| 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: ABI??? API????? Changes??? |
3 | TODO: ev_loop_wakeup |
| 4 | TODO: win32 write() to socket for signal handling |
4 | |
|
|
5 | TODO: EV_NO_THREADS/NO_SMP and/or EV_USE_PTHREADS or so. also document. |
|
|
6 | TODO: #define EV_API_DECL static? ev_default_loop_ptr? |
|
|
7 | TODO: EV_STANDALONE == NO_HASSEL (do not use clock_gettime in ev_standalone) |
|
|
8 | TODO: memory fences for clang |
|
|
9 | TODO: fix repeat=0 in http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#Be_smart_about_timeouts |
|
|
10 | 4.10 |
|
|
11 | - fix a race where the workaround against the epoll fork bugs |
|
|
12 | caused signals to not be handled anymore. |
|
|
13 | - correct backend_fudge for most backends, and implement a windows |
|
|
14 | specific workaround to avoid looping because we call both |
|
|
15 | select and Sleep, both with different time resolutions. |
|
|
16 | - document range and guarantees of ev_sleep. |
|
|
17 | - document reasonable ranges for periodics interval and offset. |
|
|
18 | - rename backend_fudge to backend_mintime to avoid future confusion :) |
|
|
19 | - change the default periodic reschedule function to hopefully be more |
|
|
20 | exact and correct even in corner cases or in the far future. |
|
|
21 | - do not rely on -lm anymore: use it when available but use our |
|
|
22 | own floor () if it is missing. This should make it easier to embed, |
|
|
23 | as no external libraries are required. |
|
|
24 | - strategically import macros from libecb and mark rarely-used functions |
|
|
25 | as cache-cold (saving almost 2k code size on typical amd64 setups). |
|
|
26 | - add Symbols.ev and Symbols.event files, that were missing. |
|
|
27 | - fix backend_mintime value for epoll (was 1/1024, is 1/1000 now). |
|
|
28 | |
|
|
29 | 4.04 Wed Feb 16 09:01:51 CET 2011 |
|
|
30 | - fix two problems in the native win32 backend, where reuse of fd's |
|
|
31 | with different underlying handles caused handles not to be removed |
|
|
32 | or added to the select set (analyzed and tested by Bert Belder). |
|
|
33 | - do no rely on ceil() in ev_e?poll.c. |
|
|
34 | - backport libev to HP-UX versions before 11 v3. |
|
|
35 | - configure did not detect nanosleep and clock_gettime properly when |
|
|
36 | they are available in the libc (as opposed to -lrt). |
|
|
37 | |
|
|
38 | 4.03 Tue Jan 11 14:37:25 CET 2011 |
|
|
39 | - officially support polling files with all backends. |
|
|
40 | - support files, /dev/zero etc. the same way as select in the epoll |
|
|
41 | backend, by generating events on our own. |
|
|
42 | - ports backend: work around solaris bug 6874410 and many related ones |
|
|
43 | (EINTR, maybe more), with no performance loss (note that the solaris |
|
|
44 | bug report is actually wrong, reality is far more bizarre and broken |
|
|
45 | than that). |
|
|
46 | - define EV_READ/EV_WRITE as macros in event.h, as some programs use |
|
|
47 | #ifdef to test for them. |
|
|
48 | - new (experimental) function: ev_feed_signal. |
|
|
49 | - new (to become default) EVFLAG_NOSIGMASK flag. |
|
|
50 | - new EVBACKEND_MASK symbol. |
|
|
51 | - updated COMMON IDIOMS SECTION. |
|
|
52 | |
|
|
53 | 4.01 Fri Nov 5 21:51:29 CET 2010 |
|
|
54 | - automake fucked it up, apparently, --add-missing -f is not quite enough |
|
|
55 | to make it update its files, so 4.00 didn't install ev++.h and |
|
|
56 | event.h on make install. grrr. |
|
|
57 | - ev_loop(count|depth) didn't return anything (Robin Haberkorn). |
|
|
58 | - change EV_UNDEF to 0xffffffff to silence some overzealous compilers. |
|
|
59 | - use "(libev) " prefix for all libev error messages now. |
|
|
60 | |
|
|
61 | 4.00 Mon Oct 25 12:32:12 CEST 2010 |
| 5 | - "PORTING FROM LIBEV 3.X TO 4.X" (in ev.pod) is recommended reading. |
62 | - "PORTING FROM LIBEV 3.X TO 4.X" (in ev.pod) is recommended reading. |
