[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAM_iQpVNGLzVd5KfwTeP9zXEUOR8yf00RsB8gd2RWn6bySOifg@mail.gmail.com>
Date: Tue, 25 Apr 2017 22:24:58 -0700
From: Cong Wang <xiyou.wangcong@...il.com>
To: Murali Karicheri <m-karicheri2@...com>
Cc: Hangbin Liu <liuhangbin@...il.com>,
"open list:TI NETCP ETHERNET DRIVER" <netdev@...r.kernel.org>,
David Miller <davem@...emloft.net>
Subject: Re: IGMP on IPv6
On Tue, Apr 25, 2017 at 9:16 AM, Murali Karicheri <m-karicheri2@...com> wrote:
> On 04/18/2017 06:37 PM, Cong Wang wrote:
>> Did you assign IPv4 and IPv6 addresses to the HSR master device?
>
> No. I just used IPv4. From the trace mld_ifc_timer_expire() -> mld_sendpack() -> ip6_output()
> do you know what is it trying to do? Is it some neighbor discovery message or something
> going over the lower interface instead of the hsr interface?
>
IPv6 is special here because it has link-local address configured
automatically for each device, so this mld_ifc_timer is armed
for link-local multicasts. See:
464 /* Join interface-local all-node multicast group */
465 ipv6_dev_mc_inc(dev, &in6addr_interfacelocal_allnodes);
466
467 /* Join all-node multicast group */
468 ipv6_dev_mc_inc(dev, &in6addr_linklocal_allnodes);
469
470 /* Join all-router multicast group if forwarding is set */
471 if (ndev->cnf.forwarding && (dev->flags & IFF_MULTICAST))
472 ipv6_dev_mc_inc(dev, &in6addr_linklocal_allrouters);
Powered by blists - more mailing lists