*BSD News Article 5560


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!uunet!sun-barr!olivea!spool.mu.edu!umn.edu!math.fu-berlin.de!news.netmbx.de!Germany.EU.net!rrz.uni-koeln.de!unidui!du9ds3!veit
From: veit@du9ds3.uni-duisburg.de (Holger Veit)
Newsgroups: comp.unix.bsd
Subject: Re: old bios's & >100Meg IDE
Message-ID: <veit.717353566@du9ds3>
Date: 24 Sep 92 16:52:46 GMT
References: <19qk5fINNsvb@aludra.usc.edu>
Reply-To: veit@du9ds3.uni-duisburg.de
Organization: Uni-Duisburg FB9 Datenverarbeitung
Lines: 56
NNTP-Posting-Host: du9ds3.uni-duisburg.de

In <19qk5fINNsvb@aludra.usc.edu> eddy@aludra.usc.edu (George Edmond Eddy) writes:

>Background:

[...]
>Problem/Question:

>Here at work i mentioned my problems to someone that in turn said
>that we had PCs that needed old bios's replaced for systems running
>IDE drive with greater than 100MG HD's  does anyone know anything
>related to this?

This is unfortunately true. After installing my 420 MB Fujitsu disk and
modifying the BIOS (see below) no DOS-EMM386 memory manager seemed to work 
together with the disk (even QEMM failed; a very strange effect, which could
be corrected by OS/2, which patches the BIOS). These flaws do not have
any effect on 386bsd which is not interested in the BIOS after loading.

>BTW - I have a noname mother board with a Pheonix bios dated 1988.
>i *think* it's revision is 1.1 (or something similar).  As matter
>of fact my drive type 47 is NOT configurable, and it does not have
>a type 48.

This is really an old BIOS, indeed. I myself also have such a beast in
my PC. It might cause lots of trouble to replace the BIOS with a new one,
as it might also require replacing the keyboard controller, and even then
it may not like your chip set, and you end up buying a new motherboard.
If you can burn EPROMs, you might consider the following way: Read out
(simply with DOS debug) the contents of the ROMs into a file and modify
any of the 47 disktab entries to match your disk parameters. Now the tricky
part: modify any other disktab entry counterwise, eg. if you increase 
the number of heads from 5 to 8, then in another entry you have to set them
from 9 to 6, for instance. Write a simple program to calculate 8 and 16bit
checksums (I found out that my Phoenix BIOS uses 16 bit checksums) and
verify that the old and new checksums match. You might find a difference of
+- 1, e.g. in my BIOS: after "proper" correction the checksum was CBFF instead
of CC00); you can adjust this by modifying an arbitrary byte in the disktab
(experiment with it). Try out the new EPROMs.

This corrects the problem with the missing configurable entry, and is the
cheapest solution. To circumvent the "100MG" (#define G B) problem, I 
recommend using OS/2... no, I recommend 386bsd ;-)

>Thanks anyone in advance!

>- rusty
>--
>eddy@usc.edu

Holger

-- 
|  |   / Dr. Holger Veit         | INTERNET: veit@du9ds3.uni-duisburg.de
|__|  /  University of Duisburg  | "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
|  | /   Dept. of Electr. Eng.   |   Sorry, the above really good fortune has
|  |/    Inst. f. Dataprocessing |      been CENSORED because of obscenity"