Blender Git Commits

Collada: removed unnecessary dup[licate matrix conversion in matrix data exporter

Collada exporter: Replaced numbers 0/1 by boolean values fasle/true for better reading

Collada Shape key export can create huge datafiles and may need some optimization. Disable export by default, user can still enable on demand

Collada enable Export of animation data by default

DRW: Fix multithreading conflict with material previews.

DRW: Refactor / Cleanup Builtin uniforms.

-Make the view and object dependant matrices calculation isolated and separated, avoiding non-needed calculation.
-Adding a per drawcall matrix cache so that we can precompute these in advance in the future.
-Replaced integer uniform location of only view dependant builtins by DRWUniforms that are only updated once per shgroup.

DRW: Do not recreate the common uniform buffer every frame.

GWN: Query builtin uniform at shader creation.

This avoids having non null entries in shaderface->builtin_uniforms and a redundant check.

Fix T54179: Assert changing layers in 2nd window

Eevee preview materials

Now that Eevee has support for offline rendering (F12) we can use it for
the Material previews.

Note: This makes the duplicated UI issue one panel worse. That happens when
Cycles if your scene engine, and Eevee is your workspace engine.

Cleanup: remove unused duplicate code

MSVC 2013 Compile Fix/Workaround for "static thread_local" vars

Apparently MSVC 2013 has trouble with stuff that's been declared
"static thread_local" (and/or maybe even the "thread_local" keyword).

https://stackoverflow.com/questions/29399494/what-is-the-current-state-of-support-for-thread-local-across-platforms

Merge branch 'master' into blender2.8

Collada: Remove unused vector of flaot[4][4] values

Was in fact causing issues on macOS, something to do with
a destructor.

Collada: Use floating point version of abs()

Avoids implicit cast of float to int.

Merge branch 'master' into blender2.8

GPU_select: correction to select-pick-finalize

Would run twice, harmless at the moment but could cause issues later.

Fix error in depth picking caused by GL contexts

Depth picking needs to read the depth buffer after drawing
since GPU_select_end runs in a different OpenGL context
reading the depth buffer wasn't working.
This caused the last object to be unelectable.

Merge branch 'master' into blender2.8

GPU_select: utility function to finalize selection

Needed for depth picking in 2.8