On 10/07/2021 22:59:13+0300, Vladimir Oltean wrote: > On Sat, Jul 10, 2021 at 12:25:56PM -0700, Colin Foster wrote: > > Indirect MDIO access is a feature that doesn't need to be specific to the > > Seville driver. Separate the feature to a common file so it can be shared. > > > > Signed-off-by: Colin Foster <colin.foster@xxxxxxxxxxxxxxxx> > > --- > > In fact, this same piece of hardware has a dedicated driver inside > drivers/net/mdio/mdio-mscc-miim.c. The only problem is that it doesn't > work with regmap, and especially not with a caller-supplied regmap. I > was too lazy to do that, but it is probably what should have been done. > > By comparison, felix_vsc9959.c was coded up to work with an internal > MDIO bus whose ops are implemented by another driver (enetc_mdio). Maybe > you could take that as an example and have mdio-mscc-miim.c drive both > seville and ocelot. That was indeed going to be my comment. -- Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
