The first part of the fix is to assign VBO/EBO's upon loading the array in the Nif reader. This avoids triggering the 'addVertexBufferObjectIfRequired' code path in osg::Geometry which has the race condition when two threads add the same Array at the same time. Essentially, we want the Arrays to be 'const' when they come out of the Nif reader.
The second part of the fix is to make sure not to create empty arrays in the Nif reader (importantly, not assigning a VBO to the empty array). This empty array would be deleted when the NIFFile is cleaned up, and the detachment of the VBO assigned to it (which is still in use by other arrays) would cause threading issues.
This rare crash bug was first introduced with commit a7c5beb7c5. When using OSG dev version 3.5 the crashes were a little more prevalent, because 'addVertexBufferObjectIfRequired' in osg::Geometry is now used even when VBO's are disabled (as part of the VAO support changes).
- Remove broken cache locking mechanism
This was supposed to unload NIFFiles after a cell transition completes, but it was never working due to a mistake on the line if (--sLockLevel), should have been if (--sLockLevel == 0). Repairing this would increase load times (NIF files would have to be reloaded more frequently), so just removed it for now.
- Decouple cache from NIFFile (now a new nifcache component)
- Add API for future background loading
- Provide a reliable way (SharedPtr) to hold on to loaded NIFFiles. This will be useful to avoid deep copies of keyframe and text key data, which is currently a performance bottleneck.
These are placed under the RootCollisionNode hierarchy, but the shapes they
hold aren't collided with. Their exact function is unknown, but seems to be
related to lava? Maybe damage avoidance info for the AI.
uramer
scrawl
scrawl
Ewan Higgs
Arthur Moore
Marc Zinnschlag
Rohit Nirmal
dteviot
Chris Robinson