Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 2 Apr 1997 21:08:17 +0200
From:      j@uriah.heep.sax.de (J Wunsch)
To:        hackers@freebsd.org
Subject:   Re: Internal clock
Message-ID:  <19970402210817.YJ08249@uriah.heep.sax.de>
In-Reply-To: <199704021757.KAA13882@phaeton.artisoft.com>; from Terry Lambert on Apr 2, 1997 10:57:17 -0700
References:  <199704021757.KAA13882@phaeton.artisoft.com>

next in thread | previous in thread | raw e-mail | index | archive | help
As Terry Lambert wrote:

> > If you had ever used it (with options DEVFS_ROOT, it's already there),
> > you knew what the bugs are.

> I've used it.  For what it's worth, prior to the Lite2 merge in 3.0,
> my system didn't crash when I unmounted things.

That's not the biggest problem.  Right now, it suffers serious
aliasing problems that even prevent the system to start at all (it
doesn't find its root filesystem entry, or rather, the vnode is being
revoked while opening the root device, resulting in an EIO when trying
to mount it).

Once i've sorted these problems out, i'll continue to make DEVFS
usable as a default.  If persistancy is once the last remaining
problem, and nobody is stepping forward to implement the all-singing-
all-dancing union layer shadowing solution, i'll probably implement a
startup/shutdown script-based solution.  If something better comes up
later, it can still be replaced.

I think FreeBSD 3.0 might be shipping with DEVFS as the default, but
it's still ``quarter o'mile'' to go.

-- 
cheers, J"org

joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970402210817.YJ08249>