Darryl Gove's blog
Friday Sep 28, 2007
Perflib and parallel regions
If perflib is called from within a parallel region, you get the serial version rather than a parallel version. This is probably better than ending up with N^2 threads when you set OMP_NUM_THREADS to be N.
Posted at 07:00PM Sep 28, 2007 by Darryl Gove in Sun | Comments[1]
Strange. I would have expected a global thread pool (in which case I can't see the issue to handle nested parallelism) or at least a heuristic of starting parallel code with a value of OMP_NUM_THREADS divided by the number of parallel threads started.
I guess there is room for improvement here...
Posted by Marc on September 30, 2007 at 10:13 AM PDT #