[Coco] CoCo gcc project

[David]

On Fri, Oct 31, 2003 at 01:21:17PM -0500, James Dessart wrote:
> If anyone would like to join the gcc-coco project, that'd be great!  In
> order to do this, join up to http://sourceforge.net/, then tell me your
> sourceforge ID.  If you plan on contributing code, or documentation, I'll
> add you as a developer.

I'm willing to participate if many people think it's worthwhile.  But
first, I think we need to get a plan of how to proceed and set up goals
and no doubt a standard.

First, what do you all think?  Would it be worthwhile to try to get a
polished product?  Would there be any application for it after it would
be finished?

Secondly, does anyone object to using this forum to fully discuss the
project?  It could be done on the sourceforge forum, but this medium
might be more covenient.

We would need to set up some sort of model to work with.  I'd like to
see several contributors working on it, and we need some sort of model
so that everyone can be going in the same direction.

I can see at least 4 targets:  6309-RSDOS, 6809-RSDOS, 6309-OS9
6809-RSDOS.  It looks like we'd need to "create" two machines - the two
processors.  Most of the methods would be similar for all platforms up
to the linking stage with minor exceptions.  Personally, I'd line to see
all OS9 code remain PIC.  RSDOS could go non-PIC, or for convenience, we
might let it be PIC.  I think a common assembler and linker could be
used.  The biggest difference would come in the linker stage.  The
linker would simply build a different module depending on whether it was
OS9 or RSDOS.

Finally, what assembler/linker?  Do we use AS or try to do a totally
rma/rlink compatible routine.  We do already have rma rebuilt as a cross
assembler (OS9 only for now), and we still have rlink to go.  If most
people prefer to go to AS, we have all the source to rebuild it to
whatever we want.

What do you all think?  I don't know how much use this thing would be
after it was done, but it might be fun and a good learning experience.