On 6/24/24 3:52 PM, Andrew Lunn wrote:
On Mon, Jun 24, 2024 at 01:52:49AM +0200, Marek Vasut wrote:
On 6/23/24 10:00 PM, Andrew Lunn wrote:
- $ref: ethernet-phy.yaml#
properties:
+ compatible:
+ enum:
+ - ethernet-phy-id0000.8201
I'm not sure that one should be listed. It is not an official ID,
since it does not have an OUI. In fact, this is one of the rare cases
where actually listing a compatible in DT makes sense, because you can
override the broken hardware and give a correct ID in realtek address
space.
Hmmm, so, shall I drop this ID or keep it ?
I generally put the PHY IDs into DT so the PHY drivers can correctly handle
clock and reset sequencing for those PHYs, before the PHY ID registers can
be read out of the PHY.
Are there any in kernel .dts files using it?
git grep ethernet-phy-id0000.8201 on current next-20240624 says no.
We could add it, if it is
needed to keep the DT validation tools are happy. But we should also
be deprecating this compatible, replacing it with one allocated from
realteks range.
I think we should drop from the bindings after all, I will prepare a V2
like that, OK ?
