Re: [PATCH v5] iio: imu: kmx61: Use guard(mutex)() over manual locking

[Jonathan Cameron <jic23@kernel.org>]

On Mon, 11 May 2026 12:27:27 -0500
Maxwell Doose <m32285159@gmail.com> wrote:

> On Mon, May 11, 2026 at 12:17 PM Jonathan Cameron <jic23@kernel.org> wrote:
> >
> > On Fri, 8 May 2026 08:03:45 -0500
> > Maxwell Doose <m32285159@gmail.com> wrote:
> >  
> > > Hi Jonathan,
> > >
> > > On Thu, May 7, 2026 at 5:33 PM Maxwell Doose <m32285159@gmail.com> wrote:  
> > > >  
> > > [snip]  
> > > > @@ -945,29 +942,26 @@ static int kmx61_write_event_config(struct iio_dev *indio_dev,
> > > >         if (state && data->ev_enable_state)
> > > >                 return 0;
> > > >
> > > > -       mutex_lock(&data->lock);
> > > > +       guard(mutex)(&data->lock);
> > > >  
> > >
> > > Correct me if I'm wrong but this looks like a TOCTOU race, was
> > > reported by sashiko at
> > > https://sashiko.dev/#/patchset/20260507223337.48437-1-m32285159%40gmail.com.
> > > If this is a confirmed race I'll get a fix out this evening.  
> > Indeed looks like a correct report.
> >
> > Take your time though. I'm swamped anyway so will be a while before
> > I get to any new patches (probably!)
> >  
> 
> Alrighty sounds good. Dunno if we merge this or not and then send the
> fix just in case or if I base it off of the original. Former seems
> preferable.

Given we know about it the fix should go first then this change to guard()
etc based on top of that.

Jonathan

> 
> best regards,
> max
> 
> 
> 
> > >
> > > best regards,
> > > max  
> >