thread switching considerably for coros that use them
337(about 40% for a BLOCK with a single assignment, so thread switching is
338still reasonably fast if the handlers are fast).
339
340These functions are best understood by an example: The following function
341will change the current timezone to "Antarctica/South_Pole", which
342requires a call to C<tzset>, but by using C<on_enter> and C<on_leave>,
343which remember/change the current timezone and restore the previous
344value, respectively, the timezone is only changed for the coro that
345installed those handlers.
346
347 use POSIX qw(tzset);
348
349 async {
350 my $old_tz; # store outside TZ value here
351
352 Coro::on_enter {
353 $old_tz = $ENV{TZ}; # remember the old value
354
355 $ENV{TZ} = "Antarctica/South_Pole";
356 tzset; # enable new value
357 };
358
359 Coro::on_leave {
360 $ENV{TZ} = $old_tz;
361 tzset; # restore old value
362 };
363
364 # at this place, the timezone is Antarctica/South_Pole,
365 # without disturbing the TZ of any other coro.
366 };
367
368This can be used to localise about any resource (locale, uid, current
369working directory etc.) to a block, despite the existance of other
370coros.
371
372Another interesting example implements time-sliced multitasking using
373interval timers (this could obviously be optimised, but does the job):
374
375 # "timeslice" the given block
376 sub timeslice(&) {
377 use Time::HiRes ();
378
379 Coro::on_enter {
380 # on entering the thread, we set an VTALRM handler to cede
381 $SIG{VTALRM} = sub { cede };
382 # and then start the interval timer
383 Time::HiRes::setitimer &Time::HiRes::ITIMER_VIRTUAL, 0.01, 0.01;
384 };
385 Coro::on_leave {
386 # on leaving the thread, we stop the interval timer again
387 Time::HiRes::setitimer &Time::HiRes::ITIMER_VIRTUAL, 0, 0;
388 };
389
390 &{+shift};
391 }
392
393 # use like this:
394 timeslice {
395 # The following is an endless loop that would normally
396 # monopolise the process. Since it runs in a timesliced
397 # environment, it will regularly cede to other threads.
398 while () { }
399 };
400
331 401
332=item killall 402=item killall
333 403
334Kills/terminates/cancels all coroutines except the currently running 404Kills/terminates/cancels all coros except the currently running one.
335one. This is useful after a fork, either in the child or the parent, as
336usually only one of them should inherit the running coroutines.
337 405
338Note that while this will try to free some of the main programs resources, 406Note that while this will try to free some of the main interpreter
407resources if the calling coro isn't the main coro, but one
339you cannot free all of them, so if a coroutine that is not the main 408cannot free all of them, so if a coro that is not the main coro
340program calls this function, there will be some one-time resource leak. 409calls this function, there will be some one-time resource leak.
341 410
342=cut 411=cut
343
344sub terminate {
345 $current->cancel (@_);
346}
347 412
348sub killall { 413sub killall {
349 for (Coro::State::list) { 414 for (Coro::State::list) {
350 $_->cancel 415 $_->cancel
351 if $_ != $current && UNIVERSAL::isa $_, "Coro"; 416 if $_ != $current && UNIVERSAL::isa $_, "Coro";
352 } 417 }
353} 418}
354 419
355=back 420=back
356 421
357=head2 COROUTINE METHODS 422=head1 CORO OBJECT METHODS
358 423
359These are the methods you can call on coroutine objects (or to create 424These are the methods you can call on coro objects (or to create
360them). 425them).
361 426
362=over 4 427=over 4
363 428
364=item new Coro \&sub [, @args...] 429=item new Coro \&sub [, @args...]
365 430
366Create a new coroutine and return it. When the sub returns, the coroutine 431Create a new coro and return it. When the sub returns, the coro
367automatically terminates as if C<terminate> with the returned values were 432automatically terminates as if C<terminate> with the returned values were
368called. To make the coroutine run you must first put it into the ready 433called. To make the coro run you must first put it into the ready
369queue by calling the ready method. 434queue by calling the ready method.
370 435
371See C<async> and C<Coro::State::new> for additional info about the 436See C<async> and C<Coro::State::new> for additional info about the
372coroutine environment. 437coro environment.
