On Thu, 8 Apr 2021 19:18:09 +0000
Stefan Chulski <stef...@marvell.com> wrote:
> >
> > Stefan, you suggest to drop this define from PHY_INTERFACE enum which we
> > can't easily do with other drivers (like NXP) also referencing this macro.
> >
> > How to continue then?
> >
> > Thanks,
> > Stefan
>
> Probably we should drop SGMII_2500 from this series, introduce "manage"
> devicetree property(that would indicate if inband supported or not).
> For example this is armada-8040-mcbin.dtsi In kernel, when in-band supported:
> &cp1_eth2 {
> /* CPS Lane 5 */
> status = "okay";
> /* Network PHY */
> phy-mode = "2500base-x";
> managed = "in-band-status";
> /* Generic PHY, providing serdes lanes */
> phys = <&cp1_comphy5 2>;
> sfp = <&sfp_eth3>;
> };
>
> If in-band not supported(for example PPv2 MAC connected to 88E2110 in 2.5G
> speed) we would use default managed = "auto" and fixed link property.
Such DTS properties should first be proposed to device-tree ML
and documented in devicetree bindings documentation.
Marek
