On Thu, 17 May 2007, Roland Dreier wrote:
> > Both ipoib_add_port() and ipoib_mcast_join_task() query the GID at > > index 0 to setup the ipoib_dev_priv structure's local_gid and the > > net_device structure's dev_addr. There does not appear to be a way for > > ipoib_mcast_join_task() to be executed before ipoib_add_port() > > completes. Therefore, the work done in ipoib_mcast_join_task() appears > > to be redundant. > > It does look like we're doing some work we don't need to do. However > ipoib_add_port() could run before an SM has brought up the local port, The same could be true for ipoib_mcast_join_task() These are both instances of the general problem that if the GID at index 0 changes, the IPoIB code is not automatically notified. Agree? > so the GID prefix might change later. > > I'm not sure what the best way to clean this up is. As an aside: Why does ipoib_add_port() treat an error return from ib_query_gid() as fatal while ipoib_mcast_join_task() only emits a warning? james _______________________________________________ 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
