On 823087885 seconds since the Beginning of the UNIX epoch der Mouse wrote:
>
>> How stupidly expensive would CPU emulation be, anyway?
>
>Depends on how much effort and code you want to pour into it. :-)
>
>> I know I've seen software 8088's on 68000 boxes before, slow though
>> they may be.
>
>I've seen a software 80286 on some recent (non-68k) Macintosh.  It's a
>commercial product called, I think, SoftPC, and I'm told a software
>80386 also exists.  And I have a PDP-11 written in C that runs on SunOS
>and probably would be easy to port to other OSes.  It could be speeded
>up immensely by compiling the -11 code to native code instead of doing
>pure interpretation, but of course that would then be sparc-specific
>instead of the current more or less MI program.
>
>					der Mouse
>
>			    mouse@collatz.mcrcim.mcgill.edu

If one could compile the native -11 code into native code, then
maybe one could just translate it into C?  Perhaps not very
efficient C code...  But then one would have a MI idea.

This has actually been bugging me for a bit, and although I
have not thought it through incredibly, I have been wondering
if there is a horrible difficulty with this approach.

(You could try to take it farther, and try to port non-unix things
this way, although I do see a lot of problems with things that
twiddle hardware/etc.  Perhaps if you considered having a 'virtual
machine' and how a few pieces of hardware worked, you could then
guess what their twiddling was trying to accomplish...)

--
/* Roland C Dowdeswell
 * http://www.imrryr.org/
 */