Blender Git Loki

New Outline: make it countour the screen.

Cycles: Correct ifdef around float3 intrinsics

Merge branch 'master' into datablock_idprops

Fix obvious mistake in logic of two functions recursively handling IDProps.

They were expecting IDP_group as parameter, but then recursiveley
calling themselves with IDProps from groups and arrays, which can be of
any type...

Fix bad auto-generated UI for ID IDProps.

`RNA_path_resolve()` tries to 'dereference' pointer properties, was not
a problem before but now that we do have pointer IDProps we want to get
property here, not it's pointed data.

Cleanup and minor changes.

No functional change expected, mostly:
* Some renaming, bit of style editing, cleanup...
* Removing some useless diff from master.

New Outlines: fix upper edge

New Grid: small modification

Fix wrong coord picked when display only one axis.
Small optimizations here and there.

Draw Engines: Make g_data struct part of the viewport storage

This makes viewport cache construction independant from each others and will allow multithread down the road.

Draw Manager: New debug timers

Both CPU time and GPU time are printed to spot bottlenecks.

GPU Timers works only if cache is enabled.

Draw Manager: Make Viewport Data passed by the manager call.

Keep base layer around for forward compatibility in 2.78

Correct for Py3.5

Merge branch 'master' into datablock_idprops

Cycles: Make shadow catcher an optional feature for OpenCL

Solves majority of speed regression on AMD OpenCL.

Cycles: Add OpenCL support for shadow catcher feature

The title says it all actually.

Cycles: Remove ccl_addr_space from RNG passed to functions

Simplifies code quite a bit, making it shorter and easier to extend.
Currently no functional changes for users, but is required for the
upcoming work of shadow catcher support with OpenCL.

Cycles: First implementation of shadow catcher

It uses an idea of accumulating all possible light reachable across the
light path (without taking shadow blocked into account) and accumulating
total shaded light across the path. Dividing second figure by first one
seems to be giving good estimate of the shadow.

In fact, to my knowledge, it's something really similar to what is
happening in the denoising branch, so we are aligned here which is good.

The workflow is following:

- Create an object which matches real-life object on which shadow is
to be catched.

- Create approximate similar material on that object.

This is needed to make indirect light properly affecting CG objects
in the scene.

- Mark object as Shadow Catcher in the Object properties.

Ideally, after doing that it will be possible to render the image and
simply alpha-over it on top of real footage.

OpenGL: remove gpuMatrixBegin3D_legacy function

It helped during the transition, but we don't need this anymore. Evolution of T49450

OpenGL: simplify initial state

Client vertex array state is deprecated, and these are the default values anyway.

No need to bind any basic shader. Let drawing code decide which shader it wants to use.

Part of T49165 (general OpenGL upgrade)