Blender Git Commits

Fix double-free happening when having proxy with motion paths

Make it so mpath is not shared between several bones now.

Fix suggested by Joshua Leung, thanks!

Recent spinlock commit made scheduling unsafe for threading

Namely, it caused nodes be adding to the pool multiple times.

Returned spin back, but use it only in cases node valency is
zero. So now valency is decreasing without any locks, then
if it's zero spin lock happens, node color (which indicates
whether node is scheduled or not) happens. Actual new task
creation happens outside of locks.

This might sound a bit complicated, but it's straightforward
code which is free from any thread synchronization latency.

fix bug with editmesh rip, active edge could be left on the unselected side of the edge loops.

bmesh utility functions for storing an editselection after another in the list.

fix own error in BM_mesh_calc_edge_groups() was clearning wrong dirty-index flag.
also remove unused transform defines.

Merging r59113 through r59129 from trunk into soc-2013-depsgraph_mt

include order change for osl, fixes gcc compile error - undefined uintptr_t

Fix for uninitialized covariance matrix

This is likely a culptrit caused #36231: Motion tracker bug in 2.68

BGE Dome: bugfix - Warp Mesh not working

Fix: to swap the drawing order for the warp mesh polygon
The code was drawing CW instead of CCW.

It would work in some cases where the drawing flags would allow for the
back faces to be visible.

More specifically the alternative fix were:
glDisable ( GL_CULL_FACE );
glFrontFace( GL_CW );

Bug originally reported in the small_planetarium mailing list.
This bug fix was a comissioned job by a group who prefer not to be
credited. Thanks regardless.

bmesh edge-net: refactor out face creation into its own function, replace array reallocation with alloca.
fix for error increasing the face tag count when the face might not be created.

utility functions to reverse and wrap arrays.

bmo_edgenet_fill_exec was building a vertex array when it didnt need to. also simplify for loops.

Typo (dimention instead of dimension).

Fix #36461: ends of curves - twist still not good

Needed correct quat for first point as well.

Maybe it's time to refactor make_minimum_twist function a bit..

Merging r59104 through r59112 from trunk into soc-2013-depsgraph_mt

Fix crash happening when clearing parent in outliner modes which doesn't support this.

use __builtin_nanf with gcc and clang.

make materials.pop() and more like pythons list.pop
- allow negative index values.
- error when invalid index value are passed in.
- remove last item if no index argument is given.

also change behavior to remove the material slot, it was only clearning by default but the list length remained the same.

Moved NAN_FLT define to BLI_math, this may come in useful.

Merged revision(s) 59035-59107 from trunk/blender into soc-2013-dingto.

Skipping 59034 (Non-Progressive on GPU).