Subject: Re: Looking for cross-compiled disklabel or bootblocks 
In-reply-to: Your message of "Sun, 30 Oct 94 12:41:51 EST."
             <9410301741.AA08652@alpha.bostic.com> 
Date: Sun, 30 Oct 1994 12:18:20 -0600
From: "Michael Graff" <explorer@vorpal.com>


>Well:
>	(1) you need an /etc/master.passwd file, in addition to that
>		passwd file,
>	(2) you need to have the password db's made, too.  (This can
>		be accomplished with 'pwd_mkdb'

And /etc/services, /etc/protocols, etc.  The miniroot has enough to boot --
it doesn't seem to be intended to help bring the machine up any further than
that.

--Michael

--
Michael Graff <explorer@vorpal.com>       NetBSD is the way to go!
PGP key on a key-server near you!         Rayshade the world!

>From owner-port-pmax  Mon Oct 31 01:47:49 1994
  (5.67a/IDA1.5hp); Mon, 31 Oct 1994 00:42:56 -0600
From: Ted Spradley <tsprad@metronet.com>
Message-Id: <199410310642.AA24259@metronet.com>
Subject: Re: Looking for cross-compiled disklabel or bootblocks
To: explorer@vorpal.com (Michael Graff)
Date: Mon, 31 Oct 1994 00:42:55 -0600 (CST)
Cc: port-pmax@sun-lamp.cs.berkeley.edu
In-Reply-To: <199410301818.MAA03319@packrat.vorpal.com> from "Michael Graff" at Oct 30, 94 12:18:20 pm
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 3215      

> >Well:
> >	(1) you need an /etc/master.passwd file, in addition to that
> >		passwd file,
> >	(2) you need to have the password db's made, too.  (This can
> >		be accomplished with 'pwd_mkdb'
> 
> And /etc/services, /etc/protocols, etc.  The miniroot has enough to boot --
> it doesn't seem to be intended to help bring the machine up any further than
> that.

The miniroot I got from gregorio.stanford.edu had /etc/services but not
/etc/protocols or /etc/hosts (I made my own of that last but it seems to
ignore it, so I used the numeric octets).  I feel I'm well on my way to
a useable system, using Jonathon's binaries, Dean's cross-disklabel, and
advice from this list.

Using Dean's disklabel under Ultrix, I created 4 partitions on the disk:
root, starting at block 16 for about 16 Mbytes, then swap (about 16
Mbytes), a 5 Mbyte partition which served as a trampoline, and the rest
for usr.  Skipping all the stuff I had to try over and over before I got
it sorted out:

First, I had put Dean's bootblock and Jonathon's miniroot on the
first partition and booted it, newfs'ed and mounted the usr partition,
and, with the advice of Dean and Chris on 'pwd_mkdb', I ifconfig'ed le0
and rcp'ed Jonathon's /usr binaries onto it, with:

    rcp -rp 192.9.200.8:/partD/netbsd/usr /

but that left me with a root filesystem 98% full on a root partition
two-thirds empty, so I dd'ed the miniroot onto the 5 Mbyte 'd' partition
where it fit snugly.  I wanted to change the disklabel, but I couldn't
run disklabel -e under netbsd because I didn't have /v/tmp/vi.recover,
or TERM, or any of that, so I booted up Ultrix and ran Dean's
cross-disklabel once more.  All I did was change the partition letters
at the beginning of the lines, so now the partition starting at block 16
was 'd' and the 5 Mbyte partition I'd just put the miniroot on was 'a'.
halt Ultrix and boot NetBSD again.

The miniroot came right up again, I newfs'ed the 16 Mbyte partition,
mounted it on /mnt, copied netbsd and /etc/* from the current root
onto it, ifconfig'ed and rcp'ed Jonathon's /bin and /sbin onto it, and
tar'ed the /dev directory from the current root:

    tar cf - /dev | (cd /mnt ; tar xvf -)


Then I halted NetBSD and booted Ultrix so I could run Dean's disklabel
one last time to change the 'a' and 'd' partition letters back, then
halted Ultrix and booted my new NetBSD root.