373 438
374=cut 439=cut
375 440
376sub _run_coro { 441sub _coro_run {
377 terminate &{+shift}; 442 terminate &{+shift};
378} 443}
379 444
380sub new {
381 my $class = shift;
382
383 $class->SUPER::new (\&_run_coro, @_)
384}
385
386=item $success = $coroutine->ready 445=item $success = $coro->ready
387 446
388Put the given coroutine into the end of its ready queue (there is one 447Put the given coro into the end of its ready queue (there is one
389queue for each priority) and return true. If the coroutine is already in 448queue for each priority) and return true. If the coro is already in
390the ready queue, do nothing and return false. 449the ready queue, do nothing and return false.
391 450
392This ensures that the scheduler will resume this coroutine automatically 451This ensures that the scheduler will resume this coro automatically
393once all the coroutines of higher priority and all coroutines of the same 452once all the coro of higher priority and all coro of the same
394priority that were put into the ready queue earlier have been resumed. 453priority that were put into the ready queue earlier have been resumed.
395 454
455=item $coro->suspend
456
457Suspends the specified coro. A suspended coro works just like any other
458coro, except that the scheduler will not select a suspended coro for
459execution.
460
461Suspending a coro can be useful when you want to keep the coro from
462running, but you don't want to destroy it, or when you want to temporarily
463freeze a coro (e.g. for debugging) to resume it later.
464
465A scenario for the former would be to suspend all (other) coros after a
466fork and keep them alive, so their destructors aren't called, but new
467coros can be created.
468
469=item $coro->resume
470
471If the specified coro was suspended, it will be resumed. Note that when
472the coro was in the ready queue when it was suspended, it might have been
473unreadied by the scheduler, so an activation might have been lost.
474
475To avoid this, it is best to put a suspended coro into the ready queue
476unconditionally, as every synchronisation mechanism must protect itself
477against spurious wakeups, and the one in the Coro family certainly do
478that.
479
396=item $is_ready = $coroutine->is_ready 480=item $is_ready = $coro->is_ready
397 481
398Return whether the coroutine is currently the ready queue or not, 482Returns true iff the Coro object is in the ready queue. Unless the Coro
483object gets destroyed, it will eventually be scheduled by the scheduler.
399 484
485=item $is_running = $coro->is_running
486
487Returns true iff the Coro object is currently running. Only one Coro object
488can ever be in the running state (but it currently is possible to have
489multiple running Coro::States).
490
491=item $is_suspended = $coro->is_suspended
492
493Returns true iff this Coro object has been suspended. Suspended Coros will
494not ever be scheduled.
495
400=item $coroutine->cancel (arg...) 496=item $coro->cancel (arg...)
401 497
402Terminates the given coroutine and makes it return the given arguments as 498Terminates the given Coro and makes it return the given arguments as
403status (default: the empty list). Never returns if the coroutine is the 499status (default: the empty list). Never returns if the Coro is the
404current coroutine. 500current Coro.
405 501
406=cut 502=cut
407 503
408sub cancel { 504sub cancel {
409 my $self = shift; 505 my $self = shift;
410 $self->{_status} = [@_];
411 506
412 if ($current == $self) { 507 if ($current == $self) {
413 push @destroy, $self; 508 terminate @_;
414 $manager->ready;
415 &schedule while 1;
416 } else { 509 } else {
417 $self->_cancel; 510 $self->{_status} = [@_];
511 Coro::State::cancel $self;
418 } 512 }
419} 513}
420 514
515=item $coro->schedule_to
516
517Puts the current coro to sleep (like C<Coro::schedule>), but instead
518of continuing with the next coro from the ready queue, always switch to
519the given coro object (regardless of priority etc.). The readyness
520state of that coro isn't changed.
521
522This is an advanced method for special cases - I'd love to hear about any
523uses for this one.
524
525=item $coro->cede_to
526
527Like C<schedule_to>, but puts the current coro into the ready
528queue. This has the effect of temporarily switching to the given
529coro, and continuing some time later.
530
531This is an advanced method for special cases - I'd love to hear about any
532uses for this one.
533
421=item $coroutine->throw ([$scalar]) 534=item $coro->throw ([$scalar])
422 535
423If C<$throw> is specified and defined, it will be thrown as an exception 536If C<$throw> is specified and defined, it will be thrown as an exception
424inside the coroutine at the next convenient point in time. Otherwise 537inside the coro at the next convenient point in time. Otherwise
425clears the exception object. 538clears the exception object.
