Blender Git Loki

Added graphviz debug output for parsed bvm node graphs.

BLender 2.76: We go 'b' series now

Don't expand toggle brush on linking

Toggle brush is more a runtime only feature, not really supposed to be used
as real ID linking as it's done for modifiers i.e.

Fix T46626: Crash generating previews

Brush.toggle_brush was allowed to be an invalid pointer,
it worked for the one operator that used it - but in general bad practice,
requiring a lookup on every access.

Ensure the pointer is kept valid now.

Fix/workaround T46622: crash w/ metas & particles

Metas are scanning all scenes duplis,
which can go into particle systems without an initialized derived-mesh.

For now just do NULL check, its not correct but real fix is not fitting well with current design.

Cleanup for the texture node compiler API functions.

Merge branch 'master' into asset-engine

Merge branch 'master' into id-remap

Merge branch 'master' into ui-align-rework

Fix warnings when compiling WITH_DECKLINK=OFF and WITH_X11_ALPHA=OFF

Removed unused code.

Use an object pointer from the evaluation data, rather than storing just indices.

dynamic fracture should be bake and saveable now, but still has rotation glitches and glitches with change in mesh island sequence.

BGE: multisample FBO supported in bge.render.offscreenCreate()

3rd optional argument to offScreenCreate() to specify the number of
multi-sample. If the GPU does not support MSAA on FBO, this is ignored.
No other change in the API.

Pointer type for passing opaque DNA references and the like through the VM.

Merge branch 'openvdb' into cvdb_ray_isect

Merge branch 'master' into openvdb

Moved the conversion of bNodeTree (DNA/pynodes) to python code for simplicity.

This has several advantages over doing the same in C/C++ code:

- The two parts of the code are clearly separate (parsing DNA settings vs. building and evaluating)
- UI constructs with modes, special cases, etc. can be done in py, heavy lifting still happens in C
- Scripters can easily add own nodes without having to deal with internals

BGE: Moto: Use float for MT_Scalar.

Test:
4104 rotating spheres.
Without moto modifications:
Logic : 5.15 : 5%
SceneGraph : 8.50 : 8%
Rasterizer : 20.90 : 21%
With:
Logic : 5.00 : 4%
SceneGraph : 8.00 : 8%
Rasterizer : 20.25 : 20%

BGE: Moto: Use MT_Scalar instead of double in MT_CmMatrix4x4.