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

Comparing Coro/Changes (file contents):
Revision 1.448 by root, Thu Nov 20 07:02:43 2008 UTC vs.
Revision 1.539 by root, Mon May 17 16:51:44 2010 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
3TODO: should explore PerlIO::coroaio (perl leaks like hell). 3TODO: should explore PerlIO::coroaio (perl leaks like hell).
4TODO: maybe implement a default message channel, very much like Erlang's 4TODO: unready_all
5 actor model (which is cool in a lot of important aspects (failures!), 5TODO: myhttpd header parsing
6 but very lacking in others (higher level ipc)). 6TODO: channel->maxsize(newsize)?
7TODO: http://www.microsoft.com/msj/archive/s2ce.aspx
7 8
85.0 95.23 Mon May 17 18:50:42 CEST 2010
10 - be more resistant to ordering changes when initialising
11 Coro::AnyEvent, Coro::EV and Coro::Event (reported by Matthias
12 Waldorf).
13 - document that perl 5.12 deliberately removed support for cloning.
14
155.22 Wed Apr 14 03:55:35 CEST 2010
16 - correctly return udnef on errors in Coro::Handle::read/write
17 (testcase by Marc Mims).
18 - convert Coro::Util into a "perl compatibility wrapper" - the functions
19 are less useful now, but are drop-in replacements for existing
20 functions, listing better alternatives in the documentation. This also
21 fixes a bug in Coro::LWP which naively substituted Socket::inet_aton
22 with Coro::Util::inet_aton.
23 - do not override $Coro::idle unconditionally in Coro.pm, as other
24 modules could have provided their own idle coro already
25 (for exmaple, Coro::AnyEvent).
26 - fix Coro::Util::gethost* functions.
27 - Coro::Timer corretcly exports it's symbols (reported by Hideki Yamamura).
28
295.21 Wed Dec 16 07:19:51 CET 2009
30 - automatically load Coro::AnyEvent when AnyEvent and Coro are used
31 together.
32 - add some examples on how to combine other event loops with Coro in
33 Coro::AnyEvent, and how to run it (and not to block). Seems to be
34 the most common source of confusion.
35 - try to catch people naively blocking in an event callback.
36 - work around the perl filehandle bug issue in conjunction with
37 older common::sense (as indirectly pointed out by ZSystem).
38 - clarify the "not from signal handlers" section.
39
405.2 Sun Oct 4 14:54:24 CEST 2009
41 - Coro::Storable destroyed the prototypes of the functions it wrapped.
42 - export rouse_cb and rouse_wait by default now.
43 - fix various prototype mismatches in Coro::AnyEvent and Coro::Handle.
44 - new method $state->swap_sv.
45 - added section on "windows process emulation" to the manpage,
46 after a not-so-fruitful (nor-friendly) "discussion" with chip
47 salzenberg (discussion implies arguments, but his only arguments
48 were ad-hominems, one wonders why he started it in the first
49 place). I hope this explains it well enough for him to understand,
50 and maybe well enough for others to understand.
51 - use common::sense everywhere now.
52 - idle callbacks are no longer supported, use idle coros instead.
53 - print a thread listing when a deadlock is detected.
54
555.17 Sat Aug 22 23:09:31 CEST 2009
56 - work around a bug in the perl debugger causing crashes
57 when running under the debugger by marking _pool_handler as nodebug.
58 - speed up Coro::async considerably.
59 - try some hacks to get netbsd to work "more often" - their broken
60 setjmp/longjmp, ucontext *and* phtreads are really hard on Coro.
61 - convert Coro to AE 5.0 API.
62
635.162 Tue Jul 28 04:04:03 CEST 2009
64 - perl 5.8.2 is now minimum requirement.
65 - skip t/19_handle.t on broken windows perls.
66
675.161 Wed Jul 22 04:47:38 CEST 2009
68 - Coro::AnyEvent::poll could have a different prototype when EV was
69 used as backend (analyzed by Tatsuhiko Miyagawa).
70 - Coro::AnyEvent errornously initialised the event loop when loaded,
71 not on demand.
72 - try to workaround rare */t/01_unblock.t failures.
73
745.16 Tue Jul 21 01:44:37 CEST 2009
75 - Coro::AnyEvent failed to hook into the event loop
76 when no threads had been readied between detecting
77 the event loop and actually running it.
78 - considerably speed up Coro::Select by taking avdantage
79 of AnyEvent > 4.8 and some other optimisations.
80 - implement paragraph readline mode in Coro::Handle
81 (based on patches by Zsbán Ambrus).
82 - replace WSAEINPROGRESS by WSAEWOULDBLOCK (reported
83 and analyzed by Yasuhiro MATSUMOTO).
