Hi,
I am testing the Intel Phi affinity with openMP. Please explain what is the default openMP affinity on Intel Phi. That is when #pragma omp parallel proc_bind(xxx) is not specifying.
The following figure shows the time spent by an algorithm when spread, close and default affinities are set. The program runs in offload mode.
I noticed the default affinity is better than spread or close. Why?
Thank you.