Blender Git Commits

Add utility function to get render-feasible tile size

Add utility function to check if given tile size can be rendered by split kernel

[CYCLES] Resolve data-flow violation

Don't pre-pone device_update() for all devices. Instead just postpone
load_kernels for SplitKernel-running-devices only.

Add comment for PATH_ITER_INC_FACTOR

Merge branch 'master' into cycles_kernel_split

Conflicts:
intern/cycles/kernel/svm/svm_light_path.h

Cycles kernel split: Fix for progressive refine

Cycles: Make sure there's no double spaces in the build options

This confuses OSX OpenCL compiler.

Cycles: Add regular license header

Cycles: Fix compilation error in release builds with strict flags

Cycles: Don't use std::set directly, use util_set instead

Cycles: Don't limit OpenCL devices by the env variable

Cycles: Compilation error fixes

Also made checks about __SPLIT_KERNEL__ srvive cases
when it's not defined.

Cycles: Make code style closer to Cycles, avoid some unneeded changes

Merge branch 'master' into cycles_kernel_split

Conflicts:
intern/cycles/device/device_opencl.cpp
intern/cycles/kernel/kernel_shader.h
intern/cycles/kernel/svm/svm_closure.h
intern/cycles/kernel/svm/svm_wavelength.h
intern/cycles/kernel/svm/svm_wireframe.h

Cycles: OpenCL kernel-splitting work

This patch contains work to split the OpenCL mega-kernel into separate kernels
to obtain better GPU utilization and therefore performance.

A description of the optimizations included in this patch is located at

https://docs.google.com/document/d/1LuXW-CV-sVJkQaEGZlMJ86jZ8FmoPfecaMdR-oiWbUY/edit?usp=sharing