@@ -21,7 +21,7 @@ __global__ void kernel(int prob_size, CudaMemAccessLogger<int> * input, CudaMemA
2121}
2222
2323int main (){
24- constexpr int prob_size = 100 ;
24+ constexpr int prob_size = 1000 ;
2525
2626 std::vector<int > h_input (prob_size);
2727 std::iota (h_input.begin (), h_input.end (), 0 );
@@ -34,8 +34,11 @@ int main(){
3434
3535 checkCudaError (cudaMemcpy (d_input, h_input.data (), sizeof (int )* prob_size, cudaMemcpyHostToDevice));
3636
37- CudaMemAccessLogger<int > input (d_input);
38- CudaMemAccessLogger<int > output (d_output);
37+ CudaMemAccessStorage<int > memAccessStorage (10000 );
38+
39+
40+ CudaMemAccessLogger<int > input (d_input, &memAccessStorage);
41+ CudaMemAccessLogger<int > output (d_output, &memAccessStorage);
3942
4043 constexpr int threads = 32 ;
4144 constexpr int blocks = (prob_size/threads)+1 ;
@@ -44,10 +47,7 @@ int main(){
4447 checkCudaError (cudaGetLastError ());
4548 cudaDeviceSynchronize ();
4649
47- auto data = input.getGlobalSettings ();
48-
49- input.analyze (" ../../../html/basic_template.html" , " ../../../out/basic_input.html" );
50- output.analyze (" ../../../html/basic_template.html" , " ../../../out/basic_output.html" );
50+ memAccessStorage.generateOutput (" ../../../html/basic_template.html" , " ../../../out/basic.html" );
5151
5252 checkCudaError (cudaMemcpy (h_output.data (), d_output, sizeof (int )*prob_size, cudaMemcpyDeviceToHost));
5353
0 commit comments