a little off topic, but..

does anyone have any ideas why when i frob the bits to set status on a 
TMSCP controller, bad things would happen? it seems that somehow this
is causing the PSL to get the 'running on interrupt stack' bit set,
(actually , it is causing a very different psl status to be set)  
which then causes a crash to the chevron with ISP ERR, from an access 
violation. it's happening while probing tmscp devices on a Qbus. I dont
have any documentation other than kernel sources for whether this should
generate an interrupt or something which might set the PSL. does this
make any sense to anybody? 
this is part of my 4.3tahoe on the 4000/600 work, i had hoped that the
Qbus, etc, would be identical to ka6[35]0.. 

Isildur