🚀 go-pugleaf

In article 11202, Theo de Raadt suggests that the registers in the PDP-11
were mapped onto memory, in the style of the PDP-10.

This was not the case.  These registers had bus addresses from which they
could be interrogated for diagnostic purposes from the front panel, 
(in a day when machines still had physical lights and switches), but
there was no practical way for programs to access these registers as 
anything but registers.

And, yes, they were implemented as registers, and not as memory, as de Raadt
implies. And no, you couldn't execute a program out of them.

BTW, I believe the Univac 1100 Series (1105/1107/1108 et al) also mapped
their registers onto the low order addresses in memory, a la the PDP-6 and
PDP-10.