Hi Jan,
On 11/8/06, Jan Wurster <[EMAIL PROTECTED]> wrote:
Athlon x2 here as well, together with a single 7800 gtx. Another
machine with Intel / nv hardware exposes the same problem ..
> I haven't tried changing drivers at all recently. The compiler is:
>
> gcc --version
> gcc (GCC) 4.0.2 20050901 (prerelease) (SUSE Linux)
yepp, same here. Standard suse 10.0, as per the rest of the system.
I use a window constructed by the system's renderthread together with
Producer::RenderSurface and Producer::Camera. The same thread does all
the resize handling and the actual rendering (via camera->frame(false)
since we do synchronized swaps).
What happens when you use standard OSG examples?
Another question: when I use gdb (stock version that came with SuSE
10.0) to try and debug into the problem, I see lots of "this=Cannot
access memory at address .." messages as soon as I try to access
anything Producer or OSG. The addresses seem valid and the code segments
that are run through seem to kind of add up, so I suppose it's one of
the usual troubles when debugging threaded applications, or is it? Not
much to go with, I know :-/
I believe this is a gcc 64 bit issue. In a thread recently a user pin
pointed the cause but I can't recall the exact mail. Check the
archives for October, might you might need to go back to September.
Maybe I should try and write a testcase that illustrates the problem
using just GL and outside of any complex system. But maybe then I won't
see the problem at all ;) *sigh*
This might be productive, but as you say it might not reveal the problem.
Robert
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
