Hi Or, Maybe I did not explain myself right. The idea is not to implement it in the mad.c code but rather to implement it at the lowest level: The problem with a new API is that a single ULP/applications which does direct umad or QP1 access will break the reference count.
Implementing at the lowest level - I.e. by sniffing QP1 packets - would be enforced for all applications/ULPs. Or Gerlitz wrote: > Eitan Zahavi wrote: > >> If the tracking (ref counting) was done at the MAD level - no change to >> IPoIB would have been required ... >> > > Maybe. > > You could also implement all the ib stack core in one module... > The openib designers have chosen not to do so and rather break it into > smaller modules namely ib_core, ib_mad, ib_sa, ib_cm, ib_addr and > rdma_cm, with the architecture at hand adding ib_multicast makes sense. > > Or. > > > _______________________________________________ > openib-general mailing list > openib-general@openib.org > http://openib.org/mailman/listinfo/openib-general > > To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general > _______________________________________________ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
