[83] in linux-scsi channel archive

home help back first fref pref prev next nref lref last post

Re: Buslogic 8GB limit

daemon@ATHENA.MIT.EDU (Michael Weller)
Wed Feb 22 14:05:44 1995

Date: Wed, 22 Feb 1995 16:59:46 +0100 (MEZ)
From: Michael Weller <eowmob@exp-math.uni-essen.de>
To: linux-scsi@vger.rutgers.edu
Cc: jered@vorlon.mit.edu
In-Reply-To: <9502221545.AA28724@pollux.exp-math.uni-essen.de>

On Wed, 22 Feb 1995, Linuxusers of the Institute wrote:

> From: jered@vorlon.mit.edu
> Date: Wed, 22 Feb 1995 09:09:21 -0500
> To: linux-scsi@vger.rutgers.edu
> Subject: Re: Buslogic
> 
> Whoa, what's that about 8 GB? I have a Buslogic 946C and was looking
> at getting a 9.1 GB drive, I wasn't aware there was a 8 GB limit.
> Do you mean, that's as high as the remapping will go? What options do

Yes, I think that is what he wanted to say. This is the same for advanced 
adaptec remapping as well. No chance to access anything except the lower 8GB 
with DOS.

> I have?

Of course this doesn't apply to linux in theory. However you'll even for 
linux want to partition the disk with a linux-fdisk. I dunno if these
fdisk's have problems with such big disks due to limits in the DOS partition
struct on the disk. If linux accepts a disk w/o valid partition info
you could also make a single filesystem on the whole disk. I vaguely remember
that the ext2fs tools use special syscalls to access > 4GB or 2GB (32bit 
offset!) disk partitions. (But check that yourself! I may be wrong.)

#define flame on

If not, it was also suggested to introduce some new llseek
sys/libcall for exactly these large disk fdisk/mkfs/chkfs problems but got
toasted by some posix/bsd/sysv compatibility until you puke whiners..

#define flame off

Maybe even DOS would accept a 9.1 gig disk as long as it's own dos 
partitions are below 8GB. Dunno (I don't care for DOS anyway).. maybe 
someone with own experience on this topic can comment on that.

> Jered
Michael.

(eowmob@exp-math.uni-essen.de or  eowmob@pollux.exp-math.uni-essen.de
Please do not use my vm or de0hrz1a accounts anymore. In case of real
problems reaching me try mat42b@aixrs1.hrz.uni-essen.de instead.)

home help back first fref pref prev next nref lref last post