Blender Git Commits

Dyntopo: minor speedup collapsing edges

Fix T42093: segfault when trying to free unallocated depths zbuffer (border-zoom in 3DView).

Thanks to jensverwiebe (jens verwiebe) for investigations too!

Quiet warning on 32bit system

Fix T41983: Array Modifier "Merge" bug?

We messed up previous fix, in 'simplified' translated merge case... :(

Cycles: Center Tile order had a slight offset to the left.

Signed-off-by: Thomas Dinges

Fix T42081, OpenCL supports float3 since the 1.1 specification, not sure why we needed this.

Cycles: Forgot to set WITH_CYCLES_DEBUG for OSL kernel

Cycles: __KERNEL_DEBUG__ wasn't set for compile-time kernels

Fix tgz script (included all files in submodules)

Cleanup: sculpt float -> bool

OSX/Ghost: little code cleanup

Cycles: Make OpenCL folks happy to use __KERNEL_DEBUG__

Quite straightforward change, the only annoying thing is that we can't use
indentation for include directive just because of the way headers inlineing
works for OpenCL.

Might do smarter job in path_source_replace_includes() but don't want to
spend time on this yet.

Cycles: Fix for debug kernel not working with CUDA

BLI_buffer: simplify buffer resize

Cycles / CUDA: Workaround to make sm_52 (Maxwell) cards work.

* sm_52 can run a sm_50 kernel, so tell runtime detection to use that until we build a dedicated sm_52 kernel.

Cycles / CUDA: Change inline rules for BVH intersection functions.

* On sm_30 and above there is no change (was not inlined already before), this just fixes a speed regression from yesterday. 6359c36ba407
* On sm_2x (tested with sm_21), I get a nice 8% speedup in the bmw scene with this. As a bonus, cubin compilation time and memory usage is significantly reduced. Regular cubin size went from 2.5MB to 2.0MB, Experimental one from 3.8MB to 2.5MB.

Previous scons fix is just wrong, there is a reason why lines were tried to be indented

Cycles: Fix wrong ifdef check around shadows record all

Fix bad RNA enum from the previous commit

Another scons fix, wrong indentation.