On 11/03/2023 12:16, Hans de Goede wrote: > > "hideep,force-native-protocol" is a good suggestion I'll prepare a new > version with that. > >>> Anyways I just realized I should have not included this at all, >>> since atm this new property is only used on X86/ACPI platforms >>> (through platform code setting a device-property), so it is not >>> used on devicetree platforms at all. >> >> Even if such properties are not documented I do not see how it will >> prevent people from using them... I guess if they validate DT they will >> be caught, but I am not sure that we can rely on this happening. > > Right, but I have beene explicitly told multiple times (1) to not document > device-properties when they are only used between x86 platform code and > drivers consuming them (and thus not actually used in any DT files > at that point in time). Comment in the code that this is not a DT property, is the most we can do now (so people won't use it later as argument to add to the binding)... Best regards, Krzysztof
