[osg-users] Are there features to help detect bad data?

OpenSceneGraph Users osg-users at lists.openscenegraph.org
Mon Mar 16 15:33:39 PDT 2020


I had for a good long time an error that sounds similar to this. We had *timeBeginPeriod 
*set pretty low, to like 1. Turns out this was causing a lot of 
dead-locking in the NVidia driver. I have also found NVIDIA Nsight to be 
pretty useful for tracking down some issues with hangs. We had another 
issue where a 3rd party lib we were using made a few unneeded glGetInt 
calls every frame. This seemed to cause some intermittent long hangs when 
we ran multiple instances of our program at the same time. I was able to 
find this one pretty quickly with Nsight.


On Friday, March 13, 2020 at 3:44:16 AM UTC-5, Alexandre Vaillancourt wrote:
>
> Hello all!
>
> We've recently been having some intermittent crashes ("hangs"--Windows 
> shuts down the process after a 2 seconds freeze in the graphics API) 
> occurring in the nVidia's OpenGL dll. Google searches revealed that this 
> can occur if we supply bad data to OpenGL. 
>
> Are there features in OSG to detect such incorrect data in our models? 
> Concurrent vertices, bad normals, bad UVs, etc.--stuff that could be 
> correct or incorrect if defaulted (the crash is intermittent)?
>
> We're using OSG 3.6.3, we read .flt files and convert them to .osgb with 
> a convert tool of our own (using OSG), then we read those .osgb in our 
> app. 
>
> Thanks!
>
> --
> Vaillancourt
>

-- 
You received this message because you are subscribed to the Google Groups "OpenSceneGraph Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to osg-users+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/osg-users/b8db1f28-9e16-406f-8239-65070a3ceaf4%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20200316/b7f3be3e/attachment.html>


More information about the osg-users mailing list