2 cutilCheckError( cutCreateTimer(&timer));
3 cutilCheckError( cutResetTimer(timer) );
4 cutilCheckError( cutStartTimer(timer) );
5 for (int ct=0; ct<1000 ; ct++)
6 kernel_ident<<< dimGrid, dimBlock, 0>>>(d_out, W);
7 cudaThreadSynchronize() ;
8 cutilCheckError( cutStopTimer(timer) );
9 cutilCheckError( cutStopTimer(timer) );
10 printf("Mean runtime: %f ms,\n", cutGetTimerValue(timer)/1000);