|
|
1 | TODO: better type checking (i.e. make sure the callback is a callback) |
|
|
2 | |
| 1 | Revision history for IO::AIO |
3 | Revision history for IO::AIO |
| 2 | |
4 | |
|
|
5 | 1.1 Sun Aug 7 05:27:23 CEST 2005 |
|
|
6 | - added configure script to detect availability of some calls. |
|
|
7 | (imperfect, as we cannot detect things like -lrt due to missing |
|
|
8 | feedback yet). |
|
|
9 | - emulate fdatasync via fsync, pread/pwrite via unsafe read/write |
|
|
10 | and readahead via manual reading on platforms lacking those. |
| 3 | - ported t/03_errors.t+framework by Brad Fitzpatrick |
11 | - ported t/03_errors.t+framework by Brad Fitzpatrick |
| 4 | from Linux::AIO. |
12 | from Linux::AIO. |
| 5 | - removed XOPEN_SOURCE because stupid solaris 9 doesn't |
13 | - removed XOPEN_SOURCE because stupid solaris 9 doesn't |
| 6 | compile with it enabled. Probably breaks others. Oh well. |
14 | compile with it enabled. Probably breaks others. Oh well. |
|
|
15 | - try to be more careful with errno on non-threaded perls. |
|
|
16 | - cygwins perl insisted on a matching open mode when morphing |
|
|
17 | the fd to an fh. |
|
|
18 | - convert sv's to "byte" encoding in aio_read/aio_write. Might |
|
|
19 | do that for pathnames, too, in the future, so make sure you |
|
|
20 | properly encode your pathnames. |
| 7 | |
21 | |
| 8 | 1.0 Fri Jul 22 10:43:30 CEST 2005 |
22 | 1.0 Fri Jul 22 10:43:30 CEST 2005 |
| 9 | - aio_open did not reliably deliver (open) filehandles. |
23 | - aio_open did not reliably deliver (open) filehandles. |
| 10 | - all aio_*-functions keep a reference to the filehandle so |
24 | - all aio_*-functions keep a reference to the filehandle so |
| 11 | it doesn't go away while the aio requests is working on it. |
25 | it doesn't go away while the aio requests is working on it. |
