ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/IO-AIO/Changes
(Generate patch)

Comparing IO-AIO/Changes (file contents):
Revision 1.197 by root, Wed Jun 17 06:33:07 2009 UTC vs.
Revision 1.234 by root, Wed Aug 4 17:16:19 2010 UTC

4TODO: aio_cptree/mvtree 4TODO: aio_cptree/mvtree
5TODO: reduce condvar fairness: schedule hot-cache-threads first? 5TODO: reduce condvar fairness: schedule hot-cache-threads first?
6TODO: splice/tee/vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) 6TODO: splice/tee/vmsplice? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/)
7TODO: aio_fcntl, at least for file-locking 7TODO: aio_fcntl, at least for file-locking
8TODO: fallocate, if it wouldn't be so useless. 8TODO: fallocate, if it wouldn't be so useless.
9TODO: aio_mincore?
10TODO: add? also linux-specific stuff?
11O_EXEC
12O_SEARCH
13O_CLOEXEC
14O_DIRECTORY
15O_DSYNC
16O_NOCTTY
17O_NOFOLLOW
18O_NONBLOCK
19O_RSYNC
20O_SYNC
21O_TTY_INIT
9 22
103.25 23O_DIRECT
24O_LARGEFILE
25O_NOATIME
26O_ASYNC
27TODO: document/test madvise, mprotect
28
29 - do not leak memory in IO::AIO::mmap when the scalar already
30 had string-data.
31 - add O_RDWR, O_APPEND and O_EXCL symbols and semi-documented
32 them.
33 - cache the result IV, for a minor speedup in the common case.
34 - croak when an mmapped-scalar changes location, to detect
35 user-errors better.
36
373.65 Wed Mar 31 02:45:05 CEST 2010
38 - actually use PATH_MAX instead of NAME_MAX for readlink, as to not
39 cut off long pathnames in aio_readlink (based on patch by
40 Rasmus Andersson).
41 - a double fork partially killed the event pipe (great testcase
42 by dormando). affects IO::AIO, BDB and Async::Interrupt.
43 - suck steve's dick and rename our symbols to avoid colliding
44 with nonstandard identifier spam on the broken os x pseudo-os.
45 affects IO::AIO and BDB.
46
473.6 Mon Jan 11 00:43:39 CET 2010
48 - (libeio) more fixes for the freebsd/apple sendfile - broken by
49 manpage standards, but apparently correct accoridng to actual
50 kernel sources.
51 - add IO::AIO::mmap/munmap support.
52 - add IO::AIO::m(un)lockall support.
53 - clean up manpage.
54
553.5 Thu Jan 7 21:25:04 CET 2010
56 - (liebio) fix freebsd sendfile (Vanilla Hsu).
57 - (libeio) also fix darwin sendfile that suffered from a similar bug.
58 - add aio_statvfs as interface to statvfs/fstatvfs.
59 - work around buggy symlink() on cygwin in the testsuite.
60 - wtf. freebsd-8.0 actually passes the testsuite! I knew
61 they could implement fork and semaphores one day! cheers!
62
633.4 Sat Jan 2 15:13:04 CET 2010
64 - (libeio) max_poll_time was not properly converted to ticks.
65 - clarify cancel_subs description.
66 - IO::AIO::sendfile did not actually return the return value
67 from the sendfile call.
68 - implement aio_msync, aio_mtouch.
69 - (libeio) tentatively support darwin in sendfile.
70
713.31 Thu Nov 12 02:14:29 CET 2009
72 - fix result status documentation of aio_copy, aio_move.
73 - speed up object creation considerably by avoiding
74 hash lookups.
75
763.3 Wed Aug 5 13:52:58 CEST 2009
77 - use common::sense.
78 - use common schmorp.h header.
79 - allow integers (file descriptors) in addition to file handles.
80 - take advantage of linux' eventfd (but it seems slower on smp?).
81 - use poll not select on posix platforms.
82
833.261 Wed Jul 1 10:11:51 CEST 2009
84 - more 0S X workarounds (patch by Tokuhiro Matsuno).
85
863.26 Tue Jun 30 09:33:26 CEST 2009
87 - 0S X of course claims to be posix 2008 but lacks posix_fadvise.
88 try to work around this horribly broken OS in a somewhat hackish
89 way. might help other os's too.
90
913.25 Sat Jun 27 05:18:26 CEST 2009
11 - added IO::AIO::fadvise and IO::AIO::sendfile. 92 - added IO::AIO::fadvise and IO::AIO::sendfile.
12 - (libeio) replaced quicksort+insertion sort by a tuned radix 93 - (libeio) replaced quicksort+insertion sort by a tuned radix
13 sort + insertion sort, resulting in comparable runtime (usually 94 sort + insertion sort, resulting in comparable runtime (usually
14 faster) to the old version, but without any ill side effects on 95 faster) to the old version, but without any ill side effects on
15 degenerated (for quicksort) data. 96 degenerated (for quicksort) data.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines