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

Comparing Coro/Changes (file contents):
Revision 1.107 by root, Sat Aug 20 01:09:19 2005 UTC vs.
Revision 1.171 by root, Sun Dec 3 21:59:52 2006 UTC

1Revision history for Perl extension Coro. 1Revision history for Perl extension Coro.
2 2
3TODO: transfer flags must be per state
4
53.0 Sun Dec 3 22:57:25 CET 2006
6 - the "FINALLY COMPLETELY STABLE" release (coming soon:
7 the "FAMOUS LAST WORDS" release).
8 - implement a new stack sharing algorithm, which uses a stack
9 pool (size currently hardcoded).
10 - make stack sharing mandatory (it no longer uses a heuristic).
11 - eval/die no longer cause weird problems under heavy use.
12 - Coro::Event could cause livelocks if it was used but
13 no Coro::Event watchers were used.
14 - Coro::Event now uses asynccheck as prepare does not
15 check for changed watchers.
16 - Coro::Event allows multiple waiting coros and will wake up one
17 per event.
18 - Coro::Event should be cleaner and more efficient now.
19 - new utility function Coro::unblock_sub.
20 - document the sad fact that Event is no longer reentrant.
21 - putting a coroutine into the ready queue twice could under
22 some circumstances lead to stack corruption.
23 - minor incompatibility: subclassing Coro::State is supported
24 directly now without going through a _coro_state member.
25 - state/coro switching is much faster now.
26 - very minor optimisations and code/documentation cleanup.
27 - avoid problems due to compiler inlining.
28 - removed timers from Coro::Timer -> use AnyEvent instead.
29 - replaced Coro::idle coroutine by (cleaner) idle handler.
30 - updated to newest libcoro.
31 - implement enhanced support for valgrind.
32 - implement is_ready and return value for ready.
33 - removed Coro::Cont, it was a misnomer (it's generators),
34 and it was somewhat annoying to get it right. Will come back
35 if somebody asks for it :)
36 - many ->wait methods and Coro::Event could return
37 spuriously without the event having happened.
38
392.5 Tue Nov 7 12:22:33 CET 2006
40 - made Coro::Util, Coro::Select, Coro::Handle and Coro::Socket
41 use AnyEvent, moved them to Coro/.
42 - added Coro::LWP which contains all the uglyness required to
43 make LWP non-blocking.
44 - should work with perl 5.9.x now (Andreas König made me do it).
45 - fixed another bug in Coro::Select when the timeout was undef.
46 - reuse PL_start_env for all coros, saves some memory per coroutine.
47 - manage PL_top_env differently, hopefully to avoid panic: top_env.
48 - timeout argument was not properly used in Coro::Socket.
49 - allow limited forms of subclassing in Coro::Handle/Coro::Socket.
50 - emulate undocumented(!) functionality of IO::Socket required
51 by LWP(!!).
52 - updated eg/lwp to work with newer lwp's.
53 - remove "FATAL: uncaught exception" prefix. Coroutines that die
54 kill the whole process, just as exceptions in the main "coroutine"
55 did already.
56
572.1 Wed Nov 1 23:01:13 CET 2006
58 - fix a long-standing bug in Coro::Select where select with
59 zero timeout would instead change the current default filehandle.
60 - use a simpler and hopefully more robust way to clone padlists
61 (uses less memory and a perl function instead of our own).
62 - coro can now create a stack guard area on many architectures.
63 - Coro::AIO properly reexports additional functions from IO::AIO.
64 - updated libcoro with a workaround for OS X,
65 pach and testing by Michael Schwern.
66
672.0 Tue Oct 24 05:47:17 CEST 2006
68 - support additional aio requests in Coro::AIO.
69
701.9 (never properly released due to a glitch)
71
721.8 Thu Feb 2 00:59:06 CET 2006
73 - applied suggested patch by SAMV to avoid problems during stupid
74 mark & sweep gc run.
75 - applied patch by Scott Walters for 5.9.3 compatibility.
76
771.7 Tue Dec 27 01:41:58 CET 2005
78 - added Coro::AIO, a thin wrapper around IO::AIO.
79 - improved Makefile.PL explanations.
80
811.6 Mon Dec 12 21:30:05 CET 2005
82 - additionally save PM_curpm might fix as-of-yet
83 unseen problems with regex matches being attributed to the
84 wrong package.
85 - add t/10_bugs.t, which currently checks against imho broken
86 perls that use 0.26MB of stack space per Perl_magic_get
87 invocation (newer linuxes) as opposed to the 0.0002MB perl
88 normally uses.
89 - make stacksize configurable for the ultimate debian experience.
90
911.51 Mon Dec 12 18:48:36 CET 2005
92 - remove debugging warn() accidentally left in Coro::Select.
93
941.5 Tue Nov 29 13:32:44 CET 2005
95 - use Coro::Event inside Coro::Select to avoid spurious deadlocks.
96 - fix Coro::Select 'select'.
97 - strict'ify some modules.
98 - libcoro errornously restored the SIGUSR2 handler to SIGUSR1.
99 - use XSLoader in selected modules.
100 - remove some 5.6 compatibility cruft.
101
1021.4 Tue Sep 6 00:11:05 CEST 2005
103 - libcoro did not take into account the trampoline on amd64,
104 when 'l'inux method was used.
105
1061.31 Tue Aug 30 23:31:33 CEST 2005
107 - some portability fixes/workarounds.
108
31.3 Sat Aug 20 03:08:56 CEST 2005 1091.3 Sat Aug 20 03:08:56 CEST 2005
110 - no code changes. module seems to work fine.
4 - improve Coro::State docs, remove reference to nonexisting 111 - improve Coro::State docs, remove reference to nonexisting
5 Coro::State::flush method. 112 Coro::State::flush method.
6 - no longer autodetect windows, present it as an option instead. 113 - no longer autodetect windows, present it as an option instead.
7 114
81.2 Mon May 16 02:00:55 CEST 2005 1151.2 Mon May 16 02:00:55 CEST 2005
151.11 Thu Mar 3 18:00:52 CET 2005 1221.11 Thu Mar 3 18:00:52 CET 2005
16 - change of contact address. 123 - change of contact address.
17 124
181.1 Tue Feb 22 20:51:16 CET 2005 1251.1 Tue Feb 22 20:51:16 CET 2005
19 - support [l]inux method on amd64. 126 - support [l]inux method on amd64.
20 - make some context switches while compiling/eval'ing work. 127 - allow some context switches while compiling/eval'ing.
21 128
221.0 Tue Aug 10 03:47:27 CEST 2004 1291.0 Tue Aug 10 03:47:27 CEST 2004
23 - Event 0.89 fixes the bug again. 130 - Event 0.89 fixes the bug again.
24 - fix the remaining(??) memleak. I didn't free the cache-AV for 131 - fix the remaining(??) memleak. I didn't free the cache-AV for
25 padlists, so one AV per created anon-sub leaked. 132 padlists, so one AV per created anon-sub leaked.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines