/**
* @brief
* main - the initialization and main loop of pbs_comm
*
* @param[in] argc - argument count.
* @param[in] argv - argument values.
*
* @return int
* @retval 0 - success
*/
main(int argc, char *argv[])
{
SC_HANDLE schManager;
SC_HANDLE schSelf;
int reg = 0;
int unreg = 0;
TCHAR szFileName[MAX_PATH];
/*the real deal or just pbs_version and exit*/
execution_mode(argc, argv);
if (argc > 1) {
if (strcmp(argv[1], "-R") == 0)
reg = 1;
else if (strcmp(argv[1], "-U") == 0)
unreg = 1;
else if (strcmp(argv[1], "-N") == 0)
stalone = 1;
}
if (reg || unreg) {
schManager = OpenSCManager(0, 0, SC_MANAGER_ALL_ACCESS);
if (schManager == 0) {
ErrorMessage("OpenSCManager");
}
if (reg) {
GetModuleFileName(0, szFileName, sizeof(szFileName)/sizeof(*szFileName));
printf("Installing service %s\n", g_PbsCommName);
schSelf =
CreateService(schManager, g_PbsCommName, __TEXT("PBS COMM"),
SERVICE_ALL_ACCESS,
SERVICE_WIN32_OWN_PROCESS,
SERVICE_AUTO_START, SERVICE_ERROR_NORMAL,
replace_space(szFileName, ""), 0, 0, 0, 0, 0);
if (schSelf) {
printf("Service %s installed succesfully!\n", g_PbsCommName);
} else {
ErrorMessage("CreateService");
}
if (schSelf != 0)
CloseServiceHandle(schSelf);
} else if (unreg) {
schSelf = OpenService(schManager, g_PbsCommName, DELETE);
if (schSelf) {
if (DeleteService(schSelf)) {
printf("Service %s uninstalled successfully!\n", g_PbsCommName);
} else {
ErrorMessage("DeleteService");
}
} else {
ErrorMessage("OpenService failed");
}
if (schSelf != 0)
CloseServiceHandle(schSelf);
}
if (schManager != 0)
CloseServiceHandle(schManager);
} else if (stalone) {
struct arg_param *pap;
int i, j;
pap = create_arg_param();
if (pap == NULL)
ErrorMessage("create_arg_param");
pap->argc = argc-1; /* don't pass the second argument */
for (i=j=0; i < argc; i++) {
if (i == 1)
continue;
pap->argv[j] = strdup(argv[i]);
j++;
}
main_thread((void *)pap);
free_arg_param(pap);
} else { /* running as a service */
SERVICE_TABLE_ENTRY rgste[] = { {(TCHAR*)g_PbsCommName, PbsCommMain },
{ 0, 0 } };
if (getenv("PBS_CONF_FILE") == NULL) {
char conf_path[80];
char *p;
char psave;
struct stat sbuf;
if (p = strstr(argv[0], "exec")) {
psave = *p;
*p = '\0';
_snprintf(conf_path, 79, "%spbs.conf", argv[0]);
*p = psave;
if (stat(conf_path, &sbuf) == 0) {
setenv("PBS_CONF_FILE", conf_path, 1);
}
}
}
if (!StartServiceCtrlDispatcher(rgste)) {
ErrorMessage("StartServiceCntrlDispatcher");
}
}
return (0);
}
int main(void) {
unsigned __CPROVER_bitvector[3] tmp_t0_r0;
unsigned __CPROVER_bitvector[3] tmp_t1_r0;
unsigned __CPROVER_bitvector[3] tmp_t2_r0;
unsigned __CPROVER_bitvector[3] tmp_t0_r1;
unsigned __CPROVER_bitvector[3] tmp_t1_r1;
unsigned __CPROVER_bitvector[3] tmp_t2_r1;
unsigned __CPROVER_bitvector[3] tmp_t0_r2;
unsigned __CPROVER_bitvector[3] tmp_t1_r2;
unsigned __CPROVER_bitvector[3] tmp_t2_r2;
unsigned __CPROVER_bitvector[3] tmp_t0_r3;
unsigned __CPROVER_bitvector[3] tmp_t1_r3;
unsigned __CPROVER_bitvector[3] tmp_t2_r3;
unsigned __CPROVER_bitvector[3] tmp_t0_r4;
unsigned __CPROVER_bitvector[3] tmp_t1_r4;
unsigned __CPROVER_bitvector[3] tmp_t2_r4;
unsigned __CPROVER_bitvector[3] tmp_t0_r5;
unsigned __CPROVER_bitvector[3] tmp_t1_r5;
unsigned __CPROVER_bitvector[3] tmp_t2_r5;
unsigned __CPROVER_bitvector[3] tmp_t0_r6;
unsigned __CPROVER_bitvector[3] tmp_t1_r6;
unsigned __CPROVER_bitvector[3] tmp_t2_r6;
unsigned __CPROVER_bitvector[3] tmp_t0_r7;
unsigned __CPROVER_bitvector[3] tmp_t1_r7;