426 539
427Coro will check for the exception each time a schedule-like-function 540Coro will check for the exception each time a schedule-like-function
428returns, i.e. after each C<schedule>, C<cede>, C<< Coro::Semaphore->down 541returns, i.e. after each C<schedule>, C<cede>, C<< Coro::Semaphore->down
429>>, C<< Coro::Handle->readable >> and so on. Most of these functions 542>>, C<< Coro::Handle->readable >> and so on. Most of these functions
431 544
432The exception object will be thrown "as is" with the specified scalar in 545The exception object will be thrown "as is" with the specified scalar in
433C<$@>, i.e. if it is a string, no line number or newline will be appended 546C<$@>, i.e. if it is a string, no line number or newline will be appended
434(unlike with C<die>). 547(unlike with C<die>).
435 548
436This can be used as a softer means than C<cancel> to ask a coroutine to 549This can be used as a softer means than C<cancel> to ask a coro to
437end itself, although there is no guarantee that the exception will lead to 550end 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 551termination, and if the exception isn't caught it might well end the whole
439program. 552program.
440 553
441You might also think of C<throw> as being the moral equivalent of 554You 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). 555C<kill>ing a coro with a signal (in this case, a scalar).
443 556
444=item $coroutine->join 557=item $coro->join
445 558
446Wait until the coroutine terminates and return any values given to the 559Wait until the coro terminates and return any values given to the
447C<terminate> or C<cancel> functions. C<join> can be called concurrently 560C<terminate> or C<cancel> functions. C<join> can be called concurrently
448from multiple coroutines, and all will be resumed and given the status 561from multiple coro, and all will be resumed and given the status
449return once the C<$coroutine> terminates. 562return once the C<$coro> terminates.
450 563
451=cut 564=cut
452 565
453sub join { 566sub join {
454 my $self = shift; 567 my $self = shift;
465 } 578 }
466 579
467 wantarray ? @{$self->{_status}} : $self->{_status}[0]; 580 wantarray ? @{$self->{_status}} : $self->{_status}[0];
468} 581}
469 582
470=item $coroutine->on_destroy (\&cb) 583=item $coro->on_destroy (\&cb)
471 584
472Registers a callback that is called when this coroutine gets destroyed, 585Registers a callback that is called when this coro gets destroyed,
473but before it is joined. The callback gets passed the terminate arguments, 586but before it is joined. The callback gets passed the terminate arguments,
474if any, and I<must not> die, under any circumstances. 587if any, and I<must not> die, under any circumstances.
475 588
476=cut 589=cut
477 590
479 my ($self, $cb) = @_; 592 my ($self, $cb) = @_;
480 593
481 push @{ $self->{_on_destroy} }, $cb; 594 push @{ $self->{_on_destroy} }, $cb;
482} 595}
483 596
484=item $oldprio = $coroutine->prio ($newprio) 597=item $oldprio = $coro->prio ($newprio)
485 598
486Sets (or gets, if the argument is missing) the priority of the 599Sets (or gets, if the argument is missing) the priority of the
487coroutine. Higher priority coroutines get run before lower priority 600coro. Higher priority coro get run before lower priority
488coroutines. Priorities are small signed integers (currently -4 .. +3), 601coro. Priorities are small signed integers (currently -4 .. +3),
489that you can refer to using PRIO_xxx constants (use the import tag :prio 602that you can refer to using PRIO_xxx constants (use the import tag :prio
490to get then): 603to get then):
491 604
492 PRIO_MAX > PRIO_HIGH > PRIO_NORMAL > PRIO_LOW > PRIO_IDLE > PRIO_MIN 605 PRIO_MAX > PRIO_HIGH > PRIO_NORMAL > PRIO_LOW > PRIO_IDLE > PRIO_MIN
493 3 > 1 > 0 > -1 > -3 > -4 606 3 > 1 > 0 > -1 > -3 > -4
494 607
495 # set priority to HIGH 608 # set priority to HIGH
496 current->prio(PRIO_HIGH); 609 current->prio (PRIO_HIGH);
497 610
498The idle coroutine ($Coro::idle) always has a lower priority than any 611The idle coro ($Coro::idle) always has a lower priority than any
499existing coroutine. 612existing coro.
500 613
501Changing the priority of the current coroutine will take effect immediately, 614Changing the priority of the current coro will take effect immediately,
502but changing the priority of coroutines in the ready queue (but not 615but changing the priority of coro in the ready queue (but not
503running) will only take effect after the next schedule (of that 616running) will only take effect after the next schedule (of that
504coroutine). This is a bug that will be fixed in some future version. 617coro). This is a bug that will be fixed in some future version.
