| 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: move some other examples to common idioms? combining watchers, |
3 | TODO: ev_loop_prefork hint? |
| 4 | thread usage, coroutine switch? |
4 | TODO: ev_loop_wakeup |
|
|
5 | TODO: confusion about threads ongiong |
|
|
6 | TODO: not_blocked flag. |
| 5 | |
7 | |
| 6 | 4.03 |
8 | TODO: ev_sleep suffers from different backend_fudge... |
|
|
9 | - correct backend_fudge for most backends, and implement a windows |
|
|
10 | specific workaround to avoid looping because we call both |
|
|
11 | select and Sleep, both with different time resolutions. |
|
|
12 | - document range and guarantees of ev_sleep. |
|
|
13 | - change the default periodic reschedule function to hopefully be more |
|
|
14 | exact and correct even in corner cases or in the far future. |
|
|
15 | - document reasonable ranges for periodics interval and offset. |
|
|
16 | - do not rely on -lm anymore: use it when available but use our |
|
|
17 | own floor () if it is missing. This should make it easier to embed, |
|
|
18 | as no external libraries are required. |
|
|
19 | |
|
|
20 | 4.04 Wed Feb 16 09:01:51 CET 2011 |
|
|
21 | - fix two problems in the native win32 backend, where reuse of fd's |
|
|
22 | with different underlying handles caused handles not to be removed |
|
|
23 | or added to the select set (analyzed and tested by Bert Belder). |
|
|
24 | - do no rely on ceil() in ev_e?poll.c. |
|
|
25 | - backport libev to HP-UX versions before 11 v3. |
|
|
26 | - configure did not detect nanosleep and clock_gettime properly when |
|
|
27 | they are available in the libc (as opposed to -lrt). |
|
|
28 | |
|
|
29 | 4.03 Tue Jan 11 14:37:25 CET 2011 |
|
|
30 | - officially support polling files with all backends. |
|
|
31 | - support files, /dev/zero etc. the same way as select in the epoll |
|
|
32 | backend, by generating events on our own. |
|
|
33 | - ports backend: work around solaris bug 6874410 and many related ones |
|
|
34 | (EINTR, maybe more), with no performance loss (note that the solaris |
|
|
35 | bug report is actually wrong, reality is far more bizarre and broken |
|
|
36 | than that). |
| 7 | - define EV_READ/EV_WRITE as macros in event.h, as some programs use |
37 | - define EV_READ/EV_WRITE as macros in event.h, as some programs use |
| 8 | #ifdef to test for them. |
38 | #ifdef to test for them. |
|
|
39 | - new (experimental) function: ev_feed_signal. |
|
|
40 | - new (to become default) EVFLAG_NOSIGMASK flag. |
|
|
41 | - new EVBACKEND_MASK symbol. |
|
|
42 | - updated COMMON IDIOMS SECTION. |
| 9 | |
43 | |
| 10 | 4.01 Fri Nov 5 21:51:29 CET 2010 |
44 | 4.01 Fri Nov 5 21:51:29 CET 2010 |
| 11 | - automake fucked it up, apparently, --add-missing -f is not quite enough |
45 | - automake fucked it up, apparently, --add-missing -f is not quite enough |
| 12 | to make it update its files, so 4.00 didn't install ev++.h and |
46 | to make it update its files, so 4.00 didn't install ev++.h and |
| 13 | event.h on make install. grrr. |
47 | event.h on make install. grrr. |
