Re: [C++ PATCH] warning about empty extern "C" structures

["Vladimir A. Merzliakov" <wanderer at rsu dot ru>]

> > On Wed, 2006-05-17 at 10:48 -0700, Andrew Pinski wrote:
> > > On May 17, 2006, at 10:34 AM, Daniel Jacobowitz wrote:
> > >
> > > > - Using it in extern "C" in the GNU dialects is obviously broken,
> > > > since
> > > > it will be given the C++ layout, but the C version it is trying to
> > > > interface will be given the C layout, and they will not match.
> > >
> > > Why? it only changes the linkage of the structs/variables/functions
> > > and never the layout.
> >
> >
> > There is smthg I'm missing. If it transforms a zero size structure into
> > a non-zeros sized one, there is a modification in the layout. Otherwise,
> > the program would not crash !!!
> >
> > A modified Roman program (first to correct errors and then to show the
> > device_t sizes from C and C++ code) gives:
> >
> > Size in C: 1
> > Size in C++: 2
> >
> > That is a layout difference....
>
> No I am saying extern "C" does not change the layout of the struct from 
> the C++ layout.
>
> Again this is not something I want warning about and I hope the kernel 
> people remove
> invalid C (though valid GNU C) from their sources if they care anything 
> about portablity.

Why not add this warning for C in -Wc++-compat ?
I think this can be helpful for C -> C++ programm migration.

Vladimir