505 618
506=item $newprio = $coroutine->nice ($change) 619=item $newprio = $coro->nice ($change)
507 620
508Similar to C<prio>, but subtract the given value from the priority (i.e. 621Similar to C<prio>, but subtract the given value from the priority (i.e.
509higher values mean lower priority, just as in unix). 622higher values mean lower priority, just as in unix).
510 623
511=item $olddesc = $coroutine->desc ($newdesc) 624=item $olddesc = $coro->desc ($newdesc)
512 625
513Sets (or gets in case the argument is missing) the description for this 626Sets (or gets in case the argument is missing) the description for this
514coroutine. This is just a free-form string you can associate with a 627coro. This is just a free-form string you can associate with a
515coroutine. 628coro.
516 629
517This method simply sets the C<< $coroutine->{desc} >> member to the given 630This method simply sets the C<< $coro->{desc} >> member to the given
518string. You can modify this member directly if you wish. 631string. You can modify this member directly if you wish.
519 632
520=cut 633=cut
521 634
522sub desc { 635sub desc {
523 my $old = $_[0]{desc}; 636 my $old = $_[0]{desc};
524 $_[0]{desc} = $_[1] if @_ > 1; 637 $_[0]{desc} = $_[1] if @_ > 1;
525 $old; 638 $old;
526} 639}
527 640
641sub transfer {
642 require Carp;
643 Carp::croak ("You must not call ->transfer on Coro objects. Use Coro::State objects or the ->schedule_to method. Caught");
644}
645
528=back 646=back
529 647
530=head2 GLOBAL FUNCTIONS 648=head1 GLOBAL FUNCTIONS
531 649
532=over 4 650=over 4
533 651
534=item Coro::nready 652=item Coro::nready
535 653
536Returns the number of coroutines that are currently in the ready state, 654Returns the number of coro that are currently in the ready state,
537i.e. that can be switched to by calling C<schedule> directory or 655i.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 656indirectly. The value C<0> means that the only runnable coro is the
539currently running one, so C<cede> would have no effect, and C<schedule> 657currently running one, so C<cede> would have no effect, and C<schedule>
540would cause a deadlock unless there is an idle handler that wakes up some 658would cause a deadlock unless there is an idle handler that wakes up some
541coroutines. 659coro.
542 660
543=item my $guard = Coro::guard { ... } 661=item my $guard = Coro::guard { ... }
544 662
545This creates and returns a guard object. Nothing happens until the object 663This function still exists, but is deprecated. Please use the
546gets destroyed, in which case the codeblock given as argument will be 664C<Guard::guard> function instead.
547executed. This is useful to free locks or other resources in case of a
548runtime error or when the coroutine gets canceled, as in both cases the
549guard block will be executed. The guard object supports only one method,
550C<< ->cancel >>, which will keep the codeblock from being executed.
551 665
552Example: set some flag and clear it again when the coroutine gets canceled
553or the function returns:
554
555 sub do_something {
556 my $guard = Coro::guard { $busy = 0 };
557 $busy = 1;
558
559 # do something that requires $busy to be true
560 }
561
562=cut 666=cut
563 667
564sub guard(&) { 668BEGIN { *guard = \&Guard::guard }
565 bless \(my $cb = $_[0]), "Coro::guard"
566}
567
568sub Coro::guard::cancel {
569 ${$_[0]} = sub { };
570}
571
572sub Coro::guard::DESTROY {
573 ${$_[0]}->();
574}
575
576 669
577=item unblock_sub { ... } 670=item unblock_sub { ... }
578 671
579This utility function takes a BLOCK or code reference and "unblocks" it, 672This utility function takes a BLOCK or code reference and "unblocks" it,
580returning a new coderef. Unblocking means that calling the new coderef 673returning a new coderef. Unblocking means that calling the new coderef
581will return immediately without blocking, returning nothing, while the 674will return immediately without blocking, returning nothing, while the
582original code ref will be called (with parameters) from within another 675original code ref will be called (with parameters) from within another
583coroutine. 676coro.
584 677
585The reason this function exists is that many event libraries (such as the 678The reason this function exists is that many event libraries (such as the
586venerable L<Event|Event> module) are not coroutine-safe (a weaker form 679venerable L<Event|Event> module) are not thread-safe (a weaker form
587of thread-safety). This means you must not block within event callbacks, 680of reentrancy). This means you must not block within event callbacks,
588otherwise you might suffer from crashes or worse. The only event library 681otherwise 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>. 682currently known that is safe to use without C<unblock_sub> is L<EV>.