unsigned __CPROVER_bitvector[3] tmp_t2_r7;
unsigned __CPROVER_bitvector[3] tmp_t0_r8;
unsigned __CPROVER_bitvector[3] tmp_t1_r8;
unsigned __CPROVER_bitvector[3] tmp_t2_r8;
unsigned __CPROVER_bitvector[3] tmp_t0_r9;
unsigned __CPROVER_bitvector[3] tmp_t1_r9;
unsigned __CPROVER_bitvector[3] tmp_t2_r9;
unsigned __CPROVER_bitvector[3] tmp_t0_r10;
unsigned __CPROVER_bitvector[3] tmp_t1_r10;
unsigned __CPROVER_bitvector[3] tmp_t2_r10;
unsigned __CPROVER_bitvector[3] tmp_t0_r11;
// round 0
thread_index = 0;
pc_cs[0] = pc[0] + tmp_t0_r0;
assume(pc_cs[0] > 0);
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r0;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[1]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r0;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[2]);
pc[2] = pc_cs[2];
}
// round 1
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r1;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r1;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r1;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 2
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r2;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r2;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r2;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 3
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r3;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r3;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r3;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 4
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r4;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r4;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r4;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 5
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r5;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r5;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r5;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 6
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r6;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r6;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r6;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 7
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r7;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r7;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r7;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 8
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r8;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r8;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r8;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 9
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r9;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r9;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r9;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
// round 10
thread_index = 0;
if (active_thread[thread_index] == 1) { // main
pc_cs[0] = pc[0] + tmp_t0_r10;
assume(pc_cs[0] <= 4);
main_thread();
pc[0] = pc_cs[0];
}
thread_index = 1;
if (active_thread[thread_index] == 1) { // thr1_0
pc_cs[1] = pc[1] + tmp_t1_r10;
assume(pc_cs[1] <= 6);
thr1_0(threadargs[thread_index]);
pc[1] = pc_cs[1];
}
thread_index = 2;
if (active_thread[thread_index] == 1) { // thr2_0
pc_cs[2] = pc[2] + tmp_t2_r10;
assume(pc_cs[2] <= 6);
thr2_0(threadargs[thread_index]);
pc[2] = pc_cs[2];
}
thread_index = 0;
if (active_thread[0] == 1) {
pc_cs[0] = pc[0] + tmp_t0_r11;
