[...] rtVAX
1000, with a KA620 CPU board, [...]
It then fits in to the category of "very
interesting". ;)
It does. The KA620 is a relatively rare spin on the VAX architecture.
It's the board used for the non-boot CPU in the only case I ever played
with multiprocessor VAXen.
> The KA620 is closely related to the KA630
(MicroVAX II) with one small but $
Yes. P0 and P1 page tables are in physical space instead of kernel
virtual space. (That is, P0BR andf P1BR contain physical addresses,
and P0/P1 PTE access goes to physical memory, rather than kernel
virtual.) Relatively minor, but enough to ensure VMS wouldn't run on
it, which as I understand it was what DEC cared about at the time.
NetBSD seems to also have some /posible/ rtVAX
support.
I can't imagine it'd be all _that_ hard to tweak the VAX pmap as needed
for the KA620. Of course, that's coming from someone who's never tried
it, so it's pure speculation and probably worth what you paid for it.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at
rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B