First off, thanks for all the replies - this is a nice friendly group of
people (once one establishes that one either knows or wishes to know what
one is doing :-)).  Secondly, I have verified that if the kernel is
compiled with FIFO defined, FIFOs are handled quite nicely.  (I was just
kind of concerned that if it WASN'T defined, fifos panicked the kernel
with a Kernel Data Fault.  I'll toss out a bug report soon...)

I've been doing some experimentation with the fs_state flag inside of
ufs_sync() and fsck.  I suspect some stuff still needs to be done to
actually set fs_state &= FSOKAY, since fsck -p still finds all the
systems stable even after I've just done a kernel build.

Apparently the mfs driver is written but has no user-level support yet
(mount_ufs just echoes "exec: <mount-args>" and mount prints "mfs on /tmp").

Unless, of course, I'm doing something wrong, but there are as yet no
sources for mount_mfs present in the last tar snapshot I grabbed (our site
doesn't yet do "sup" due to the way the firewall is implemented -- I hope
to change this).


--

Make a hacker happy.  Use a *real* OS.
--

 ________ _____ ____  ________ _____WHO: Greywolf (my nameplate even says so)
/ ___\ _ \ __\ V / \  / /__ \| | __/WHAT: UNIX System Mangler...er, Admin
\ \| |   < _| ` ' \ '` / \/ /|_| _/ WHERE: Autodesk, Inc.  3 Harbor Dr.
 \___|_|\_\__\|_|  \/\/ \__/___/_|  Sausalito, CA 94965 (415) 332-2344 x4219
Friends don't let friends use System V.

------------------------------------------------------------------------------

to: port-sparc@sun-lamp.cs.berkeley.edu