assume(pc_cs[0] <= 4);
main_thread();
}
return 0;
}
int main(void)
{
unsigned __CPROVER_bitvector[5] __cs_tmp_t0_r0 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t1_r0 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[4] __cs_tmp_t2_r0 = (unsigned __CPROVER_bitvector[4])nondet_uint();
unsigned __CPROVER_bitvector[4] __cs_tmp_t3_r0 = (unsigned __CPROVER_bitvector[4])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t4_r0 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t0_r1 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t1_r1 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[4] __cs_tmp_t2_r1 = (unsigned __CPROVER_bitvector[4])nondet_uint();
unsigned __CPROVER_bitvector[4] __cs_tmp_t3_r1 = (unsigned __CPROVER_bitvector[4])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t4_r1 = (unsigned __CPROVER_bitvector[5])nondet_uint();
unsigned __CPROVER_bitvector[5] __cs_tmp_t0_r2 = (unsigned __CPROVER_bitvector[5])nondet_uint();
/* round 0 */
__CPROVER_assume(__cs_tmp_t0_r0 > 0);
__cs_thread_index = 0;
__cs_pc_cs[0] = __cs_pc[0] + __cs_tmp_t0_r0;
__CPROVER_assume(__cs_pc_cs[0] > 0);
__CPROVER_assume(__cs_pc_cs[0] <= 16);
main_thread();
__cs_pc[0] = __cs_pc_cs[0];
if (__cs_active_thread[1] == 1)
{
__cs_thread_index = 1;
__cs_pc_cs[1] = __cs_pc[1] + __cs_tmp_t1_r0;
__CPROVER_assume(__cs_pc_cs[1] <= 22);
P0_0(__cs_threadargs[1]);
__cs_pc[1] = __cs_pc_cs[1];
}
if (__cs_active_thread[2] == 1)
{
__cs_thread_index = 2;
__cs_pc_cs[2] = __cs_pc[2] + __cs_tmp_t2_r0;
__CPROVER_assume(__cs_pc_cs[2] <= 14);
P1_0(__cs_threadargs[2]);
__cs_pc[2] = __cs_pc_cs[2];
}
if (__cs_active_thread[3] == 1)
{
__cs_thread_index = 3;
__cs_pc_cs[3] = __cs_pc[3] + __cs_tmp_t3_r0;
__CPROVER_assume(__cs_pc_cs[3] <= 12);
P2_0(__cs_threadargs[3]);
__cs_pc[3] = __cs_pc_cs[3];
}
if (__cs_active_thread[4] == 1)
{
__cs_thread_index = 4;
__cs_pc_cs[4] = __cs_pc[4] + __cs_tmp_t4_r0;
__CPROVER_assume(__cs_pc_cs[4] <= 27);
P3_0(__cs_threadargs[4]);
__cs_pc[4] = __cs_pc_cs[4];
}
/* round 1 */
if (__cs_active_thread[0] == 1)
{
__cs_thread_index = 0;
__cs_pc_cs[0] = __cs_pc[0] + __cs_tmp_t0_r1;
__CPROVER_assume(__cs_pc_cs[0] >= __cs_pc[0]);
__CPROVER_assume(__cs_pc_cs[0] <= 16);
main_thread();
__cs_pc[0] = __cs_pc_cs[0];
}
if (__cs_active_thread[1] == 1)
{
__cs_thread_index = 1;
__cs_pc_cs[1] = __cs_pc[1] + __cs_tmp_t1_r1;
__CPROVER_assume(__cs_pc_cs[1] >= __cs_pc[1]);
__CPROVER_assume(__cs_pc_cs[1] <= 22);
P0_0(__cs_threadargs[__cs_thread_index]);
__cs_pc[1] = __cs_pc_cs[1];
}
if (__cs_active_thread[2] == 1)
{
__cs_thread_index = 2;
__cs_pc_cs[2] = __cs_pc[2] + __cs_tmp_t2_r1;
__CPROVER_assume(__cs_pc_cs[2] >= __cs_pc[2]);
__CPROVER_assume(__cs_pc_cs[2] <= 14);
P1_0(__cs_threadargs[__cs_thread_index]);
__cs_pc[2] = __cs_pc_cs[2];
}
if (__cs_active_thread[3] == 1)
{
__cs_thread_index = 3;
__cs_pc_cs[3] = __cs_pc[3] + __cs_tmp_t3_r1;
__CPROVER_assume(__cs_pc_cs[3] >= __cs_pc[3]);
__CPROVER_assume(__cs_pc_cs[3] <= 12);
P2_0(__cs_threadargs[__cs_thread_index]);
__cs_pc[3] = __cs_pc_cs[3];
}
if (__cs_active_thread[4] == 1)
{
__cs_thread_index = 4;
__cs_pc_cs[4] = __cs_pc[4] + __cs_tmp_t4_r1;
__CPROVER_assume(__cs_pc_cs[4] >= __cs_pc[4]);
__CPROVER_assume(__cs_pc_cs[4] <= 27);
P3_0(__cs_threadargs[__cs_thread_index]);
__cs_pc[4] = __cs_pc_cs[4];
}
if (__cs_active_thread[0] == 1)
{
__cs_thread_index = 0;
__cs_pc_cs[0] = __cs_pc[0] + __cs_tmp_t0_r2;
__CPROVER_assume(__cs_pc_cs[0] >= __cs_pc[0]);
__CPROVER_assume(__cs_pc_cs[0] <= 16);
main_thread();
