Blender Git Commits

Fix for typos in array size

Fix for mask slide operators didn't work with LMB selection

Mask spline slide didn't respect auto-key

Mask spline slide: holding Ctrl will make handles free

Scons/cycles: fix header distribution for geom

Followup to the previous commit: forgot to tweak handle type

Mask spline segment slide improvement

This implements weighted slide of second handle, just the
same exact way as it works in Gimp.

Optimization for mempool initial chunk allocation

Almost all pools allocated 2 chunks on initialization,
every element needed to be added to the free-list which
would never be used for small pools.

Now allocate only one, gives minor speedup for some bmesh operations.

Mask point slide: clean-up of SlidePointData structure

Maks point slide: get rid of jump when you press shift

Mask point slide: remove commented out code

The code wasn't needed for quite a while already, so
let's get rid of it.

Mask curvature slide: reduce jumps on slide

Also correct wrong logic around tweaking handles type.

Fix crash happening in DAG_pose_sort() due to threading issues

This function used ugly hack with static variable which was
preventing some type checks in DAG nodes. Using this variable
form multiple threads is not considered safe, apparently.

Solved by moving this variable inside the DAGForest structure.
so it's global for the graph now, but different graphs does not
run into conflicts.

This required passing the forest to some functions, which doesn't
look so much nice, but don't want to spend time on making this
code look beautiful because it is really to be replaced by the
new dependency graph.

This is really bad bug actually which is must go to 'a'.

Code cleanup: no need to use calloc when memory is initialized after

also replace AT with __func__ since AT expands the full pathname

Cloth: replace EdgeHash with EdgeSet

Clicking outside menus wasn't closing when opened from non menu popups

Cycles: Disable Branched Path on all GPUs for now, until we separate the cubins.

SM_20 fails now as well, reported by Zanqdo in IRC.

Cycles volume: detect homogeneous volume automatically in common cases.

In practice this means that if you don't connect a texture to your volume nodes
it will figure that out and render the node faster, rather than you having to
specify it manually.

Main weakness is custom OSL nodes where we have to assume it is heterogeneous
because we don't know what kind of data the node accesses.

Cycles: SSE optimization for Voronoi cells texture

Gives 5-6% speedup for Caterpillar_PatazStudio.blend.

Reviewed By: brecht, dingto

Differential Revision: https://developer.blender.org/D419

I18N: add missing lines

Reviewed By: mont29

Differential Revision: https://developer.blender.org/D328