--- IO-AIO/Changes 2007/06/01 13:25:51 1.124 +++ IO-AIO/Changes 2008/05/11 01:00:03 1.148 @@ -3,20 +3,80 @@ TODO: better autoconf.pm that can return LIBS etc. TODO: aio_cptree/mvtree TODO: reduce condvar fairness: schedule hot-cache-threads first? -TODO: fix setsig TODO: splice/tee/vmsplice/sync_file_range? (http://kerneltrap.org/node/6505 http://lwn.net/Articles/178199/) +TODO: aio_fcntl, at least for file-locking -2.4 - - fix aio_busy w.r.t. fractional delays. - - add aio_chmod, aio_chown, aio_utime (linux successfully demonstrated - that you can block on futimes...). + - (libeio) make it compile on systems without readahead. + - (libeio) improve configure check for readahead availability. + - do not try to link against -lrt. + - use a separate configure script for IO::AIO (not the libeio one). + +3.0 Sun May 11 00:57:14 CEST 2008 + - added bin/treescan to the distribution. + - switched to using libeio. + - LOTS OF INCOMPATIBLE CHANGES: + - remove signal functionality, it is not worth the effort. + - max_outstanding no longer returns the previous number of requests. + - poll_cb no longer returns number of requests. + +2.62 Sat Apr 26 13:59:33 CEST 2008 + - port to solaris perls configured for old posix + (analysed by Jost Krieger). + - keep a reference to the perl filehandle object in aio_close, + so it doesn't get closed prematurely, leading to ugly races. + +2.61 Wed Apr 16 18:45:02 CEST 2008 + - fix treescan output duplication and improve output ordering. + also display files given on the commandline. + - use a different algorithm for aio_close that is faster + and probably has even lower chances of blocking. + - do our own stack memory management for threads - linux + allocates outrageous amounts of VM (not so bad unless you + use mlockall...), which severely limits the number of threads + on 32-bit arches: stack size is the larger of PTHREAD_STACK_MIN + and 4096 * sizeof (long) (usually this is 16..64k). + +2.6 Sun Mar 30 08:28:11 CEST 2008 + - added aio_sync. + - added aio_pathsync. + - fix prototypes of (void) functions. + +2.51 Sat Oct 6 16:04:54 CEST 2007 + - perlio isn't generally threadsafe, so aio_close could not work reliably. + aio_close now tries a safe hack that might not be that asynchronous + (see the manpage for details). + - discard callback return values as to not grow the stack endlessly + when poll_cb handles many requests without returning. + - minor code cleanups. + +2.5 Thu Oct 4 14:49:08 CEST 2007 + - replaced _fd2fh with faster xs code. + - aio_close will now try to do "the right thing" and thus might + work sensibly for the very first time. + +2.41 Mon Sep 24 21:28:21 CEST 2007 + - after fork, invest some work to make sure that the poll_fileno + stays stable (by dup2'ing the new fd over the old one), to make + it easier for programs/libs that don't control their forking + behaviour and cannot recreate their watchers. + +2.4 Sun Aug 5 18:44:22 CEST 2007 + - add aio_truncate, aio_chmod, aio_chown, aio_utime (linux + successfully demonstrated that you can block on futimes...). - allow undef as fileoffset for aio_read/write and use read/write(2) internally (useful for sockets or O_APPEND handles). + - allow undef for length in aio_write. + - negative dataoffsets work as expected now in aio_read/aio_write. - use NV instead of UV for 32 bit perls and file offsets, as NVs - have a large range then. + have a larger range then. - shared code between BDB and IO::AIO. + - aio_busy was completely broken. now fixed. + - readahead emulation now returns something. - correctly set errno to ENOSYS on unimplemented functions (should never happen, though). + - large changes to make it partially compile and run on win32, + but win32 lacks too much functionality, and perl overrides way + too many functions with crashing versions. 2.33 Tue Jan 23 23:55:41 CET 2007 - fix install path (Andreas J. Koenig).