[5035] in linux-scsi channel archive
Re: DC390/AM53C974 driver test
daemon@ATHENA.MIT.EDU (Kurt Garloff)
Sat Oct 31 19:21:31 1998
Date: Fri, 30 Oct 1998 22:22:47 +0100
From: Kurt Garloff <garloff@kg1.ping.de>
To: "Pursel, Frank" <frank_pursel@merck.com>,
"'K.Garloff@ping.de'" <K.Garloff@ping.de>,
"'linux-scsi@vger.rutgers.edu'" <linux-scsi@vger.rutgers.edu>
Mail-Followup-To: "Pursel, Frank" <frank_pursel@merck.com>,
"'K.Garloff@ping.de'" <K.Garloff@ping.de>,
"'linux-scsi@vger.rutgers.edu'" <linux-scsi@vger.rutgers.edu>
In-Reply-To: <199810302030.VAA01779@kg1.ping.de>; from Pursel, Frank on Fri, Oct 30, 1998 at 02:40:42PM -0500
On Fri, Oct 30, 1998 at 02:40:42PM -0500, Pursel, Frank wrote:
> Kurt,
>
> Ok, it's clearly part my fault. I'm running a patched (only with 2.0a)
> Linux 2.0.35 on a COMPAQ XL 550, so I know the PCI BIOS is mucked up. The
> question seems to be, how do I specify my IOBASE and IRQ for my AM53C974 so
> that it can be found at bootup. The tmscsim=-1 doesn't contain any of this
> info and since the driver isn't found on boot, there isn't a useful
> /proc/scsi/tmscsi/7 entry to read or write to.
>
> I have read that if I go into DOS first I can execute a MEMOVE procedure
> that remaps the PCI BIOS and may fix this problem. I don't want to do that,
> it's ugly. The old AM53C974 version 0.5 driver didn't require this to find
> the host adapter. I am hoping the new one won't require it either.
>
> Any suggestions?
Hi Frank,
thanks for your report.
Unfortunately, I don't know what 's the problem with a Compaq XL 550. From
your description I conclude that the PCI BIOS is broken.
Normally the PCI BIOS should configure your PCI devices, i.e. assign IRQs,
I/O space and mmap regions. And it should provide an interface to the
devices (PCI configuration space).
I don't know exactly what's wrong with you BIOS, here.
What are exactly the messages from your kernel, espec. the ones from the PCI
and DC390 and SCSI subsystem?
The driver relies on the kernel's PCI functions to access the AM53C974. If
the kernel's PCI functions aren't able to find any AM53C974 PCI chip, the
driver won't work. Have a look at the detect() function of the tmscsim.c
driver. Also have a look at the AM53C974 driver. They use almost the same
mechanism. So, I'm really astonished that your AM53C974 driver found the
adpater and the tmscsim driver doesn't. Strange!
Note that with 2.1 kernel you can select between direct accesses to the PCI
config space and the access via BIOS. Maybe selecting one of them will help
you. (If you want a kernel recommendation: 2.1.125 is great.)
Hope this helps you a little bit.
--
Kurt Garloff <K.Garloff@ping.de> (Dortmund, FRG)
PGP key on http://student.physik.uni-dortmund.de/homepages/garloff
Unix IS user friendly - it's just selective about who its friends are!
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
