Sergey Sharybin 7b356a8565 Cycles: Reduce amount of malloc() calls from the kernel ...

This commit makes it so malloc() is only happening once per volume and
once per transparent shadow query (per thread), improving scalability of
the code to multiple CPU cores.

Hard to measure this with a low-bottom i7 here currently, but from quick
tests seems volume sampling gave about 3-5% speedup.

The idea is to store allocated memory in kernel globals, which are per
thread on CPU already.

Reviewers: dingto, juicyfruit, lukasstockner97, maiself, brecht

Reviewed By: brecht

Subscribers: Blendify, nutel

Differential Revision: https://developer.blender.org/D1996

2016-05-18 10:14:24 +02:00

build_files

install_deps: Avoid conflicts on Arch-based systems when gcc-multilib is installed

2016-05-12 18:17:14 -04:00

doc

BGE: Fix KX_LightObject python API documentation.

2016-05-15 14:56:40 +00:00

extern

Cleanup: simplify checks calculating tangents

2016-05-17 02:18:11 +10:00

intern

Cycles: Reduce amount of malloc() calls from the kernel

2016-05-18 10:14:24 +02:00

release

Fix: "Whole Character" Keying Set should not include Location on bones with "connected" joint

2016-05-18 16:29:18 +12:00

source

Fix T48395: Grease Pencil, pressing Ekey to sculpt don't work for left click configurations

2016-05-18 17:23:16 +12:00

tests

Fix bl_load_py_modules test

2016-05-02 21:06:15 +10:00

.arcconfig

…

.gitignore

Git ignore BUILD_NOTES.txt from install_deps.sh.

2015-11-23 20:17:02 +01:00

.gitmodules

Remove SCons building system

2016-01-04 14:20:48 +05:00

CMakeLists.txt

Cleanup: cmake, indentation, line length

2016-05-18 11:58:08 +10:00

COPYING

…

GNUmakefile

Fix T48078: incorrect OS X command for retrieving number of CPUs on in Makefile.

2016-04-09 00:25:18 +02:00

make.bat

Fix make.bat checking 64bit systems

2016-02-10 00:26:12 +11:00

Description

No description provided

926 MiB

Languages

C++ 78.1%

Python 15%

C 2.8%

GLSL 1.8%

CMake 1.2%

Other 0.9%