On 2022-09-02 11:28 -0700, Jay Vosburgh wrote:
> Benjamin Poirier <bpoirier@xxxxxxxxxx> wrote:
>
> Repeating a couple of questions that I suspect were missed the
> first time around:
Thanks for repeating, I did miss the other questions, sorry.
[...]
> >@@ -2171,12 +2169,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
> > dev_uc_sync_multiple(slave_dev, bond_dev);
> > netif_addr_unlock_bh(bond_dev);
> >
> >- if (BOND_MODE(bond) == BOND_MODE_8023AD) {
> >- /* add lacpdu mc addr to mc list */
> >- u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR;
> >-
> >+ if (BOND_MODE(bond) == BOND_MODE_8023AD)
> > dev_mc_add(slave_dev, lacpdu_multicast);
> >- }
> > }
>
> Just to make sure I'm clear (not missing something in the
> churn), the above changes regarding lacpdu_multicast have no functional
> impact, correct? They appear to move lacpdu_multicast to global scope
> for use in the change just below.
Yes, that's right - no functional impact. I'll split that to a separate
patch to make it clearer.
> > bond->slave_cnt++;
> >@@ -4211,6 +4205,9 @@ static int bond_open(struct net_device *bond_dev)
> > /* register to receive LACPDUs */
> > bond->recv_probe = bond_3ad_lacpdu_recv;
> > bond_3ad_initiate_agg_selection(bond, 1);
> >+
> >+ bond_for_each_slave(bond, slave, iter)
> >+ dev_mc_add(slave->dev, lacpdu_multicast);
> > }
>
> After this change, am I understanding correctly that both
> bond_enslave() and bond_open() will call dev_mc_add() for
> lacpdu_multicast? Since dev_mc_add() -> __dev_mc_add() calls
> __hw_addr_add_ex() with sync=false and exclusive=false, could that allow
> us to end up with two references for lacpdu_multicast?
You are correct once again. When enslaving to an up bond (case in the
selftest), it is ok, but when enslaving to a down bond and then setting
it up, there is a double add.
Thanks for the review. I'll send a v3.
