#include "distribution_test.cuh"

#include "functions_library.cuh"

distribution_test::distribution_test()

{

functions_library function = functions_library();

cout << "Listing all CUDA capable devices:" << endl;

// int nDevices;

// cudaGetDeviceCount(&nDevices);

// for (int i = 0; i < nDevices; i++)

// {

// cout << endl;

// cudaDeviceProp prop;

// cudaGetDeviceProperties(&prop, i);

// cout << "GPU number\t: " << i << endl;

// cout << "GPU name\t: " << prop.name << endl;

// size_t l_free = 0;

// size_t l_Total = 0;

// cudaError_t error_id = cudaMemGetInfo(&l_free, &l_Total);

// cout << "GPU memory (GB)\t: " << l_Total / (1000 * 1000 * 1000) << endl;

// cout << "GPU number of multiprocessor(s)\t: " << prop.multiProcessorCount << endl;

// cout << "GPU block(s) per multiprocessor\t: " << prop.maxBlocksPerMultiProcessor << endl;

// cout << "GPU thread(s) per block\t: " << prop.maxThreadsPerBlock << endl;

// cout << endl;

// cudaError_t err = cudaGetLastError();

// if (err != cudaSuccess)

// {

// printf("CUDA Error: %s\n", cudaGetErrorString(err));

// }

// }

function.print_Cuda_device(0, this->tot_Blocks, this->tot_ThreadsperBlock);

}

__global__ void setup(curandState_t *d_states)

{

int tid = threadIdx.x + blockIdx.x * blockDim.x;

while (tid < (100 * 100))

{

curand_init(tid + 7, tid, 0, &d_states[tid]);

tid += blockDim.x * gridDim.x;

}

}

__global__ void normal(curandState_t *d_states, double *values)

{

int tid = threadIdx.x + blockIdx.x * blockDim.x;

while (tid < (100 * 100))

{

curandState_t localState = d_states[tid];

double z = (double)(curand_poisson(&localState, 2));

values[tid] = z;

tid += blockDim.x * gridDim.x;

}

}

void distribution_test::ingress()

{

this->gpu_Limit = 1000;

this->CPU_cores = 5;

functions_library function = functions_library(this->tot_Blocks, this->tot_ThreadsperBlock, gpu_Limit, CPU_cores);

random_device rd;

mt19937 gen(rd());

cout << "DT testing" << endl

<< endl;

int number = 200005;

// float *normal_distribution = function.normal_distribution_CUDA(number, 100, 2);

// int *poisson_distribution = function.poisson_distribution_CUDA(number, 0.039);

int *neg_binomial_distribution = function.negative_binomial_distribution_CUDA(number, 3, 5);

fstream results;

results.open("result_test_dis.csv", ios::out);

for (size_t i = 0; i < number; i++)

{

// cout << neg_binomial_distribution[i] << "\n";

results << neg_binomial_distribution[i] << "\n";

}

results.close();

free(neg_binomial_distribution);

cout << endl;

}