| 1 | Revision history for IO::AIO |
1 | Revision history for IO::AIO |
| 2 | |
2 | |
| 3 | TODO: better autoconf.pm that can return LIBS etc. |
3 | TODO: better autoconf.pm that can return LIBS etc. |
| 4 | TODO: aio_cptree/mvtree |
4 | TODO: aio_cptree/mvtree |
| 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/sync_file_range? (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 | |
8 | |
|
|
9 | - (libeio) pwrite emulation was even more flawed than intended and did |
|
|
10 | not restore the file offset. |
|
|
11 | - unfortunately, utime, chmod, chown on an open file that has just |
|
|
12 | been written can easily block, which caused aio_copy to block the |
|
|
13 | process. no more! |
|
|
14 | - no longer rely on dst path in aio_copy when futime is available. |
|
|
15 | |
|
|
16 | 3.19 Tue Apr 21 22:05:21 CEST 2009 |
|
|
17 | - more perl 5.10 workarounds for aio_read and write. |
|
|
18 | - aio_write no longer modifies the sv (if possible). |
|
|
19 | - aio_read now works correctly with magic values. |
|
|
20 | |
|
|
21 | 3.18 Sun Apr 19 21:17:32 CEST 2009 |
|
|
22 | - better diagnostics when some aio_* functions get passed an illegal fh. |
|
|
23 | - try to avoid crashes due to incompatible 5.10 API changes. grmbl. |
|
|
24 | |
|
|
25 | 3.17 Thu Nov 20 08:45:36 CET 2008 |
|
|
26 | - (libeio) added aio_sync_file_range (untested). |
|
|
27 | - add aio_busy to @IO::AIO::AIO_REQ. |
|
|
28 | |
|
|
29 | 3.16 Wed Oct 22 18:28:01 CEST 2008 |
|
|
30 | - use SvREFCNT_inc instead of SvREFCNT_inc_NN in a non-speed critical |
|
|
31 | part to improve portability to perl 5.8 (reported by szymon). |
|
|
32 | |
|
|
33 | 3.15 Mon Oct 13 00:39:55 CEST 2008 |
|
|
34 | - automatic removal of feeders was broken. |
|
|
35 | - (libeio) use a more robust method to detect whether a feeder |
|
|
36 | has added something to the group or not. |
|
|
37 | |
|
|
38 | 3.1 Thu Oct 2 13:34:40 CEST 2008 |
|
|
39 | - pre-deref the passed callback object, for increased speed |
|
|
40 | and decreased memory usage. |
| 9 | - call on_next_submit callback even fro aio_group - impact unknown, |
41 | - call on_next_submit callback even for aio_group - impact unknown, |
| 10 | but seems more correct. |
42 | but seems more correct. |
|
|
43 | - $req->cb now returns the original callback. |
| 11 | - (libeio) pass EIO_GROUP/aio_group requests through the queue without |
44 | - (libeio) pass EIO_GROUP/aio_group requests through the queue without |
| 12 | requiring a thread switch. |
45 | requiring a thread switch. |
|
|
46 | - (libeio) status code of aio_futime and aio_utime was always passed 0. |
| 13 | - do some other µ-optimisations. |
47 | - do some other µ-optimisations. |
| 14 | |
48 | |
| 15 | 3.07 Sat Aug 2 16:06:13 CEST 2008 |
49 | 3.07 Sat Aug 2 16:06:13 CEST 2008 |
| 16 | - do not include LIBS in autoconf tests. |
50 | - do not include LIBS in autoconf tests. |
| 17 | |
51 | |
