From owner-freebsd-questions  Wed Mar 13 07:14:46 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id HAA02970
          for questions-outgoing; Wed, 13 Mar 1996 07:14:46 -0800 (PST)
Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id HAA02964
          for <freebsd-questions@freebsd.org>; Wed, 13 Mar 1996 07:14:42 -0800 (PST)
Message-Id: <199603131514.HAA02964@freefall.freebsd.org>
Received: by cheops.anu.edu.au
	(1.37.109.16/16.2) id AA133220196; Thu, 14 Mar 1996 02:16:36 +1100
From: Darren Reed <avalon@coombs.anu.edu.au>
Subject: Re: DDS DAT Drive and 2.1.0-R ?
To: msmith@atrad.adelaide.edu.au (Michael Smith)
Date: Thu, 14 Mar 1996 02:16:36 +1100 (EDT)
Cc: freebsd-questions@freebsd.org
In-Reply-To: <199603131432.BAA03121@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Mar 14, 96 01:02:05 am
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

In some mail from Michael Smith, sie said:
> 
> Darren Reed stands accused of saying:
[...]
> Your settings are fine.  You're looking at either a driver bug/shortcoming
> or hardware limitation.

Hmmm.  Well, in testing, "dd if=foo of=/dev/nrst0 bs=16k" seems to work
fine, but after that, no.  I can't seem to make dump use small blocks,
I suspect, or it does other werid stuff.

> > Mar 14 00:38:56 freebsd /kernel.scsi: data length underflow
> > Mar 14 00:38:56 freebsd /kernel.scsi: nca0/4/0 data length underflow
> > Mar 14 00:38:56 freebsd last message repeated 130 times
> > Mar 14 00:40:42 freebsd /kernel.scsi: nca0/4/0 (st0) timed out
> 
> What does the driver mean? Have you looked at the source?

I don't understand why it would occur, ie I'm not a scsi driver expert.

> > Also, when I tried a kernel config with the following, st0 didn't work:
> >
> > controller scbus0 at nca0
> > disk sd? at scbus? target ?
> > disk sd? at scbus? target ?
> > tape st? at scbus? target ?
> > device cd? at scbus? target ?
> 
> That's because the systax is all wrong?  Whay makes you think that's the
> way to do it?  Look at GENERIC :
> 
> controller scbus0
> device sd0
> device od0
> device st0
> device cd0

Ahh, I assumed the wild carding worked differently.

thanks,
darren