Re: [PATCHv3 1/5] net: dsa: Support internal phy on 'cpu' port

[Andrew Lunn <andrew@lunn.ch>]

>  int dsa_port_fixed_link_register_of(struct dsa_port *dp)
>  {
>  	struct device_node *dn = dp->dn;
> @@ -305,6 +354,10 @@ int dsa_port_fixed_link_register_of(struct dsa_port *dp)
>  			ds->ops->adjust_link(ds, port, phydev);
>  
>  		put_device(&phydev->mdio.dev);
> +	} else {
> +		err = dsa_port_setup_phy_of(dp, true);
> +		if (err)
> +			return err;

Hi Sebastian

First off, i tend to agree with Florian. I'm not sure how reliable
this is. There is normally a state machine moving the PHY between
different states. But in order to do that, i think you need a netdev.
Have you tried multiple down/up of the other MAC/PHY? Does it always
work?

But, at the moment, we don't have much better.

What i don't like is having this code inside
dsa_port_fixed_link_register_of(). This has nothing to do with a fixed
link. Please export functions from port.c and call them directly from
dsa_port_setup() and dsa_port_teardown().

	 Andrew