| 1 | Revision history for IO::AIO |
1 | Revision history for IO::AIO |
|
|
2 | |
|
|
3 | TODO: better autoconf.pm that can return LIBS etc. |
|
|
4 | TODO: aio_cp/mvtree/rmtree |
|
|
5 | TODO: aio_utime, aio_chmod, aio_chown? |
|
|
6 | |
|
|
7 | TODO: expose max_idle |
|
|
8 | - doc update. |
|
|
9 | - fix a bug in aio_move where it would fail everytime |
|
|
10 | it would have to copy files manually. |
|
|
11 | - start threads only on (very low) demand. |
|
|
12 | - end superfluous threads automatically after 10s idling. |
|
|
13 | - some tuning applied to random places. |
|
|
14 | - add aio_mknod. |
|
|
15 | |
|
|
16 | 2.0 Thu Oct 26 18:27:58 CEST 2006 |
|
|
17 | - minor incompatibility: max_outstanding semantics have changed. |
|
|
18 | - major rewrite, consider all functionality EXPERIMENTAL. |
|
|
19 | - aio requests can be cancelled. |
|
|
20 | - callbacks can be (re-)set later. |
|
|
21 | - aio requests can choose between 9 priorities. |
|
|
22 | - add aio_group: aio requests can be grouped |
|
|
23 | into composite requests. |
|
|
24 | - generator interface for groups. |
|
|
25 | - add IO::AIO::poll_some. |
|
|
26 | - major documentation improvements. |
|
|
27 | - very minor bugfixes. |
|
|
28 | - add aio_nop for dummy requests. |
|
|
29 | - add aio_busy mainly for benchmarking and debugging. |
|
|
30 | - use "fast" mutexes on linux. |
|
|
31 | - use dynamic (and larger) buffers instead of |
|
|
32 | putting them onto the stack. |
|
|
33 | - optimise x86 and amd64 a bit. |
|
|
34 | - better error checking when ENOMEM. |
|
|
35 | - hopefully fix mem and dirp leaks on fork. |
|
|
36 | - warn about broken perl malloc. |
|
|
37 | - compiles on cygwin (but pread is broken on cygwin |
|
|
38 | which is not detected by autoconf, so you have to |
|
|
39 | edit autoconf/config.h yourself to enable emulation). |
|
|
40 | |
|
|
41 | 1.8 Sat Jun 24 22:20:44 CEST 2006 |
|
|
42 | - add and/or document aio_rename, aio_link, aio_symlink and aio_move. |
|
|
43 | |
|
|
44 | 1.73 Wed Mar 1 22:49:32 CET 2006 |
|
|
45 | - codename "jost" |
|
|
46 | - AIX/Solaris(?) aren't being helpful again, hardcode a value |
|
|
47 | for NAME_MAX and drive with your eyes closed (reported by |
|
|
48 | wcooley@nakedape.cc). |
|
|
49 | |
|
|
50 | 1.72 Thu Feb 2 00:40:44 CET 2006 |
|
|
51 | - another minor bug in aio_scandir, again, the callback was called |
|
|
52 | twice when the directory can't be stat'ed. |
|
|
53 | |
|
|
54 | 1.71 Mon Dec 26 20:20:03 CET 2005 |
|
|
55 | - grr, another minor bug in aio_scandir, this time resulting |
|
|
56 | in a runtime error. |
|
|
57 | |
|
|
58 | 1.7 Mon Dec 26 19:29:48 CET 2005 |
|
|
59 | - aio_scandir errornously called callback twice in some cases. |
|
|
60 | - added AnyEvent example. |
|
|
61 | |
|
|
62 | 1.61 Wed Sep 7 19:40:42 CEST 2005 |
|
|
63 | - no longer include $Config{libs}, as distros include too much |
|
|
64 | garbage in there that is not installed on typical systems |
|
|
65 | (for example, debian links perl against -lgdbm and many more |
|
|
66 | libraries despite perl not needing it, slowing down startup and |
|
|
67 | wasting memory). |
| 2 | |
68 | |
| 3 | 1.6 Tue Aug 30 17:44:44 CEST 2005 |
69 | 1.6 Tue Aug 30 17:44:44 CEST 2005 |
| 4 | - added aio_readdir, aio_scandir. |
70 | - added aio_readdir, aio_scandir. |
| 5 | - added aio_sendfile, including sendfile emulation when sendfile |
71 | - added aio_sendfile, including sendfile emulation when sendfile |
| 6 | isn't available. Linux, FreeBSD and HP-UX might work (sendfilev |
72 | isn't available. Linux, FreeBSD and HP-UX might work (sendfilev |
| … | |
… | |
| 12 | - somebody nagged about C89 compatibility, so this release should |
78 | - somebody nagged about C89 compatibility, so this release should |
| 13 | be C89 compatible, but no guarantees for future ones. |
79 | be C89 compatible, but no guarantees for future ones. |
| 14 | - abort when the configure script fails. |
80 | - abort when the configure script fails. |
| 15 | |
81 | |
| 16 | 1.4 Wed Aug 17 08:07:27 CEST 2005 |
82 | 1.4 Wed Aug 17 08:07:27 CEST 2005 |
| 17 | - forget to recreate the pipe after forking, this could cause |
83 | - forgot to recreate the pipe after forking, this could cause |
| 18 | deadlocks. |
84 | deadlocks. |
| 19 | |
85 | |
| 20 | 1.3 Wed Aug 17 07:25:54 CEST 2005 |
86 | 1.3 Wed Aug 17 07:25:54 CEST 2005 |
| 21 | - properly propagate exceptions from callbacks, instead of silently |
87 | - properly propagate exceptions from callbacks, instead of silently |
| 22 | eating them. |
88 | eating them. |
