… | |
… | |
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? (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 | 3.21 Fri Jun 12 18:45:53 CEST 2009 |
|
|
10 | - new options --dirs and --files for treescan. |
|
|
11 | - install bin/treescan by default. |
|
|
12 | - (libeio) aio_readdir can now be cancelled while executing. |
|
|
13 | - fix a printf format string for 64 bit systems (could lead |
|
|
14 | to problems on big endian 64 bit systems). |
|
|
15 | - do not use qsort() but our own algorithm: glibc initialises |
|
|
16 | the whole locale and I/O subsystem inside qsort, causing |
|
|
17 | a stack overflow on 32 bit machines. The new sort uses much less |
|
|
18 | stack and is more than twice as fast in typical situations. |
|
|
19 | |
|
|
20 | 3.2 Sun Jun 7 20:30:05 CEST 2009 |
9 | - (libeio) pwrite emulation was even more flawed than intended and did |
21 | - (libeio) pwrite emulation was even more flawed than intended and did |
10 | not restore the file offset. |
22 | not restore the file offset. |
11 | - add aio_readdirx, which can return inode and filetype and sort |
23 | - add aio_readdirx, which can return inode and filetype and sort |
12 | the names in various ways. |
24 | the names in various ways. |
13 | - unfortunately, utime, chmod, chown on an open file that has just |
25 | - unfortunately, utime, chmod, chown on an open file that has just |