This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: libjava status on Solaris 8/Intel and IRIX 6.5
Bryce McKinlay writes:
> Andrew Haley wrote:
>
> > > >Even with this I think it makes sense for us to make the change. It
> > > >will take a lot of time and will be hard to maintain, but it seems
> > > >necessary. I'd prefer to change g++, but I'd guess that will be hard
> > > >to sell.
> > >
> > > I disagree. We should not change CNI.
> >
> >Because?
>
> We shouldn't make such a major, incompatible change to the CNI syntax
> because CNI is simple and works very well.
Oh right, it's not that you've something against the syntax or
whatever, just that the gain isn't worth the pain of changing. I see
what you mean.
> But, I accept that something needs to be done if we really want to
> support MMU-less systems. Why wouldn't overloading operator "->"
> for Java types work?
I don't think so, because a C++ pointer to a type is not part of the
type.
> > > As for platforms without memory management, well Java just isn't an
> > > appropriate language.
> >
> >That's outrageous.
> >
> Not as outrageous as using -fcheck-references on a platform that has
> perfectly good hardware support for detecting null references! OK, I'll
> retract that comment, but please lets keep -fcheck-references only for
> platforms that really need it.
Oh, of course. The nice thing about smart pointers is that in most
cases the code generated would be exactly the same, and
-fcheck-references only gets used on appropriate platforms.
Andrew.