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

Comparing IO-AIO/Changes (file contents):
Revision 1.231 by root, Wed Aug 4 14:02:58 2010 UTC vs.
Revision 1.264 by root, Tue Jul 5 18:59:28 2011 UTC

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? 9TODO: aio_mincore?
10TODO: posix_madvise. 10TODO: getxattr etc.?
11TODO: add? also linux-specific stuff? 11TODO: MS_xxx, posix_fadvise, posix_madvise
12O_EXEC 12TODO: realpath
13O_SEARCH
14O_CLOEXEC
15O_DIRECTORY
16O_DSYNC
17O_NOCTTY
18O_NOFOLLOW
19O_NONBLOCK
20O_RSYNC
21O_SYNC
22O_TTY_INIT
23 13
14TODO: remove fork support, eio_init can be called multiple times maybe
15TODO: "this might change"
16 - (libeio) added realpath.
17
183.93 Wed Jun 29 23:44:18 CEST 2011
19 - ECB.H WAS MISSING.
20
213.92 Wed Jun 29 14:45:41 CEST 2011
22 - ecb.h was missing.
23
243.91 Wed Jun 29 13:24:42 CEST 2011
25 - (libeio) work around a Linux (and likely FreeBSD and other
26 kernels) bug where sendfile would not transfer all the requested
27 bytes on large transfers, using a heuristic.
28 - buggy sendfile caused aio_move/copy to sometimes fail for big
29 files (fortunately it checks that the whole file has been
30 transferred...)
31 - use libecb for higher performance and higher portability.
32 - (libeio) disable sendfile on darwin, it's too broken.
33 - disable fork tests on !linux, as only linux supports mixing
34 pthread and fork in perl.
35 - document the fact that fork doesn't work anymore when using
36 this module, on anything !GNU/Linux.
37 - increase timeout in t/04_fork.t, as too many CPAN-tester setups
38 run on a 0.1MHz cpu. Or so.
39
403.9 Fri May 27 02:43:47 CEST 2011
41 - (libeio) fix memory corruption in aio_readdirx for the flags
42 combination READDIR_STAT_ORDER | READDIR_DIRS_FIRST.
43 - add lots of newre POSIX and GNU/Linux-specific open
44 flags.
45
463.8 Sun Mar 27 12:25:33 CEST 2011
47 - use nonstandard but maybe-working-on-bsd fork technique.
48 - support a max_idle value of 0.
49 - support setting of idle timeout value (IO::AIO::idle_timeout).
50
513.72 Fri Feb 11 04:25:38 CET 2011
52 - use _POSIX_MEMLOCK_RANGE to detect mlock/munlock.
53 - aio_mknod always used a dev_t value of 0.
54 - new treescan option: --grep.
55 - add more S_IF macros, and major/minor/makedev "macros".
56
573.71 Thu Dec 30 08:18:46 CET 2010
58 - the numerical result value passed to callbacks did not stringify
59 correctly, due to internal reuse and failure to reset the sv flags.
60 - actually test for posix_[mf]advise, as at least uClibc defines
61 _POSIX_ADVISORY_INFO without actually having any of the required
62 funcitonality. ugh.
63
643.7 Mon Nov 1 23:00:34 CET 2010
65 - implement/add madvise, mmap, munmap, aio_mlockall, munlockall,
66 aio_mlock, munlock, msync and mtouch to @EXPORT_OK.
67 - document the sad state of affairs w.r.t. pthread on many bsds.
68 - do not enable mmap on systems without _POSIX_MAPPED_FILES
69 (openbsd 4.8).
24 - do not leak memory in IO::AIO::mmap when the scalar already 70 - do not leak memory in IO::AIO::mmap when the scalar already
25 had string-data. 71 had string-data.
26 - add O_RDWR, O_APPEND and O_EXCL symbols and semi-documented 72 - add O_RDWR, O_APPEND and O_EXCL symbols and semi-document
27 them. 73 them.
28 - cache the result IV, for a minor speedup in the common case. 74 - cache the result IV, for a minor speedup in the common case.
29 - croak when an mmapped-scalar changes location, to detect 75 - croak when an mmapped-scalar changes location, to detect
30 user-errors better. 76 user-errors better.
77 - fix aio_readlink prototype.
31 78
323.65 Wed Mar 31 02:45:05 CEST 2010 793.65 Wed Mar 31 02:45:05 CEST 2010
33 - actually use PATH_MAX instead of NAME_MAX for readlink, as to not 80 - actually use PATH_MAX instead of NAME_MAX for readlink, as to not
34 cut off long pathnames in aio_readlink (based on patch by 81 cut off long pathnames in aio_readlink (based on patch by
35 Rasmus Andersson). 82 Rasmus Andersson).

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines