Oh yeah, one more slight glitch in the MAD API. It turns out that if a 0-hop DR SMP is passed to ib_post_send_mad(), the client's recv_handler will be called back directly from the same context. This means that the client has to be very careful to avoid deadlocking by taking the same lock in both the send posting code and the receive handling code.
I fixed up the locking in user_mad.c to handle this but we may want to think about changing the MAD code to avoid this case. - R. _______________________________________________ openib-general mailing list [EMAIL PROTECTED] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
