[1834] in linux-scsi channel archive
Re[2]: scsi error
daemon@ATHENA.MIT.EDU (Barrie Spence)
Fri May 9 14:25:33 1997
From: Barrie Spence <barrie@calvin.demon.co.uk>
Date: Fri, 9 May 1997 18:25:33 +0100 (BST)
To: linux-scsi@vger.rutgers.edu
In-Reply-To: <Pine.A32.3.95.970509123152.107652C-100000@werner.exp-math.uni-essen.de>
Michael Weller <eowmob@exp-math.uni-essen.de> wrote:
> On Wed, 7 May 1997, Robert wrote:
>
> > aha1740[18]_mbxout wait!
> > aha1740[13]_mbxout wait!
> > aha1740[1]_mbxout wait!
> > aha1740[25]_mbxout wait!
> > aha1740[21]_mbxout wait!
> > aha1740[12]_mbxout wait!
> >
> > Any ideas on what is causing this error??? I would love to stop
> > having my messages file filling up with this.
>
> Well, I understand that apert from filling your msgs files it does not do
> any harm to you.
>
> Well, the aha1740 driver is a bit dusted and at the times I worked on
> it Adaptec claimed one would never see the mbxout flag set because the
> card was so fast except when it locked up due to a programming error. And
> yet, it was perfectly right at that time. Alas, I claim you use a very
> fast clocked machine s.t. it questions the aha card too fast, making the
> driver emit these warning messages (s.t. you have a cluu if the machine
> *would* lockup after).
>
> Anyway, a patch to enhance the driver in a few other areas plus extending
> the timeout loops here is out. Alas, I do not know if it was accepted for
> the official kernel and if so from which version onwards it is included.
>
> Certainly someone on the list knows it?
The patch is certainly in 2.0.30 - I don't get the message for disk access,
though I think that I can still tickle the time-out for i/o to a DAT drive.
I played with increasing the time-out, but it didn't appear to make much
difference.
Barrie
--
Barrie Spence Sanity Clause? There is no Sanity Clause
Home: barrie@calvin.demon.co.uk Telephone +44 1506 442304
Play: barrie@sqf.hp.com Telephone +44 131 331 7103
