Blender Git Loki

Merge branch 'blender2.8' into greasepencil-object

Fix T57835: Textured-fill layer opacity not working

This was an unsupported feature.

Fix T57896: Crash updating grease pencil material preview

New customdata types for base hair curve data.

Merge HairPattern into HairCurveData, no reason to keep these separate any more.

Keymaps: make click event use position on button press rather than release.

This may improve reliability with left click select and pen input, assuming
that the place where the pen first touched the surface is closer to the
intended location than where it was released from the surface.

I'm not sure if this will make a significant difference in practice, but it
seems worth a try.

Introduce curve and vertex customdata for hair.

GP: Set Rim Blur default to 0

Keymap: preference to swap space/shift-space keys

This option is for animators who may not use the tool-system much
(mostly staying in object/pose-mode with the transform-tool active).

Keymap: add user-prefs to the default keyconfig

Currently the only option is to use select-all as a toggle.

Keymap: Default pref for select-all to toggle

Access in the UI next.

UI: show active keymap user preferences

PyAPI: add bpy.utils.execfile

Wraps `importlib.util`, avoids duplicate preset loading calls.

Cleanup: tweak to keymap layout args

Needed to support user-keyconfig draw call.
No functional changes.

GP: Set Thickness default to 2

New modifiers have thsi default value.

Eevee: Optimize Color Ramp node for common cases.

GP: Drawing Engine use only one batch

This is part of T57829.

Reduce the number of used batches to only one by shader type. This reduces GPU overhead and increase a lot the FPS. As the number of batches is small, the time to allocate and free memory was reduced in 90% or more.

Also the code has been simplified and all batch management has been removed because this is not necessary. Now, all shading groups are created after all vertex buffer data for all strokes has been created using DRW_shgroup_call_range_add().

All batch cache data has been moved to the Object runtime struct and not as before where some parts (derived data) were saved inside GPD datablock.

For particles, now the code is faster and cleaner and gets better FPS.

Thanks to Cl�ment Foucault for his help and advices to improve speed.

Eevee: Fix SSR/SSS when the bsdf nodes are inside groups

Add 'nomain' hair creation functions for making hair systems.

GPU: Cleanup / Opti : Vectorize + MADD and remove unused function