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.? |
10 | TODO: F_DUPFD_CLOEXEC |
9 | TODO: F_DUPFD_CLOEXEC |
|
|
10 | TODO: emulation for splice? |
|
|
11 | TODO: eio_mmap|mlock|munmap|splice... |
|
|
12 | |
|
|
13 | - (libeio) implement aio_realpath for win32. |
|
|
14 | - (xthread) work around compile time bugs in ptw32. |
|
|
15 | |
|
|
16 | 4.19 Sun Jan 6 12:47:26 CET 2013 |
|
|
17 | - avoid endless loop in fiemap with some XFS files. |
|
|
18 | - in aio_rename and aio_rmdir, specialcase the case of [$wd, "."] |
|
|
19 | and call rename/rmdir instead of renameat/unlinkat. |
|
|
20 | |
|
|
21 | 4.18 Thu Oct 11 07:01:26 CEST 2012 |
|
|
22 | - fix unintended xthread_create by intentionalising it :) |
|
|
23 | |
|
|
24 | 4.17 Thu Oct 11 05:19:47 CEST 2012 |
|
|
25 | - rename aio_fallocate to aio_allocate, to match documentation. |
|
|
26 | - add list of linux fsid values to aio_statvfs docs. |
|
|
27 | - work around a bug in btrfs' FIEMAP ioctl implementation. |
|
|
28 | - work around AIX bug: statvfs.f_fsid is a struct, not unsigned long |
|
|
29 | as per unix spec. |
|
|
30 | |
|
|
31 | 4.16 Tue Aug 14 05:39:03 CEST 2012 |
|
|
32 | - aio_statvfs was wromgly marked for wd emulation, causing |
|
|
33 | it to malfunction for paths. |
|
|
34 | - fix a crash in aio_fiemap, when used on ranges without any |
|
|
35 | extents. |
|
|
36 | - work around linux kernel bug (at least in 3.2): kernel might |
|
|
37 | trash fiemap input arguments. |
|
|
38 | - work around linux kernel bug (at least in 3.2): kernel does not |
|
|
39 | set FIEMAP_EXTENT_LAST on last segment. |
|
|
40 | - work around linux kernel bug (at least in 3.2): kernel silently |
|
|
41 | truncates segment count if a file. |
|
|
42 | - make fh argument of IO::AIO::mmap optional, also add mmap/munmap |
|
|
43 | to quick overview. |
|
|
44 | - splice can corrupt data. |
|
|
45 | - (libeio) remove pread/pwrite emulation. |
|
|
46 | - do not grow mmaped scalars in aio_read. |
|
|
47 | - add FALLOC_FL_PUNCH_HOLE and document fallocate. |
|
|
48 | - provide pread/pwrite for win32, making it link again on native win32. |
|
|
49 | |
|
|
50 | 4.15 Tue Apr 10 06:59:00 CEST 2012 |
|
|
51 | - always include linux/types.h for fiemap, for compatibility to |
|
|
52 | ancient systems (Paul Howarth). |
|
|
53 | - experimental support for IO::AIO::splice and ::tee (no aio_...). |
|
|
54 | - provide SEEK_HOLE and SEEK_DATA, if available. |
|
|
55 | - work around (again!) an immensely stupid bug in RHEL, defining |
|
|
56 | autoconf macros in linux system headers (analysed by Paul Howarth). |
|
|
57 | |
|
|
58 | 4.14 Sat Apr 7 02:45:18 CEST 2012 |
|
|
59 | - fix stat structure usage on windows, which caused bogus stat results. |
|
|
60 | - (libeio) make readahead emulation behave more like actual readahead by never failing. |
|
|
61 | - new request aio_seek. |
|
|
62 | - new request aio_fiemap. |
|
|
63 | - autogenerate the #ifdef/#define 0 blocks for symbols we export. |
11 | |
64 | |
12 | 4.12 Fri Dec 30 08:51:25 CET 2011 |
65 | 4.12 Fri Dec 30 08:51:25 CET 2011 |
13 | - realpath would return a random value if the name was too long, |
66 | - realpath would return a random value if the name was too long, |
14 | instead of -1. |
67 | instead of -1. |
15 | - port to c89. |
68 | - port to c89. |