Re: [PATCH net-next 3/6] net: dsa: add support for retrieving the interface mode

[Andrew Lunn <andrew@lunn.ch>]

> > The way I understand what you're saying is that there is no guarantee
> > that the DSA master and CPU port will agree whether to use in-band
> > autoneg or not here (and implicitly, there is no guarantee that this
> > link will work):
> > 
> > 	&eth0 {
> > 		phy-mode = "2500base-x";
> > 		managed = "in-band-status";
> > 	};
> > 
> > 	&switch_cpu_port {
> > 		ethernet = <&eth0>;
> > 		phy-mode = "25000base-x";
> 
> I'll assume that 25000 is a typo.
> 
> > 		managed = "in-band-status";
> > 	};
> 
> Today, there is no guarantee - because it depends on how people have
> chosen to implement 2500base-X, and whether the hardware requires the
> use of in-band AN or prohibits it.

In practice, a Marvell MAC and a Marvell switch are likely to work,
since Marvell produce and tested both ends. I would expect this to be
true for any vendor. It is only going to be a problem when you have
devices from different vendors. And they have different
interpretations of what 2500Base-X is.

So far, mixed vendor systems tend to be

1) Freescale FEC and Marvell switches, and the FEC it still only 1G RGMII
2) The smaller simpler devices which are 1G and do not yet use a SERDES.

So this might be a future problem we will have, as more devices start
supporting 2.5G and 5G, but hopefully those future devices follow the
standard.

	Andrew