590 683
591This function allows your callbacks to block by executing them in another 684This function allows your callbacks to block by executing them in another
592coroutine where it is safe to block. One example where blocking is handy 685coro where it is safe to block. One example where blocking is handy
593is when you use the L<Coro::AIO|Coro::AIO> functions to save results to 686is when you use the L<Coro::AIO|Coro::AIO> functions to save results to
594disk, for example. 687disk, for example.
595 688
596In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when 689In short: simply use C<unblock_sub { ... }> instead of C<sub { ... }> when
597creating event callbacks that want to block. 690creating event callbacks that want to block.
598 691
599If your handler does not plan to block (e.g. simply sends a message to 692If 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), 693another coro, or puts some other coro into the ready queue), there is
601there is no reason to use C<unblock_sub>. 694no reason to use C<unblock_sub>.
602 695
603Note that you also need to use C<unblock_sub> for any other callbacks that 696Note 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 697are 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 698use 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 699provides callbacks that are the result of some event callback, then you
615# return immediately and can be reused) and because we cannot cede 708# return immediately and can be reused) and because we cannot cede
616# inside an event callback. 709# inside an event callback.
617our $unblock_scheduler = new Coro sub { 710our $unblock_scheduler = new Coro sub {
618 while () { 711 while () {
619 while (my $cb = pop @unblock_queue) { 712 while (my $cb = pop @unblock_queue) {
620 # this is an inlined copy of async_pool 713 &async_pool (@$cb);
621 my $coro = (pop @async_pool) || new Coro \&pool_handler;
622 714
623 $coro->{_invoke} = $cb;
624 $coro->ready;
625 cede; # for short-lived callbacks, this reduces pressure on the coro pool 715 # for short-lived callbacks, this reduces pressure on the coro pool
716 # as the chance is very high that the async_poll coro will be back
717 # in the idle state when cede returns
718 cede;
626 } 719 }
627 schedule; # sleep well 720 schedule; # sleep well
628 } 721 }
629}; 722};
630$unblock_scheduler->{desc} = "[unblock_sub scheduler]"; 723$unblock_scheduler->{desc} = "[unblock_sub scheduler]";
638 } 731 }
639} 732}
640 733
641=item $cb = Coro::rouse_cb 734=item $cb = Coro::rouse_cb
642 735
643Create and return a "rouse callback". That's a code reference that, when 736Create and return a "rouse callback". That's a code reference that,
644called, will save its arguments and notify the owner coroutine of the 737when called, will remember a copy of its arguments and notify the owner
645callback. 738coro of the callback.
646 739
647See the next function. 740See the next function.
648 741
649=item @args = Coro::rouse_wait [$cb] 742=item @args = Coro::rouse_wait [$cb]
650 743
651Wait for the specified rouse callback (or the last one tht was created in 744Wait for the specified rouse callback (or the last one that was created in
652this coroutine). 745this coro).
653 746
654As soon as the callback is invoked (or when the calback was invoked before 747As soon as the callback is invoked (or when the callback was invoked
655C<rouse_wait>), it will return a copy of the arguments originally passed 748before C<rouse_wait>), it will return the arguments originally passed to
656to the rouse callback. 749the rouse callback. In scalar context, that means you get the I<last>
750argument, just as if C<rouse_wait> had a C<return ($a1, $a2, $a3...)>
751statement at the end.
657 752
658See the section B<HOW TO WAIT FOR A CALLBACK> for an actual usage example. 753See the section B<HOW TO WAIT FOR A CALLBACK> for an actual usage example.
659 754
660=back 755=back
661 756
663 758
6641; 7591;
665 760
666=head1 HOW TO WAIT FOR A CALLBACK 761=head1 HOW TO WAIT FOR A CALLBACK
667 762
668It is very common for a coroutine to wait for some callback to be 763It is very common for a coro to wait for some callback to be
669called. This occurs naturally when you use coroutines in an otherwise 764called. This occurs naturally when you use coro in an otherwise
670event-based program, or when you use event-based libraries. 765event-based program, or when you use event-based libraries.
