*BSD News Article 3969


Return to BSD News archive

Path: sserve!manuel!munnari.oz.au!uunet!dtix!darwin.sura.net!jvnc.net!yale.edu!yale!gumby!destroyer!ncar!noao!amethyst!organpipe.uug.arizona.edu!afthree.as.arizona.edu!tom
From: tom@afthree.as.arizona.edu (Thomas J. Trebisky)
Newsgroups: comp.unix.bsd
Subject: Re: 680x0 version of 386BSD?? [Was: Re: Mac version of 386BSD??]
Message-ID: <1992Aug20.173817.21681@organpipe.uug.arizona.edu>
Date: 20 Aug 92 17:38:17 GMT
Sender: news@organpipe.uug.arizona.edu
Organization: University of Arizona, Tucson, AZ
Lines: 48

gab10@griffincd.amdahl.com (Gary A Browning) writes:
>It is not just drivers that need to be rewritten.  Have you though about
>changes required for the different hardware interrupt stack frame,

Yes

>interrupt masking structure, the MMU (BTW, the 68000 machines do not
>usually have one of these.

Any 680x0 system you would want to run unix on will have an MMU. It is
just not on chip (unless you have an 030 or 040)  Motorola made the
68851 mmu coprocessor for the 68020, but at lot of folks (like sun and HP)
built their own custom MMU's external to the cpu.  You can even run unix
without an MMU on a vanilla 68000, but I won't go into the horrid schemes
involved in doing that.

The 680x0 interrupt structure is just the other way around, it is on-chip
instead of the off chip PIC situation with the 80x86.

>This makes the porting task much harder, not easier.

I am not sure what you mean.  But I never claimed it was easy.  :-)

>You should read the series of articles from Bill and Lynne in Dr Dobb's
>Journal about the porting of 386BSD.  You are basically doing the same
>task for the 680x0 architecture as he has for the 80[34]86.

>After reading the articles,
>you may change your mind and decide to wait and start with BSD4.4 which
>already has some support for various 680x0 machines.

NO way!  When the going gets tough, the tough get going!  Besides, there
is no need to wait, the hp300 stuff in the Net2 code supports the
HP 68020 and 68030 workstations, so I (We) have a lot more to go with than
Jolitz did for the 386.

In fact if I had an hp300, I could just about open the can, heat and eat.
I get the impression the guys at Berkeley actually had hp300's on their
desks, but I could just be starting a rumor.

But this doesn't mean it will be easy.  If it was easy, I would drop it
and go find something hard to do.  :-)

--
	Tom Trebisky	ttrebisky@as.arizona.edu
....."There's no sense in being precise when you don't even
..... know what you're talking about." 
                       - John von Neumann