On Wed, Nov 20, 2013 at 01:50:42PM -0800, Joe Stringer wrote: > On 20 November 2013 11:15, Ben Pfaff <b...@nicira.com> wrote: > > I think that it is a little surprising to use a netdev sequence number > > to track changes to things that are not network devices. If we are > > going to do that, then I think we need to document it somewhere to make > > it harder to surprise developers. Another alternative would be to add a > > sequence number for each object type (cfm, bfd, lacp, stp?). (If we did > > that, then it would also solve the ordering dependency issue I mentioned > > above.) > > I agree that it is surprising. From a code readability standpoint, the > name doesn't really match "something told us that a port has gone > up/down, whether locally or remote". A better name and some > documentation could fix this. Perhaps "connectivity_change_seq"? And > this could be moved to somewhere less netdev-specific, but still > accessible to bridge and ofproto.
I think that's a good option. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
