blender/intern
Brecht Van Lommel a3abb020e3 Fix Cycles CUDA performance on CUDA 8.0.
Mostly this is making inlining match CUDA 7.5 in a few performance critical
places. The end result is that performance is now better than before, possibly
due to less register spilling or other CUDA 8.0 compiler improvements.

On benchmarks scenes, there are 3% to 35% render time reductions. Stack memory
usage is reduced a little too.

Reviewed By: sergey

Differential Revision: https://developer.blender.org/D2269
2016-10-03 22:15:25 +02:00
..
atomic Fix building on NetBSD 2016-07-19 22:04:14 +10:00
audaspace Audapsace: Make it work with Jack and no libjack-dev package installed 2016-08-01 20:39:54 +02:00
container Remove SCons building system 2016-01-04 14:20:48 +05:00
cycles Fix Cycles CUDA performance on CUDA 8.0. 2016-10-03 22:15:25 +02:00
decklink Cleanup: comment blocks 2016-07-02 10:08:33 +10:00
dualcon Remove SCons building system 2016-01-04 14:20:48 +05:00
eigen Remove SCons building system 2016-01-04 14:20:48 +05:00
elbeem Fix fluid sim build error with MSVC. 2016-10-03 22:15:24 +02:00
ffmpeg Fix compilation error with FFmpeg and Ubuntu Trusty 2016-02-19 22:05:06 +01:00
ghost Fix Windows mouse wheel scroll speed 2016-09-28 17:22:02 -06:00
glew-mx Remove SCons building system 2016-01-04 14:20:48 +05:00
gpudirect Cleanup: comment blocks 2016-07-02 10:08:33 +10:00
guardedalloc Cleanup: comment blocks 2016-07-15 02:39:45 +10:00
iksolver CMake: add missing headers 2016-04-12 11:22:30 +10:00
itasc Cleanup: Reduce amount of misleading indentation 2016-09-01 12:14:16 +02:00
libmv fix comparison of identicals 2016-07-21 18:16:08 -04:00
locale Remove SCons building system 2016-01-04 14:20:48 +05:00
memutil Remove SCons building system 2016-01-04 14:20:48 +05:00
mikktspace Fix T49384: crash in tangent space calculation with NaN mesh vertices. 2016-09-18 13:38:20 +02:00
moto BGE: Various render improvements. 2016-06-11 22:05:20 +02:00
opencolorio Cleanup: glsl style 2016-05-26 18:53:09 +10:00
opensubdiv OpenSubdiv: Remove redundant section define from shader compilation 2016-09-16 12:28:35 +02:00
openvdb Fix T48236: OpenVDB smoke cache glitches out with smokeless flames. 2016-04-22 11:46:02 +02:00
rigidbody Remove SCons building system 2016-01-04 14:20:48 +05:00
smoke Fix T43220, T47551: collider scaling or rotation causes smoke to explode. 2016-08-02 19:47:31 +03:00
string Remove SCons building system 2016-01-04 14:20:48 +05:00
utfconv Remove SCons building system 2016-01-04 14:20:48 +05:00
CMakeLists.txt BGE: DeckLink card support for video capture and streaming. 2016-06-11 22:26:05 +02:00