[TUHS] Re: 4.3 BSD version in the Unix Archive

Markus Weber jmbw at nather.com
Sat Sep 27 06:24:27 AEST 2003


Stand/copy from tms(0,1) to ra(0,1) fails with a 'disk unlabeled' diagnostic
on a cold disk image. It turns out that NetBSD's disklabel(8) will create a
template disk image file that makes stand/copy happy, although I haven't
verified that the label is all that it should be.

The boot of the miniroot's GENERIC kernel does indeed fail. If only ra0 is
configured for simh, it takes the kernel a long time to decide that ra1,
ra2, and ra3 are offline. Having said that, I do not know how long this
takes on actual hardware. This part can be speeded up by supplying empty
image files for the remaing three disks, but in either case the kernel (and
not simh) subsequently segfaults (trap type 8, code = c0000200, pc =
8003fe16).

I haven't yet investigated what this diagnostic signifies. There are two
likely scenarios, the kernel either crashes late in the ra driver
initialization or while probing for the next device in line. At this point
I'm not prepared to argue whether this is primarily an emulation error or a
kernel bug.

Owing to the lack of Vax hardware or at least a compatible environment, you
have me at a disadvantage. Which function does pc = 8003fe16 correspond to?

Markus

-----Original Message-----
From: Michael Sokolov [mailto:msokolov at ivan.Harhan.ORG]
Sent: Friday, September 26, 2003 1:36 PM
To: jmbw at nather.com; tuhs at tuhs.org
Subject: RE: [TUHS] Re: 4.3 BSD version in the Unix Archive


Markus Weber <jmbw at nather.com> wrote:

> Clearly, stand boots. I'm hopeful that 4.3-Quasijarus0a will indeed run on
> simh.

Just be warned that I have heard from others that the kernel won't boot on
SIMH
due to sucky emulation.

> For some follow-up questions... Standalone format supports hp and up
disks,
> but simh only emulates RL and RQ controllers - which means that the Ultrix
> utilities to label a cold disk must be ported or recreated to the simh
host
> platform. Of course, it may well be that applying a hex editor to the disk
> image will suffice.

Arghh, why don't people just f-ing get it!  Standalone format is what is
known
as low-level format in the PeeCee world (magnetic level stuff, certainly not
needed on an emulator), and it has absolutely nothing to do with disk
labeling.
4.3BSD-Quasijarus0[a] does not have standalone disklabel, but RA82 is known
to
the compiled-in tables and thus will work unlabeled.  Also Ultrix' disk
labels
are not useful for 4.3BSD-Tahoe/Quasijarus, since the labels are stored in a
different place in a different format.  (Berkeley and DEC developed pack
labels
independently.  BSD stores them in the boot block, Ultrix stores them in the
root fs superblock, which has the disadvantage of requiring you to create
some
dummy root fs before you can write a label, even if you want a root fs of
different size.)

> Am I correct that Quasijarus supports RA82 disks (using the ra type) and
the
> TK50 tape (as a tms)?

Yes.

MS


---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.520 / Virus Database: 318 - Release Date: 9/18/2003



More information about the TUHS mailing list