| … | |
… | |
| 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 | TODO: aio_mincore? |
9 | TODO: aio_mincore? |
| 10 | TODO: add? also linux-specific stuff? |
10 | TODO: getxattr etc.? |
|
|
11 | TODO: MS_xxx, posix_fadvise, posix_madvise |
|
|
12 | TODO: realpath |
| 11 | |
13 | |
| 12 | const_iv (O_ASYNC) |
14 | TODO: remove fork support, eio_init can be called multiple times maybe |
| 13 | const_iv (O_DIRECT) |
15 | TODO: "this might change" - no wrklock for malloc, no automatic free in child. |
| 14 | const_iv (O_NOATIME) |
16 | - (libeio) fix a deadlock where a wakeup signal could be missed when |
|
|
17 | a timeout occured at the same time. |
|
|
18 | - (libeio) added realpath. |
|
|
19 | - (libeio) added fallocate. |
|
|
20 | - (libeio) disabling sendfile on darwin unfortunately broke the emulation. |
|
|
21 | - (libeio) do not acquire any locks when forking. |
|
|
22 | - (libeio) use fewer time() syscalls when waiting for new requests. |
| 15 | |
23 | |
| 16 | const_iv (O_CLOEXEC) |
24 | 3.93 Wed Jun 29 23:44:18 CEST 2011 |
| 17 | const_iv (O_NOCTTY) |
25 | - ECB.H WAS MISSING. |
| 18 | const_iv (O_NOFOLLOW) |
|
|
| 19 | const_iv (O_NONBLOCK) |
|
|
| 20 | const_iv (O_EXEC) |
|
|
| 21 | const_iv (O_SEARCH) |
|
|
| 22 | const_iv (O_DIRECTORY) |
|
|
| 23 | const_iv (O_DSYNC) |
|
|
| 24 | const_iv (O_RSYNC) |
|
|
| 25 | const_iv (O_SYNC) |
|
|
| 26 | const_iv (O_TTY_INIT) |
|
|
| 27 | |
26 | |
|
|
27 | 3.92 Wed Jun 29 14:45:41 CEST 2011 |
|
|
28 | - ecb.h was missing. |
|
|
29 | |
|
|
30 | 3.91 Wed Jun 29 13:24:42 CEST 2011 |
|
|
31 | - (libeio) work around a Linux (and likely FreeBSD and other |
|
|
32 | kernels) bug where sendfile would not transfer all the requested |
|
|
33 | bytes on large transfers, using a heuristic. |
|
|
34 | - buggy sendfile caused aio_move/copy to sometimes fail for big |
|
|
35 | files (fortunately it checks that the whole file has been |
|
|
36 | transferred...) |
|
|
37 | - use libecb for higher performance and higher portability. |
|
|
38 | - (libeio) disable sendfile on darwin, it's too broken. |
|
|
39 | - disable fork tests on !linux, as only linux supports mixing |
|
|
40 | pthread and fork in perl. |
|
|
41 | - document the fact that fork doesn't work anymore when using |
|
|
42 | this module, on anything !GNU/Linux. |
|
|
43 | - increase timeout in t/04_fork.t, as too many CPAN-tester setups |
|
|
44 | run on a 0.1MHz cpu. Or so. |
|
|
45 | |
|
|
46 | 3.9 Fri May 27 02:43:47 CEST 2011 |
|
|
47 | - (libeio) fix memory corruption in aio_readdirx for the flags |
|
|
48 | combination READDIR_STAT_ORDER | READDIR_DIRS_FIRST. |
|
|
49 | - add lots of newer POSIX and GNU/Linux-specific open |
|
|
50 | flags. |
| 28 | |
51 | |
| 29 | 3.8 Sun Mar 27 12:25:33 CEST 2011 |
52 | 3.8 Sun Mar 27 12:25:33 CEST 2011 |
| 30 | - use nonstandard but maybe-working-on-bsd fork technique. |
53 | - use nonstandard but maybe-working-on-bsd fork technique. |
| 31 | - support a max_idle value of 0. |
54 | - support a max_idle value of 0. |
| 32 | - support setting of idle timeout value (IO::AIO::idle_timeout). |
55 | - support setting of idle timeout value (IO::AIO::idle_timeout). |
