On Feb 11, 2014, at 4:40 PM, Ben Pfaff <b...@nicira.com> wrote: > On Tue, Feb 11, 2014 at 04:30:44PM -0800, Jarno Rajahalme wrote: >> Reading the hmap count for determining if it is empty or not is thread >> safe, so avoid locking when not necessary. >> >> Signed-off-by: Jarno Rajahalme <jrajaha...@nicira.com> > > It might be worth adding an atomic_thread_fence(memory_order_acquire); > before the hmap_is_empty() checks, so that we can be sure to get "fresh" > results in case that some other CPU has modified the hmap. (On x86, > with GCC 4.7, this generates no code.)
Would this be best placed in the hmap_is_empty() itself? Jarno _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
