Re: [PATCH net v2] lan966x: Fix crash when adding interface under a lag

From: Horatiu Vultur
Date: Mon Feb 12 2024 - 03:33:16 EST


The 02/12/2024 09:10, Horatiu Vultur wrote:
> The 02/09/2024 13:52, Jakub Kicinski wrote:
>
> Hi Jakub,
>
> >
> > On Tue, 6 Feb 2024 13:30:54 +0100 Horatiu Vultur wrote:
> > > for (lag = 0; lag < lan966x->num_phys_ports; ++lag) {
> > > - struct net_device *bond = lan966x->ports[lag]->bond;
> > > + struct lan966x_port *port = lan966x->ports[lag];
> > > int num_active_ports = 0;
> > > + struct net_device *bond;
> > > unsigned long bond_mask;
> > > u8 aggr_idx[16];
> > >
> > > - if (!bond || (visited & BIT(lag)))
> > > + if (!port || !port->bond || (visited & BIT(lag)))
> > > continue;
> > >
> > > + bond = port->bond;
> > > bond_mask = lan966x_lag_get_mask(lan966x, bond);
> > >
> > > for_each_set_bit(p, &bond_mask, lan966x->num_phys_ports) {
> > > struct lan966x_port *port = lan966x->ports[p];
> > >
> > > + if (!port)
> > > + continue;
> >
> > Why would lan966x_lag_get_mask() set a bit for a port that doesn't
> > exist? Earlier check makes sense. This one seems too defensive.
>
> You are right, the lan966x_lag_get_mask() will not set a bit for a port
> that doesn't exist[1]. Therefore this check is not needed.
>
> [1] https://elixir.bootlin.com/linux/latest/source/drivers/net/ethernet/microchip/lan966x/lan966x_lag.c#L354

While trying to rebase on net, the next version of this patch, I have seen that
actually this version was accepted even though it was marked as "Changes
Requested".
The commit sha is: 15faa1f67ab405d47789d4702f587ec7df7ef03e

How do you prefer to go forward from here?
- do you want to revert this and then I will send a new version?
- should I send a patch that just removes this unneeded check?
- any other suggestion?

>
> > --
> > pw-bot: cr
>
> --
> /Horatiu

--
/Horatiu