search

StreamHPC / gromacs

OpenCL porting of the GROMACS molecular simulation toolkit
http://www.gromacs.org
Other
25 stars 4 forks source link

Check the OpenCL implementation for warp vote function #37

Closed ancahamuraru closed 9 years ago

ancahamuraru commented 9 years ago

The CUDA kernels use the __any warp vote function.

The OpenCL kernels have implemented this function using local memory. Search for: warp_any

Verify the OpenCL implementation, test it and fix it, if any problems are found.

ancahamuraru commented 9 years ago

Done. See the implementation and comments for __any warp vote function here 72458c9c7bdd641ee02e2e6ca6cbeb031977ddb7