Re: [PATCH net v2] bonding: fix slave_cnt leak on XDP error paths

[Nikolay Aleksandrov]

On 03/09/2026 12:27, Hangbin Liu wrote:
> On Thu, Sep 03, 2026 at 11:18:32AM +0300, Nikolay Aleksandrov wrote:
> > > diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
> > > index 947d92a669b6..f3290aee50d1 100644
> > > --- a/drivers/net/bonding/bond_main.c
> > > +++ b/drivers/net/bonding/bond_main.c
> > > @@ -2284,7 +2284,6 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
> > >    		}
> > >    	}
> > > -	WRITE_ONCE(bond->slave_cnt, bond->slave_cnt + 1);
> > >    	netdev_compute_master_upper_features(bond->dev, true);
> > >    	bond_set_carrier(bond);
> > > @@ -2339,6 +2338,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
> > >    	bond_xdp_set_features(bond_dev);
> > > +	WRITE_ONCE(bond->slave_cnt, bond->slave_cnt + 1);
> > > +
> > >    	slave_info(bond_dev, slave_dev, "Enslaving as %s interface with %s link\n",
> > >    		   bond_is_active_slave(new_slave) ? "an active" : "a backup",
> > >    		   new_slave->link != BOND_LINK_DOWN ? "an up" : "a down");
> > >
> > > ---
> > > base-commit: 70f3995830d3f1e79faa14eb0605914f778feca9
> > > change-id: 20260807-bond_slave_cnt-88e78c5f0ab9
> > >
> > > Best regards,
> >
> > Thanks, looks good to me. One minor nit if there's another version - mention
> > also the WRITE_ONCE() in the commit message or better yet do it in a separate
> > commit that properly annotates slave_cnt everywhere.
>
> The WRITE_ONCE() change is made by Eric's recent fix
> bc93419130bb ("net: bonding: annotate lockless writes with WRITE_ONCE()").
>
> Do you mean add a comment for all the slave_cnt usage?
>
> Thanks
> Hangbin

Oops, my tree was older, forgot to sync it, my bad.
All looks good.