I noticed some old change in FreeBSD tree, which made async mounted
filesystems to do even less stuff synchronously, thus speed metadata
writes up to linux ext2fs level (at least so the commit log claimed).

Would we be interested in such a change in our tree? Asynchronous
filesystems would be even more likely to lose data in case of
system failure, but the speed advantage might be worth the risk.
Asynchronously mounted filesystems are not meant for important
data anyway.

The changes I noticed are these:

dyson       95/11/05 13:01:17

  Modified:    sys/kern  vfs_syscalls.c
               sys/ufs/ufs  ufs_readwrite.c
               sys/ufs/ffs  ffs_inode.c
  Log:
  Make MNT_ASYNC more effective for UFS.  It should not be too much more
  dangerous than the original MNT_ASYNC.  There might be some minor
  security considerations due to data writes not being posted as promptly
  as before.  Meta-data operations are still not quite as fast as Linux.

and 

FreeBSD ufs/ufs/ufs_lookup.c rev 1.8: """
This commit causes UFS to perform at Linux EXT2FS metadata rates.  After
earlier discussions with DG, and a recent email exchange with SEF, I
decided to allow UFS to run wide-open on an experimental basis.  We
will probably support eventually multiple async modes, and this is
the fastest the we can expect.  Just use the -o async flag on the
UFS mount.  Good luck...
"""

Jaromir
-- 
Jaromir Dolecek <jdolecek@NetBSD.org>            http://www.NetBSD.org/
-=- We should be mindful of the potential goal, but as the tantric    -=-
-=- Buddhist masters say, ``You may notice during meditation that you -=-
-=- sometimes levitate or glow.   Do not let this distract you.''     -=-