Re: [PATCH RFC net-next 0/3] net: dsa: initial support for MaxLinear MxL862xx switches

[Daniel Golle]

On Thu, Dec 04, 2025 at 01:02:14AM +0000, Russell King (Oracle) wrote:
> On Wed, Dec 03, 2025 at 11:23:11PM +0000, Daniel Golle wrote:
> > On Wed, Dec 03, 2025 at 10:26:05PM +0200, Vladimir Oltean wrote:
> > > Hi Daniel,
> > > 
> > > On Tue, Dec 02, 2025 at 11:37:13PM +0000, Daniel Golle wrote:
> > > > Hi,
> > > > 
> > > > This series adds very basic DSA support for the MaxLinear MxL86252
> > > > (5 PHY ports) and MxL86282 (8 PHY ports) switches. The intent is to
> > > > validate and get feedback on the overall approach and driver structure,
> > > > especially the firmware-mediated host interface.
> > > > 
> > > > MxL862xx integrates a firmware running on an embedded processor (Zephyr
> > > > RTOS). Host interaction uses a simple API transported over MDIO/MMD.
> > > > This series includes only what's needed to pass traffic between user
> > > > ports and the CPU port: relayed MDIO to internal PHYs, basic port
> > > > enable/disable, and CPU-port special tagging.
> > > > 
> > > > Thanks for taking a look.
> > > 
> > > I see no phylink_mac_ops in your patches.
> > 
> 
> As you didn't respond to Vladimir's statement here, I will also echo
> this. Why do you have no phylink_mac_ops ?
> 
> New DSA drivers are expected to always have phylink_mac_ops, and not
> rely on the legacy fallback in net/dsa/port.c

All three phylink_mac_ops functions are no-ops for the internal PHYs,
see also

https://github.com/frank-w/BPI-Router-Linux/blob/6.18-rc/drivers/net/dsa/mxl862xx/mxl862xx.c#L3242

The exception in the reference driver are the SerDes PCS ports, and for
those I'd rather use .pcs_config than setting the interface mode in
.phylink_mac_config.
Hence I was planing to introduce phylink_mac_ops together with support
for the SerDes ports, and it will only have a .mac_select_pcs op.