Re: [PATCH net-next v1 2/7] dt-bindings: net: phy: add PoDL PSE property
flat view
From: Andrew Lunn <andrew@lunn.ch>
Date: 2022-08-22 19:35:17
Also in:
linux-devicetree, linux-doc, lkml
On Mon, Aug 22, 2022 at 01:45:34PM -0500, Rob Herring wrote:
On Fri, Aug 19, 2022 at 02:01:04PM +0200, Oleksij Rempel wrote:quoted
Add property to reference node representing a PoDL Power Sourcing Equipment. Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de> --- Documentation/devicetree/bindings/net/ethernet-phy.yaml | 6 ++++++ 1 file changed, 6 insertions(+)diff --git a/Documentation/devicetree/bindings/net/ethernet-phy.yaml b/Documentation/devicetree/bindings/net/ethernet-phy.yaml index ed1415a4381f2..49c74e177c788 100644 --- a/Documentation/devicetree/bindings/net/ethernet-phy.yaml +++ b/Documentation/devicetree/bindings/net/ethernet-phy.yaml@@ -144,6 +144,12 @@ properties: Mark the corresponding energy efficient ethernet mode as broken and request the ethernet to stop advertising it. + ieee802.3-podl-pse: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Specifies a reference to a node representing a Power over Data Lines + Power Sourcing Equipment.Ah, here is the consumer. Why do you anything more than just a -supply property here for the PoE/PoDL supply? The only reason I see is you happen to want a separate driver for this and a separate node happens to be a convenient way to instantiate drivers in Linux. Convince me otherwise.
The regulator binding provides a lot of very useful properties, which look to do a good job describing the regulator part of a PoE/PeDL supplier side. What however is missing is the communication part, the power provider and the power consumer communicate with each other, via a serial protocol. They negotiate the supply of power, a sleep mode where power is reduced, but not removed, etc. So a Power Sourcing Equipment driver is very likely to have a regulator embedded in it, but its more than a regulator. Andrew