Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!zephyr.ens.tek.com!tektronix!reed!busker!f8.n30112.z1.FIDONET.ORG!Howard.Spindel
From: Howard.Spindel@f8.n30112.z1.FIDONET.ORG (Howard Spindel)
Newsgroups: comp.sys.ibm.pc
Subject: Re: AT BIOS's supporting > 1024 cylinders
Message-ID: <1120.254FF5A5@busker.FIDONET.ORG>
Date: 1 Nov 89 17:01:43 GMT
Sender: ufgate@busker.FIDONET.ORG (newsout1.26)
Organization: FidoNet node 1:30112/8 - Spindel's House, Portland OR
Lines: 43

> chris@zorin.UUCP (Christopher Nielsen) write:
> Date: 30 Oct 89 06:30:39 GMT
> Organization: Zorin Data Systems, Inc., Santa Monica, CA
> 
> Interesting thing though... I tried to do with with a
> Microport installation.
> It seemed to work ok... until one day I filled up the disk
> enough to go
> (apperently) above 1024 cyls. Suddenly, things went berzerk.
> Fdisk then
> though I had 1-199 cyls in use (1224 - 1024?). I then used
> another disk
> of the same type, creating a 1024 cyl. partition. Then
> copied the kernel
> over and over until 0 blocks. no problem...
> 
> Anyway, I don't think the controller has any problems, cause
> I've seen it
> done. I was thinking either the BIOS, but then thought it
> unlikely, since
> there is a drive type supporting >1024 cyls. Could it be the
> partition
> table? (unix) that gave me problems? Don't know...
If your controller does not support > 1024 cylinders and your disk
has 1224 cylinders (which incidentally is the same Maxtor I use and
the same problem I encountered) what happens is that when your
software tries to access cylinders above 1024 the cylinder number
wraps mod 1024 and you wind up accessing cylinders 0-199 when you
wanted 1024-1223.  You then of course wind up trying to store two
sectors worth of information in a single sector.  It definitely
sounds like one of the sectors you overwrote was the partition table.
A controller which does not support > 1024 cylinders only has 10 bits
of cylinder address to play with and cannot generate a cylinder
address >= 1024.
Many of the original IBM AT with the WD1002 controller have this
problem.  I had to upgrade to a WD1003 to get my Maxtor drive to
work.


--  
Howard Spindel - via FidoNet node 1:105/14
	    UUCP: ...!{uunet!oresoft, tektronix!reed}!busker!30112!8!Howard.Spindel
	    ARPA: Howard.Spindel@f8.n30112.z1.FIDONET.ORG