| 6 | - ev_embed_stop did not correctly stop the watcher (very good |
63 | - ev_embed_stop did not correctly stop the watcher (very good |
| 7 | testcase by Vladimir Timofeev). |
64 | testcase by Vladimir Timofeev). |
|
|
65 | - ev_run will now always update the current loop time - it erroneously |
|
|
66 | didn't when idle watchers were active, causing timers not to fire. |
|
|
67 | - fix a bug where a timeout of zero caused the timer not to fire |
|
|
68 | in the libevent emulation (testcase by Péter Szabó). |
| 8 | - applied win32 fixes by Michael Lenaghan (also James Mansion). |
69 | - applied win32 fixes by Michael Lenaghan (also James Mansion). |
| 9 | - replace EV_MINIMAL by EV_FEATURES. |
70 | - replace EV_MINIMAL by EV_FEATURES. |
|
|
71 | - prefer EPOLL_CTL_ADD over EPOLL_CTL_MOD in some more cases, as it |
|
|
72 | seems the former is *much* faster than the latter. |
|
|
73 | - linux kernel version detection (for inotify bug workarounds) |
|
|
74 | did not work properly. |
|
|
75 | - reduce the number of spurious wake-ups with the ports backend. |
| 10 | - remove dependency on sys/queue.h on freebsd (patch by Vanilla Hsu). |
76 | - remove dependency on sys/queue.h on freebsd (patch by Vanilla Hsu). |
|
|
77 | - do async init within ev_async_start, not ev_async_set, which avoids |
|
|
78 | an API quirk where the set function must be called in the C++ API |
|
|
79 | even when there is nothing to set. |
| 11 | - add (undocumented) EV_ENABLE when adding events with kqueue, |
80 | - add (undocumented) EV_ENABLE when adding events with kqueue, |
|
|
81 | this might help with OS X, which seems to need it despite documenting |
| 12 | this might help (helpfully pointed out by Tilghman Lesher). |
82 | not to need it (helpfully pointed out by Tilghman Lesher). |
|
|
83 | - do not use poll by default on freebsd, it's broken (what isn't |
|
|
84 | on freebsd...). |
|
|
85 | - allow to embed epoll on kernels >= 2.6.32. |
| 13 | - configure now prepends -O3, not appends it, so one can still |
86 | - configure now prepends -O3, not appends it, so one can still |
| 14 | override it. |
87 | override it. |
|
|
88 | - ev.pod: greatly expanded the portability section, added a porting |
|
|
89 | section, a description of watcher states and made lots of minor fixes. |
| 15 | - disable poll backend on AIX, the poll header spams the namespace |
90 | - disable poll backend on AIX, the poll header spams the namespace |
| 16 | and it's not worth working around dead platforms (reported |
91 | and it's not worth working around dead platforms (reported |
| 17 | and analyzed by Aivars Kalvans). |
92 | and analyzed by Aivars Kalvans). |
| 18 | - improve header file compatibility of the standalone eventfd code |
93 | - improve header file compatibility of the standalone eventfd code |
| 19 | in an obscure case. |
94 | in an obscure case. |
| … | |
… | |
| 22 | no sscanf dependency). |
97 | no sscanf dependency). |
| 23 | - new EV_CHILD_ENABLE and EV_SIGNAL_ENABLE configurable settings. |
98 | - new EV_CHILD_ENABLE and EV_SIGNAL_ENABLE configurable settings. |
| 24 | - update libev.m4 HAVE_CLOCK_SYSCALL test for newer glibcs. |
99 | - update libev.m4 HAVE_CLOCK_SYSCALL test for newer glibcs. |
| 25 | - add section on accept() problems to the manpage. |
100 | - add section on accept() problems to the manpage. |
| 26 | - rename EV_TIMEOUT to EV_TIMER. |
101 | - rename EV_TIMEOUT to EV_TIMER. |
| 27 | - rename ev_loop_count/depth/verify. |
102 | - rename ev_loop_count/depth/verify/loop/unloop. |
|
|
103 | - remove ev_default_destroy and ev_default_fork. |
| 28 | - switch to two-digit minor version. |
104 | - switch to two-digit minor version. |
|
|
105 | - work around an apparent gentoo compiler bug. |
|
|
106 | - define _DARWIN_UNLIMITED_SELECT. just so. |
|
|
107 | - use enum instead of #define for most constants. |
|
|
108 | - improve compatibility to older C++ compilers. |
|
|