671 766
672These typically register a callback for some event, and call that callback 767These typically register a callback for some event, and call that callback
673when the event occured. In a coroutine, however, you typically want to 768when the event occured. In a coro, however, you typically want to
674just wait for the event, simplyifying things. 769just wait for the event, simplyifying things.
675 770
676For example C<< AnyEvent->child >> registers a callback to be called when 771For example C<< AnyEvent->child >> registers a callback to be called when
677a specific child has exited: 772a specific child has exited:
678 773
679 my $child_watcher = AnyEvent->child (pid => $pid, cb => sub { ... }); 774 my $child_watcher = AnyEvent->child (pid => $pid, cb => sub { ... });
680 775
681But from withina coroutine, you often just want to write this: 776But from within a coro, you often just want to write this:
682 777
683 my $status = wait_for_child $pid; 778 my $status = wait_for_child $pid;
684 779
685Coro offers two functions specifically designed to make this easy, 780Coro offers two functions specifically designed to make this easy,
686C<Coro::rouse_cb> and C<Coro::rouse_wait>. 781C<Coro::rouse_cb> and C<Coro::rouse_wait>.
687 782
688The first function, C<rouse_cb>, generates and returns a callback that, 783The first function, C<rouse_cb>, generates and returns a callback that,
689when invoked, will save it's arguments and notify the coroutine that 784when invoked, will save its arguments and notify the coro that
690created the callback. 785created the callback.
691 786
692The second function, C<rouse_wait>, waits for the callback to be called 787The second function, C<rouse_wait>, waits for the callback to be called
693(by calling C<schedule> to go to sleep) and returns the arguments 788(by calling C<schedule> to go to sleep) and returns the arguments
694originally passed to the callback. 789originally passed to the callback.
709you can roll your own, using C<schedule>: 804you can roll your own, using C<schedule>:
710 805
711 sub wait_for_child($) { 806 sub wait_for_child($) {
712 my ($pid) = @_; 807 my ($pid) = @_;
713 808
714 # store the current coroutine in $current, 809 # store the current coro in $current,
715 # and provide result variables for the closure passed to ->child 810 # and provide result variables for the closure passed to ->child
716 my $current = $Coro::current; 811 my $current = $Coro::current;
717 my ($done, $rstatus); 812 my ($done, $rstatus);
718 813
719 # pass a closure to ->child 814 # pass a closure to ->child
735 830
736=item fork with pthread backend 831=item fork with pthread backend
737 832
738When Coro is compiled using the pthread backend (which isn't recommended 833When Coro is compiled using the pthread backend (which isn't recommended
739but required on many BSDs as their libcs are completely broken), then 834but required on many BSDs as their libcs are completely broken), then
740coroutines will not survive a fork. There is no known workaround except to 835coro will not survive a fork. There is no known workaround except to
741fix your libc and use a saner backend. 836fix your libc and use a saner backend.
742 837
743=item perl process emulation ("threads") 838=item perl process emulation ("threads")
744 839
745This module is not perl-pseudo-thread-safe. You should only ever use this 840This module is not perl-pseudo-thread-safe. You should only ever use this
746module from the same thread (this requirement might be removed in the 841module from the first thread (this requirement might be removed in the
747future to allow per-thread schedulers, but Coro::State does not yet allow 842future to allow per-thread schedulers, but Coro::State does not yet allow
748this). I recommend disabling thread support and using processes, as having 843this). I recommend disabling thread support and using processes, as having
749the windows process emulation enabled under unix roughly halves perl 844the windows process emulation enabled under unix roughly halves perl
750performance, even when not used. 845performance, even when not used.
751 846
752=item coroutine switching not signal safe 847=item coro switching is not signal safe
753 848
754You must not switch to another coroutine from within a signal handler 849You must not switch to another coro from within a signal handler
755(only relevant with %SIG - most event libraries provide safe signals). 850(only relevant with %SIG - most event libraries provide safe signals).
756 851
757That means you I<MUST NOT> call any function that might "block" the 852That means you I<MUST NOT> call any function that might "block" the
758current coroutine - C<cede>, C<schedule> C<< Coro::Semaphore->down >> or 853current coro - C<cede>, C<schedule> C<< Coro::Semaphore->down >> or
759anything that calls those. Everything else, including calling C<ready>, 854anything that calls those. Everything else, including calling C<ready>,
760works. 855works.
