| Message ID | 20240228-net-v6-7-6d78d3d598c1@outlook.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | net: hisi-femac: add support for Hi3798MV200, remove unmaintained compatibles | expand |
On 28/02/2024 10:02, Yang Xiwen via B4 Relay wrote: > From: Yang Xiwen <forbidden405@outlook.com> > > The old property "phy-mode" should be replaced with the latest > "phy-connection-type". Why? The old property is not deprecated. Best regards, Krzysztof
On 2/29/2024 4:08 PM, Krzysztof Kozlowski wrote: > On 28/02/2024 10:02, Yang Xiwen via B4 Relay wrote: >> From: Yang Xiwen <forbidden405@outlook.com> >> >> The old property "phy-mode" should be replaced with the latest >> "phy-connection-type". > > Why? The old property is not deprecated. The old property is just a $ref to the new property. While at it, i would like to update it as well. Since there is no real user in mainline yet, this update does not need to keep dts sync. I would expect it(the old property) to be deprecated in the future. > > > > Best regards, > Krzysztof >
On 29/02/2024 09:12, Yang Xiwen wrote: > On 2/29/2024 4:08 PM, Krzysztof Kozlowski wrote: >> On 28/02/2024 10:02, Yang Xiwen via B4 Relay wrote: >>> From: Yang Xiwen <forbidden405@outlook.com> >>> >>> The old property "phy-mode" should be replaced with the latest >>> "phy-connection-type". >> >> Why? The old property is not deprecated. > > The old property is just a $ref to the new property. While at it, i > would like to update it as well. It does not mean which property is preferred. Please provide some context why this change should be done. > > Since there is no real user in mainline yet, this update does not need > to keep dts sync. I would expect it(the old property) to be deprecated > in the future. Best regards, Krzysztof
On Wed, Feb 28, 2024 at 05:02:31PM +0800, Yang Xiwen via B4 Relay wrote: > From: Yang Xiwen <forbidden405@outlook.com> > > The old property "phy-mode" should be replaced with the latest > "phy-connection-type". In practice, phy-mode is typically used, and phy-connection-type is not used much. All new user tend to be phy-mode. ~/linux$ grep -r phy-mode arch/* | wc 1561 6258 113122 ~/linux$ grep -r phy-connection-type arch/* | wc 372 1489 28981 Most of the phy-connection-type appear in PowerPC. ~/linux$ grep -r phy-connection-type arch/powerpc/* | wc 247 990 18335 It was the early adopter of DT, long before ARM. The DT standard of the time, which PowerPC followed, said to use phy-connection-type. When ARM started using DT, it did not follow the standard too well, and phy-mode got used a lot. Officially, phy-mode is deprecated because it is not part of the standard. But practice does not always follow the standard. So both are O.K, all the generic code accepts both, and there is no real reason to change. Andrew
diff --git a/Documentation/devicetree/bindings/net/hisilicon,hisi-femac.yaml b/Documentation/devicetree/bindings/net/hisilicon,hisi-femac.yaml index ba207f2c9ae4..6c5888897876 100644 --- a/Documentation/devicetree/bindings/net/hisilicon,hisi-femac.yaml +++ b/Documentation/devicetree/bindings/net/hisilicon,hisi-femac.yaml @@ -66,7 +66,7 @@ required: - clocks - resets - reset-names - - phy-mode + - phy-connection-type - phy-handle - hisilicon,phy-reset-delays-us @@ -83,7 +83,7 @@ examples: resets = <&crg 0xec 0>, <&crg 0xec 3>; reset-names = "mac", "phy"; mac-address = [00 00 00 00 00 00]; - phy-mode = "mii"; + phy-connection-type = "mii"; phy-handle = <&fephy>; hisilicon,phy-reset-delays-us = <10000 20000 20000>; };