> The attach QP only occurs in the context of the multicast callback thread. > ib_sa_free_multicast() blocks until the callback returns, which ensures that > the > detach check/call (which is now done after ib_sa_free_multicast) cannot race > with the attach call.
OK, makes sense. Do we have confirmation that this fixes the original problem? If so I'll queue this up... - R. _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
