On Mon, Feb 04, 2019 at 03:36:21PM -0800, Florian Fainelli wrote: > Hi all, > > Based on discussion with Ido and feedback from Jakub there are clearly > two classes of users that implement SWITCHDEV_ATTR_ID_PORT_PARENT_ID: > > - PF/VF drivers which typically only implement return the port's parent > ID, yet have to implement switchdev_port_attr_get() just for that > > - Ethernet switch drivers: mlxsw, ocelot, DSA, etc. which implement more > attributes which we want to be able to eventually veto in the context > of the caller, thus making them candidates for using a blocking notifier > chain Florian, thanks for working on this. I think you're missing implementation over stacked devices. At least 8021q, bond and team need to implement this ndo as well. For bond and team we can return an error if not all the slaves have the same parent ID. It would be consistent with what switchdev_port_attr_get() is doing. _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
