> -----Original Message----- > From: Aisheng Dong > Sent: Thursday, May 16, 2019 6:40 PM > To: Anson Huang <[email protected]>; [email protected]; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; Daniel Baluta > <[email protected]>; Peng Fan <[email protected]>; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; > [email protected]; [email protected]; linux-arm- > [email protected]; [email protected] > Cc: dl-linux-imx <[email protected]> > Subject: RE: [PATCH V12 3/5] thermal: imx_sc: add i.MX system controller > thermal support > > [...] > > > > > + if (!sensor) > > > > + return -ENOMEM; > > > > + > > > > + ret = thermal_zone_of_get_sensor_id(child, > > > > + &sensor_specs, > > > > > > It looks a bit strange why need sensor_specs as you even did not use it. > > > > It is because parsing the phandle arguments needs it, such as the > > thermal_zone_of_sensor_register() function needs this pointer to be > > passed back to check the elements are valid or NOT, so the API does > > NOT define it as local structure. The API NOT just return the sensor > > id, but also the of_phandle_args address, although we do NOT need it at all > here. > > The main problem is this one introduced in 2/5 which needs get ack from > maintainer. > For me, I would suggest to make sensor_specs transparent to user if it's > really not needed by users. > > Please try to make it either optional or hide to core users as well. > And I guess the later one is doable after a glance at the core code.
OK, I will hide it to user if no other different comments. Anson. > > Regards > Dong Aisheng