}
return 0;
}
// Main ------------------------------------------------------------------------------------------
int main(int argc, char **argv) {
const Params p(argc, argv);
CUDASetup setcuda(p.device);
Timer timer;
cudaError_t cudaStatus;
// Allocate
timer.start("Allocation");
int n_flow_vectors = read_input_size(p);
int best_model = -1;
int best_outliers = n_flow_vectors;
#ifdef CUDA_8_0
flowvector *h_flow_vector_array;
cudaStatus = cudaMallocManaged(&h_flow_vector_array, n_flow_vectors * sizeof(flowvector));
int *h_random_numbers;
cudaStatus = cudaMallocManaged(&h_random_numbers, 2 * p.max_iter * sizeof(int));
int *h_model_candidate;
cudaStatus = cudaMallocManaged(&h_model_candidate, p.max_iter * sizeof(int));
int *h_outliers_candidate;
cudaStatus = cudaMallocManaged(&h_outliers_candidate, p.max_iter * sizeof(int));
float *h_model_param_local;
cudaStatus = cudaMallocManaged(&h_model_param_local, 4 * p.max_iter * sizeof(float));
std::atomic_int *h_g_out_id;
cudaStatus = cudaMallocManaged(&h_g_out_id, sizeof(std::atomic_int));
flowvector * d_flow_vector_array = h_flow_vector_array;
int * d_random_numbers = h_random_numbers;
int * d_model_candidate = h_model_candidate;
int * d_outliers_candidate = h_outliers_candidate;
float * d_model_param_local = h_model_param_local;
std::atomic_int *d_g_out_id = h_g_out_id;
std::atomic_int * worklist;
cudaStatus = cudaMallocManaged(&worklist, sizeof(std::atomic_int));
#else
flowvector * h_flow_vector_array = (flowvector *)malloc(n_flow_vectors * sizeof(flowvector));
int * h_random_numbers = (int *)malloc(2 * p.max_iter * sizeof(int));
int * h_model_candidate = (int *)malloc(p.max_iter * sizeof(int));
int * h_outliers_candidate = (int *)malloc(p.max_iter * sizeof(int));
float * h_model_param_local = (float *)malloc(4 * p.max_iter * sizeof(float));
std::atomic_int *h_g_out_id = (std::atomic_int *)malloc(sizeof(std::atomic_int));
flowvector * d_flow_vector_array;
cudaStatus = cudaMalloc((void**)&d_flow_vector_array, n_flow_vectors * sizeof(flowvector));
int * d_random_numbers;
cudaStatus = cudaMalloc((void**)&d_random_numbers, 2 * p.max_iter * sizeof(int));
int * d_model_candidate;
cudaStatus = cudaMalloc((void**)&d_model_candidate, p.max_iter * sizeof(int));
int * d_outliers_candidate;
cudaStatus = cudaMalloc((void**)&d_outliers_candidate, p.max_iter * sizeof(int));
float * d_model_param_local;
cudaStatus = cudaMalloc((void**)&d_model_param_local, 4 * p.max_iter * sizeof(float));
int *d_g_out_id;
cudaStatus = cudaMalloc((void**)&d_g_out_id, sizeof(int));
ALLOC_ERR(h_flow_vector_array, h_random_numbers, h_model_candidate, h_outliers_candidate, h_model_param_local, h_g_out_id);
#endif
CUDA_ERR();
cudaDeviceSynchronize();
timer.stop("Allocation");
timer.print("Allocation", 1);
// Initialize
timer.start("Initialization");
const int max_gpu_threads = setcuda.max_gpu_threads();
read_input(h_flow_vector_array, h_random_numbers, p);
cudaDeviceSynchronize();
timer.stop("Initialization");
timer.print("Initialization", 1);
#ifndef CUDA_8_0
// Copy to device
timer.start("Copy To Device");
