Re: [PATCH v2] iio: adc: ad7280a: use cleanup helpers guard() and scoped_guard() for mutex locking

[Jonathan Cameron <jic23@kernel.org>]

On Fri, 17 Apr 2026 11:05:35 +0300
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Wed, Apr 15, 2026 at 01:08:38PM -0300, Lucas Ivars Cadima Ciziks wrote:
> > Replace open-coded mutex_lock/unlock pairs with the cleanup-based
> > guard() and scoped_guard() helpers in ad7280a_write_thresh(),
> > ad7280_show_balance_timer(), ad7280_store_balance_sw(),
> > ad7280_store_balance_timer() and ad7280_read_raw().
> > 
> > This removes the need for the err_unlock label and explicit
> > mutex_unlock() calls, as the lock is now automatically released
> > when the function returns or the guarded scope exits, regardless
> > of the exit path.  
> 
> ...
> 
> > static ssize_t ad7280_show_balance_timer(struct iio_dev *indio_dev,  
> 
> >  	unsigned int msecs;
> >  	int ret;
> >  
> > -	mutex_lock(&st->lock);
> > +	guard(mutex)(&st->lock);  
> 
> This will add some rather heavy operation into the critical section.
> Why do we need that? Please, use scoped_guard() in such cases.
> 

Even better. Look at how ad7280_read_reg() is used and consider should the lock
actually be external to that?

There might be a strong enough consistency argument for scoped guard out here.
I haven't really looked at it closely.
> >  	ret = ad7280_read_reg(st, chan->address >> 8,
> >  			      (chan->address & 0xFF) + AD7280A_CB1_TIMER_REG);
> > -	mutex_unlock(&st->lock);
> >  
> >  	if (ret < 0)
> >  		return ret;  
>