84 - clarified libcoro license and copyright.
85 - someone stole my EXTRA_META!!!
86 - implement Coro::Select::patch_pp_sselect and it's brother,
87 for hardcode select overriding.
88
895.151 Mon Jul 6 05:41:57 CEST 2009
90 - backport to windows process emulation code again (patch by
91 Yasuhiro MATSUMOTO).
92 - slightly update Coro::MakeMaker.
93
945.15 Tue Jun 30 10:28:06 CEST 2009
95 - deprecate Coro::Socket, document how to get ipv6 support via
96 AnyEvent::Socket instead.
97 - implement signal->wait ($cb) interface, similar to semaphores.
98 - work around SvOK not supporting getmagic, so we have to getmagic
99 to test for undef :/ (reported by Matthias Waldorf).
100 - load Coro::AnyEvent in all modules using AnyEvent.
101 - work around perl corrupting our internal data structures,
102 reported by Tokuhiro Matsuno.
103 - enable per-coroutine real and cpu time gathering
104 (Coro::State::enable_times).
105
1065.14 Wed Jun 24 01:37:48 CEST 2009
107 - provide explicit functions to "cede" to the event loop in Coro::AnyEvent,
108 as this seems to have been a difficult concept (poll, sleep, idle,
109 idle_upto).
110 - add Coro::AnyEvent::readable/writable functions.
111 - clarify Coro::EV/Event/AnyEvent manpages.
112 - free per-thread global scalars in the thread calling ->cancel, to
113 avoid crashes when $_, $@ etc., are magical but some of those
114 had already been freed.
115 - "unexperimentalise" the callback interface for Coro::Semaphore.
116 - speed up ready queue management/context switching by using a linked
117 list instead of an array (~5-10%).
118 - implement "watch" command in Coro::Debug shells.
119 - for fun, implement time-slicing as an example in the manpage.
120 - if AnyEvent detects EV or Event, but we don't have Coro::EV or
121 Coro::Event, use the normal AnyEvent handling instead of dieing
122 (the same is true for Coro::Handle).
123 - properly document Coro::EV::timed_io_once.
124 - avoid unneccessary ->cancel calls in Coro::Handle.
125 - maybe make it work on mingw32 with win32 backend
126 (based on patch by Yasuhiro Matsumoto).
127
1285.132 Fri May 29 09:00:39 CEST 2009
129 - do not keep a reference to the argument itself in
130 Coro::Semaphore::guard, as it could change later.
131 - support SO_RCVBUF/SO_SNDBUF nonstandard Coro::Socket options,
132 should support a prepare callback.
133
1345.131 Mon Mar 16 23:20:37 CET 2009
135 - implement and document Coro->suspend, Coro->resume.
136 - fix Coro::Select implementation to not (often) close
137 the passed file descriptors (testcase provided by pippijn).
138
1395.13 Mon Dec 15 21:51:42 CET 2008
140 - EXPERIMENTAL: implement dynamic winds (on_enter/on_leave).
141 - don't set diehook to C<undef> but instead to NULL, to avoid
142 spurious warnings.
143 - fix a lot of bugs in Coro::SemaphoreSet.
144 - Coro::SemaphoreSet will less often create a semaphore needlessly.
145 - add Coro::SemaphoreSet::count and wait methods.
146 - take advantage of the new Guard module.
147 - deprecate Coro::guard.
148 - try to fix the dreaded 01_unblock tests once more. I hate it when
149 testsuites need more fixing than the code they are supposed to test.
150 - croak in more cases when a required callback isn't resolvable.
151 - fix some minor issues in Coro::State->call/eval.
152 - use current coroutine context instead of a temporary one
153 when temporarily switching to another coroutine.
154 - do not call C-level on_destroy handlers during global destruction,
155 to avoid needless segfaults.
156
1575.12 Sun Dec 7 13:30:38 CET 2008
158 - add default config for MirOS, which seems to be a bug-to-bug
159 compatible fork of openbsd ("world domination by releasing
160 openbsd cvs before the openbsd folks do it" or so :).
161 - free_padlist did destroy the names pad, not good, but didn't
162 seem to bother perl - this could fix issues such as eval ""
163 inside a function called from multiple coroutines.
164 - use a different method to detect destruction time.
165 - be more careful when freeing padlists just before global
166 destruction.
167 - fixed and expanded the call/cc example.
168 - renamed _terminate to _coro_run.
169 - new method Coro::Channel->shutdown.
170 - try pthreads on openbsd <4.4 (broken sigaltstack, will
171 pthreads fare better?).
172 - be less picky about destroying "a" running coroutine.
