Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!rochester!udel!burdvax!sdcrdcf!CAM.UNISYS.COM!mrc
From: mrc@CAM.UNISYS.COM.UUCP
Newsgroups: comp.sys.ibm.pc
Subject: Re: Question about Xebec HD controller
Message-ID: <387@sdcjove.CAM.UNISYS.COM>
Date: Thu, 12-Nov-87 16:18:50 EST
Article-I.D.: sdcjove.387
Posted: Thu Nov 12 16:18:50 1987
Date-Received: Sat, 14-Nov-87 16:51:16 EST
References: <879@csun.UUCP>
Distribution: na
Organization: Unisys McLean Research Center
Lines: 43
Keywords: Xebec HD
Summary: OOPS

In article <879@csun.UUCP>, abcscnge@csun.UUCP (Scott Neugroschl) writes:
> 
> We are trying to install a second hard disk in a true-blue IBM XT.  We
> have 2 10 MB hard disks, 2 Xebec controllers.  We would like to install
> both HD's on one controller, but we don't have the proper cabling, so we
> are going to use 2 controller boards.  Our problem is this:  our in-house
> technical support at work doesn't have the documentation, so when we install
> both boards into our PC, they are both at the same I/O address and ROM address,
> thereby causing an obvious contention lockup when drive C is accessed.  How
> do we set the ROM ADDR, I/O ADDR and OPTION jumpers to configure one drive
> and controller as C and the other as D?
> 
> Thanks in advance...
> -- 
> Scott "The Pseudo-Hacker" Neugroschl
> UUCP: {litvax,humboldt,sdcrdcf,rdlvax,ttidca,}\_ csun!abcscnge
>       {psivax,csustan,nsc-sca,trwspf         }/

WHOA, WHOOPS, OOPS, and LOGICALGOTOHELL!

You are up against more than a matter of plugging jumpers or twiddling
dipswitches...   You are up against the fundamentals of the IBM-PC
architecture and ROMBIOS.

You cannot! have two hard-disk controllers, not even two floppy-disk
controllers, in any PC or PClone.  The BIOS provides for ONLY ONE
hard-disk-start interrupt vector and ONLY ONE floppy-disk-start
interrupt vector.  That is, only one each of the magic addresses in
low-end memory which tell bios where to call with the parameters for a low-level
I/O operation.
 
Given that you do not want to rewrite the whole bloomin' BIOS, you can't
get there ( 2 hd-controllers) from here (IBM PC).

There are also other issues, like controller-alert-to-CPU interrupt numbers
and vectors, and BIOS Parameter Block vectors, which might be manageable
--which are managed by single-controller multi-type and multi-drive
setups--but the controller-request vectors are the killers.

OH Well...

regardz,
Ken