[880] in linux-scsi channel archive
Re: MO media with 2048 byte sectors
daemon@ATHENA.MIT.EDU (Eric Youngdale)
Tue Nov 5 17:35:26 1996
Date: Mon, 4 Nov 1996 23:34:29 -0500
From: Eric Youngdale <eric@sub2317.jic.com>
To: Ronald.Wahl@informatik.tu-chemnitz.de
CC: linux-scsi@vger.rutgers.edu
CC: ruud.s@pi.net
In-reply-to:
<Pine.LNX.3.95.961011192503.2354C-100000@goliath.csn.tu-chemnitz.de>
(message from Ronald Wahl on Fri, 11 Oct 1996 19:31:47 +0200 (CEST))
A number of people have been asking for patches to allow MO media with
2048 byte sectors. I have some test patches which should get things working,
and I need some guinea pigs to try it out to make sure that it works.
Unfortunately I have absolutely no way at all of testing this right now.
In fact, I have to candidly admit that I haven't even tried to compile it,
so there may be minor syntactical errors which I have missed.
Part of the trick was that the partition table reading code needed to
use a larger blocksize for these devices. I believe that other people had
tried to add support for these devices, but without changes to the partition
table handling code, it still wouldn't work properly.
At the moment, I believe that ext2 is still broken with respect to
the larger sector sizes. Right now I just want to verify that we got the
basic handling of things at the block device level done correctly, and then
I want to turn it over to Remy or Ted to find out what the most appropriate
way of fixing ext2 would be. I don't think that the kernel should ever
panic, but I believe that it would refuse to mount anything from the large
sector discs.
The patches are up for anonymous ftp on
ftp://sub2317.jic.com/pub/scsi/2048.diff
-Eric
