Roland, I'd like to verify with you an issue related to the logic applied by ipoib about when/why it decided to join the multicast group used for IPv4 broadcast (eg ARP) and when/why its sets the carrier bit.
Probing ipoib and ifconfing ib0 to be UP, i see that ib0 is RUNNING, has an IPv6 address and is joined to the IPv4 broascast group (ie to MGID ff12:401b:ffff:0:0:0:ffff:ffff), see the exact sequence of operations below. I wonder if you can shed some light on: 1) what is the exact reason that ib0 is running here, is it as of this "magic" configuration of the IPv6 addr that caused it to join to the IPv4 and IPv6 broascast groups? 2) is it well defined what conditions should hold s.t IPoIB will be RUNNING 3) just to make sure: RUNNING <--> ipoib called netif_carrier_on(), correct? i see that latter is called by ipoib_mcast_join_task(), is it when "joining everything we want to join to" holds or you can somehow refine the predicate? thanks, Or. $ modprobe ib_ipoib $ ifconfig ib0 up $ ifconfig ib0 ib0 Link encap:UNSPEC HWaddr 00-04-04-04-FE-80-00-00-00-00-00-00-00-00-00-00 inet6 addr: fe80::208:f104:396:51dd/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:2044 Metric:1 RX packets:35 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:128 RX bytes:2760 (2.6 Kb) TX bytes:456 (456.0 b) $ ip addr show ib0 10: ib0: <BROADCAST,MULTICAST,UP,10000> mtu 2044 qdisc pfifo_fast qlen 128 link/[32] 00:04:04:04:fe:80:00:00:00:00:00:00:00:08:f1:04:03:96:51:dd brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff inet6 fe80::208:f104:396:51dd/64 scope link valid_lft forever preferred_lft forever $ cat /sys/kernel/debug/ipoib/ib0_mcg GID: ff12:401b:ffff:0:0:0:ffff:ffff created: 4294982836 queuelen: 0 complete: yes send_only: no GID: ff12:601b:ffff:0:0:0:0:1 created: 4294982838 queuelen: 0 complete: yes send_only: no GID: ff12:601b:ffff:0:0:0:0:2 created: 4294983192 queuelen: 0 complete: yes send_only: yes GID: ff12:601b:ffff:0:0:0:0:16 created: 4294982840 queuelen: 0 complete: yes send_only: yes GID: ff12:601b:ffff:0:0:1:ff96:51dd created: 4294982838 queuelen: 0 complete: yes send_only: no $ dmesg ib0: bringing up interface ib0: starting multicast thread ADDRCONF(NETDEV_UP): ib0: link is not ready ib0: joining MGID ff12:401b:ffff:0000:0000:0000:ffff:ffff ib0: restarting multicast task ib0: stopping multicast thread ib0: waiting for MGID ff12:401b:ffff:0000:0000:0000:ffff:ffff ib0: join completion for ff12:401b:ffff:0000:0000:0000:ffff:ffff (status -4) ib0: starting multicast thread ib0: joining MGID ff12:401b:ffff:0000:0000:0000:ffff:ffff ib0: join completion for ff12:401b:ffff:0000:0000:0000:ffff:ffff (status 0) ib0: Created ah ffff810034630380 ib0: MGID ff12:401b:ffff:0000:0000:0000:ffff:ffff AV ffff810034630380, LID 0xc000, SL 0 ib0: successfully joined all multicast groups ADDRCONF(NETDEV_CHANGE): ib0: link becomes ready ib0: restarting multicast task ib0: stopping multicast thread ib0: waiting for MGID ff12:401b:ffff:0000:0000:0000:ffff:ffff ib0: adding multicast entry for mgid ff12:601b:ffff:0000:0000:0001:ff96:51dd ib0: adding multicast entry for mgid ff12:601b:ffff:0000:0000:0000:0000:0001 ib0: starting multicast thread ib0: joining MGID ff12:601b:ffff:0000:0000:0001:ff96:51dd ib0: join completion for ff12:601b:ffff:0000:0000:0001:ff96:51dd (status 0) ib0: Created ah ffff810034630d00 ib0: MGID ff12:601b:ffff:0000:0000:0001:ff96:51dd AV ffff810034630d00, LID 0xc006, SL 0 ib0: joining MGID ff12:601b:ffff:0000:0000:0000:0000:0001 ib0: join completion for ff12:601b:ffff:0000:0000:0000:0000:0001 (status 0) ib0: Created ah ffff8100346305c0 ib0: MGID ff12:601b:ffff:0000:0000:0000:0000:0001 AV ffff8100346305c0, LID 0xc003, SL 0 ib0: successfully joined all multicast groups ib0: setting up send only multicast group for ff12:601b:ffff:0000:0000:0000:0000:0016 ib0: no multicast record for ff12:601b:ffff:0000:0000:0000:0000:0016, starting join ib0: Created ah ffff81003d669400 ib0: MGID ff12:601b:ffff:0000:0000:0000:0000:0016 AV ffff81003d669400, LID 0xc004, SL 0 ib0: setting up send only multicast group for ff12:601b:ffff:0000:0000:0000:0000:0002 ib0: no multicast record for ff12:601b:ffff:0000:0000:0000:0000:0002, starting join ib0: Created ah ffff81002009f7c0 ib0: MGID ff12:601b:ffff:0000:0000:0000:0000:0002 AV ffff81002009f7c0, LID 0xc005, SL 0 ib0: no IPv6 routers present ib0: neigh_destructor for ffffff ff12:601b:ffff:0000:0000:0000:0000:0002 ib0: neigh_destructor for ffffff ff12:601b:ffff:0000:0000:0001:ff96:51dd _______________________________________________ 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