Hi Ewe,
If you run cmake using the provided ./configure it'll automatically
set the release build.
W.r.t the optimization bug, I really don't want to start populating
the code with #ifdef compile_with_bug_version as we end with
unmangable code pretty quickly.
Testing -O2 would be interesting.
BTW, do you see the bug on any of the standard examples?
Robert.
On 6/19/07, Uwe Woessner <[EMAIL PROTECTED]> wrote:
Hello,
Back to front sorting in the transparent bin did not work in our
application on linux since quite a while.
Once we used CMake to build OpenSceneGraph, it worked correctly but our
application was three times slower than before.
It turned out, that I forgot to add -DCMAKE_BUILD_TYPE=RELEASE to the
cmake options and thus got a non optimized osg build.
With the optimized build I got decent framerates again but geometry
sorting did not work anymore.
It turned out to be CullVisitor.cpp in osgUtil which causes the bug to
appear if compiled with -O3 and disappear if compiled with -g.
The solution to the problem is not to inline the distance function.
inline CullVisitor::value_type distance(const osg::Vec3& coord,const
osg::Matrix& matrix)
we had this issue with
gcc-4.0.2
gcc-4.1.2
gcc-4.1.3
and other versions.
Robert, could you add a #ifndef __GNUC__ to the inline declaration or
convert it to a macro?
Uwe
--
\\\|/// *HLRS, High Performance Computing Center Stuttgart*
_I_ ( o o ) *Visualization/VR* _I_
([EMAIL
PROTECTED])--oo0O--(_)--O0oo------------------------------------------([EMAIL
PROTECTED])
| | Uwe Woessner [EMAIL PROTECTED] | |
| | .ooo0 http://www.hlrs.de/people/woessner/ | |
|_| ( ) Oooo. Phone: +49-711-6856-5790 or ...-5970 |_|
([EMAIL PROTECTED])-------\ (---(
)-----------------------------------------([EMAIL PROTECTED])
I \_) ) / I
(_/
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/