> And one other thing worth noting, your output says:
>
> OpenGL version string: 1.3 Mesa 6.4.2
>
> and if that is really the case then there is no real advantage to using
> OpenGL over Tk.
Not neccessarily true. From Marco's lspci output you can see that he is
using an Intel graphics adaptor for which direct rendering is suported in
by the open source Mesa drivers. This is confirmed by the
direct rendering: Yes
line in his glxinfo output.
> A couple of common OpenGL problems. There are two flags in
> environment.txt that can cause problems, GLUT_NEED_INIT and GL_FLAG.
> Unfortunately OpenGL doesn't respond gracefully if you choose the wrong
> options here, it just crashes. Each of those has two options so in theory
> there are four options total to try (via "make clean" and "make").
certainly worth trying.
> But that might not be the problem. It could be that a different glut is
> being picked up at runtime than was used at compile time.
which you can check using
/usr/sbin/lsof -p <pid of analysis python process>
Another possibility with your graphics adapter, which I guess is one of
those critters with a limited amount of its own memory that then steals
from the main RAM, is that you are hitting some problem to do with that
that has not been seen yet. Could be worth trying to start analysis with
the memory usage throttled right down (-m flag if I remember) - this would
not be a long term solution, but might help to identify the problem.
--
Dr. Brian O. Smith ---------------------- B Smith at bio gla ac uk
Division of Biochemistry & Molecular Biology,
Institute Biomedical & Life Sciences,
Joseph Black Building, University of Glasgow, Glasgow G12 8QQ, UK.
Tel: 0141 330 5167/6459/3089 Fax: 0141 330 8640
|