As long as I'm willing to type numbers rather than names I can mount_nfs
filesystems from my server, and telnet to it, and most everything I've
tried works fine, within the limitations of what I know is missing.

Let me restate that.  I don't claim to _know_ what's missing, but I do
know that I should have a lot more stuff in /etc, I don't have a /var at
all, I don't even have /usr/include.

All this finally comes around to my question:  What next?  What's the
best way to proceed to flesh this thing out?

Thanks again, everybody.  When I finally got around to trying this stuff
everything came together MUCH easier than I imagined!



-- 
Ted Spradley                tsprad@metronet.com             +1-214-484-5356
Brisco: "...the more I learn the less I know."
Bowler: "At the rate we're learning things we won't know nothing in no time."

>From owner-port-pmax  Mon Oct 31 11:54:31 1994
	id AA02313; Mon, 31 Oct 94 09:52:47 MST
Message-Id: <9410311652.AA02313@fsa.ca>
To: Ted Spradley <tsprad@metronet.com>
Cc: port-pmax@netbsd.org
Subject: Re: Looking for cross-compiled disklabel or bootblocks 
In-Reply-To: Your message of "Mon, 31 Oct 1994 00:42:55 CST."
             <199410310642.AA24259@metronet.com> 
Date: Mon, 31 Oct 1994 09:52:19 -0700
From: Dean Huxley <dean@fsa.ca>

> but that left me with a root filesystem 98% full on a root partition
> two-thirds empty, so I dd'ed the miniroot onto the 5 Mbyte 'd' partition
> where it fit snugly.  I wanted to change the disklabel, but I couldn't
> run disklabel -e under netbsd because I did't have /v/tmp/vi.recover,
> or TERM, or any of that

There is another way to do this in the miniroot:

	mount /
	cd /tmp
	disklabel -r -r rz0 > foo
	ed foo
	disklabel -W rz0
	disklabel -R -r rz0 foo

I guess this information is a little late to help you, but I thought I would
let others know.  Actually, once the bootblocks and the miniroot are on the
disk, it is possible never to have to run Ultrix again.  Resizing the 'a'
partition requires a few disklabel/dd/reboot sequences, but it is possible ;-)

Dean.

>From owner-port-pmax  Wed Nov  2 12:31:34 1994
Message-Id: <199411021731.KAA00310@sidney.novatel.ca>
To: port-pmax@pain.lcs.mit.edu
Subject: ds3100.
From: Herb Peyerl <hpeyerl@novatel.ca>
Date: Wed, 02 Nov 1994 10:31:25 -0700


I took my disk over to Dean's house and we dd'd a decent disklabel and miniroot
onto my disk and tested it before unplugging. It booted.

So; this morning I come in; plug my disk into my ds3100 assuming it would work
and I got:

Excptn: <vtr=NRML>
Excptn pc: 0xbfc06aec
 etc ...

Is anyone else running this on a ds3100? If so, are you running it with the
colour display or serial? I'm going to try it with a serial console later 
today sometime just in case that's it...

Oh; also:

My bootrom comes up and says "KN01 V7.02" just in case that's useful information
to anyone..

Please be gentle, I'm a pmax newbie.

H.

>From owner-port-pmax  Wed Nov  2 13:28:42 1994
Message-Id: <199411021827.KAA21912@rurapenthe.ipd.wellsfargo.com>
To: Herb Peyerl <hpeyerl@novatel.ca>
Cc: port-pmax@netbsd.org
Subject: Re: ds3100. 
In-Reply-To: Your message of "Wed, 02 Nov 1994 10:31:25 MST."
             <199411021731.KAA00310@sidney.novatel.ca> 
Date: Wed, 02 Nov 1994 10:27:50 -0800
From: Ted Lemon <mellon@ipd.wellsfargo.com>


Does Dean also have a 3100?

