Blender Git Commits

Merge branch 'master' into soc-2013-depsgraph_mt

Conflicts:
release/datafiles/splash.png
release/scripts/startup/bl_ui/properties_render.py

Merging r60854 through r60868 from trunk into soc-2013-depsgraph_mt

Merging r60847 through r60853 from trunk into soc-2013-depsgraph_mt

Avoid spawning local evaluation contexts

Instead of creating local one, make sure either context
from Render or G.main is used.

Make evaluatio ncontext a part of render strcture

That's how it's supposed to be actually.

Merging r60471 through r60846 from trunk into soc-2013-depsgraph_mt

Merging r60468 through r60470 from trunk into soc-2013-depsgraph_mt

Merging r60198 through r60467 from trunk into soc-2013-depsgraph_mt

Merging r60106 through r60197 from trunk into soc-2013-depsgraph_mt

Simplify code around detailed analysis in scene.c

Merging r60069 through r60105 from trunk into soc-2013-depsgraph_mt

Mark some CCG funcrions as inlined

This seems to be giving speedup up to 10%
and in fact perhaps the change need to go
to the trunk.

Second attempt at fixing windows + guardedalloc issues in main()

Although previous fix compiled, it would crash on startup, since
the guardedalloc changes need to read from the args passed in.
Now use plain malloc for allocating the offending array, so that
we can grab the args before they're needed.

Another windows compile fix for r.59942

Moved var declarations for unicode args array to start of main().

The fix used in the original commit won't work since the args
defined in that block won't be available for later steps.
However, they must be defined before the block of code used
to initialise the mem allocators

scons + mingw Compile Fixes for r.59942

* SCons files for guardedalloc module requires all the source files to be listed out, like in CMake.

* mingw32 (or at least gcc 4.6.2) seems to follow the Windows convention of using _msize() instead of having a "malloc_usable_size()"

Merging r60012 through r60068 from trunk into soc-2013-depsgraph_mt

Merging r59941 through r60011 from trunk into soc-2013-depsgraph_mt

Lock-free memory allocator

Release builds will now use lock-free allocator by
default without any internal locks happening.

MemHead is also reduces to as minimum as it's possible.
It still need to be size_t stored in a MemHead in order
to make us keep track on memory we're requesting from
the system, not memory which system is allocating. This
is probably also faster than using a malloc's usable
size function.

Lock-free guarded allocator will say you whether all
the blocks were freed, but wouldn't give you a list
of unfreed blocks list. To have such a list use a
--debug or --debug-memory command line arguments.

Debug builds does have the same behavior as release
builds. This is so tools like valgrind are not
screwed up by guarded allocator as they're currently
are.

Merging r59874 through r59940 from trunk into soc-2013-depsgraph_mt

Smoke modifier is now expected to use proper render flags

Not so much up to using smoke simulations, only did quick
tests. Would be nice to have some more intense testing some
day later.