Age | Commit message (Collapse) | Author |
|
There were couple of reasons why it wasn't safe for usage from
multiple threads.
First of all, it was trying to cache BVH in derived mesh, which
wasn't safe because multiple threads might have requested BVH
tree and simultaneous reading and writing to the cache became a
big headache.
Solved this with RW lock so now access to BVH cache is safe.
Another issue is causes by the fact that it's not guaranteed
DM to have vert/edge/face arrays pre-allocated and when one
was calling functions like getVertDataArray() array could
have been allocated and marked as temporary. This is REALLY
bad, because NO ONE is ever allowed to modify data which
doesn't belong to him. This lead to situations when multiple
threads were using BVH tree and they run into race condition
with this temporary allocated arrays.
Now bvhtree owns allocated arrays and keeps track of them, so
no race condition happens with temporary data stored in the
derived mesh. This solved threading issues and likely wouldn't
introduce noticeable slowdown. Even when DM was keeping track
of this arrays, they were re-allocated on every BVH creation
anyway, because those arrays were temporary and were freed
with dm->release() call.
We might re-consider this a bit and make it so BVH trees are
allocated when DM itself is being allocated based on the DAG
layout, but that i'd consider an optimization and not something
we need to do 1st priority.
Fixes crash happening with 05_4g_track.blend from Mango after
the threaded object update landed to master.
|
|
Developer note:
BVHTREE_FROM_FACES was being used for both edit-mesh and derived-mesh
bvh-trees, this could cause index lookup errors in editmode.
Fix by adding a new type for editmesh so theres no confusion.
|
|
|
|
|
|
|
|
|
|
without the underscores these clogged up the namespace for autocompleation which was annoying.
|
|
|
|
|
|
* More code changes pointed by Brecht in codereview.
* Some user interface improvements.
* Updating brush settings now also updates canvas preview.
|
|
http://markmail.org/message/fp7ozcywxum3ar7n
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The goodies:
* Curves can be used as normal dynamic effectors too with
the new "curve" field shape.
* Group visualization has optional duplication counts for
each object in the specified group.
* Object & group visualizations, which are done without
taking the dupliobject's global position into account
(unless the whole group is used). This is much nicer than
the previous behavior, but I added a "Use Global Location"
option for those who want to use it the old way.
* The active particle system's particles are now drawn a
with theme coloured outline instead of pure white.
* Added object aligned velocity factors (buttons categorized
and re-organized too).
Bug fixes:
* Absorption didn't work as the ui toggle button was forgotten.
* Some other force field ui tweaks.
* Crash after adding children and changing trails count.
* Display types "cross" and "axis" crashed.
* Particles weren't drawn with correct coloring.
* Billboards didn't update properly in viewport to camera
location changes.
* Particle rotation wasn't recreated correctly from point cache.
* Changing particles amount crashed sometimes.
* Some files with child hair crashed on loading.
* Compiler warning fixes.
* Adding boids crashed on frame 1;
|
|
+bvhtree cache (if the derived model doenst gets destroyed then the same BVHtree can be used)
this was needed to allow shrinkwrap constraint to be usable.
It has been ready for a long time.. but only got merged now, for 2.49.
|
|
neighbour, non-recursive now, faster than kdtree.c implementation normaly, divided into 2 sources: generla structure in blenlib, mesh/derivedmesh depending interface stuff in blenkernel
|