Subject: Re: UNIBUS and Q-bus
To: None <port-vax@NetBSD.ORG>
From: Allison J Parent <allisonp@world.std.com>
List: port-vax
Date: 03/23/1998 16:11:50
<A second ball of hair is the J-11. If you have a full 4Mb, you can
<address the last 8K as instruction without frobbing BSS7. Only
<D-space references assert BSS7, or sould I say, only references
<reloceated by D-space PAR/PDRs assert BSS7. This is usefull,
<a it give you another 4kw of instuctions for your driver!
<Does nothing for data or poll space alas.

This is where the "system" has to be written defensively and any device 
probes done before turning on the mapping hardware.  It likely applies 
to 11/70s and possibly VAX via UBAs (or the qbus equivilent) as well.

This the sort of stuff DEC would call possible but unsupported as it's
possble to utilize it in a working system but don't ask them to figure 
it out(under contract or warrenty) if it doesn't work or stops working.

It's also a booby trap supporting older hardware as RLV11 is real old 
and offers nothing over the RLV12.  This would apply to a few other 
peripherals that come from the PDP-11 U/Q space.  For netBSD supporting 
RLV11 is pointless.  For the rare user that needs RL01/2, the RLV12 is 
the route and maybe a driver lofted from PDP-11 unix.  

Getting full MSCP, VS2000, and good SCSI support makes more sense and 
better use of effort for VAX.  There are plenty of things that need work 
without hunting for old bombs.


Allison