On Fri, 2016-03-04 at 19:53 +0100, Bjørn Mork wrote:
> Bjørn Mork <bjorn@xxxxxxx> writes:
>
> > > +static void iwl_mvm_thermal_zone_unregister(struct iwl_mvm *mvm)
> > > +{
> > > + if (!iwl_mvm_is_tt_in_fw(mvm))
> > > + return;
> > > +
> > > + if (mvm->tz_device.tzone) {
> > > + IWL_DEBUG_TEMP(mvm, "Thermal zone device
> > > unregister\n");
> > > + thermal_zone_device_unregister(mvm-
> > > >tz_device.tzone);
> >
> > Won't that ERR_PTR blow up when dereferenced by
> > thermal_zone_device_unregister() ?
>
> To answer myself: No, it won't. I was tricked by the oddly placed
> tzp = tz->tzp;
> line here, but we will return before that becomes a problem:
>
> void thermal_zone_device_unregister(struct thermal_zone_device *tz)
> {
> int i;
> const struct thermal_zone_params *tzp;
> struct thermal_cooling_device *cdev;
> struct thermal_zone_device *pos = NULL;
>
> if (!tz)
> return;
>
> tzp = tz->tzp;
>
> mutex_lock(&thermal_list_lock);
> list_for_each_entry(pos, &thermal_tz_list, node)
> if (pos == tz)
> break;
> if (pos != tz) {
> /* thermal zone device not found */
> mutex_unlock(&thermal_list_lock);
> return;
> }
>
>
> Still think it's unwise to leave ERR_PTR's around though.
The main point is that we don't want to fail the entire driver loading
if the thermal zone (or cooling device) registration fails. We have a
fallback for that, which is what was used before this new
implementation (namely, the thermal throttling code in the driver).
IIRC Emmanuel has a patch in the queue to clean this up a bit...
--
Cheers,
Luca.��.n��������+%������w��{.n�����{���zW����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
