MiniCL includes a cross-platform run-time frontend based on pthreads, Win32 Threads, or libspe2 for Cell SPU.
It is there, to bridge the gap until OpenCL is more widely available.
See Bullet/Demos/VectorAdd, influenced by NVidia OpenCL Jumpstart Guide:
http://developer.download.nvidia.com/OpenCL/NVIDIA_OpenCL_JumpStart_Guide.pdf
It is disabled by default, see Demos/Gpu2dDemo/btGpuDemoDynamicsWorld.h for instructions how to enable it
Cleaned up Extras/CUDA folder and libbulletcuda project
For example, to only allow linear motion in the X-Z plane, and only rotation around Y axis use:
body->setLinearFactor(btVector3(1,0,1));
body->setAngularFactor(btVector3(0,1,0));
Fix build issues with CodeBlocks, when generating projectfiles using CMake 2.6:
${OPENGL_glU_LIBRARY} should be ${OPENGL_glu_LIBRARY}
Fix build issue with CodeBlocks, comment out xmlfree in Extras/COLLADA_DOM/src/modules/LIBXMLPlugin/daeLIBXMLPlugin.cpp (will leak memory)
Revert definition for ATTRIBUTE_ALIGNED16, and try to force sizeof(btSolverConstraint) by using unions with btScalar, for non-btScalar data types.
Use btAssert and not assert.
Don't access btAlignedObjectArray elements, for zero sets
1) Add fast branchless SIMD support for constraint solver (Windows only until we get other contributions).
See resolveSingleConstraintRowGenericSIMD in Bullet/src/BulletDynamics/ConstraintSolver/btSequentialImpulseConstraintSolver.cpp
resolveSingleConstraintRowGenericSIMD can be used for all constraints, including contact, point 2 point, hinge, generic etc.
2) During this refactoring, all constraints support the obsolete 'solveConstraintObsolete' while we add 'getInfo1' and 'getInfo2' support.
This interface is almost identical interface to Open Dynamics Engine, to make it easier to port Dantzig LCP solver.
3) Some minor refactoring to reduce huge constructor overhead in math classes.
+ Share some more code between SpuParallelSolver and btSequentialImpulseConstraintSolver (both use btSolverBody now)
Lack of warmstarting reported here: http://bulletphysics.com/Bullet/phpBB3/viewtopic.php?f=9&t=2858
+ allow btBoxBoxDetector to execute on SPU.
Disable libxml LIBXML_MODULE_EXTENSION
Fixed include path in btConeTwistConstraint (it should be relative to src folder, not using ../../ )
Thanks to ejtttje, http://code.google.com/p/bullet/issues/detail?id=129
Add CMake install support for libbulletmath,libbulletcollision,libbulletdynamics,libbulletsoftbody,and toplevel include files
Options to enable/disable BUILD_DEMOS and BUILD_ETRAS
Fixed compile issue in BulletMultiThreaded
Fixed double-precision issues with btMatrix3x3::getEulerZYX
to build, go to Bullet/src/ibm_sdk and run make
then, go to Bullet/src/BulletMultiThreaded/ibm_sdk
make -f Makefile.original spu
make -f Makefile.original ppu
btQuantizedBvh has a version number, memory layout might be different now (due to aligned btVector3)
reorganized some data members of some classes, to reduce memory footprint
BulletMultiThreaded/NarrowPhaseCollision makes use of this boxBoxDistance.
Cache some values in src/BulletMultiThreaded/SpuContactManifoldCollisionAlgorithm.cpp, to avoid DMA transfers
2) Added btConvexSeparatingDistanceUtil: this allows caching of separating distance/vector as early-out to avoid convex-convex collision detection.
btConvexSeparatingDistanceUtil is used in src/BulletCollision/CollisionDispatch/btConvexConvexAlgorithm.cpp and can be controlled by btDispatcherInfo.m_useConvexConservativeDistanceUtil/m_convexConservativeDistanceThreshold
3) Use BulletMultiThreaded/vectormath/scalar/cpp/vectormath/scalar/cpp/vectormath_aos.h as fallback for non-PlayStation 3 Cell SPU/PPU platforms (used by boxBoxDistance).
Note there are other implementations in Extras/vectormath folder, that are potentially faster for IBM Cell SDK 3.0 SPU (libspe2)
