On 7/31/2020 9:53 PM, Vikas Singh wrote:
> Hi Andrew,
>
> As i have already mentioned that this patch is based on
> https://www.spinics.net/lists/netdev/msg662173.html,
> <https://www.spinics.net/lists/netdev/msg662173.html>
>
> When MDIO bus gets registered itself along with devices on it , the
> function mdiobus_register() inside of_mdiobus_register(), brings
> up all the PHYs on the mdio bus and attach them to the bus with the help
> of_mdiobus_link_mdiodev() inside mdiobus_scan() .
> Additionally it has been discussed with the maintainers that the
> mdiobus_register() function should be capable of handling both ACPI &
> DTB stuff
> without any change to existing implementation.
> Now of_mdiobus_link_mdiodev() inside mdiobus_scan() see if the
> auto-probed phy has a corresponding child in the bus node, and set the
> "of_node" pointer in DT case.
> But lacks to set the "fwnode" pointer in ACPI case which is resulting in
> mdiobus_register() failure as an end result theoretically.
>
> Now this patch set (changes) will attempt to fill this gap and
> generalise the mdiobus_register() implementation for both ACPI & DT with
> no duplicacy or redundancy.
Please reply in plain text and do not top-post, thank you.
--
Florian