109 | - (experimental) ev_run/ev_default_loop/ev_break/ev_loop_new have now |
|
|
110 | default arguments when compiled as C++. |
|
|
111 | - enable automake dependency tracking. |
|
|
112 | - ev_loop_new no longer leaks memory when loop creation failed. |
|
|
113 | - new ev_cleanup watcher type. |
| 29 | |
114 | |
| 30 | 3.9 Thu Dec 31 07:59:59 CET 2009 |
115 | 3.9 Thu Dec 31 07:59:59 CET 2009 |
| 31 | - signalfd is no longer used by default and has to be requested |
116 | - signalfd is no longer used by default and has to be requested |
| 32 | explicitly - this means that easy to catch bugs become hard to |
117 | explicitly - this means that easy to catch bugs become hard to |
| 33 | catch race conditions, but the users have spoken. |
118 | catch race conditions, but the users have spoken. |
| 34 | - point out the unspecified signal mask in the documentation, and |
119 | - point out the unspecified signal mask in the documentation, and |
| 35 | that this is a race condition regardless of EV_SIGNALFD. |
120 | that this is a race condition regardless of EV_SIGNALFD. |
| 36 | - backport inotify code to C89. |
121 | - backport inotify code to C89. |
| 37 | - inotify file descriptors could leak into child processes. |
122 | - inotify file descriptors could leak into child processes. |
| 38 | - ev_stat watchers could keep an errornous extra ref on the loop, |
123 | - ev_stat watchers could keep an erroneous extra ref on the loop, |
| 39 | preventing exit when unregistering all watchers (testcases |
124 | preventing exit when unregistering all watchers (testcases |
| 40 | provided by ry@tinyclouds.org). |
125 | provided by ry@tinyclouds.org). |
| 41 | - implement EV_WIN32_HANDLE_TO_FD and EV_WIN32_CLOSE_FD configuration |
126 | - implement EV_WIN32_HANDLE_TO_FD and EV_WIN32_CLOSE_FD configuration |
| 42 | symbols to make it easier for apps to do their own fd management. |
127 | symbols to make it easier for apps to do their own fd management. |
| 43 | - support EV_IDLE_ENABLE being disabled in ev++.h |
128 | - support EV_IDLE_ENABLE being disabled in ev++.h |
| … | |
… | |
| 49 | - changed minimum glibc requirement from glibc 2.9 to 2.7, for |
134 | - changed minimum glibc requirement from glibc 2.9 to 2.7, for |
| 50 | signalfd. |
135 | signalfd. |
| 51 | - add missing string.h include (Denis F. Latypoff). |
136 | - add missing string.h include (Denis F. Latypoff). |
| 52 | - only replace ev_stat.prev when we detect an actual difference, |
137 | - only replace ev_stat.prev when we detect an actual difference, |
| 53 | so prev is (almost) always different to attr. this might |
138 | so prev is (almost) always different to attr. this might |
| 54 | have caused the probems with 04_stat.t. |
139 | have caused the problems with 04_stat.t. |
| 55 | - add ev::timer->remaining () method to C++ API. |
140 | - add ev::timer->remaining () method to C++ API. |
| 56 | |
141 | |
| 57 | 3.8 Sun Aug 9 14:30:45 CEST 2009 |
142 | 3.8 Sun Aug 9 14:30:45 CEST 2009 |
| 58 | - incompatible change: do not necessarily reset signal handler |
143 | - incompatible change: do not necessarily reset signal handler |
| 59 | to SIG_DFL when a sighandler is stopped. |
144 | to SIG_DFL when a sighandler is stopped. |
| 60 | - ev_default_destroy did not properly free or zero some members, |
145 | - ev_default_destroy did not properly free or zero some members, |
| 61 | potentially causing crashes and memory corruption on repated |
146 | potentially causing crashes and memory corruption on repeated |
| 62 | ev_default_destroy/ev_default_loop calls. |
147 | ev_default_destroy/ev_default_loop calls. |
| 63 | - take advantage of signalfd on GNU/Linux systems. |
148 | - take advantage of signalfd on GNU/Linux systems. |
| 64 | - document that the signal mask might be in an unspecified |
149 | - document that the signal mask might be in an unspecified |
| 65 | state when using libev's signal handling. |
