Blender Git Loki

Actual hair point size support for collision and, eventually, rendering.

Use BLI_task for some split normals code - wrong solution

Looks like tasks are not suited for many little elements
(this code is at least two times slower as mono-thread previous one!).
And probably things like calling `BLI_stack_new()` in each task instead of
once globally does not help either. :/

Will refactor into a few worker tasks feeded with a queue...

Fix remaining floats for clnors data :/

Error with force_grid_size fixed. Direction still incorrect

Simple air drag term for hair.

This avoids ceaseless movement and is physically quite plausible. It
doesn't replace real hair-hair collision though.

Fix for restitution force scaling: This has to use the overall timestep
size, since collision only happens once during this step.

Otherwise the simulation explodes due to forces being a factor
num_force_steps too large.

Merge branch 'master' into testbuild

Merge branch 'hair_system_squash' into testbuild

Revert "Hair System test build."

This reverts commit 4b7f793d1ddac4e761d1555c84470247464a575f.

Hair System test.

Fix for threading crash: Collision contact filtering has to happen
outside the thread functions, or Bullet will have a race condition.

Merge branch 'master' into opensubdiv-modifier

Expose substeps for the force and damping loops in the UI.

Split the integrator loop into 2 parts, so we can actually perform the
collision response in the correct place.

Make hair debug functions threadsafe by passing explicit per-thread
debug data to them, instead of using static global data.

Style and related cleanups

Merge branch 'hair_system_squash' into testbuild

Hair System test build.

NURBS UV Editor

Merge branch 'master' into testbuild

forces included in manta simulation on per-basis step