This is a multi-part message in MIME format.
--------------010009060107010304050302
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Hmm. It appears there is something causing conflicts in the Generic kernel.
I built one with only the specifics for the machine I have, and that 
kernel detects everything and boots fine.
The previous 3.99.15 kernel was also specifically tailored to the 
hardware, so this might be a problem that have been around for a long time.

Another thing I noticed was that boot have become broken. It causes a 
severe machine check followed by a full cpu snapshot generated. The boot 
from 3.99.15 works fine.

	Johnny


Johnny Billquist skrev:
> Yes, it's a real 8650. And yes, I have problems. :)
> I've been keeping the machine off for some time, but recently decided to 
> play with it again. Built a new kernel and booted it, but here comes 
> trouble.
> 
> The hardware is as follows:
> 
> 3 Unibuses.
> uba0 have one UDA50 and one DELUA.
> uba1 have nothing
> uba2 have one UDA50 and one TU81.
> 
> The TU81 don't have a drive attached to it, so it won't initialize.
> The machine boots 3.99.15 just fine. It detects the UDA50 (both of 
> them), and the DELUA.
> -current from a week ago don't boot fine. It fires up and everything 
> looks good. However, during the autoconf stage it fails to identify the 
> UDA50 on uba0. It identifies everything else just fine. However, since 
> the system boot disk is on that UDA50 you can understand that it's not a 
> happy system.
> 
> I have been poking around a little, but perhaps someone who knows the 
> innards of a few things here can help me out. Otherwise I'll have to dig 
> through a lot more of the innards I'm not that familiar with.
> 
> What I have observed is that the configuration scanner notice that the 
> UDA50 is there, and it starts the initialization steps on it. Slightly 
> later I get an error on the unibus. The unibus error register says that 
> I accessed an invalid map register. So far I'm following it. If we're 
> trying to play with the UDA50 via an invalid unibus map register of 
> course it will fail. Now, where are the unibus map registers set up, and 
> why isn't it setup right? Since the UDA50 on uba2 works, it can't be a 
> outright error, but something a bit more magic. Also, what Unibus access 
> are we talking about? The first access to the CSR works fine, so it's 
> something else. Interrupt?
> 
> Ragge, you remember how this works, and feel like debugging? Anyone 
> else? I can setup remote access to the console if that would help.
> 
>     Johnny
> 

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt@softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol

--------------010009060107010304050302
Content-Type: text/x-vcard; charset=utf-8;
 name="bqt.vcf"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
 filename="bqt.vcf"

begin:vcard
fn:Johnny Billquist
n:Billquist;Johnny
email;internet:bqt@softjar.se
tel;cell:+46 70-8641486
x-mozilla-html:FALSE
version:2.1
end:vcard


--------------010009060107010304050302--