[468] in linux-scsi channel archive
Re: Generic Scsi Driver Bug Report and comments.
daemon@ATHENA.MIT.EDU (Michael Neuffer)
Thu Aug 3 20:07:16 1995
Date: Thu, 3 Aug 1995 15:18:35 +0200 (MET DST)
From: Michael Neuffer <neuffer@goofy.zdv.uni-mainz.de>
To: "H. Peter Anvin" <Peter.Anvin@linux.org>
cc: submit-linux-dev-scsi@ratatosk.yggdrasil.com
In-Reply-To: <hpa.301f02d9.Linux.is.free@asgard.yggdrasil.com>
On 2 Aug 1995, H. Peter Anvin wrote:
> Also, it still doesn't solve what I think is the #1 problem with SCSI
> as we know it, which is that swapping out one card with a different
> type in a machine with several controllers can cause the device
> numbers to reverse. Hence, there will always be an element of
> dynamicity no matter how hard we try to avoid it.
With a dirty hack and little "command-line magic" we could add a
possibility to modify the sequence in which the drivers are called for
probing.
Then a "scsisequence=eata_dma,NCR53c710xx" addition to the kernel
command line would cause the eata_dma driver and it's hosts to be probed
and the devices registered before the the ones on the NCR driver.
Mike
