Re: [gtk-vnc-devel] [Q] gvncviewer with gthread

[Anthony Liguori <anthony codemonkey ws>]

Daniel P. Berrange wrote:
> On Wed, Feb 06, 2008 at 10:16:44PM -0300, Jonh Wendell wrote:
>   
> > Em Seg, 2008-02-04 às 21:52 -0600, Anthony Liguori escreveu:
> >     
> > > Hi Atsushi,
> > >
> > > Atsushi SAKAI wrote:
> > >       
> > > > Hi, Anthony
> > > >
> > > > I have a question about Authenication window 
> > > > in sample program gvncviewer with gthread.
> > > >
> > > > If I select --with-coroutine=ucontext,
> > > > Authenication window appeared.
> > > >
> > > > But If I select --with-coroutine=gthread,
> > > > Authenication window appeared but just white blank.
> > > >
> > > > Is there any reason exists?
> > > >   
> > > >         
> > > It looks like it's an issue with gtk_dialog_run().  This function enters 
> > > a recursive main loop which is a generally unsafe thing to do in gtk.  
> > > It's unfortunately really that it does this.
> > >
> > >       
> > Can't it be solved by dropping gtk_dialog_run() call, and replace it
> > with something like this:
> >     
>
> Yes, but the point is that applications shouldn't have to worry about
> the fact that GTK-VNC uses threads internally - we explicitly used
> coroutines so apps wouldn't have this complication. They should be able
> to use any GTK calls they like with GTK-VNC. So we have todo a better
> job of hiding the fact that we use threads by emitting signals from the
> main thread.
>   

Which really means we should emit signals from the system coroutine.  
This should be pretty easy to do, I'll see what I can hack up.

Regards,

Anthony Liguori

> Regards,
> Dan.
>