You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
btPersistentManifold uses globals for contact processing callbacks, which
crashes in multithreaded scenarios (stepping two worlds at once). Some
programmers here are using it, but can't find more references to it in the
Bullet code base, maybe it's debug/dead code? (see also
http://www.bulletphysics.org/mediawiki-1.5.8/index.php?title=Collision_Callbacks
_and_Triggers#gContactAddedCallback )
File btPersistentManifold.cpp:
btScalar gContactBreakingThreshold = btScalar(0.02);
ContactDestroyedCallback gContactDestroyedCallback = 0;
ContactProcessedCallback gContactProcessedCallback = 0;
///gContactCalcArea3Points will approximate the convex hull area using 3 points
///when setting it to false, it will use 4 points to compute the area: it is
more accurate but slower
bool gContactCalcArea3Points = true;
Original issue reported on code.google.com by daan.nus...@gmail.com on 11 Feb 2014 at 12:47
The text was updated successfully, but these errors were encountered:
By the way, I will probably need to fix this myself so I could make and submit
a patch. However, where to put the vars (and the interface to get/set them)?
- in the world object?
- in the btPersistentManifold object?
Original comment by daan.nus...@gmail.com on 11 Feb 2014 at 12:51
I'll look into fixing this before next release. I will create a bug-fix release
pretty soon.
Note that development moved over to github, for the Bullet 2 and Bullet 3
merge. This merge is going to take time, and during this merge you can just
Bullet 2 as usual.
http://github.com/erwincoumans/bullet3
Original comment by erwin.coumans on 19 Feb 2014 at 6:27
Original issue reported on code.google.com by
daan.nus...@gmail.com
on 11 Feb 2014 at 12:47The text was updated successfully, but these errors were encountered: