[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20171227082847.GB10517@lunn.ch>
Date: Wed, 27 Dec 2017 09:28:47 +0100
From: Andrew Lunn <andrew@...n.ch>
To: Florian Fainelli <f.fainelli@...il.com>
Cc: jassisinghbrar@...il.com, netdev@...r.kernel.org,
devicetree@...r.kernel.org, davem@...emloft.net,
arnd.bergmann@...aro.org, ard.biesheuvel@...aro.org,
robh+dt@...nel.org, mark.rutland@....com,
masami.hiramatsu@...aro.org,
Jassi Brar <jaswinder.singh@...aro.org>
Subject: Re: [PATCHv3 1/3] dt-bindings: net: Add DT bindings for Socionext
Netsec
On Tue, Dec 26, 2017 at 11:34:26AM -0800, Florian Fainelli wrote:
> On 12/21/2017 05:55 AM, Andrew Lunn wrote:
> >> +- mdio device tree subnode: When the Netsec has a phy connected to its local
> >> + mdio, there must be device tree subnode with the following
> >> + required properties:
> >> +
> >> + - compatible: Must be "socionext,snq-mdio".
> >
> > Is there a need for a compatible string? Is there different versions
> > of the MDIO bus hardware? If it was an independent MDIO bus driver,
> > then yes, you need a compatible string. But since it is embedded in
> > the MAC driver, there should not be a need.
>
> I don't see a problem with doing that though, it may be extra
> information, but if we ever have a standalone MDIO bus controller that
> happens to be supported by the same HW/driver, and the Ethernet driver
> delegates the MDIO management to this MDIO bus driver, then having the
> compatible string is kind of mandatory for proper matching/identification.
Hi Florian
I does not create a problem. It does however make it different to
every other embedded MDIO bus.
My preference is not to have the compatible string, but i can live
with it.
Andrew
Powered by blists - more mailing lists