Gerrit Voß wrote:
> Hi,
>
>
> On Thu, 2009-04-30 at 15:12 +0200, Marcus Lindblom wrote:
>> Hi,
>>
>> After re-loading a scene, we get a std::bad_alloc (with 3 gigs) from the
>> following place (line 601 in OSGShaderProgram.cpp):
>>
>> if(iStatus == 0)
>> {
>> Char8 *szDebug;
>> GLint iDebugLength;
>>
>> osgGlGetShaderiv( uiShader,
>> GL_INFO_LOG_LENGTH,
>> &iDebugLength);
>>
>> szDebug = new Char8[iDebugLength];
>> ^^^^^^^^
>>
>> Looking at the values in a debugger, iDebugLength is stupidly large
>> (1768036544) so I think the glgetshaderiv fails and never writes to this
>> value. (Is it guaranteed to do that always?)
>>
>
> hmm that is strange, the function should either return a useful value or
> 0 if not log is available. I admit that we don't check the length 0 case
> but as this is the compilation failed path I would expect the log buffer
> to contain something. The only thing I could think of is that there is
> an OpenGL error. You could add a gl error check after/before the
> osgGlGetShaderiv call to see if this is happening. AFAIK this could only
> happen if uiShader is not a valid OpenGL shader object or shaders are
> not supported at all.
Makes sense. Hm. Time to install a gldebugger again. :)
I'll check and see what happens in that case. We are
destroying/recreating set of shaders here, so it's possible something
goes wrong.
We also use your vaglo.patch, btw, applied to SVN HEAD from yesterday.
Cheers,
/Marcus
------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations
Conference from O'Reilly Media. Velocity features a full day of
expert-led, hands-on workshops and two days of sessions from industry
leaders in dedicated Performance & Operations tracks. Use code vel09scf
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
