[967] in linux-scsi channel archive
Re: Adaptec -> NCR
daemon@ATHENA.MIT.EDU (Gerard Roudier)
Mon Nov 18 22:24:08 1996
Date: Mon, 18 Nov 1996 22:41:57 +0000 (GMT)
From: Gerard Roudier <groudier@club-internet.fr>
To: "Leonard N. Zubkoff" <lnz@dandelion.com>
cc: stephen@it.com.au, linux-scsi@vger.rutgers.edu
In-Reply-To: <199611172321.PAA01642@dandelion.com>
On Sun, 17 Nov 1996, Leonard N. Zubkoff wrote:
> I'n not sure precisely what you're saying... Moving a disk from Adaptec to
> BusLogic will always work, since the Adaptec translation set is a subset of the
> BusLogic set, and the BusLogic BIOS will adopt. Moving a disk that's >= 1GB
> and < 2GB from BusLogic to Adaptec won't be correct if it was formatted with
> BusLogic's 128/32 translation. Personally, I always turn off the > 1GB switch
> and create partitions with the 64/32 translation since it's the most common
> across all SCSI host adapters.
Leonard,
Thanks for you accurate explanations about BusLogic BIOS mappings.
The scsicam_bios_param() is used at least by the following Linux scsi
low-level drivers:
- eata
- AM53C974
- 53c7,8xx and ncr53c8xx
- aha152x with some changes
and perhaps others if I missed my "grep" command.
This routine looks like a _general_ service of the linux scsi stuff.
For the moment, I did not understand why scsicam_bios_param() may fail
to find the geometry from the partition table when BusLogic specific stuff
is able to succeed.
Have you some explanations of that?
Are such failures ncr specific?
I am not very knowledgeable about MS/DOS bogosities and I will probably
need time to understand that code.
I just hope that the problem is not due to a bug, being given the number
of involved low-level drivers.
Regards, Gerard.
