btGenerateInternalEdgeInfo double vertex buffer locking

note173 · **Joined:** Wed Jan 25, 2012 6:32 pm **Posts:** 16

It seems that btGenerateInternalEdgeInfo locks vertex buffer of striding mesh interface twice:

Before the (triangleIndex, numFaces) loop:

Code:

...
meshInterface->getLockedReadOnlyVertexIndexBase(&vertexbase,numverts,   type,stride,&indexbase,indexstride,numfaces,indicestype,partId);
...

then, in the end of the loop:

Code:

    trimeshShape->processAllTriangles(&connectivityProcessor,aabbMin,aabbMax);
}

which tries to lock the same buffer again during callback:

Code:

      virtual void processNode(int nodeSubPart, int nodeTriangleIndex)
      {
         const unsigned char *vertexbase;
         int numverts;
         PHY_ScalarType type;
         int stride;
         const unsigned char *indexbase;
         int indexstride;
         int numfaces;
         PHY_ScalarType indicestype;
         

         m_meshInterface->getLockedReadOnlyVertexIndexBase(
            &vertexbase,
            numverts,
            type,
            stride,
            &indexbase,
            indexstride,
            numfaces,
            indicestype,
            nodeSubPart);

I'm using ogre, and in desktop GL hardware buffers implementation it throws an exception while trying to lock an already locked buffer.
I also don't see where btGenerateInternalEdgeInfo unlocks the buffer.

Physics Simulation Forum

btGenerateInternalEdgeInfo double vertex buffer locking

Who is online