1 | Revision history for IO::AIO |
1 | Revision history for IO::AIO |
2 | |
2 | |
3 | 1.1 |
3 | TODO: better autoconf.pm that can return LIBS etc. |
|
|
4 | |
|
|
5 | 1.73 Wed Mar 1 22:49:32 CET 2006 |
|
|
6 | - codename "jost" |
|
|
7 | - AIX/Solaris(?) aren't being helpful again, hardcode a value |
|
|
8 | for NAME_MAX and drive with your eyes closed (reported by |
|
|
9 | wcooley@nakedape.cc). |
|
|
10 | |
|
|
11 | 1.72 Thu Feb 2 00:40:44 CET 2006 |
|
|
12 | - another minor bug in aio_scandir, again, the callback was called |
|
|
13 | twice when the directory can't be stat'ed. |
|
|
14 | |
|
|
15 | 1.71 Mon Dec 26 20:20:03 CET 2005 |
|
|
16 | - grr, another minor bug in aio_scandir, this time resulting |
|
|
17 | in a runtime error. |
|
|
18 | |
|
|
19 | 1.7 Mon Dec 26 19:29:48 CET 2005 |
|
|
20 | - aio_scandir errornously called callback twice in some cases. |
|
|
21 | - added AnyEvent example. |
|
|
22 | |
|
|
23 | 1.61 Wed Sep 7 19:40:42 CEST 2005 |
|
|
24 | - no longer include $Config{libs}, as distros include too much |
|
|
25 | garbage in there that is not installed on typical systems |
|
|
26 | (for example, debian links perl against -lgdbm and many more |
|
|
27 | libraries despite perl not needing it, slowing down startup and |
|
|
28 | wasting memory). |
|
|
29 | |
|
|
30 | 1.6 Tue Aug 30 17:44:44 CEST 2005 |
|
|
31 | - added aio_readdir, aio_scandir. |
|
|
32 | - added aio_sendfile, including sendfile emulation when sendfile |
|
|
33 | isn't available. Linux, FreeBSD and HP-UX might work (sendfilev |
|
|
34 | support is unavailable on my solaris 8 machine, but preliminary |
|
|
35 | code is there. What's lacking is configury stuff and testing...). |
|
|
36 | |
|
|
37 | 1.5 Sat Aug 20 02:32:19 CEST 2005 |
|
|
38 | - create the aio threads on demand only (good after forking). |
|
|
39 | - somebody nagged about C89 compatibility, so this release should |
|
|
40 | be C89 compatible, but no guarantees for future ones. |
|
|
41 | - abort when the configure script fails. |
|
|
42 | |
|
|
43 | 1.4 Wed Aug 17 08:07:27 CEST 2005 |
|
|
44 | - forget to recreate the pipe after forking, this could cause |
|
|
45 | deadlocks. |
|
|
46 | |
|
|
47 | 1.3 Wed Aug 17 07:25:54 CEST 2005 |
|
|
48 | - properly propagate exceptions from callbacks, instead of silently |
|
|
49 | eating them. |
|
|
50 | - use a different fork algorithm that avoids recursive callback |
|
|
51 | invocations and reduces fork latency in busy processes. |
|
|
52 | - fix a bug where the buffer scalar in aio_read got assigned the wrong |
|
|
53 | length after reading. |
|
|
54 | - fix a bug where the child would process some outstanding requests |
|
|
55 | from the parent. |
|
|
56 | - make a copy of the callback scalar, to avoid some very uncommon |
|
|
57 | (but valid) usages where the callback scalar changes. |
|
|
58 | - the callback was never freed in aio_read/aio_write. |
|
|
59 | - aio_read/aio_write will now set the readonly flag on the data |
|
|
60 | scalar for better error reporting. |
|
|
61 | |
|
|
62 | 1.2 Wed Aug 17 01:22:58 CEST 2005 |
|
|
63 | - verified on cygwin, linux/amd64+ia32+alpha, solaris. |
|
|
64 | - use SvPVbyte for filenames now. You *need* to encode your strings to |
|
|
65 | the external filename encoding first. |
|
|
66 | - fix the bug where filehandles could leak (again). |
|
|
67 | - clarify freebsd situation: make it work and tell me how, or go away |
|
|
68 | with your underdocumented, broken and outdated OS. |
|
|
69 | - fork'ing should now be officially supported (where linux and |
|
|
70 | posix contradict each other, I followed posix). |
|
|
71 | - only call read() once per poll, not twice (speedup). |
|
|
72 | - new functions: aio_rmdir, aio_symlink. |
|
|
73 | - force callback arguments to be undef or a reference, for better |
|
|
74 | error catching. |
|
|
75 | |
|
|
76 | 1.1 Sun Aug 7 05:27:23 CEST 2005 |
4 | - added configure script to detect availability of some calls. |
77 | - added configure script to detect availability of some calls. |
5 | (imperfect, as we cannot detect things like -lrt due to missing |
78 | (imperfect, as we cannot detect things like -lrt due to missing |
6 | feedback yet). |
79 | feedback yet). |
7 | - emulate fdatasync via fsync, pread/pwrite via unsafe read/write |
80 | - emulate fdatasync via fsync, pread/pwrite via unsafe read/write |
8 | and readahead via manual reading on platforms lacking those. |
81 | and readahead via manual reading on platforms lacking those. |
… | |
… | |
11 | - removed XOPEN_SOURCE because stupid solaris 9 doesn't |
84 | - removed XOPEN_SOURCE because stupid solaris 9 doesn't |
12 | compile with it enabled. Probably breaks others. Oh well. |
85 | compile with it enabled. Probably breaks others. Oh well. |
13 | - try to be more careful with errno on non-threaded perls. |
86 | - try to be more careful with errno on non-threaded perls. |
14 | - cygwins perl insisted on a matching open mode when morphing |
87 | - cygwins perl insisted on a matching open mode when morphing |
15 | the fd to an fh. |
88 | the fd to an fh. |
|
|
89 | - convert sv's to "byte" encoding in aio_read/aio_write. Might |
|
|
90 | do that for pathnames, too, in the future, so make sure you |
|
|
91 | properly encode your pathnames. |
16 | |
92 | |
17 | 1.0 Fri Jul 22 10:43:30 CEST 2005 |
93 | 1.0 Fri Jul 22 10:43:30 CEST 2005 |
18 | - aio_open did not reliably deliver (open) filehandles. |
94 | - aio_open did not reliably deliver (open) filehandles. |
19 | - all aio_*-functions keep a reference to the filehandle so |
95 | - all aio_*-functions keep a reference to the filehandle so |
20 | it doesn't go away while the aio requests is working on it. |
96 | it doesn't go away while the aio requests is working on it. |