On 05/18/11 17:02, Guenter Roeck wrote:
> On Wed, May 18, 2011 at 11:39:31AM -0400, Jonathan Cameron wrote:
>> From: Fabien Marteau <fabien.marteau-d2DlULPkwbNWk0Htik3J/w@xxxxxxxxxxxxxxxx>
>>
>> Signed-off-by: Jonathan Cameron <jic23-KWPb1pKIrIJaa/9Udqfwiw@xxxxxxxxxxxxxxxx>
>
> Hi Jonathan,
>
> nice job. Thanks a lot for the effort.
>
> [ ... ]
>
>> +static int as1531_read_raw(struct iio_dev *indio_dev,
>> + struct iio_chan_spec const *chan,
>> + int *val,
>> + int *val2,
>> + long m)
>> +{
>> +
> Extra blank line
>
>> + int status = 0;
>> + int ret_value = 0;
>> + struct as1531_state *st = iio_priv(indio_dev);
>
> Move it to here, maybe ?
>
>> + if (mutex_lock_interruptible(&st->lock))
>> + return -ERESTARTSYS;
>> +
>> + status = as1531_message(st->spi,
>> + AS1531_START_BIT | chan->address |
>> + AS1531_RANGE_0_TO_VREF | AS1531_MODE_COM |
>> + AS1531_POWER_NORMAL,
>> + &ret_value);
>> + mutex_unlock(&st->lock);
>> + if (status < 0)
>> + goto out;
>> +
>> + *val = ret_value*2500/4096;
>> +
>> + return IIO_VAL_INT;
>> +out:
>> + mutex_unlock(&st->lock);
>
> Mutex was unlocked above already. Maybe just return status above ?
Hmm.. Sometimes working without testing shows just how bad first versions
of code really can be ;)
Thanks for the other review as well. Have fixed up both.
Thanks.
Jonathan
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
