[8407] in linux-scsi channel archive
Re: AIC-7890 SCSI problems
daemon@ATHENA.MIT.EDU (Dr. Michael Weller)
Sun Mar 19 07:32:58 2000
Date: Sun, 19 Mar 2000 13:27:50 +0100 (MEZ)
From: "Dr. Michael Weller" <eowmob@exp-math.uni-essen.de>
To: "G. Hugh Song" <hugh@lucia.kjist.ac.kr>
Cc: linux-kernel@vger.rutgers.edu, linux-scsi@vger.rutgers.edu
In-Reply-To: <Pine.LNX.4.10.10003191554040.14695-100000@lucia.kjist.ac.kr>
Message-Id: <Pine.A32.3.95.1000319132031.33318A-100000@werner.exp-math.uni-essen.de>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Sun, 19 Mar 2000, G. Hugh Song wrote:
> On Sat, 18 Mar 2000, Dr. Michael Weller wrote:
>
> > On Fri, 17 Mar 2000, G. Hugh Song wrote:
[...]
> > BTW, I had some major problems with the aic7xxx driver on a aha2940u2w in
> > 2.2.14. It seems the mode sense handling was pretty much screwed and
> > caused havoc with more elaborate scsi devices (tapes, removables) that
> > need more than plain SCSI read/write commands.
>
> Ah-ha! I've got a SCSI CDRewritable drive from Sony. That probably
> caused timing erors. I guess I need to take out that fancy rewritable
> and put a plain IDE cdrom.
>
> > My problem (fortunately, knock on wood) seems to be cured by the 5.1.28
> > release. From my problems I'd claim that anyone trying to use a tape or
> > similar device would get plenty of problems w/o it.
>
> Without what?
W/o the 5.1.28 patch (and predecessors) applied to 2.2.14..
> > A new 2.2 release with
> > the updated aic7xxx driver is almost unavoidable.
>
> Are you talking about version 5.2.? of aic7xxx? The aic7xxx homepage does
> not have it yet. http://people.redhat.com/dledford/aic7xxx.html
Sorry, it seems I made myself not clear enough. I meant, IMHO, a 2.2.15
release of the kernel with the 5.1.28 aic7xxx included is absolutely
necessary, s.t. it ends up on linux distribs as a default. Otherwise I'd
suspect people with aic7xxx cards using some SCSI removables (mod/tape)
for backup like I do will dangerous problems.
For example, I use the mo disk on another linux system since ages. Thus I
did not expect a single problem and I have some important data archived on
MOD's. If they haven't been write protected (I'm not that stupid...), I'd
completely lost that data due to media change misdetection...
So, I mean, the aic7xxx has DANGEROUS bugs in it.
--
Michael Weller: eowmob@exp-math.uni-essen.de, eowmob@ms.exp-math.uni-essen.de,
or even mat42b@spi.power.uni-essen.de. If you encounter an eowmob account on
any machine in the net, it's very likely it's me.
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