I'm also having problems getting NetBSD to boot on a machine on which
I haven't tried to boot it before, so I suspect a subtle bug somewhere
in the machine initialization code, but haven't tracked it down yet.

			       _MelloN_


--
Ted Lemon		      Wells Fargo Bank, Information Protection Division
mellon@ipd.wellsfargo.com					+1 415 477 5045

>From owner-port-pmax  Wed Nov  2 14:39:08 1994
Message-Id: <199411021937.MAA00553@sidney.novatel.ca>
To: Ted Lemon <mellon@ipd.wellsfargo.com>
Cc: port-pmax@netbsd.org
Subject: Re: ds3100. 
From: Herb Peyerl <hpeyerl@novatel.ca>
Date: Wed, 02 Nov 1994 12:36:59 -0700

On Wed, 02 Nov 1994 10:27:50 -0800   Ted Lemon <mellon@ipd.wellsfargo.com> wrote:
 > Does Dean also have a 3100?

We're not sure... 

 > I'm also having problems getting NetBSD to boot on a machine on which
 > I haven't tried to boot it before, so I suspect a subtle bug somewhere
 > in the machine initialization code, but haven't tracked it down yet.

Well; as soon as I set mine to a serial console, it went along and booted.

There is a problem in that cons.c doesn't appear to know what the console
device really is because I get this:

>> boot -f rz(0,2,0)/netbsd
Boot: rz(0,2,0)/netbsd
Size: 772096+71536+88416
Starting at 0x80030000

And then at that point, the console switches back to the bitmapped display
and keyboard... But at least it's going... I might take a boo at that
later as soon as I manage to get the machine up enough to rebuild the 
kernel and add some swap space.



>From owner-port-pmax  Wed Nov  2 15:03:09 1994
Date: Wed, 2 Nov 1994 13:02:59 -0700
From: "Terry R. Friedrichsen" <terry@venus.sunquest.com>
Message-Id: <199411022002.NAA10205@venus.sunquest.com>
To: port-pmax@netbsd.org
Subject: re:  ds3100.


> >> boot -f rz(0,2,0)/netbsd
> Boot: rz(0,2,0)/netbsd
> Size: 772096+71536+88416
> Starting at 0x80030000

> And then at that point, the console switches back to the bitmapped display
> and keyboard...

FYI, this does *not* happen on a DECstation 5000/25.  The NetBSD kernel
quite happily talks to a serial console the whole way up, until it gives
me ENXIO errors while looking on the disk for init ...

This is using an Ultrix file system and an ECOFF /netbsd built from -current
plus Jonathan Stone's August patch kit.

Terry R. Friedrichsen

terry@venus.sunquest.com  (Internet)
uunet!sunquest!terry      (Usenet)
terry@sds.sdsc.edu        (alternate address; I live in Tucson)

>From owner-port-pmax  Wed Nov  2 19:31:30 1994
  (5.67a/IDA1.5hp); Wed, 2 Nov 1994 18:20:41 -0600
From: Ted Spradley <tsprad@metronet.com>
Message-Id: <199411030020.AA06346@metronet.com>
Subject: Re: ds3100.
To: hpeyerl@novatel.ca (Herb Peyerl)
Date: Wed, 2 Nov 1994 18:20:39 -0600 (CST)
Cc: port-pmax@sun-lamp.cs.berkeley.edu
In-Reply-To: <199411021731.KAA00310@sidney.novatel.ca> from "Herb Peyerl" at Nov 2, 94 10:31:25 am
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 924       

> Is anyone else running this on a ds3100? If so, are you running it with the
> colour display or serial? I'm going to try it with a serial console later 
> today sometime just in case that's it...

I'm using a DECstation 3100, with a monochrome display, 16 MB ram, RZ55
with Ultrix 4.2 + another Micropolis 1578 taken out of a Data General
machine with NetBSD.  I haven't tried anything with serial ports yet.

> My bootrom comes up and says "KN01 V7.02" just in case that's useful information
> to anyone..

