Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 2 Apr 1997 13:16:38 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        karl@Mcs.Net (Karl Denninger)
Cc:        terry@lambert.org, ache@nagual.ru, dyson@FreeBSD.ORG, bde@zeta.org.au, current@FreeBSD.ORG
Subject:   Re: ufs lock panic in -current
Message-ID:  <199704022016.NAA14329@phaeton.artisoft.com>
In-Reply-To: <19970402131508.59503@Jupiter.Mcs.Net> from "Karl Denninger" at Apr 2, 97 01:15:08 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> Am I correct in understanding that this isn't fixed yet, and that also the
> NFS exec() problems with paging are also still open?

Yes.

I believe, however, that Doug Rabson was replacing the stubbed routines
in the NFS case to respect the lock used by vclean.  This is a sort
of kludge soloution, but should fix the NFS vnode problems.

The paging problems were mentioned by John Dyson as something he
was actively looking at; they are, it seems, related to the VM object
alising issues that the Lite2 unionfs also suffers under (from my
minimal reading of John's posting and the affected code in both
FS's).


> I'm sorry about not being more up-to-date on this; I've been INSANELY busy
> the last few weeks, and haven't had time to dig into the NFS layer.  Is it
> still as it was?
> 
> (BTW, its easy to test -- a loopback mount from the same machine will fail
> just as hard as one over a real wire).

I believe the work discussed above may be sufficient to mask both these
problems (and solve them, in one of the cases).


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704022016.NAA14329>