I also missed the meeting, did not have it my calendar, sorry. Jarno
> On Jan 24, 2017, at 5:52 AM, Jiri Benc <[email protected]> wrote: > > On Wed, 18 Jan 2017 09:53:18 +0000, Jan Scheurich wrote: >> Please be invited to the next sync meeting. > > Sorry, won't make the meeting today. Too busy with DevConf.cz > preparations. > >> Actions Points: >> AP-1 (Jarno): Coordinate review of Yi's backported net-next patches >> AP-2 (Jiri) Check the ability of the kernel datapath to match on the >> presence of Ethernet header w/o matching on Ethernet addresses. >> Formating of such DP flows? > > The OVS_KEY_ATTR_ETHERNET netlink attribute contains source and > destination MAC address. It can be masked. > > If the OVS_KEY_ATTR_ETHERNET attribute is present, only Ethernet frames > will match. The MAC addresses may be masked (wildcarded). The > OVS_KEY_ATTR_ETHERTYPE attribute indicates the ethertype to match and > cannot be masked (it's always an exact match). If it's not present, > ETH_P_802_2 is assumed. > > If the OVS_KEY_ATTR_ETHERNET in not present, the OVS_KEY_ATTR_ETHERTYPE > attribute is mandatory and indicates the ethertype to match. Again, it > cannot be masked. > > To answer the question, it's currently possible to match on Ethernet > header without matching on Ethernet addresses but not without matching > on ethertype. I think it can be relaxed with some work if needed. > >> AP-3 (Jiri) Provide an update on the status and ETA of the RTNETLINK API >> patches > > RFC patchset posted last week: > https://mail.openvswitch.org/pipermail/ovs-dev/2017-January/327749.html > > Jiri _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
