At 09:42 PM 12/26/00 +0100, Johnny wrote:
>That being said, I'm still having problems with my machine. I get hangings
>in biowait which never wakes up again. I suspect it might be a problem
>with my CMD CQD-220 SCSi controller, but I'm also suspicious about the
>MSCP driver...

Yup, this was a problem from the 1.4 days on my uVAX II with the Sigma 
RQD11D controller. There is a bug in the MSCP code whereby the state 
machine in the MSCP code has either missed an interrupt or over written 
some buffer somewhere and it goes south. Once it hits processes start dying 
(especially if you need to swap a lot) until the system is dead.

I've looked at the MSCP code a couple of times (and initially plugged my 
neoDSSI code into that module rather than figure out how config worked :-) 
but changed controllers which took it off my "hot" list.

I'll probably end up re-implementing MSCP for the DSSI code if someone 
doesn't beat me to it.

--Chuck