Re: [PATCH net-next v2 5/5] net: dsa: realtek: rtl8365mb: handle PHY interface modes correctly

[Alvin Šipraga]

On Fri, Jun 10, 2022 at 05:36:39PM +0100, Russell King (Oracle) wrote:
> On Fri, Jun 10, 2022 at 05:38:29PM +0200, Alvin Šipraga wrote:
> > Finally, rtl8365mb_phylink_get_caps() is fixed up to return supported
> > capabilities based on the external interface properties described above.
> > This allows for ports with an external interface to be treated as DSA
> > user ports, and for ports with an internal PHY to be treated as DSA CPU
> > ports.
> 
> I've needed to read that a few times... and I'm still not sure. You seem
> to be saying that:
> - ports with an internal PHY (which presumably provide baseT connections?)
>   are used as DSA CPU ports.

With this change they now _can_ be used as CPU ports. In practice I think most
application would use ports with internal PHY as DSA user ports. But the
hardware can treat any port as a CPU port, and as Vladimir pointed out in the
thread I added in the Link:, the previous test (dsa_is_user_port()) was
spurious.

> - ports with an external interface supporting a range of RGMII, SGMII and
>   HSGMII interface modes are DSA user ports.

Same as above: previously they could not be configured as DSA user ports. Now
they can be. The utility is up for debate, but at least the code is correct.

> 
> With Marvell switches, it's the other way around - the ports with an
> internal PHY are normally DSA user ports. Other ports can be a user,
> inter-switch or CPU port.
> 
> So, I'm slightly confused by your description.

After reading the above, please let me know if you are still confused.

Kind regards,
Alvin