Missing synchronization for time measurements with CUDA
Showing
- apps/benchmarks/FlowAroundSphereCodeGen/CMakeLists.txt 4 additions, 4 deletionsapps/benchmarks/FlowAroundSphereCodeGen/CMakeLists.txt
- apps/benchmarks/PhaseFieldAllenCahn/CMakeLists.txt 2 additions, 2 deletionsapps/benchmarks/PhaseFieldAllenCahn/CMakeLists.txt
- apps/tutorials/codegen/CMakeLists.txt 2 additions, 3 deletionsapps/tutorials/codegen/CMakeLists.txt
- src/CMakeLists.txt 1 addition, 3 deletionssrc/CMakeLists.txt
- src/core/mpi/Datatype.h 3 additions, 9 deletionssrc/core/mpi/Datatype.h
- src/core/mpi/MPIWrapper.h 10 additions, 5 deletionssrc/core/mpi/MPIWrapper.h
- src/core/timing/CMakeLists.txt 1 addition, 0 deletionssrc/core/timing/CMakeLists.txt
- src/core/timing/DeviceSynchronizePolicy.h 84 additions, 0 deletionssrc/core/timing/DeviceSynchronizePolicy.h
- src/core/timing/Timer.h 5 additions, 1 deletionsrc/core/timing/Timer.h
- src/core/timing/TimingNode.cpp 1 addition, 0 deletionssrc/core/timing/TimingNode.cpp
- src/core/timing/TimingNode.h 1 addition, 0 deletionssrc/core/timing/TimingNode.h
- src/core/timing/TimingPool.cpp 1 addition, 0 deletionssrc/core/timing/TimingPool.cpp
- src/core/timing/TimingPool.h 1 addition, 0 deletionssrc/core/timing/TimingPool.h
- src/core/timing/TimingTree.cpp 1 addition, 0 deletionssrc/core/timing/TimingTree.cpp
- src/core/timing/TimingTree.h 1 addition, 0 deletionssrc/core/timing/TimingTree.h
- src/gpu/AlignedAllocation.cpp 13 additions, 2 deletionssrc/gpu/AlignedAllocation.cpp
- src/gpu/CMakeLists.txt 1 addition, 0 deletionssrc/gpu/CMakeLists.txt
- src/gpu/DeviceSelectMPI.cpp 31 additions, 28 deletionssrc/gpu/DeviceSelectMPI.cpp
- src/gpu/DeviceSelectMPI.h 1 addition, 0 deletionssrc/gpu/DeviceSelectMPI.h
- src/gpu/DeviceWrapper.h 292 additions, 0 deletionssrc/gpu/DeviceWrapper.h
Please register or sign in to comment