New stuff:
* bugs in handling binding of non-directories fixed (opening with
O_CREAT opened the mountpoint instead of the bound file, unlink() and
non-directory rename() happily blew the mountpoints our instead of
returning -EBUSY). These bugs are common with the official tree, fixes
sent to Alan.
* due to rather nasty attack scenarios and popular demand bind is
root-only for the time being. I think that I know how to make it user-safe
but it will definitely take additional thinking/work.
* race in d_lookup() fixed.
* dput() made inline, dcache_lock not taken on the fast path (suggested
by Ingo).
* oopsable SMP race in do_proc_acct() fixed (patch sent to Alan).
* big lock shifted in sys_msync() - not that much, but anyway...
Please, try it. Thing is on ftp.math.psu.edu/pub/viro/dcache-ac7-J.gz.
Unless I'm seroiusly mistaken, big lock ceased to be the main source of
contention (and no, it's not because other sources had grown ;-) Results
of testing with different loads are more than appreciated, indeed.
Cheers,
Al
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:17 EST