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

Comparing EV/Changes (file contents):
Revision 1.159 by root, Thu Nov 20 00:35:16 2008 UTC vs.
Revision 1.180 by root, Sun Jul 19 20:39:54 2009 UTC

1Revision history for Perl extension EV 1Revision history for Perl extension EV
2 2
3 - implement $loop->signal/signal_ns.
4
53.7 Fri Jul 17 16:49:16 CEST 2009
6 - add EV::loop_verify and EV::loop_depth.
7 - use output filehandle for i/o watchers waiting
8 for EV_WRITE, otherwise input filehandle.
9 - use common schmorp.h header.
10 - add EV::Timer->remaining.
11 - allow for subclassing of EV::Loop (see the exciting
12 EV::Loop::Async module for an example).
13 - added EV::invoke_pending and EV::pending_count.
14
15 - (libev) ev_unloop and ev_loop wrongly used a global variable to
16 exit loops, instead of using a per-loop variable.
17 - (libev) the ev_set_io_collect_interval interpretation has changed.
18 - add new functionality: EV::invoke_pending, EV::pending_count.
19 - add $timer->remaining.
20 - add EV::loop_depth.
21 - (libev) calling unloop in fork/prepare watchers will no longer poll
22 for new events.
23 - (libev) use GetSystemTimeAsFileTime instead of _timeb on windows,
24 for slightly higher accuracy.
25 - (libev) actually 0-initialise struct sigaction when installing
26 signals.
27
283.6 Tue Apr 28 02:50:37 CEST 2009
29 - keepalive(0) could decrease the refcount of the loop
30 permanently.
31 - add ev_suspend/ev_resume, also make ev_now_update
32 accessible via the XS API.
33 - most EV::<eventtype> constants were missing and have been
34 added :/.
35 - add EV::VERSION_MAJOR/VERSION_MINOR constants.
36 - (libev) multiple timers becoming ready within an event loop
37 iteration will be invoked in the "correct" order now.
38 - (libev) do not leave the event loop early just because we have
39 no active watchers, fixing a problem when embedding a kqueue
40 loop that has active kernel events but no registered watchers
41 (reported by blacksand blacksand).
42 - (libev) correctly zero the idx values for arrays, so destroying
43 and reinitialising the default loop actually works (patch by
44 Malek Hadj-Ali).
45 - (libev) new EV::CUSTOM revents flag for use by applications.
46 - (libev) add documentation section about priorites.
47 - (libev) add a glossary to the dcoumentation.
48 - (libev) extend the ev_fork description slightly.
49 - (libev) optimize a jump out of call_pending.
50 - t/03_keepalive could fail when there was no actual error.
51
523.53 Sun Feb 15 02:38:20 CET 2009
53 - (libev) on win32, the event loop creation could randomly fail
54 due to an initialised variable having the wrong value.
55 - (libev) probe for CLOCK_REALTIME support at runtime as well and
56 fall back to gettimeofday if there is an error, to support older
57 operating systems with newer header files/libraries.
58 - prefer gettimeofday over clock_gettime by default.
59
603.52 Wed Jan 7 21:46:14 CET 2009
61 - (libev) fix some issues in the select backend when in fd_set mode.
62 - (libev) due to a thinko, instead of disabling everything but
63 select on the borked OS X platform, everything but select was
64 allowed (reported by Emanuele Giaquinta).
65 - (libev) actually verify that local and remote port are matching in
66 libev's socketpair emulation, which makes denial-of-service
67 attacks harder (but not impossible - it's windows). Make sure
68 it even works under vista, which thinks that getpeer/sockname
69 should return fantasy port numbers.
70
713.51 Wed Dec 24 23:01:59 CET 2008
72 - do not cache the arguments passed to callbacks if the refcount
73 indicates that the callback has stolen them.
3 - (libev) try to avoid librt on GNU/Linux. 74 - (libev) try to avoid librt on GNU/Linux.
75 - (libev) check port_getn return value dfferently, might potentially
76 avoid problems.
77 - (libev) fix a bug with stat watchers possibly causing freezes.
78 - (libev) work around OS X 10.5 breaking poll, now select is the
79 only viable choice left on that pile of garbage.
80 - play tester whore: disable some tests that typically fail only
81 on cpan tester machines.
4 82
53.49 Wed Nov 19 11:26:53 CET 2008 833.49 Wed Nov 19 11:26:53 CET 2008
6 - fix typos in manpage (Alex Efros). 84 - fix typos in manpage (Alex Efros).
7 - increase timing even further, for the ever-overloaded cpan-tester 85 - increase timing even further, for the ever-overloaded cpan-tester
8 machines, and to the detriment of everybody else who wants a fast 86 machines, and to the detriment of everybody else who wants a fast

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines