| … | |
… | |
| 5 | TODO: reduce condvar fairness: schedule hot-cache-threads first? |
5 | TODO: reduce condvar fairness: schedule hot-cache-threads first? |
| 6 | TODO: splice/tee/vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) |
6 | TODO: splice/tee/vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) |
| 7 | TODO: aio_fcntl, at least for file-locking |
7 | TODO: aio_fcntl, at least for file-locking |
| 8 | TODO: fallocate, if it wouldn't be so useless. |
8 | TODO: fallocate, if it wouldn't be so useless. |
| 9 | |
9 | |
|
|
10 | 3.31 Thu Nov 12 02:14:29 CET 2009 |
|
|
11 | - fix result status documentation of aio_copy, aio_move. |
|
|
12 | - speed up object creation considerably by avoiding |
|
|
13 | hash lookups. |
|
|
14 | |
|
|
15 | 3.3 Wed Aug 5 13:52:58 CEST 2009 |
| 10 | - use common::sense. |
16 | - use common::sense. |
| 11 | - use common schmorp.h header. |
17 | - use common schmorp.h header. |
| 12 | - allow integers (file descriptors) in addition to file handles. |
18 | - allow integers (file descriptors) in addition to file handles. |
| 13 | - take advantage of linux' eventfd (but it seems slower on smp?). |
19 | - take advantage of linux' eventfd (but it seems slower on smp?). |
| 14 | - use poll not seelct on posix platforms. |
20 | - use poll not select on posix platforms. |
| 15 | |
21 | |
| 16 | 3.261 Wed Jul 1 10:11:51 CEST 2009 |
22 | 3.261 Wed Jul 1 10:11:51 CEST 2009 |
| 17 | - more OS X workarounds (patch by Tokuhiro Matsuno). |
23 | - more 0S X workarounds (patch by Tokuhiro Matsuno). |
| 18 | |
24 | |
| 19 | 3.26 Tue Jun 30 09:33:26 CEST 2009 |
25 | 3.26 Tue Jun 30 09:33:26 CEST 2009 |
| 20 | - os x of course claims posix 2008 but lacks posix_fadvise. try to work |
26 | - 0S X of course claims to be posix 2008 but lacks posix_fadvise. |
| 21 | around this horribly broken OS in a somewhat hackish way. might help |
27 | try to work around this horribly broken OS in a somewhat hackish |
| 22 | other os's too. |
28 | way. might help other os's too. |
| 23 | |
29 | |
| 24 | 3.25 Sat Jun 27 05:18:26 CEST 2009 |
30 | 3.25 Sat Jun 27 05:18:26 CEST 2009 |
| 25 | - added IO::AIO::fadvise and IO::AIO::sendfile. |
31 | - added IO::AIO::fadvise and IO::AIO::sendfile. |
| 26 | - (libeio) replaced quicksort+insertion sort by a tuned radix |
32 | - (libeio) replaced quicksort+insertion sort by a tuned radix |
| 27 | sort + insertion sort, resulting in comparable runtime (usually |
33 | sort + insertion sort, resulting in comparable runtime (usually |
