[PATCH net-next] net: dsa: mt7530: support OF-based registration of switch MDIO bus

Simon Horman horms at kernel.org
Mon Jan 8 03:15:03 PST 2024


On Mon, Jan 08, 2024 at 01:00:00PM +0200, Vladimir Oltean wrote:
> On Mon, Jan 08, 2024 at 01:22:18PM +0300, Arınç ÜNAL wrote:
> > > > diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
> > > > index 391c4dbdff42..39d7e7ad7154 100644
> > > > --- a/drivers/net/dsa/mt7530.c
> > > > +++ b/drivers/net/dsa/mt7530.c
> > > > @@ -2153,17 +2153,25 @@ mt7530_free_irq(struct mt7530_priv *priv)
> > > >   static int
> > > >   mt7530_setup_mdio(struct mt7530_priv *priv)
> > > >   {
> > > > +	struct device_node *mnp, *np = priv->dev->of_node;
> > > >   	struct dsa_switch *ds = priv->ds;
> > > >   	struct device *dev = priv->dev;
> > > >   	struct mii_bus *bus;
> > > >   	static int idx;
> > > > -	int ret;
> > > > +	int ret = 0;
> > > > +
> > > > +	mnp = of_get_child_by_name(np, "mdio");
> > > > +
> > > > +	if (mnp && !of_device_is_available(mnp))
> > > > +		goto out;
> > > 
> > > nit: I think it would easier on the eyes to simply
> > > 
> > > 		return 0;
> 
> Actually "return 0" leaks "mnp". An of_node_put() is needed.

Yes, sorry for not noticing that.



More information about the Linux-mediatek mailing list