761 856
762=back 857=back
763 858
764 859
860=head1 WINDOWS PROCESS EMULATION
861
862A great many people seem to be confused about ithreads (for example, Chip
863Salzenberg called me unintelligent, incapable, stupid and gullible,
864while in the same mail making rather confused statements about perl
865ithreads (for example, that memory or files would be shared), showing his
866lack of understanding of this area - if it is hard to understand for Chip,
867it is probably not obvious to everybody).
868
869What follows is an ultra-condensed version of my talk about threads in
870scripting languages given onthe perl workshop 2009:
871
872The so-called "ithreads" were originally implemented for two reasons:
873first, to (badly) emulate unix processes on native win32 perls, and
874secondly, to replace the older, real thread model ("5.005-threads").
875
876It does that by using threads instead of OS processes. The difference
877between processes and threads is that threads share memory (and other
878state, such as files) between threads within a single process, while
879processes do not share anything (at least not semantically). That
880means that modifications done by one thread are seen by others, while
881modifications by one process are not seen by other processes.
882
883The "ithreads" work exactly like that: when creating a new ithreads
884process, all state is copied (memory is copied physically, files and code
885is copied logically). Afterwards, it isolates all modifications. On UNIX,
886the same behaviour can be achieved by using operating system processes,
887except that UNIX typically uses hardware built into the system to do this
888efficiently, while the windows process emulation emulates this hardware in
889software (rather efficiently, but of course it is still much slower than
890dedicated hardware).
891
892As mentioned before, loading code, modifying code, modifying data
893structures and so on is only visible in the ithreads process doing the
894modification, not in other ithread processes within the same OS process.
895
896This is why "ithreads" do not implement threads for perl at all, only
897processes. What makes it so bad is that on non-windows platforms, you can
898actually take advantage of custom hardware for this purpose (as evidenced
899by the forks module, which gives you the (i-) threads API, just much
900faster).
901
902Sharing data is in the i-threads model is done by transfering data
903structures between threads using copying semantics, which is very slow -
904shared data simply does not exist. Benchmarks using i-threads which are
905communication-intensive show extremely bad behaviour with i-threads (in
906fact, so bad that Coro, which cannot take direct advantage of multiple
907CPUs, is often orders of magnitude faster because it shares data using
908real threads, refer to my talk for details).
909
910As summary, i-threads *use* threads to implement processes, while
911the compatible forks module *uses* processes to emulate, uhm,
912processes. I-threads slow down every perl program when enabled, and
913outside of windows, serve no (or little) practical purpose, but
914disadvantages every single-threaded Perl program.
915
916This is the reason that I try to avoid the name "ithreads", as it is
917misleading as it implies that it implements some kind of thread model for
918perl, and prefer the name "windows process emulation", which describes the
919actual use and behaviour of it much better.
920
765=head1 SEE ALSO 921=head1 SEE ALSO
766 922
767Event-Loop integration: L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>. 923Event-Loop integration: L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
768 924
769Debugging: L<Coro::Debug>. 925Debugging: L<Coro::Debug>.
770 926
771Support/Utility: L<Coro::Specific>, L<Coro::Util>. 927Support/Utility: L<Coro::Specific>, L<Coro::Util>.
772 928
773Locking/IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>, L<Coro::SemaphoreSet>, L<Coro::RWLock>. 929Locking and IPC: L<Coro::Signal>, L<Coro::Channel>, L<Coro::Semaphore>,
930L<Coro::SemaphoreSet>, L<Coro::RWLock>.
774 931
775IO/Timers: L<Coro::Timer>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::AIO>. 932I/O and Timers: L<Coro::Timer>, L<Coro::Handle>, L<Coro::Socket>, L<Coro::AIO>.
776 933
777Compatibility: L<Coro::LWP>, L<Coro::BDB>, L<Coro::Storable>, L<Coro::Select>. 934Compatibility with other modules: L<Coro::LWP> (but see also L<AnyEvent::HTTP> for
935a better-working alternative), L<Coro::BDB>, L<Coro::Storable>,
936L<Coro::Select>.
778 937
779XS API: L<Coro::MakeMaker>. 938XS API: L<Coro::MakeMaker>.
780 939
781Low level Configuration, Coroutine Environment: L<Coro::State>. 940Low level Configuration, Thread Environment, Continuations: L<Coro::State>.
782 941
783=head1 AUTHOR 942=head1 AUTHOR
784 943
785 Marc Lehmann <schmorp@schmorp.de> 944 Marc Lehmann <schmorp@schmorp.de>
786 http://home.schmorp.de/ 945 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines