Re: WANTED: Ada-to-C Translator

[Corey Minyard <minyard@metronet.com>]

dewar@merv.cs.nyu.edu (Robert Dewar) writes:

> 
> Frank said
> 
> "I got plenty of opinion about what a dumb idea this is, but nobody
> told me where I can get one of these compilers -- dumb or not."
> 

I would not consider this a bad idea.  The idea of using "C" as a
low-level language (much like most compilers use assembly as a low
level language) woud probably work well (much like Cfront).  Of
course, it opens up a few new abuses (such as hacking the output of
the compiler), but those really aren't big issues.

> 
> Note that another path to an Ada compiler for the 8051 is to use GNAT.
> Is there already a GCC port for this chip? If not, buidling one would
> not be that har.
> 

Spoken by one who has probably never ported GCC to a new architecture :-).
Believe me, I've done it for the TMS320C40 DSP and it is not easy.  It
looks easy on the surface, but you really have to understand what the
compiler is doing on the inside to write good machine descriptions.
You end up spending a lot of time looking through the code wondering
"Why is this expression doing what it is doing?"  I doubt that GCC
would produce very good code for the 8051 processor, either, because
it is targeted more towards 32-bit or better machines.  That's just a
gut feeling, though, I can't think of anything specific that would
make this true, so you can take that for what it is worth :-).

With that said, porting the Ada libraries to the 8051 might even be
harder.  You have all that tasking stuff and all the machine-specific
stuff to do.  Maybe it's not that bad, I'm speaking outside of
experience here.

Plus, if you if needed floating point, you would have to write all
those floating point routines.  Yuck.

-- 
Corey Minyard               Internet:  minyard@metronet.com
  Work: minyard@nortel.ca       UUCP:  minyard@wf-rch.cirr.com
-- 
Corey Minyard               Internet:  minyard@metronet.com
  Work: minyard@nortel.ca       UUCP:  minyard@wf-rch.cirr.com