On 21/09/15 14:15, Noel Chiappa wrote:
From: tony
duell
In some cases it should be possible to write a
machine code program
that executes on 2 processors with wildly different instruciton sets.
I have this bit set that I was told (or something, the memory is _very_
vague) that early versions of the KL-10 had this hack; the root block on the
disk was the boot block both the PDP-10 and the PDP-11 front end machine, and
the first instruction or two was very cleverly construced and sent the two
machines different ways. Alas, I looked in the front-end PDP-11 code (in the
KLDCP; directory) and saw no signs of this, so maybe it was an urban legend?
I can't find a definitive reference right now, but I *think* that the
ODS-1 disk format
was first used on the PDP-11 and then later used in early versions of
VMS. I *think*
that it was arranged such that a PDP-11 booting and a VMS system booting
could
be done from the same disk by arranging for each to interpret the boot
block in
a way that each was happy with.
Antonio
arcarlini at
iee.org