Re: [PATCH net-next 3/4] net: dsa: mv88e6xxx: check hardware VLAN in use

[Sergei Shtylyov]

On 02/13/2016 10:53 PM, Vivien Didelot wrote:

> > > The DSA drivers now have access to the VLAN prepare phase and the bridge
> > > net_device. It is easier to check for overlapping bridges from within
> > > the driver. Thus add such check in mv88e6xxx_port_vlan_prepare.
> > >
> > > Signed-off-by: Vivien Didelot <vivien.didelot@xxxxxxxxxxxxxxxxxxxx>
> > > ---
> > >    drivers/net/dsa/mv88e6xxx.c | 64 +++++++++++++++++++++++++++++++++++++++++++++
> > >    1 file changed, 64 insertions(+)
> > >
> > > diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
> > > index 2e515e8..685dcb0 100644
> > > --- a/drivers/net/dsa/mv88e6xxx.c
> > > +++ b/drivers/net/dsa/mv88e6xxx.c
> > > @@ -1471,14 +1471,78 @@ static int _mv88e6xxx_vlan_init(struct dsa_switch *ds, u16 vid,
> > >    	return 0;
> > >    }
> > >
> > > +static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port,
> > > +					u16 vid_begin, u16 vid_end)
> > > +{
> > > +	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
> > > +	struct mv88e6xxx_vtu_stu_entry vlan;
> > > +	int i, err;
> > > +
> > > +	if (!vid_begin)
> > > +		return -EOPNOTSUPP;
> > > +
> > > +	mutex_lock(&ps->smi_mutex);
> > > +
> > > +	err = _mv88e6xxx_vtu_vid_write(ds, vid_begin - 1);
> > > +	if (err)
> > > +		goto unlock;
> > > +
> > > +	do {
> > > +		err = _mv88e6xxx_vtu_getnext(ds, &vlan);
> > > +		if (err)
> > > +			goto unlock;
> >
> >      Why are you not using *break*?
>
> I use goto for explicit error handling, and break for expected flow.

   Thought you'd say so. :-)
   I still think *break* is preferable...

> > > +
> > > +		if (!vlan.valid)
> > > +			break;
> > > +
> > > +		if (vlan.vid > vid_end)
> > > +			break;
> > > +
> > > +		for (i = 0; i < ps->num_ports; ++i) {
> > > +			if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i))
> > > +				continue;
> > > +
> > > +			if (vlan.data[i] ==
> > > +			    GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
> > > +				continue;
> > > +
> > > +			if (ps->ports[i].bridge_dev ==
> > > +			    ps->ports[port].bridge_dev)
> > > +				break; /* same bridge, check next VLAN */
> > > +
> > > +			netdev_warn(ds->ports[port],
> > > +				    "hardware VLAN %d already used by %s\n",
> > > +				    vlan.vid,
> > > +				    netdev_name(ps->ports[i].bridge_dev));
> > > +			err = -EOPNOTSUPP;
> > > +			goto unlock;
> > > +		}
> >
> >      Why not *break*?
>
> Because here it would only break the for loop, and not the while loop.

   Oops, I overlooked the *for* loop. Sorry about that.

> > > +	} while (vlan.vid < vid_end);
> > > +
> > > +unlock:
> > > +	mutex_unlock(&ps->smi_mutex);
> > > +
> > > +	return err;
> > > +}
> > > +
> > [...]
>
> Thanks,
> -v

MBR, Sergei