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

Comparing Coro/Changes (file contents):
Revision 1.84 by pcg, Sat Nov 15 03:53:10 2003 UTC vs.
Revision 1.153 by root, Fri Nov 24 00:31:21 2006 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
30.9 3TODO: check wether stack-sharing is still effective
4TODO: more effective stack sharing by having one "default stack" for "simple" coroutines?
5TODO: verify exception frame changes and force stack copies?
6
7 - Coro::Event could cause livelocks if it was used but
8 Event watchers were used directly.
9 - Coro::Event should be cleaner and more efficient now.
10 - minor documentation cleanup.
11 - very minor optimisations and cleanups.
12 - avoid problems due to compiler inlining.
13
142.5 Tue Nov 7 12:22:33 CET 2006
15 - made Coro::Util, Coro::Select, Coro::Handle and Coro::Socket
16 use AnyEvent, moved them to Coro/.
17 - added Coro::LWP which contains all the uglyness required to
18 make LWP non-blocking.
19 - should work with perl 5.9.x now (Andreas König made me do it).
20 - fixed another bug in Coro::Select when the timeout was undef.
21 - reuse PL_start_env for all coros, saves some memory per coroutine.
22 - manage PL_top_env differently, hopefully to avoid panic: top_env.
23 - timeout argument was not properly used in Coro::Socket.
24 - allow limited forms of subclassing in Coro::Handle/Coro::Socket.
25 - emulate undocumented(!) functionality of IO::Socket required
26 by LWP(!!).
27 - updated eg/lwp to work with newer lwp's.
28 - remove "FATAL: uncaught exception" prefix. Coroutines that die
29 kill the whole process, just as exceptions in the main "coroutine"
30 did already.
31
322.1 Wed Nov 1 23:01:13 CET 2006
33 - fix a long-standing bug in Coro::Select where select with
34 zero timeout would instead change the current default filehandle.
35 - use a simpler and hopefully more robust way to clone padlists
36 (uses less memory and a perl function instead of our own).
37 - coro can now create a stack guard area on many architectures.
38 - Coro::AIO properly reexports additional functions from IO::AIO.
39 - updated libcoro with a workaround for OS X,
40 pach and testing by Michael Schwern.
41
422.0 Tue Oct 24 05:47:17 CEST 2006
43 - support additional aio requests in Coro::AIO.
44
451.9 (never properly released due to a glitch)
46
471.8 Thu Feb 2 00:59:06 CET 2006
48 - applied suggested patch by SAMV to avoid problems during stupid
49 mark & sweep gc run.
50 - applied patch by Scott Walters for 5.9.3 compatibility.
51
521.7 Tue Dec 27 01:41:58 CET 2005
53 - added Coro::AIO, a thin wrapper around IO::AIO.
54 - improved Makefile.PL explanations.
55
561.6 Mon Dec 12 21:30:05 CET 2005
57 - additionally save PM_curpm might fix as-of-yet
58 unseen problems with regex matches being attributed to the
59 wrong package.
60 - add t/10_bugs.t, which currently checks against imho broken
61 perls that use 0.26MB of stack space per Perl_magic_get
62 invocation (newer linuxes) as opposed to the 0.0002MB perl
63 normally uses.
64 - make stacksize configurable for the ultimate debian experience.
65
661.51 Mon Dec 12 18:48:36 CET 2005
67 - remove debugging warn() accidentally left in Coro::Select.
68
691.5 Tue Nov 29 13:32:44 CET 2005
70 - use Coro::Event inside Coro::Select to avoid spurious deadlocks.
71 - fix Coro::Select 'select'.
72 - strict'ify some modules.
73 - libcoro errornously restored the SIGUSR2 handler to SIGUSR1.
74 - use XSLoader in selected modules.
75 - remove some 5.6 compatibility cruft.
76
771.4 Tue Sep 6 00:11:05 CEST 2005
78 - libcoro did not take into account the trampoline on amd64,
79 when 'l'inux method was used.
80
811.31 Tue Aug 30 23:31:33 CEST 2005
82 - some portability fixes/workarounds.
83
841.3 Sat Aug 20 03:08:56 CEST 2005
85 - no code changes. module seems to work fine.
86 - improve Coro::State docs, remove reference to nonexisting
87 Coro::State::flush method.
88 - no longer autodetect windows, present it as an option instead.
89
901.2 Mon May 16 02:00:55 CEST 2005
91 - included libcoro.c earlier in State.xs, might improve portability.
92 - use a faster and possibly more stable Coro::Cont implementation.
93 - accept x86_64 in addition to amd64 for optimized linux-amd64 support.
94 - fix bugs Coro::SemaphoreSet that could cause locks never to be freed.
95 - fix bugs in CoroAPI.h and document it in Coro::MakeMaker.
96
971.11 Thu Mar 3 18:00:52 CET 2005
98 - change of contact address.
99
1001.1 Tue Feb 22 20:51:16 CET 2005
101 - support [l]inux method on amd64.
102 - allow some context switches while compiling/eval'ing.
103
1041.0 Tue Aug 10 03:47:27 CEST 2004
105 - Event 0.89 fixes the bug again.
106 - fix the remaining(??) memleak. I didn't free the cache-AV for
107 padlists, so one AV per created anon-sub leaked.
108
1090.97 Fri May 14 15:23:32 CEST 2004
110 - remove stupid gcc extension that was an experiment only anyway.
111
1120.96 Thu May 13 18:09:29 CEST 2004
113 - tell people if they have yet another broken Event variant.
114 - made Coro to work _better_ with threads (it seems that, as long
115 as you use Coros in one thread only everything will work fine.
116 Looking at the perl source the problem might be missing locking
117 between Coro and some perl internal routines. That's not easily
118 fixable).
119 - (hopefully) got rid of the static xnext inside transfer.
120
1210.95 Sun Feb 15 03:20:28 CET 2004
122 - removed Coro::State::flush, implemented a new and much
123 faster caching mechanism by attaching magic to the CV.
124 - a memleak with (real) closures remains.
125 - dramatically reduced size of stacks to 65536 (linux/x86), after
126 observing that even myhttpd never had more than a kilobyte
127 of stack in use (stack is still only physically allocated in
128 pagesize increments for systems suporting mmap).
129 - fix for a "cygwin-64" architecture added, proposed by Stefan Traby.
130 - fix for threaded perl.
131
1320.9 Sun Nov 23 04:32:39 CET 2003
4 - fix a few very hard to track down but (of course) stupid 133 - fix a few very hard to track down but (of course) stupid
5 memory leaks. 134 memory leaks.
6 - stringify version number to avoid locale problems :(. 135 - stringify version number to avoid locale problems :(.
136 - perl 5.9.0 does not have SvPADBUSY, reported by Scott Walters,
137 hopefully ignoring it "just works".
7 138
80.8 Wed Nov 5 19:38:40 CET 2003 1390.8 Wed Nov 5 19:38:40 CET 2003
9 - port to perl5.005_03, but only 5.8.x+ is supported! 140 - port to perl5.005_03, but only 5.8.x+ is supported!
10 - honor LocalAddr even without LocalPort in Coro::Socket. 141 - honor LocalAddr even without LocalPort in Coro::Socket.
11 142

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines