Re: Dispatch of GObject virtual functions in GtkMM

["Matt Hoosier" <matt hoosier gmail com>]

On 1/10/07, Murray Cumming <murrayc murrayc com> wrote:
> On Wed, 2007-01-10 at 10:27 -0600, Matt Hoosier wrote:
> > > You might also try compiling glibmm and gtkmm with disabled vfuncs
> > and
> > > disabled default signal handlers to see if this makes much
> > difference.
> > >
> >
> > If it turns out that this make a difference, would I be forgoing the
> > ability to do useful overriding of things like expose-event handlers?
>
> You'd be forgoing the ability to do it easily. You'll always be able to
> use the C API to do it the slightly harder way.
>
> You'll save some code size and memory too.
>
> I'll be interested in your results. If the vfuncs and default signal
> handlers significantly slow anything down then I'd like to know why
> and/or optimize the problem away.

I'm having trouble executing the problem when linked against a version
of {glib,gtk}mm with the default signal handlers removed. The call to
Container::add() appears never to return, and the backtrace is trashed
(every frame listed has '??' as its symbol name). These symptoms
appear almost identical to the ones I got with G_GNUC_NORETURN bug in
glibmm, but I don't find that annotation anywhere else in either the
glibmm or gtkmm source trees.

Any hints?

> --
> Murray Cumming
> murrayc murrayc com
> www.murrayc.com
> www.openismus.com