> > Moreover, that commit also allowed path resolution to be triggered > > for an invalid path; if that path resolution failed, old_ah would > > be freed outside priv->lock violating the assumption that dropping > > references inside the lock are guaranteed not to reach zero > > reference.
> and what would happen next, I understand its the oops shown in > https://bugs.openfabrics.org/show_bug.cgi?id=1194, I guess you want to > mention that in the change log So is this patch fixing a crash or just avoiding a WARN_ON? I'm not going to send a change that just affects warnings for 2.6.27, but if it's an oops that triggers in real life then that's different. _______________________________________________ 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
