20   using namespace std::chrono;
 
   22   std::pair<std::chrono::microseconds, std::chrono::microseconds> ret;
 
   25   for (
int i = 0; i < 5; i++) {
 
   26      auto start = steady_clock::now();
 
   28      ret.first += duration_cast<microseconds>(steady_clock::now() - start);
 
   29      start = steady_clock::now();
 
   31      ret.second += duration_cast<microseconds>(steady_clock::now() - start);
 
virtual void * cudaMalloc(size_t)
 
virtual void memcpyToCPU(void *, const void *, size_t, cudaStream_t *=nullptr)
 
virtual void cudaFree(void *)
 
virtual void * cudaMallocHost(size_t)
 
virtual void memcpyToCUDA(void *, const void *, size_t, cudaStream_t *=nullptr)
 
virtual void cudaFreeHost(void *)
 
R__EXTERN RooBatchComputeInterface * dispatchCUDA
 
std::pair< std::chrono::microseconds, std::chrono::microseconds > memcpyBenchmark(std::size_t nBytes)
Measure the time for transfering data from host to device and vice-versa.