On Tue, Nov 04, 2025 at 05:22:59AM +0000, Jacky Chou wrote:

quoted

quoted

diff --git
a/Documentation/devicetree/bindings/net/faraday,ftgmac100.yaml
b/Documentation/devicetree/bindings/net/faraday,ftgmac100.yaml
index d14410018bcf..de646e7e3bca 100644

--- a/Documentation/devicetree/bindings/net/faraday,ftgmac100.yaml
+++ b/Documentation/devicetree/bindings/net/faraday,ftgmac100.yaml

@@ -19,6 +19,12 @@ properties:
               - aspeed,ast2500-mac
               - aspeed,ast2600-mac

I don't know if it is possible, but it would be good to mark aspeed,ast2600-mac
as deprecated.

I also think some comments would be good, explaining how
aspeed,ast2600-mac01 and aspeed,ast2600-mac23 differ from
aspeed,ast2600-mac, and why you should use them.

Thanks for the suggestion.
We keep "aspeed,ast2600-mac" in the compatible list mainly for backward compatibility.
There are already many existing device trees and systems using this string.
Removing or deprecating it right now might break those setups.

I'm not saying remove it. I'm just saying mark it as deprecated. For
properties you add an extra attribute, e.g.

https://elixir.bootlin.com/linux/v6.17.7/source/Documentation/devicetree/bindings/net/snps,dwmac.yaml#L433

but for a compatible, i've no idea if YAML supports it. However,
snps,dwmac.yam actually places st,spear600-gmac at the end of the list
after a comment. Maybe that is the best way to do this, and in the
comment you can explain what it gets wrong and why you should not use
it.

In the future, if someone submits a new DTS for an AST2600-based platform,
I think they should add the new compatible string and properly describe the TX/RX delay and 
phy-mode properties in their DTS.

Yes, i agree. It would be good if you can keep on out of for such
patches, and make review comments. I assume you will also update the
vendor documentation with this recommendation.

	Andrew