#pragma omp parallel
indicates that the following block shall be executed by all the threads.
int omp_get_num_threads (void)
: returns the number of the threads working on the parallel region (aka team of threads).
int omp_get_thread_num (void)
: returns the identifier of the calling thread (ranges from 0 to N-1 where N is bounded to omp_get_num_threads()
).
You can use the OMP_NUM_THREADS
environment variable or the num_threads
directive within the #pragma parallel
to indicate the number of executing threads for the whole application or for the specified region, respectively.