1 | Revision history for IO::AIO |
1 | Revision history for IO::AIO |
2 | |
2 | |
3 | TODO: better autoconf.pm that can return LIBS etc. |
|
|
4 | TODO: aio_cptree/mvtree |
3 | TODO: aio_cptree/mvtree |
5 | TODO: reduce condvar fairness: schedule hot-cache-threads first? |
4 | TODO: reduce condvar fairness: schedule hot-cache-threads first? |
6 | TODO: splice/tee/vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) |
5 | TODO: vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) |
7 | TODO: aio_fcntl, at least for file-locking |
6 | TODO: aio_fcntl, at least for file-locking |
8 | TODO: aio_mincore? |
7 | TODO: aio_mincore? |
9 | TODO: getxattr etc.? |
8 | TODO: getxattr etc.? |
|
|
9 | TODO: F_DUPFD_CLOEXEC |
|
|
10 | TODO: emulation for splice? |
|
|
11 | TODO: eio_mmap|mlock|munmap|splice... |
10 | |
12 | |
|
|
13 | - fix a crash in aio_fiemap, when used on ranges without any |
|
|
14 | extents. |
|
|
15 | - work around linux kernel bug (at least in 3.2): kernel might |
|
|
16 | trash fiemap input arguments. |
|
|
17 | - work around linux kernel bug (at least in 3.2): kernel does not |
|
|
18 | set FIEMAP_EXTENT_LAST on last segment. |
|
|
19 | - work around linux kernel bug (at least in 3.2): kernel silently |
|
|
20 | truncates segment count if a file. |
|
|
21 | - make fh argument of IO::AIO::mmap optional, also add mmap/munmap |
|
|
22 | to quick overview. |
|
|
23 | - splice can corrupt data. |
|
|
24 | - (libeio) remove pread/pwrite emulation. |
|
|
25 | - do not grow mmaped scalars in aio_read. |
|
|
26 | - add FALLOC_FL_PUNCH_HOLE and document fallocate. |
|
|
27 | |
|
|
28 | 4.15 Tue Apr 10 06:59:00 CEST 2012 |
|
|
29 | - always include linux/types.h for fiemap, for compatibility to |
|
|
30 | ancient systems (Paul Howarth). |
|
|
31 | - experimental support for IO::AIO::splice and ::tee (no aio_...). |
|
|
32 | - provide SEEK_HOLE and SEEK_DATA, if available. |
|
|
33 | - work around (again!) an immensely stupid bug in RHEL, defining |
|
|
34 | autoconf macros in linux system headers (analysed by Paul Howarth). |
|
|
35 | |
|
|
36 | 4.14 Sat Apr 7 02:45:18 CEST 2012 |
|
|
37 | - fix stat structure usage on windows, which caused bogus stat results. |
|
|
38 | - (libeio) make readahead emulation behave more like actual readahead by never failing. |
|
|
39 | - new request aio_seek. |
|
|
40 | - new request aio_fiemap. |
|
|
41 | - autogenerate the #ifdef/#define 0 blocks for symbols we export. |
|
|
42 | |
|
|
43 | 4.12 Fri Dec 30 08:51:25 CET 2011 |
|
|
44 | - realpath would return a random value if the name was too long, |
|
|
45 | instead of -1. |
|
|
46 | - port to c89. |
|
|
47 | |
|
|
48 | 4.11 Mon Oct 10 00:24:11 CEST 2011 |
|
|
49 | - libeio didn't compile on !linux (Paul Howarth). |
|
|
50 | |
|
|
51 | 4.1 Sun Oct 9 10:24:11 CEST 2011 |
|
|
52 | - IO::AIO did access uninitialised memory on unsuccessful stats. |
11 | - (libeio) added syncfs syscall wrapper. |
53 | - (libeio) added syncfs syscall wrapper. |
12 | - (libeio) set thread name on linux (ps -L/Hcx, top, gdb). |
54 | - (libeio) set thread name on linux (ps -L/Hcx, top, gdb). |
|
|
55 | - (libeio) support multiple indepenent working directories. |
13 | - applied speling corrections by Nicholas Bamber. |
56 | - applied speling corrections by Nicholas Bamber. |
|
|
57 | - tune treescan #threads and #outstanding requests a bit. |
|
|
58 | - reduce the number of file descriptors in use at any one time |
|
|
59 | in treescan, to avoid running out of them (and using too much ram). |
|
|
60 | - take advantage of working directoriy abstractions |
|
|
61 | in aio_scandir and treescan. |
|
|
62 | - reduce compiled size by ~10% by not inlining some key functionality. |
|
|
63 | - added --progress switch to treescan. |
14 | |
64 | |
15 | 4.0 Mon Jul 18 05:01:10 CEST 2011 |
65 | 4.0 Mon Jul 18 05:01:10 CEST 2011 |
16 | - INCOMPATIBLE CHANGE: fork is no longer supported (indeed, it never was), |
66 | - INCOMPATIBLE CHANGE: fork is no longer supported (indeed, it never was), |
17 | see FORK BEHAVIOUR in manpage for details. |
67 | see FORK BEHAVIOUR in manpage for details. |
18 | - passes testsuite on win32 now and may actually work |
68 | - passes testsuite on win32 now and may actually work |