| 1 | Revision history for libev, a high-performance and full-featured event loop. |
1 | Revision history for libev, a high-performance and full-featured event loop. |
| 2 | |
2 | |
|
|
3 | TODO: ev_time_freeze, ev_time_thaw |
|
|
4 | TODO: ev_walk |
|
|
5 | TODO: ev_stop_all |
|
|
6 | TODO: fix signal handling(?) under win32 |
|
|
7 | - correctly zero the idx values for arrays, so destroying and |
|
|
8 | reinitialising the default loop actually works (patch by |
|
|
9 | Malek Hadj-Ali). |
|
|
10 | |
|
|
11 | 3.53 Sun Feb 15 02:38:20 CET 2009 |
|
|
12 | - fix a bug in event pipe creation on win32 that would cause a |
|
|
13 | failed assertion on event loop creation (patch by Malek Hadj-Ali). |
|
|
14 | - probe for CLOCK_REALTIME support at runtime as well and fall |
|
|
15 | back to gettimeofday if there is an error, to support older |
|
|
16 | operating systems with newer header files/libraries. |
|
|
17 | - prefer gettimeofday over clock_gettime with USE_CLOCK_SYSCALL |
|
|
18 | (default most everywhere), otherwise not. |
|
|
19 | |
|
|
20 | 3.52 Wed Jan 7 21:43:02 CET 2009 |
|
|
21 | - fix compilation of select backend in fd_set mode when NFDBITS is |
|
|
22 | missing (to get it to compile on QNX, reported by Rodrigo Campos). |
|
|
23 | - better select-nfds handling when select backend is in fd_set mode. |
|
|
24 | - diagnose fd_set overruns when select backend is in fd_set mode. |
|
|
25 | - due to a thinko, instead of disabling everything but |
|
|
26 | select on the borked OS X platform, everything but select was |
|
|
27 | allowed (reported by Emanuele Giaquinta). |
| 3 | - actually verify that local and remote port are matching in |
28 | - actually verify that local and remote port are matching in |
| 4 | libev's socketpair emulation, which makes denial-of-service |
29 | libev's socketpair emulation, which makes denial-of-service |
| 5 | attacks harder (but not impossible - it's windows). Make sure |
30 | attacks harder (but not impossible - it's windows). Make sure |
| 6 | it even works under vista, which thinks that getpeer/sockname |
31 | it even works under vista, which thinks that getpeer/sockname |
| 7 | should return fantasy port numbers. |
32 | should return fantasy port numbers. |
|
|
33 | - include "libev" in all assertion messages for potentially |
|
|
34 | clearer diagnostics. |
|
|
35 | - event_get_version (libevent compatibility) returned |
|
|
36 | a useless string instead of the expected version string |
|
|
37 | (patch by W.C.A. Wijngaards). |
| 8 | |
38 | |
| 9 | 3.51 Wed Dec 24 23:00:11 CET 2008 |
39 | 3.51 Wed Dec 24 23:00:11 CET 2008 |
| 10 | - fix a bug where an inotify watcher was added twice, causing |
40 | - fix a bug where an inotify watcher was added twice, causing |
| 11 | freezes on hash collisions (reported and analysed by Graham Leggett). |
41 | freezes on hash collisions (reported and analysed by Graham Leggett). |
| 12 | - new config symbol, EV_USE_CLOCK_SYSCALL, to make libev use |
42 | - new config symbol, EV_USE_CLOCK_SYSCALL, to make libev use |
