Re: [RFC PATCH net-next 12/24] of: phy: Only register a phy device for phys

On Mon, Jan 04, 2016 at 12:01:04PM -0800, Florian Fainelli wrote:

On 04/01/16 09:36, Andrew Lunn wrote:

quoted

We will soon support devices other than phys on the mdio bus. Look at
a child's compatibility string to determine if it is a phy, before
registering a phy device.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---

[snip]

quoted

+	if (of_get_phy_id(child, &phy_id) != -EINVAL)
+		return true;
+
+	if (of_device_is_compatible(child, "ethernet-phy-ieee802.3-c45"))
+		return true;

Is there a particular reason not to check for
"ethernet-phy-ieee802.3-c22 here?

Humm interesting. Didn't even know it was a valid compatible string.
Also, Linux does not know it is a valid compatible string:

~/linux$ grep -r "ethernet-phy-ieee802.3-c22" ./drivers
~/linux$

It is however used in a number of DT files and is documented in
Documentation/devicetree/bindings/net/phy.txt.

I will add it here as well.
 

quoted

+
+	if (!of_find_property(child, "compatible", NULL))
+		return true;

That seems reasonable, we should probably make the compatible string for
PHY Device Tree nodes not an optional property anymore though, but
that's for a separate change and topic.

Yes, printing a warning here would be good.

     Andrew