cudaStatus = cudaMemcpy(d_flow_vector_array, h_flow_vector_array, n_flow_vectors * sizeof(flowvector), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_random_numbers, h_random_numbers, 2 * p.max_iter * sizeof(int), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_model_candidate, h_model_candidate, p.max_iter * sizeof(int), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_outliers_candidate, h_outliers_candidate, p.max_iter * sizeof(int), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_model_param_local, h_model_param_local, 4 * p.max_iter * sizeof(float), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_g_out_id, h_g_out_id, sizeof(int), cudaMemcpyHostToDevice);
cudaDeviceSynchronize();
CUDA_ERR();
timer.stop("Copy To Device");
timer.print("Copy To Device", 1);
#endif
for(int rep = 0; rep < p.n_warmup + p.n_reps; rep++) {
// Reset
memset((void *)h_model_candidate, 0, p.max_iter * sizeof(int));
memset((void *)h_outliers_candidate, 0, p.max_iter * sizeof(int));
memset((void *)h_model_param_local, 0, 4 * p.max_iter * sizeof(float));
#ifdef CUDA_8_0
h_g_out_id[0].store(0);
if(p.alpha < 0.0 || p.alpha > 1.0) { // Dynamic partitioning
worklist[0].store(0);
}
#else
h_g_out_id[0] = 0;
cudaStatus = cudaMemcpy(d_model_candidate, h_model_candidate, p.max_iter * sizeof(int), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_outliers_candidate, h_outliers_candidate, p.max_iter * sizeof(int), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_model_param_local, h_model_param_local, 4 * p.max_iter * sizeof(float), cudaMemcpyHostToDevice);
cudaStatus = cudaMemcpy(d_g_out_id, h_g_out_id, sizeof(int), cudaMemcpyHostToDevice);
CUDA_ERR();
#endif
cudaDeviceSynchronize();
if(rep >= p.n_warmup)
timer.start("Kernel");
// Launch GPU threads
// Kernel launch
if(p.n_gpu_blocks > 0) {
assert(p.n_gpu_threads <= max_gpu_threads &&
"The thread block size is greater than the maximum thread block size that can be used on this device");
cudaStatus = call_RANSAC_kernel_block(p.n_gpu_blocks, p.n_gpu_threads, n_flow_vectors, p.max_iter,
p.error_threshold, p.convergence_threshold, p.max_iter, p.alpha, d_model_param_local,
d_flow_vector_array, d_random_numbers, d_model_candidate, d_outliers_candidate, (int*)d_g_out_id,
sizeof(int)
#ifdef CUDA_8_0
+ sizeof(int), (int*)worklist
#endif
);
CUDA_ERR();
}
// Launch CPU threads
std::thread main_thread(run_cpu_threads, h_model_candidate, h_outliers_candidate, h_model_param_local,
h_flow_vector_array, n_flow_vectors, h_random_numbers, p.max_iter, p.error_threshold,
p.convergence_threshold, h_g_out_id, p.n_threads, p.max_iter, p.alpha
#ifdef CUDA_8_0
,
worklist);
#else
);
#endif
cudaDeviceSynchronize();
main_thread.join();
if(rep >= p.n_warmup)
timer.stop("Kernel");
#ifndef CUDA_8_0
// Copy back
if(rep >= p.n_warmup)
timer.start("Copy Back and Merge");
int d_candidates = 0;
if(p.alpha < 1.0) {
cudaStatus = cudaMemcpy(&d_candidates, d_g_out_id, sizeof(int), cudaMemcpyDeviceToHost);
cudaStatus = cudaMemcpy(&h_model_candidate[h_g_out_id[0]], d_model_candidate, d_candidates * sizeof(int), cudaMemcpyDeviceToHost);
cudaStatus = cudaMemcpy(&h_outliers_candidate[h_g_out_id[0]], d_outliers_candidate, d_candidates * sizeof(int), cudaMemcpyDeviceToHost);
CUDA_ERR();
}
h_g_out_id[0] += d_candidates;
cudaDeviceSynchronize();
if(rep >= p.n_warmup)
timer.stop("Copy Back and Merge");
#endif
// Post-processing (chooses the best model among the candidates)
if(rep >= p.n_warmup)
timer.start("Kernel");
for(int i = 0; i < h_g_out_id[0]; i++) {
if(h_outliers_candidate[i] < best_outliers) {
best_outliers = h_outliers_candidate[i];
best_model = h_model_candidate[i];
}
}
if(rep >= p.n_warmup)
timer.stop("Kernel");
}