Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!cs.utexas.edu!samsung!zaphod.mps.ohio-state.edu!rpi!uupsi!sunic!tut!jt19840
From: jt19840@tut.fi (Tuomi Jyrki Juhani)
Newsgroups: comp.sys.ibm.pc
Subject: Re: How to reset CMOS configuration that prevents booting?
Message-ID: <3777@tutor.tut.fi>
Date: 30 May 90 06:37:49 GMT
References: <9005271315.AA06648@jade.berkeley.edu> <1990May30.015641.15522@bach.amd.com>
Reply-To: jt19840@tutor.tut.fi (Tuomi Jyrki Juhani)
Organization: Tampere University of Technology, Finland
Lines: 34

In article <1990May30.015641.15522@bach.amd.com> phil@pepsi.amd.com (Phil Ngai) writes:
>In article <9005271315.AA06648@jade.berkeley.edu> DLV@CUNYVMS1.BITNET writes:
>|As Jeff pointed out, C&T memory is separate from the configuration CMOS, so if
>|I had cleared that by disconnecting the battery, it would not have helped at
>|all---good thing I did not.
>
>Say what! I did not know this. Are you sure?

The term 'C&T memory' is a bit vague in this context.  The C&T chipsets have
various configuration *registers* that can affect thins like bus timing and
memory an I/O wait states.  These registers have certain *initial values*
a power-on reset.  A BIOS for C&T-based 386 motherboard may initialize
(and usually does) initialize these registers to some other values.
The BIOS looks at certain locations in CMOS RAM to see if there are valid
initial values for C&T registers.  If not, then the BIOS initialization code
loads the C&T registers from a table stored in the BIOS ROM.

How do the C&T initial values get stored in the CMOS RAM?  Well, usually
a BIOS has a special setup for C&T registers (or you can use a C&T program
called SETUP386), which accepts user's modifications and writes them in
CMOS along with a checksum.

I have deducted these facts after disassembling two different AMI BIOSes
for C&T/386 motherboards, and I suspect strongly that they are valid also
for Phoenix, Award, etc.

So, if you could have invalidated the CMOS contents (e.g. by
disconnecting the battery), the BIOS would not have used the defaults
stored in CMOS, and the system would have booted all right.

-- 

Jyrki Tuomi
Internet:  jt19840@tutor.tut.fi    UUCP:  ..mcvax!tut!jt19840