150 | state when using libev's signal handling. |
| 66 | - take advantage of some GNU/Linux calls to set cloexec/nonblock |
151 | - take advantage of some GNU/Linux calls to set cloexec/nonblock |
| … | |
… | |
| 101 | - correctly zero the idx values for arrays, so destroying and |
186 | - correctly zero the idx values for arrays, so destroying and |
| 102 | reinitialising the default loop actually works (patch by |
187 | reinitialising the default loop actually works (patch by |
| 103 | Malek Hadj-Ali). |
188 | Malek Hadj-Ali). |
| 104 | - implement ev_suspend and ev_resume. |
189 | - implement ev_suspend and ev_resume. |
| 105 | - new EV_CUSTOM revents flag for use by applications. |
190 | - new EV_CUSTOM revents flag for use by applications. |
| 106 | - add documentation section about priorites. |
191 | - add documentation section about priorities. |
| 107 | - add a glossary to the dcoumentation. |
192 | - add a glossary to the dcoumentation. |
| 108 | - extend the ev_fork description slightly. |
193 | - extend the ev_fork description slightly. |
| 109 | - optimize a jump out of call_pending. |
194 | - optimize a jump out of call_pending. |
| 110 | |
195 | |
| 111 | 3.53 Sun Feb 15 02:38:20 CET 2009 |
196 | 3.53 Sun Feb 15 02:38:20 CET 2009 |
| … | |
… | |
| 188 | do similar changes for event.c. |
273 | do similar changes for event.c. |
| 189 | |
274 | |
| 190 | 3.45 Tue Oct 21 21:59:26 CEST 2008 |
275 | 3.45 Tue Oct 21 21:59:26 CEST 2008 |
| 191 | - disable inotify usage on linux <2.6.25, as it is broken |
276 | - disable inotify usage on linux <2.6.25, as it is broken |
| 192 | (reported by Yoann Vandoorselaere). |
277 | (reported by Yoann Vandoorselaere). |
| 193 | - ev_stat errornously would try to add inotify watchers |
278 | - ev_stat erroneously would try to add inotify watchers |
| 194 | even when inotify wasn't available (this should only |
279 | even when inotify wasn't available (this should only |
| 195 | have a performance impact). |
280 | have a performance impact). |
| 196 | - ev_once now passes both timeout and io to the callback if both |
281 | - ev_once now passes both timeout and io to the callback if both |
| 197 | occur concurrently, instead of giving timeouts precedence. |
282 | occur concurrently, instead of giving timeouts precedence. |
| 198 | - disable EV_USE_INOTIFY when sys/inotify.h is too old. |
283 | - disable EV_USE_INOTIFY when sys/inotify.h is too old. |
| … | |
… | |
| 266 | - enable inlining with gcc versions 2 and 3. |
351 | - enable inlining with gcc versions 2 and 3. |
| 267 | - work around broken poll implementations potentially |
352 | - work around broken poll implementations potentially |
| 268 | not clearing revents field in ev_poll (Brandon Black) |
353 | not clearing revents field in ev_poll (Brandon Black) |
| 269 | (no such systems are known at this time). |
354 | (no such systems are known at this time). |
| 270 | - work around a bug in realloc on openbsd and darwin, |
355 | - work around a bug in realloc on openbsd and darwin, |
| 271 | also makes the errornous valgrind complaints |
356 | also makes the erroneous valgrind complaints |
| 272 | go away (noted by various people). |
357 | go away (noted by various people). |
| 273 | - fix ev_async_pending, add c++ wrapper for ev_async |
358 | - fix ev_async_pending, add c++ wrapper for ev_async |
| 274 | (based on patch sent by Johannes Deisenhofer. |
359 | (based on patch sent by Johannes Deisenhofer). |
| 275 | - add sensible set method to ev::embed. |
360 | - add sensible set method to ev::embed. |
| 276 | - made integer constants type int in ev.h. |
361 | - made integer constants type int in ev.h. |
| 277 | |
362 | |
| 278 | 3.2 Wed Apr 2 17:11:19 CEST 2008 |
363 | 3.2 Wed Apr 2 17:11:19 CEST 2008 |
| 279 | - fix a 64 bit overflow issue in the select backend, |
364 | - fix a 64 bit overflow issue in the select backend, |
