Re: [MiNT] MiNTLib 0.51

["Guido Flohr" <gufl0000@stud.uni-sb.de>]

Hi!

On Tue, May 18, 1999 at 12:33:52AM +0200, Michael Schwingen wrote:
> On Tue, May 18, 1999 at 03:01:12AM +0200, Guido Flohr wrote:
> > 
> > P.S.: To return to the question, the code is ugly because I wanted to
> > simplify things as much as possible.  With gcc it works because gcc
> > internally doesn't care if an argument or return value is really a pointer
> > or a long (or unsigned long).
> 
> This *will* change as soon as I get function parameters in registers
> working - then you have the same situation as with PureC.

Be prepared for a lot of complaints.  There are some hacks floating around
that depend on the gcc calling conventions, I think the CAB ovl is one of
them.

Is that really worth the while? I mean, the code has to do extra work to
spread the arguments on the registers and there is less space for local
register variables.
 
> I think for a library that should work on multiple compilers on the TOS
> platform (there are more than gcc and PureC), we should try to keep such
> hacks down to the necessary minimum.

I agree. But in fact "multiple" for the MiNTLib means two.  And I don't
think that we can stick forever with PureC unless it gets further
developped.

Ciao

Guido
-- 
http://stud.uni-sb.de/~gufl0000
mailto:gufl0000@stud.uni-sb.de