[4598] in linux-scsi channel archive
Re: [(patch?)] W95 can handle SCSI errors - Linux can't
daemon@ATHENA.MIT.EDU (Eric Youngdale)
Mon Aug 24 00:08:01 1998
Date: Mon, 24 Aug 1998 00:06:04 -0400 (EDT)
From: Eric Youngdale <eric@andante.jic.com>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
cc: Andries.Brouwer@cwi.nl, linux-kernel@vger.rutgers.edu,
linux-scsi@vger.rutgers.edu, lnz@dandelion.com
In-Reply-To: <m0z8k07-000aNFC@the-village.bc.nu>
On Tue, 18 Aug 1998, Alan Cox wrote:
> > upon use of a good CDROM after the bad one.
> > The reason is that scsi_allocate_device() expects to be called
> > while io_request_lock is held, but sr_do_ioctl() calls it without
> > first claiming this lock.
>
> That may well explain the crashes I get if Im using the cd under load
> (the buslogic driver seems solid enough but I get double scsi free's
> or an instantanous reboot)
>
> > P.S. Is detailed information on the Adaptec 1542 available today?
> > For example, it would be nice to have an abort routine that
> > really worked.
>
> Ask Adaptec - they are supposed to be being Linux friendly now 8)
I was never able to get abort to work - the best I was ever able
to do was to not make matters worse. I did manage to crash the firmware a
couple of times, however. I always considered the firmware to be buggy in
the 1542, but I guess it wouldn't hurt to ask Adaptec. Then again, they
might not have anyone left who would even know.
FWIW, I still have some docs for the 1542 online. The
specifications for abort were *extremely* vague - it wasn't at all clear
what an abort would really do, and it never stated whether the host would
be notified in any way whether/if the abort succeeds.
-Eric
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
