Hi Guenter, > > > > > > My understanding is that once you removed the sysfs files associated with a > > > core, > > > it is safe to remove its tdata entry since sysfs access to the data > associated > > > with the core is no longer possible. So a lock should not be needed here. > > > > > > I am curious - did you experience a scenario which made you believe > > > that you need the lock ? > > > > > > > No I did not experience. While offlining this core, we can possibly unload > the whole module.. > > That's the thought with which I included the lock. > > > Protecting tdata won't help in that case, though. You would need to protect > pdata. > If I understand you correctly, you are concerned that coretemp_remove_core() > can be called from both coretemp_remove() and from put_core_offline(). Yes..Exactly this is my concern. > > I don't think that can happen, since the notifier function is removed before > the platform devices are unregistered. ...got it...Shall remove this unnecessary mutex and roll out v7 soon.. Thanks, Durga _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
