On Mon, Nov 1, 2021 at 12:49 PM Hans de Goede <hdegoede@xxxxxxxxxx> wrote: > On 11/1/21 11:46, Andy Shevchenko wrote: > > On Mon, Nov 1, 2021 at 12:44 PM Andy Shevchenko > > <andy.shevchenko@xxxxxxxxx> wrote: ... > > Hmm... But then in the original code and with this proposal the > > acpi_dev_put() seems a bit strange to me. > > If we are fine (no error code returned) why would the caller (note > > _er_) go different paths? > > We always need to get the dev to get the name, but some callers are > only interested in the name, so they pass NULL for sensor_adev_ret, > this helps to keep the code calling this clean, which is the whole > idea of having a helper for this. OK. (Not that I'm very happy with this, but... the function needs a comment about this alternative behaviour, I forgot if it already has, though) -- With Best Regards, Andy Shevchenko