173
1745.11 Tue Nov 25 21:49:05 CET 2008
175 - DEBUGGING in 5.10.0 is a mess: it constantly flags perfectly
176 working code with failed assertions, introducing more bugs than
177 it fixes, requiring elaborate workarounds :(
178
1795.1 Mon Nov 24 08:54:59 CET 2008
180 - wrote a small introductory tutorial - Coro::Intro.
181 - convert Coro::Timer, Coro::Select and Coro::Util to rouse API.
182 - Coro::Select did errornously dup the file descriptors
183 and didn't work with all AnyEvent backends.
184 - Coro::Select wasn't imported correctly form Coro::LWP, causing blocking
185 LWP data transfers.
186 - disassociate c contexts from coro objects - this is agruably more
187 correct, but mostly allows sharing of cctxs between coro and state
188 objects, for added memory savings and speed increases.
189 - bumped $Coro::POOL_RSS up to 32kb by default.
190 - no longer set the optype to OP_CUSTOM, as B::* understandably
191 doesn't like this very much (and we *are* a type of entersub).
192 - implement state cloning, just to prove that call/cc can be done.
193 - automatically load Coro::AnyEvent in Coro::Handle.
194 - wrap ->cancel calls in eval inside Coro::Handle as EV watchers
195 do not have this method (and don't need it either).
196 - speed up generic anyevent methods in Coro::Handle by using rouse
197 callbacks.
198 - allow coroutines in $Coro::IDLE, speeding up Coro::AnyEvent and
199 others. It also makes the debugger happier, as you can trace
200 through the idle threads now.
201 - add comppad_name* and hints ($^H, %^H) to per-thread variables.
202 - eg/event was pretty broken.
203 - better 5.8.6 compatibility.
204
2055.0 Thu Nov 20 10:35:05 CET 2008
9 - NEW ARCHITECTURE: use the latest 4.x version if you experience 206 - NEW ARCHITECTURE: use the latest 4.x version if you experience
10 stability issues. 207 stability issues.
11 - bump API version to 7 - all dependents must be recompiled. 208 - bump API version to 7 - all dependents must be recompiled.
12 - removed timed_* functions - they were not being used anyways 209 - removed timed_* functions - they were not being used anyways
13 and should be replaced by a more generic mechanism - 210 and should be replaced by a more generic mechanism -
14 and were annoying to support anyways :) 211 and were annoying to support anyways :)
15 - removed SemaphoreSet's waiter method - use sem instead. 212 - removed SemaphoreSet's waiter method - use sem method instead.
16 - Coro::Semaphore->adjust didn't correctly wake up enough waiters. 213 - Coro::Semaphore->adjust didn't correctly wake up enough waiters.
17 - async_pool did free a scalar value twice 214 - async_pool did free a scalar value twice
18 ("Attempt to unreference..."). 215 ("Attempt to unreference...").
19 - fix a disastrous bug in the readline optimisation
20 introduced in 4.801.
21 - fix a longstanding bug where calling terminate on a coro that 216 - fix a longstanding bug where calling terminate on a coro that
22 was waiting for a semaphore that was just becoming available 217 was waiting for a semaphore that was just becoming available
23 would cause a deadlock (semaphore would get into a state where 218 would cause a deadlock (semaphore would get into a state where
24 it was available but waiters were still blocked). 219 it was available but waiters were still blocked).
25 - calling throw on a coroutine that is waiting for a semaphore will 220 - calling throw on a coroutine that is waiting for a semaphore will
… …
67 - use new rouse functions to speed up and simplify Coro::BDB. 262 - use new rouse functions to speed up and simplify Coro::BDB.
68 - make "prefer perl native functions" work with threaded perls. 263 - make "prefer perl native functions" work with threaded perls.
69 - condense Coro::Debug ps output, hint at v and w flags. 264 - condense Coro::Debug ps output, hint at v and w flags.
70 - (libcoro) lots of minor cleanups and portability improvements. 265 - (libcoro) lots of minor cleanups and portability improvements.
71 266
2674.914 Wed Nov 19 12:54:18 CET 2008
268 - fix a disastrous bug in the readline optimisation
269 introduced in 4.801.
270
724.913 Sat Nov 15 07:58:28 CET 2008 2714.913 Sat Nov 15 07:58:28 CET 2008
73 - async_pool did free a scalar value twice 272 - async_pool did free a scalar value twice
74 ("Attempt to unreference..."). 273 ("Attempt to unreference...").
75 274
764.912 Thu Nov 13 18:31:23 CET 2008 2754.912 Thu Nov 13 18:31:23 CET 2008

Diff Legend

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