>>>>> "Michael" == Michael van Elst <mlelstv@serpens.rhein.de> writes:

Michael> dgilbert@jaywon.pci.on.ca writes:
>> No such luck.  They had some form of MMU on it.  The 68000 was
>> designed to run things like unix.

Michael> No. The 68000 does not support (transparent) virtual memory
Michael> because a page fault cannot be handled for every user-mode
Michael> instruction.  The 68010 extended the bus-error exception so
Michael> that any user-mode instruction could be continued where it
Michael> was aborted.

Michael> I believe the Sun-2 used 68010s.

Michael> It is possible to do VM on the 68000 if you avoid certain
Michael> instructions. You also need to emulate some instructions in
Michael> the page-fault handler.

Well with 2 68000, running in parallel with one of them an instruction
behind, page-faults just started the up the lagging 68000, and
reversed roles of the CPUs.  Apollo used this technique I seem to
recall.

Niklas

Niklas Hallqvist       Phone: +46-(0)31-40 75 00  Home: +46-(0)31-41 93 95
Applitron Datasystem   Fax:   +46-(0)31-83 39 50  Home: +46-(0)31-41 93 96
Molndalsvagen 95       Email: niklas@appli.se     GSM:  +46-(0)70-714 10 35
S-412 63  GOTEBORG     WWW:   Here
Sweden		       IRC:   niklas (#NetBSD)    ICB:  niklas (netbsd)