Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 9 Jan 95 10:31:43 MST
From:      terry@cs.weber.edu (Terry Lambert)
To:        amy@physics.su.OZ.AU (Shaun Amy)
Cc:        questions@FreeBSD.org, amy@physics.su.OZ.AU
Subject:   Re: FreeBSD 2.0R and ESDI drives
Message-ID:  <9501091731.AA02173@cs.weber.edu>
In-Reply-To: <199501091102.AA10442@physics.su.OZ.AU> from "Shaun Amy" at Jan 9, 95 10:02:44 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> 
> I am hoping that someone else may have seen this problems and the solution has
> already been found.  Essentially I cannot get 2.0R to install onto an ESDI
> drive in an Olivetti XP9 (386 based).  I was quite willing to believe it was
> some form of incompatability with the Olivetti and FreeBSD and ESDI until I
> decided to try installing 1.1.5.1 which installed first go and ran without
> problems.  Thus it appears that something has broken in 2.0R which worked in
> 1.1.5.1.

[ ... ]
>   Disk 0 -  Type 31 (304MB)
>             Cylinders                       1409             1412      813
> 
>   Disk 0 -  Type 33 (136MB)
>             Cylinders                       1221             1224      819

Both of these drives are in excess of 1024 physical cylinders.

You are supposed to use the BIOS geometry to install the disklabel to make
the boot happy, since BIOS can't handle more than 1024 cylinders.

Since you note that you have a WD1007 interface, check to see if it is
jumpered for "perfect media" or not.

The WD1007 comes in several flavors, and one of these uses a non-linear
translation (ie: it reserves replacement blocks in the middle of the
media to make it look perfect).

On top of that, there is the problem of BAD144, the bad tracking algorithm
(which doesn't work in 2.0).

The final possiblity is that 2.0 computes it's disklabel geometry from
an existing DOS partition on the drive -- this is kinda bogus, unless
your C/H/S values are all prime numbers -- which may result in bogus
offsets when the protected mode code attempts to mount root and it's
not at the expected absolute sector offset.  This is also a change from
the 1.x code.


Any one of these could be your problem.


					Terry Lambert
					terry@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9501091731.AA02173>