kern/99979: Get Ready for Kernel Module in C++

pfgshield-freebsd at pfgshield-freebsd at
Wed Jul 12 14:10:48 UTC 2006

--- Peter Jeremy <peterjeremy at> ha scritto: 
> I think the general concensus is that it's up to one of the proponents
> of this to actually implement it and demonstrate that it works and has
> no undesirable side-effects.

I only wanted to point out that Darwin modules are not the only port candidates
that want to use C++. While existing code will not be revamped to C++, we must
weight exactly what we find acceptable for use in the kernel, and I'm glad the
people doing the port brought this up before expecting to commit undesired

> >
> ...
> >- Nicer code 
> >- Easier to maintain 
> These are both very subjective.  For someone who isn't comfortable with
> C++, I doubt either are true.

Yes. it's subjective. I admitedly prefer C over C++, and I'm glad to have
kobj() but it remains to be seen if it can really replace C++ for all our
needs. C++ is the de-facto standard for OO: a lot of people know how to use it
and since it was always meant to be an extension to C, C programs are expected
to build just the same (I know ... C99 broke some of this).



Chiacchiera con i tuoi amici in tempo reale!* 

More information about the freebsd-hackers mailing list