Re: [PATCH 5/9] thermal/core: Introduce locked version of thermal_zone_device_update

[Guenter Roeck <linux@roeck-us.net>]

On Thu, Nov 10, 2022 at 02:01:49PM +0100, Rafael J. Wysocki wrote:
> On Thu, Nov 10, 2022 at 1:25 AM Guenter Roeck <linux@roeck-us.net> wrote:
> >
> > On Wed, Nov 09, 2022 at 08:15:17PM +0100, Rafael J. Wysocki wrote:
> > > On Mon, Oct 17, 2022 at 3:09 PM Guenter Roeck <linux@roeck-us.net> wrote:
> > > >
> > > > In thermal_zone_device_set_mode(), the thermal zone mutex is released only
> > > > to be reacquired in the subsequent call to thermal_zone_device_update().
> > > >
> > > > Introduce __thermal_zone_device_update() as locked version of
> > >
> > > Did you mean "unlocked"?
> > >
> > No, I did mean "locked", as in "must be called with thermal zone device
> > mutex acquired".
> >
> > locked:
> >
> > void __thermal_zone_device_update(struct thermal_zone_device *tz,
> >                                   enum thermal_notify_event event)
> > {
> >         ...
> > }
> >
> > unlocked:
> >
> > void thermal_zone_device_update(struct thermal_zone_device *tz,
> >                                 enum thermal_notify_event event)
> > {
> >         mutex_lock(&tz->lock);
> >         if (device_is_registered(&tz->device))
> >                 __thermal_zone_device_update(tz, event);
> >         mutex_unlock(&tz->lock);
> > }
> 
> Thanks for the explanation.
> 
> > Should I phrase or explain it differently ?
> 
> I would rather say "bare" or something like that so it is all clear to
> people like me, but it is your call.

I updated the commit description to use "must be called with thermal
device mutex held". I kept 'locked' in the subject; I don't think using
'bare' there would add any clarity. Hope that is ok.

Thanks,
Guenter