Blender Git Loki

Immediate Mode: remove commented out code from drawgpencil.c

Some of this was "tagged" for removal by Dec 2016.

Use FTOCHAR rather than CLAMP of uchar output

This makes no sense to clamp after value was assigned to uchar already.
Proper way to do so is to use FTOCHAR.

OpenGL: Converted gpu_compositing.c to use batches.

Merge branch 'master' into blender2.8

Fix strict compiler warning in the previous commit

Fix T50938: Cache not being reset when changing simulation settings with new depsgraph

The thing i'm really starting to hate is the requirement to specify both
operation code and node type. Seems to be duplicated enums without real
need for that.

BGE: Fix blenderplayer stub.

Add dummy definition of WM_operator_is_repeat.

Merge branch 'master' into blender2.8

WM: free operators when repeating

Needed since the active operator isn't ensured to be the last.

fix D2552: Collada - Follow up change to complete the fix in rBda6cd7762810 (use unique id for bones with same name in different armatures)

Collada simplify: avoid duplicate negation in boolean

WM: store operators with undo flag

This is needed so we can tell if operators are executed repeatedly.

Update path_menu for recent API change

WM: add Operator.is_repeat() check for C & Py

This addresses an issue raised by D2453 -
that there was no way to check if operators are run
multiple times in a row.

Actions are still ignored that don't cause an UNDO event.

Cycles: Cleanup, indentation

Fix T50926: python crashes with path containing utf8 characters.

Default text encoding is platform-dependent in python, and windows
usually do not use utf-8 as default...

fix D2552: Collada - use unique id for bones with same name but in different armatures. Co-authored-by: Gaia <gaia.clary@machiniamtrix.org>

CLNor: rework threaded computation.

Was using some threaded queue on top of task pool, tssk...

Now using properly task pool directly to crunch chunks of smooth fans.

No noticable changes in speed.

Tried to completely get rid of the 'no threading with few loops' code,
but even just creating/freeing the task pool, without actually pushing
any task, is enough to make code 50% slower in worst case scenario (i.e.
few thousands of simple cube objects).

CLNor code: use averaged debug timing.

Fix T50876: Cycles Crash - Cycles crashes before sampling when certain meshes have autosmooth enabled.

The root of the issue was in custom normal code, so far it assumed that
we could only have one cyclic smooth fan around each vertex, which is...
blatantly wrong (again, the two cones sharing same vertex tip e.g.).

This required a rather deep change in how smooth fans/clnor spaces are processed,
took me some time to find a 'good' solution.

Note that new code is slightly slower than previous one (maybe about 5%),
not much to be done here, am afraid.

Tested against all older report files I could find, seems OK.