On Tue, Feb 9, 2021 at 10:55 PM Florian Fainelli <f.fainelli@xxxxxxxxx> wrote: > > On 2/9/21 9:23 AM, Vladimir Oltean wrote: > > On Tue, Feb 09, 2021 at 11:18:12AM -0600, George McCollister wrote: > >> On Tue, Feb 9, 2021 at 10:33 AM Souptick Joarder <jrdr.linux@xxxxxxxxx> wrote: > >>> I think, this can defined under #ifdef CONFIG_OF. > >> > >> Was this robot reporting this warning for other DSA drivers? A lot of > >> them appear to use of_device_id in the same manner (unless I'm missing > >> something). Do we need to do something DSA wide to deal with this? > > > > Probably not DSA-wide since we still have drivers that probe on platform > > data and not OF (dsa_loop), but individual drivers should have "depends > > on OF" in Kconfig as far as I can tell. > > Not necessarily, there is a runtime dependency on OF being available for > probing but as far as building goes there are stubs provided for when > CONFIG_OF=n which allow the driver to continue to build. > > Adding either an #ifdef CONFIG_OF or __maybe_unused should silence those > warnings. Agree. Some of the similar warning in other drivers was fixed in the same way. > -- > Florian
