> The original rtVAX used a special implementation of the MicroVAX II.
> As I understand it (and I've not seen any technical specs on the
> processor, so this could just be lunatic rantings), the change made
> for the original rtVAX was to not allow user-mode page tables to be
> paged.

Yeah, this sounds like the KA620, which was just a KA630 tweaked so
that user page table addresses (the things computed from the P0BR and
P1BR by adding user space page frame numbers) were taken as being
physical addresses, rather than system-space virtual addrseses.

As someone who once wrote a (severely stripped-down) kernel to run on a
KA620, I think, based both on documentation and experience, that this
was the only difference between the KA620 and KA630.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B