Hi Christophe,
Thanks for the stack trace. This is exception is perfectly fine - it
just shows that too much memory has been allocated by the imagery
itself.
Reducing the number of OpenGL objects via the -g option to 7000 or
less should at least avoid this OSG/C++ memory error. The other thing
to try is reduce the dimensions of the texture so it uses less memory.
One thing that is a bit odd is why you are running out of memory on a
3GB machine while others are succeeding. My little shuttle with 1GB
of main memory and 128Mb graphics card is able to allocate all 10,000
Image/Texture and gets to apply 1450 of them before issuing an OpenGL
error. The compute is using up plenty of swap space at to do this,
but is at least able to allocate all the imagery in main memory.
Could you try a 128x128 texture dimension? This will reduce memory
consumption to 1/4rd of the 256x256 test.
osgmemorytest --window --texture 128 128 -g 10000
Robert.
On Wed, Nov 12, 2008 at 6:02 PM, christophe loustaunau
<[EMAIL PROTECTED]> wrote:
> Hi Robert,
>
>
>> The exception on the texture allocation is concerning, could you run
>> it in a debugger and found out what the stack trace is.
>
> Here is the stack trace :
>
> msvcr80d.dll!_unlock(int locknum=8) Line 376
> msvcr80d.dll!_unlockexit() Line 760 + 0x7 bytes
> msvcr80d.dll!_CxxThrowException(void * pExceptionObject=0x00dbf170,
> const _s__ThrowInfo * pThrowInfo=0x68fb3300) Line 161
> msvcr80d.dll!operator new(unsigned int size=262144) Line 64
> osg50-osgd.dll!operator new[](unsigned int count=262144) Line 7 + 0x9
> bytes
> osg50-osgd.dll!osg::Image::allocateImage(int s=256, int t=256, int r=1,
> unsigned int format=6408, unsigned int type=5121, int packing=1) Line 545 +
> 0x9 bytes
> osgmemorytestd.exe!TextureTest::allocate() Line 150 + 0x2e bytes
> osgmemorytestd.exe!main(int argc=2, char * * argv=0x0247c828) Line 441
> + 0x2d bytes
> osgmemorytestd.exe!__tmainCRTStartup() Line 597 + 0x19 bytes
> osgmemorytestd.exe!mainCRTStartup() Line 414
>
> It fails when it try to allocate the data for the new image. It always
> happend around the 7781, 7782 or 7783 image.
> When it fails in debug 2.74 of my 3.0 Go ram is used .
> When it fails in release 0.9 of my 3.0 Go ram is used .
>
> I don't really know what to think about it.
>
>
> Regards.
> --
> Christophe Loustaunau.
>
> _______________________________________________
> osg-users mailing list
> osg-users@lists.openscenegraph.org
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>
>
_______________________________________________
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
