Subject: Re: NetBSD crashes on boot
To: None <tm@squonk.net>
From: Chris G Demetriou <Chris_G_Demetriou@ux2.sp.cs.cmu.edu>
List: port-alpha
Date: 11/03/1996 01:16:50
Sorry for the delay in replying, but i've been out of town.  Anyway...

> I'm trying to get NetBSD running on my Alpha Multia.  I'm able to load
> NetBSD, but it crashes soon after loading.  :-(

Hmm.  I see that you've got one of the 233MHz Multias.  Is anybody
else running NetBSD in one?  (All of the multias have the same CPU,
the 21066, right?  I've got a 166MHz one here, but can't crack it open
right now to double-check...)

Have you updated your firmware to the latest revision?  It (along with
instructions) should be somewhere under:

	ftp://ftp.digital.com/pub/DEC/Alpha/firmware

I seem to recall somebody (John Birrell, maybe?) running into problems
like these a long time ago...  If you've not tried to upgrade your
firmware, i'd strongly suggest that you do ASAP.  You should be able
to put a firmware image on to a floppy with another UN*X machine or a
DOS system.

Please tell me (or the list) the results of booting with the new
firmware.


> The following is a log of what happens.  BTW, if it makes a difference,
> it's a headless system.  Any ideas on what's wrong?  Also, is there 
> anything that can be done about the unknown vendor complaints?

Headlessness shouldn't make a difference.  See above for one idea on
what's wrong, but I dunno why that would cause the symptoms you're
seeing.

All of the 'weird' kernel messages (from the unkonwn 0x200d/0x00c2
devices to the apparent pcivga recognition) seem bogus.  Also, a 'tga'
should be been recognized, since the Multia includes a TGA frame
buffer (and it should have shown up even if you're not using it).
Note also that the thing which claimed to be a PCI VGA board
apperently wasn't responding to cursor-position-query I/O port
accesses in a very 'normal' manner (which makes sense, since there
wasn't any real PCI VGA board there! 8-).

The many lines of register dumps and machine checks, etc., are caused
by the firmware losing.  This lends (a bit of) credence to the theory
that the firmware is to blame.


Try the latest firmware, please, and get back to me/us...


chris