Hello Christoph,
Christoph Schäfer wrote:
> Carsten Neumann schrieb:
>> is it possible to do conditional breakpoints and then set the
>> condition to something like glGetError() != GL_NO_ERROR ? That should
>> be close to the point where the real error occurs. Alternatively it
>> would be nice if the debugger could just check glGetError() after each
>> OpenGL function call as Marcus suggested, that should pinpoint the
>> location.
> Unfortunately I can't set conditions in the debugger. I told it to stop
> on every OpenGL error - works fine with the shipped examples but not
> with you aspect1render example. I've created debug output like last time
> in the end of this mail.
are those complete? The context 0 log shows no OpenGL calls (which is
what we'd want, but still) and the context 1 log ends in the middle of a
function name (see below).
>>> If you need further information just let me know where to set
>>> breakpoints etc.
>>
>> hm, it is a bit strange that all the context creation calls seem to
>> happen on context 0 and then all the OpenGL calls are on context 1.
>> I've noticed that I made a mistake with the test program I sent, it
>> creates the SSM in the wrong thread, attached is an updated version.
>> However, that one occasionally prints a warning:
>> WARNING: Attachment::unlinkParent: Child <-> Parent link inconsistent.
BTW those warnings are unrelated and they should be fixed in latest trunk.
>> I'll have to look into what is causing that.
> I tried the new version but on vista 64 it still won't run. The error
> changed, it is now (within the mentioned while loop in above mails)
>
> WARNING: (..\..\..\Source\System\State\Base\OSGLightChunk.cpp,136):
> light:activate:precheck failed: Der Vorgang ist ung³ltig. (0x502)
>
> I think this is from osglightchunk.cpp line 136:
> glErr("light:activate:precheck");
yes.
>
> ////////////////////////////////////////////////////////////
> // This File contain an OpenGL calls log
> // Application: opensg2_mt_tests
> // Generation date: Wednesday, January 28, 2009
> // Generation time: 18:12:38
> // Context id: 0
> //
> // Generated by gDEBugger - an OpenGL Debugger and Profiler
> // www.gremedy.com <http://www.gremedy.com>
> ////////////////////////////////////////////////////////////
>
>
>
> ------------------------------------------------------------------------
>
>
> ////////////////////////////////////////////////////////////
> // This File contain an OpenGL calls log
> // Application: opensg2_mt_tests
> // Generation date: Wednesday, January 28, 2009
> // Generation time: 18:12:38
> // Context id: 1
> //
> // Generated by gDEBugger - an OpenGL Debugger and Profiler
> // www.gremedy.com <http://www.gremedy.com>
> ////////////////////////////////////////////////////////////
>
>
> wglDescribePixelFormat(0xC30110AC, 2, 40, 0x466FB18)
> glPixelStorei(GL_UNPACK_ALIGNMENT, 1)
> glPixelStorei(GL_PACK_ALIGNMENT, 1)
> glDepthFunc(GL_LEQUAL)
> glEnable(GL_DEPTH_TEST)
> glEnable(GL_NORMALIZE)
> glLightfv(GL_LIGHT0, GL_DIFFUSE, {0, 0, 0, 0})
> glLightfv(GL_LIGHT0, GL_SPECULAR, {0, 0, 0, 0})
interesting, except for the wglDescribePixelFormat this is the sequence
of calls from Window::setupGL (called from PassiveWindow::init).
Strangely enough this time there is no call to wglCreateContext in
either log.
> glGetString(GL_VERSION)
> glGetString(GL_EXTENSIONS)
> wglGetProcAddress(glBlendColorEXT)
> wglGetProcAddress(glBlendFuncSeparateEXT)
[SNIP - more wglGetProcAddress calls)
> wglGetProcAddress(glGenQueriesARB)
> wglGetProcAddress(glBeginQueryARB)
> wg
this is the end of the log, is it complete?
I'll grab a linux version of gDEBugger and see what it logs here.
Perhaps comparing that with your logs gives a hint.
Cheers,
Carsten
------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