Mine says "KN01 Vsomething".  I'll note it next time I take 'er down to
boot NetBSD.

> Please be gentle, I'm a pmax newbie.

I'm a *BSD newbie.  The closest I've been is trying to make sense out of
Ultrix.


-- 
Ted Spradley                tsprad@metronet.com             +1-214-484-5356
Brisco: "...the more I learn the less I know."
Bowler: "At the rate we're learning things we won't know nothing in no time."

>From dean@fsa.ca  Fri Nov  4 10:39:13 1994
Date: Fri, 4 Nov 1994 08:39:11 -0700
From: Herb Peyerl <Herb.Peyerl@sidney.novatel.ca>
Message-Id: <199411041539.IAA05939@sidney.novatel.ca>
To: port-pmax@pain.lcs.mit.edu
Subject: my machine.


Well; I got my machine up and running with proper partitions defined and
all.  I mounted a usr/src from a stupid i386 box and started to build myself
a kernel that has some swap defined in it.

After kludging a bunch of generic gunk, I get to the linking phase and get
tons of this sort of thing:

locore.o(.text+0x38): undefined reference to `mach_init'
locore.o(.text+0x54): undefined reference to `cpu'
locore.o(.text+0x58): undefined reference to `cpu'
locore.o(.text+0x5c): undefined reference to `fpu'
locore.o(.text+0x60): undefined reference to `fpu'
locore.o(.text+0x64): undefined reference to `main'
locore.o(.text+0x4bc): relocation truncated to fit: WDISP16 copyerr
locore.o(.text+0x4c4): undefined reference to `copystr'
locore.o(.text+0x4e8): relocation truncated to fit: WDISP16 copyerr

in a bunch of files.  I figure it's got to be a problem with the 'ld' in
the snapshot generated by Jonathan... All of these symbols exist and are
properly defined... 

The 'ld' I have says:

# ld -v
ld version 2.4 (with BFD 2.4)

Does anyone have any suggestions?

>From dean@fsa.ca  Fri Nov  4 13:12:22 1994
From: Patrick Schulz <patrick@eas.iis.fhg.de>
Date: Fri, 4 Nov 94 19:11:10 +0100
Message-Id: <9411041811.AA04725@samuel.eas.iis.fhg.de>
To: port-pmax@sun-lamp.cs.berkeley.EDU
Subject: can I use NetBSD on my machine ???
Cc: patrick@eas.iis.fhg.de
Reply-To: schulz%freia.inf.tu-dresden.de.patrick@eas.iis.fhg.de

Hi there,
Here at work I use a decstation 5000/200 as an X-terminal.
The machine has 8 MByte RAM ( really ) and runs ULTRIX 4.1.
The old Xserver isn't very stable and the system often swaps.

Yesterday I read some facts about NetBSD. Now I hope to use
NetBSD instead of ULTRIX. The smaller kernel ( I hope it's 
smaller then 4 MB ) and the shared libs could make work with
8 Mbyte RAM easier. Am I right ???

So what's the current status of the pmax port ? How many diskspace
will I need ?  ( I have a 285 MB disk ) Is there a port of X11R5
( with a server ) ? 

note: our firm won't upgrade this decstation, because we're using 
      suns. So I have to work with it as is.


Hope You may help,

Patrick.

 ----------------------------------------------------------
|Patrick Schulz; Hochschulstrasse 50 /0401;  01069 Dresden |
 ----------------------------------------------------------
  If you like a state of the art system - try NeXTSTEP !!!
 ----------------------------------------------------------
|         email: schulz@freia.inf.tu-dresden.de            |
|                ps3@irz.inf.tu-dresden.de                 |
|                patrick@eas.iis.fhg.de                    |
|                                                          |
|                No NeXT Mail, please !!!    		   |
| 				                           |
|           PLEASE: USE ONLY THIS EMAIL ADRESSES !         |
|              (other email could be ignored)              |
 ----------------------------------------------------------

>From dean@fsa.ca  Fri Nov  4 14:03:14 1994