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

Comparing cvsroot/Coro/Coro.pm (file contents):
Revision 1.30 by root, Sat Aug 11 19:59:19 2001 UTC vs.
Revision 1.38 by root, Wed Oct 3 01:09:56 2001 UTC

… …
32 32
33=cut 33=cut
34 34
35package Coro; 35package Coro;
36 36
37no warnings qw(uninitialized);
38
37use Coro::State; 39use Coro::State;
38 40
39use base Exporter; 41use base Exporter;
40 42
41$VERSION = 0.45; 43$VERSION = 0.51;
42 44
43@EXPORT = qw(async cede schedule terminate current); 45@EXPORT = qw(async cede schedule terminate current);
44@EXPORT_OK = qw($current); 46%EXPORT_TAGS = (
47 prio => [qw(PRIO_MAX PRIO_HIGH PRIO_NORMAL PRIO_LOW PRIO_IDLE PRIO_MIN)],
48);
49@EXPORT_OK = @{$EXPORT_TAGS{prio}};
45 50
46{ 51{
47 my @async; 52 my @async;
48 my $init; 53 my $init;
49 54
… …
111}; 116};
112 117
113# this coroutine is necessary because a coroutine 118# this coroutine is necessary because a coroutine
114# cannot destroy itself. 119# cannot destroy itself.
115my @destroy; 120my @destroy;
121my $manager;
116my $manager = new Coro sub { 122$manager = new Coro sub {
117 while() { 123 while() {
118 delete ((pop @destroy)->{_coro_state}) while @destroy; 124 # by overwriting the state object with the manager we destroy it
125 # while still being able to schedule this coroutine (in case it has
126 # been readied multiple times. this is harmless since the manager
127 # can be called as many times as neccessary and will always
128 # remove itself from the runqueue
129 (pop @destroy)->{_coro_state} = $manager->{_coro_state} while @destroy;
119 &schedule; 130 &schedule;
120 } 131 }
121}; 132};
122 133
123# static methods. not really. 134# static methods. not really.
… …
226=cut 237=cut
227 238
228sub cancel { 239sub cancel {
229 push @destroy, $_[0]; 240 push @destroy, $_[0];
230 $manager->ready; 241 $manager->ready;
242 &schedule if $current == $_[0];
243}
244
245=item $oldprio = $process->prio($newprio)
246
247Sets the priority of the process. Higher priority processes get run before
248lower priority processes. Priorities are smalled signed integer (currently
249-4 .. +3), that you can refer to using PRIO_xxx constants (use the import
250tag :prio to get then):
251
252 PRIO_MAX > PRIO_HIGH > PRIO_NORMAL > PRIO_LOW > PRIO_IDLE > PRIO_MIN
253 3 > 1 > 0 > -1 > -3 > -4
254
255 # set priority to HIGH
256 current->prio(PRIO_HIGH);
257
258The idle coroutine ($Coro::idle) always has a lower priority than any
259existing coroutine.
260
261Changing the priority of the current process will take effect immediately,
262but changing the priority of processes in the ready queue (but not
263running) will only take effect after the next schedule (of that
264process). This is a bug that will be fixed in some future version.
265
266=cut
267
268sub prio {
269 my $old = $_[0]{prio};
270 $_[0]{prio} = $_[1] if @_ > 1;
271 $old;
272}
273
274=item $newprio = $process->nice($change)
275
276Similar to C<prio>, but subtract the given value from the priority (i.e.
277higher values mean lower priority, just as in unix).
278
279=cut
280
281sub nice {
282 $_[0]{prio} -= $_[1];
231} 283}
232 284
233=back 285=back
234 286
235=cut 287=cut
236 288
2371; 2891;
238 290
239=head1 BUGS/LIMITATIONS 291=head1 BUGS/LIMITATIONS
240 292
241 - could be faster, especially when the core would introduce special 293 - you must make very sure that no coro is still active on global destruction.
242 support for coroutines (like it does for threads). 294 very bad things might happen otherwise (usually segfaults).
243 - there is still a memleak on coroutine termination that I could not
244 identify. Could be as small as a single SV.
245 - this module is not well-tested.
246 - if variables or arguments "disappear" (become undef) or become
247 corrupted please contact the author so he cen iron out the
248 remaining bugs.
249 - this module is not thread-safe. You must only ever use this module from 295 - this module is not thread-safe. You must only ever use this module from
250 the same thread (this requirement might be loosened in the future to 296 the same thread (this requirement might be loosened in the future to
251 allow per-thread schedulers, but Coro::State does not yet allow this). 297 allow per-thread schedulers, but Coro::State does not yet allow this).
252 298
253=head1 SEE ALSO 299=head1 SEE ALSO

Diff Legend

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