Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!bionet!csd4.milw.wisc.edu!leah!rpi!crdgw1!ge-dab!peora!rtmvax!bilver!bill
From: bill@bilver.UUCP (Bill Vermillion)
Newsgroups: comp.unix.xenix
Subject: Re: 3.5" drive on a 386/compatible
Message-ID: <243@bilver.UUCP>
Date: 1 Jul 89 15:47:40 GMT
References: <24AA7624.213@marob.masa.com>
Reply-To: bill@bilver.UUCP (Bill Vermillion)
Organization: W. J. Vermillion, Winter Park, FL
Lines: 43

In article <24AA7624.213@marob.masa.com> daveh@marob.masa.com (Dave Hammond) writes:
>I'm interested in adding a 3.5" drive to my vanilla 386 (running SCO 2.3.1),
>to both read and write 3.5" disks.  I've checked through the 2.3.1 release
>notes, but this topic appears not to be covered.
>
>Does Xenix know about 3.5" drives (ie does the 3.5" use the standard floppy
>driver), or must the manufacturer supply a driver?
>
Check the HW section of the manual under FD.
I have added 3.5" disks to an AT to be able to exchange disks with 3.5"
IBM-80's at the same site.  Your CMOS must support the drive under setup.  ( I
had to get a later version of the OS for the AT machine).  Then just do the
mknod as shown under (HW) fd.

>On a similar note -- does anyone know if standard Xenix binaries run on
>the PS/2?  For that matter, do any differences between the PS/2 and
>standard 386's affect user software?
>
All I did was copy the Xenix 286 stuff to the 3.5" disks and run them.  The
only apparent difference between the PS/2 and the standard 386 is the portion
of code that talks to the hardware.  If you look at a full distribution for
the 386, the only disks that are marked 386/PS2 are in the runtime package.
All the others are marked n86 (for any of the iNTEL chips) or k286 for the
graphics package.  A large part of the code is 286 "stuff" or 8086 "stuff".

One comment.  I moved a 286 package to the 386 side and noticed a program that
took 12 hours to complete did it in 5 hours.  Got the '386 version of the
software and it went to 2.5.  (This was a one-time only re-crunch of a fairly
large data base).
((It was SUPPOSED to be one time - but  --- that's how I found out the
different timings.  We actually went through it about 5 time - you know the
old shooting at a moving target - as those who got the output suddenly set
"but we also need ..."  ))

bill
>--
>Dave Hammond
>daveh@marob.masa.com


-- 
Bill Vermillion - UUCP: {uiucuxc,hoptoad,petsd}!peora!rtmvax!bilver!bill
                      : bill@bilver.UUCP