Blender Git Loki

Merge branch 'blender-v2.91-release'

Merge branch 'blender-v2.91-release'

Merge branch 'blender-v2.91-release'

PyAPI: set sys.executable to None when Python can't be found

Without this, sys.executable may be set to Blender's
executable instead.

Cleanup: Fix typo in comment

Fix T82089 Boolean on empty mesh with collection fails.

For the fast solver, there was an optimization carried over
from the non-collection case for empty meshes which did not
work in the chained boolean code for collection operands.
Removed that optimization in the collection case.

Fix T82089 Boolean on empty mesh with collection fails.

For the fast solver, there was an optimization carried over
from the non-collection case for empty meshes which did not
work in the chained boolean code for collection operands.
Removed that optimization in the collection case.

Fix new boolean performance bug.

The code that decided to use a faster double version of plane
side testing forgot to take an absolute value, so half the time
the exact code was being used when it was unnecessary.

Fix new boolean performance bug.

The code that decided to use a faster double version of plane
side testing forgot to take an absolute value, so half the time
the exact code was being used when it was unnecessary.

Cycles: Fix debug compilation after tile stealing commit

GPencil: Change maximum limit of Join stroke

Also, clarify comments.

Merge branch 'blender-v2.91-release'

Fix T82265: GPencil interpolate crash when next frame was NULL

If the next frame of the layer was NULL and it was not the active layer crashed.

Cycles: Implement tile stealing to improve CPU+GPU rendering performance

While Cycles already supports using both CPU and GPU at the same time, there
currently is a large problem with it: Since the CPU grabs one tile per thread,
at the end of the render the GPU runs out of new work but the CPU still needs
quite some time to finish its current times.

Having smaller tiles helps somewhat, but especially OpenCL rendering tends to
lose performance with smaller tiles.

Therefore, this commit adds support for tile stealing: When a GPU device runs
out of new tiles, it can signal the CPU to release one of its tiles.
This way, at the end of the render, the GPU quickly finishes the remaining
tiles instead of having to wait for the CPU.

Thanks to AMD for sponsoring this work!

Differential Revision: https://developer.blender.org/D9324

Cleanup: clang-format

Previous check seems to have a race condition, resulting in Wintab not being enabled.

Add explanation for dequing Wintab PACKETs during WT_PACKET event into a local queue to consume during WM_* mouse events.

Commented assumption was incorrect, using tickCountToMillis causes underflow when compared to current time.

Cleanup: rename functions for consistency with other process*Events functions.

Cleanup: Use C style comments, punctuation, and capitalization.

Added missing function documentation.