cf5028d55b
* Added CUDA support * Performance improvements * API improvements * Added "depth" parameter to GEMM * Android cross-compilation
57 lines
1.5 KiB
C++
57 lines
1.5 KiB
C++
#include "isaac/array.h"
|
|
#include "isaac/tools/timer.hpp"
|
|
|
|
#include <vector>
|
|
|
|
namespace ad = isaac;
|
|
|
|
#ifdef BENCH_CUBLAS
|
|
__global__ void dummy(){}
|
|
#endif
|
|
|
|
|
|
int main()
|
|
{
|
|
for(ad::driver::queues_type::data_type::const_iterator it = ad::driver::queues.data().begin() ; it != ad::driver::queues.data().end() ; ++it)
|
|
{
|
|
cl::CommandQueue queue = it->second[0];
|
|
cl::Context context = it->first;
|
|
cl::Device device = queue.getInfo<CL_QUEUE_DEVICE>();
|
|
cl::Program program(context,"__kernel void dummy(){}");
|
|
program.build();
|
|
cl::Kernel kernel(program, "dummy");
|
|
|
|
cl::NDRange offset = cl::NullRange;
|
|
cl::NDRange global(1);
|
|
cl::NDRange local(1);
|
|
|
|
cl::Event event;
|
|
std::cout << "Device: " << device.getInfo<CL_DEVICE_NAME>() << std::endl;
|
|
std::cout << "-------------------------" << std::endl;
|
|
|
|
queue.enqueueNDRangeKernel(kernel, offset, global, local, NULL, &event);
|
|
queue.flush();
|
|
queue.finish();
|
|
|
|
{
|
|
long time = event.getProfilingInfo<CL_PROFILING_COMMAND_END>() - event.getProfilingInfo<CL_PROFILING_COMMAND_START>();
|
|
std::cout << "Kernel launch overhead: " << time << std::endl;
|
|
}
|
|
|
|
#ifdef BENCH_CUBLAS
|
|
float time;
|
|
cudaEvent_t start, stop;
|
|
cudaEventCreate(&start);
|
|
cudaEventCreate(&stop);
|
|
cudaEventRecord(start);
|
|
dummy<<<1, 1>>>();
|
|
cudaEventRecord(stop);
|
|
cudaEventSynchronize(stop);
|
|
cudaEventElapsedTime(&time, start, stop);
|
|
std::cout << "CUDA Kernel launch overhead: " << time << std::endl;
|
|
#endif
|
|
std::cout << "-------------------------" << std::endl;
|
|
}
|
|
|
|
}
|