[RFC] New cpu --- Morpho ms1

[Doug Evans]

Frank Ch. Eigler writes:
 > These are reasonable concerns in the abstract.  Let's not insist on
 > extra maintenance work on this old code until there is an indication
 > that another of those 20 different apps shows up, or that gdb/sim will
 > model delayed writes too.

I disagree.  There is value in avoiding obviously bad code, especially
instances that set a bad precedent.
I seriously doubt a reasonable solution is non-trivial.