There was an error fetching the commit references. Please try again later.
OpenCL: thread interleaving
If two threads are using the same GPU device the start time of each hash round is optimized based on the average time needed to calculate a bunch of hashes. This way to optimize the hash rate was first introduced by @SChernykh. This implementation based on the implementation in xmrig but differen in the details. - introduce a new config option `interleave` - implement thread interleaving
Showing
- xmrstak/backend/amd/amd_gpu/gpu.cpp 116 additions, 0 deletionsxmrstak/backend/amd/amd_gpu/gpu.cpp
- xmrstak/backend/amd/amd_gpu/gpu.hpp 17 additions, 2 deletionsxmrstak/backend/amd/amd_gpu/gpu.hpp
- xmrstak/backend/amd/autoAdjust.hpp 1 addition, 1 deletionxmrstak/backend/amd/autoAdjust.hpp
- xmrstak/backend/amd/config.tpl 8 additions, 2 deletionsxmrstak/backend/amd/config.tpl
- xmrstak/backend/amd/jconf.cpp 20 additions, 1 deletionxmrstak/backend/amd/jconf.cpp
- xmrstak/backend/amd/jconf.hpp 1 addition, 0 deletionsxmrstak/backend/amd/jconf.hpp
- xmrstak/backend/amd/minethd.cpp 5 additions, 0 deletionsxmrstak/backend/amd/minethd.cpp
Loading
Please register or sign in to comment