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

Comparing Coro/Changes (file contents):
Revision 1.88 by pcg, Sat Feb 14 12:39:41 2004 UTC vs.
Revision 1.156 by root, Sun Nov 26 03:22:50 2006 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
30.95 Fri Dec 12 20:50:57 CET 2003 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 - reimplement a new stack sharing algorithm.
8 - make stack sharing mandatory (it no longer uses a heuristic).
9 - Coro::Event could cause livelocks if it was used but
10 Event watchers were used directly.
11 - Coro::Event should be cleaner and more efficient now.
12 - minor incompatibility: subclassing is supported directly now
13 without going through a _coro_state member.
14 - state/coro switching is much faster now.
15 - very minor optimisations and code/documentation cleanup.
16 - avoid problems due to compiler inlining.
17 - removed timers from Coro::Timer -> use AnyEvent instead.
18 - replaced Coro::idle coroutine by (cleaner) idle handler.
19 - updated to newest libcoro.
20
212.5 Tue Nov 7 12:22:33 CET 2006
22 - made Coro::Util, Coro::Select, Coro::Handle and Coro::Socket
23 use AnyEvent, moved them to Coro/.
24 - added Coro::LWP which contains all the uglyness required to
25 make LWP non-blocking.
26 - should work with perl 5.9.x now (Andreas König made me do it).
27 - fixed another bug in Coro::Select when the timeout was undef.
28 - reuse PL_start_env for all coros, saves some memory per coroutine.
29 - manage PL_top_env differently, hopefully to avoid panic: top_env.
30 - timeout argument was not properly used in Coro::Socket.
31 - allow limited forms of subclassing in Coro::Handle/Coro::Socket.
32 - emulate undocumented(!) functionality of IO::Socket required
33 by LWP(!!).
34 - updated eg/lwp to work with newer lwp's.
35 - remove "FATAL: uncaught exception" prefix. Coroutines that die
36 kill the whole process, just as exceptions in the main "coroutine"
37 did already.
38
392.1 Wed Nov 1 23:01:13 CET 2006
40 - fix a long-standing bug in Coro::Select where select with
41 zero timeout would instead change the current default filehandle.
42 - use a simpler and hopefully more robust way to clone padlists
43 (uses less memory and a perl function instead of our own).
44 - coro can now create a stack guard area on many architectures.
45 - Coro::AIO properly reexports additional functions from IO::AIO.
46 - updated libcoro with a workaround for OS X,
47 pach and testing by Michael Schwern.
48
492.0 Tue Oct 24 05:47:17 CEST 2006
50 - support additional aio requests in Coro::AIO.
51
521.9 (never properly released due to a glitch)
53
541.8 Thu Feb 2 00:59:06 CET 2006
55 - applied suggested patch by SAMV to avoid problems during stupid
56 mark & sweep gc run.
57 - applied patch by Scott Walters for 5.9.3 compatibility.
58
591.7 Tue Dec 27 01:41:58 CET 2005
60 - added Coro::AIO, a thin wrapper around IO::AIO.
61 - improved Makefile.PL explanations.
62
631.6 Mon Dec 12 21:30:05 CET 2005
64 - additionally save PM_curpm might fix as-of-yet
65 unseen problems with regex matches being attributed to the
66 wrong package.
67 - add t/10_bugs.t, which currently checks against imho broken
68 perls that use 0.26MB of stack space per Perl_magic_get
69 invocation (newer linuxes) as opposed to the 0.0002MB perl
70 normally uses.
71 - make stacksize configurable for the ultimate debian experience.
72
731.51 Mon Dec 12 18:48:36 CET 2005
74 - remove debugging warn() accidentally left in Coro::Select.
75
761.5 Tue Nov 29 13:32:44 CET 2005
77 - use Coro::Event inside Coro::Select to avoid spurious deadlocks.
78 - fix Coro::Select 'select'.
79 - strict'ify some modules.
80 - libcoro errornously restored the SIGUSR2 handler to SIGUSR1.
81 - use XSLoader in selected modules.
82 - remove some 5.6 compatibility cruft.
83
841.4 Tue Sep 6 00:11:05 CEST 2005
85 - libcoro did not take into account the trampoline on amd64,
86 when 'l'inux method was used.
87
881.31 Tue Aug 30 23:31:33 CEST 2005
89 - some portability fixes/workarounds.
90
911.3 Sat Aug 20 03:08:56 CEST 2005
92 - no code changes. module seems to work fine.
93 - improve Coro::State docs, remove reference to nonexisting
94 Coro::State::flush method.
95 - no longer autodetect windows, present it as an option instead.
96
971.2 Mon May 16 02:00:55 CEST 2005
98 - included libcoro.c earlier in State.xs, might improve portability.
99 - use a faster and possibly more stable Coro::Cont implementation.
100 - accept x86_64 in addition to amd64 for optimized linux-amd64 support.
101 - fix bugs Coro::SemaphoreSet that could cause locks never to be freed.
102 - fix bugs in CoroAPI.h and document it in Coro::MakeMaker.
103
1041.11 Thu Mar 3 18:00:52 CET 2005
105 - change of contact address.
106
1071.1 Tue Feb 22 20:51:16 CET 2005
108 - support [l]inux method on amd64.
109 - allow some context switches while compiling/eval'ing.
110
1111.0 Tue Aug 10 03:47:27 CEST 2004
112 - Event 0.89 fixes the bug again.
113 - fix the remaining(??) memleak. I didn't free the cache-AV for
114 padlists, so one AV per created anon-sub leaked.
115
1160.97 Fri May 14 15:23:32 CEST 2004
117 - remove stupid gcc extension that was an experiment only anyway.
118
1190.96 Thu May 13 18:09:29 CEST 2004
120 - tell people if they have yet another broken Event variant.
121 - made Coro to work _better_ with threads (it seems that, as long
122 as you use Coros in one thread only everything will work fine.
123 Looking at the perl source the problem might be missing locking
124 between Coro and some perl internal routines. That's not easily
125 fixable).
126 - (hopefully) got rid of the static xnext inside transfer.
127
1280.95 Sun Feb 15 03:20:28 CET 2004
4 - removed Coro::State::flush, implemented a new and much 129 - removed Coro::State::flush, implemented a new and much
5 faster caching mechanism by attaching magic to the CV. 130 faster caching mechanism by attaching magic to the CV.
6 - a memleak with (real) closures remains. 131 - a memleak with (real) closures remains.
7 - dramatically reduced size of stacks to 65536 (linux/x86), after 132 - dramatically reduced size of stacks to 65536 (linux/x86), after
8 observing that even myhttpd never had more than a kilobyte 133 observing that even myhttpd never had more than a kilobyte
9 of stack in use (stack is still only physically allocated in 134 of stack in use (stack is still only physically allocated in
10 pagesize increments for systems suporting mmap). 135 pagesize increments for systems suporting mmap).
136 - fix for a "cygwin-64" architecture added, proposed by Stefan Traby.
137 - fix for threaded perl.
11 138
120.9 Sun Nov 23 04:32:39 CET 2003 1390.9 Sun Nov 23 04:32:39 CET 2003
13 - fix a few very hard to track down but (of course) stupid 140 - fix a few very hard to track down but (of course) stupid
14 memory leaks. 141 memory leaks.
15 - stringify version number to avoid locale problems :(. 142 - stringify version number to avoid locale problems :(.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines