On 11.06.2022 20:54, Jonathan Cameron wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
>
> On Thu, 9 Jun 2022 11:32:04 +0300
> Claudiu Beznea <claudiu.beznea@xxxxxxxxxxxxx> wrote:
>
>> Simplify a bit the code in at91_adc_read_info_raw() by reducing the
>> number of lines of code.
>>
>> Signed-off-by: Claudiu Beznea <claudiu.beznea@xxxxxxxxxxxxx>
>
> I'm not convinced this is worth while, but there are some lesser
> steps visible in this patch that probably are.
>
> Given your earlier reorg to move at01_adc_adjust_val_osr() under the locks,
> you can now move the locks to the caller, thus not needing to handle them
> separately in all the exit paths.
OK, I'll give it a try. With this, would you prefer to still keep this patch?
>
>> ---
>> drivers/iio/adc/at91-sama5d2_adc.c | 35 +++++++++---------------------
>> 1 file changed, 10 insertions(+), 25 deletions(-)
>>
>> diff --git a/drivers/iio/adc/at91-sama5d2_adc.c b/drivers/iio/adc/at91-sama5d2_adc.c
>> index b52f1020feaf..fbb98e216e70 100644
>> --- a/drivers/iio/adc/at91-sama5d2_adc.c
>> +++ b/drivers/iio/adc/at91-sama5d2_adc.c
>> @@ -1576,6 +1576,7 @@ static int at91_adc_read_info_raw(struct iio_dev *indio_dev,
>> struct iio_chan_spec const *chan, int *val)
>> {
>> struct at91_adc_state *st = iio_priv(indio_dev);
>> + int (*fn)(struct at91_adc_state *, int, u16 *) = NULL;
>> u16 tmp_val;
>> int ret;
>>
>> @@ -1583,29 +1584,18 @@ static int at91_adc_read_info_raw(struct iio_dev *indio_dev,
>> * Keep in mind that we cannot use software trigger or touchscreen
>> * if external trigger is enabled
>> */
>> - if (chan->type == IIO_POSITIONRELATIVE) {
>> - ret = iio_device_claim_direct_mode(indio_dev);
>> - if (ret)
>> - return ret;
>
> You can drop this out of the if statements as it happens in all paths.
> Or even better, move it to the caller..
>
>> - mutex_lock(&st->lock);
>> -
>> - ret = at91_adc_read_position(st, chan->channel,
>> - &tmp_val);
>
> huh? ret not checked?
Yep, this should have been missed...
>
>> - *val = tmp_val;
>> - ret = at91_adc_adjust_val_osr(st, val);
> Sure this is duplicated, but meh it's only a few lines.
>
>
>> - mutex_unlock(&st->lock);
>> - iio_device_release_direct_mode(indio_dev);
>
> this early release (compared to the long path) is the only bit really
> gets duplicated in all paths..
>
>> + if (chan->type == IIO_POSITIONRELATIVE)
>> + fn = at91_adc_read_position;
>> + if (chan->type == IIO_PRESSURE)
>> + fn = at91_adc_read_pressure;
>>
>> + ret = iio_device_claim_direct_mode(indio_dev);
>> + if (ret)
>> return ret;
>> - }
>> - if (chan->type == IIO_PRESSURE) {
> this should always have been an else if () as the chan->type couldn't be both.
>
>> - ret = iio_device_claim_direct_mode(indio_dev);
>> - if (ret)
>> - return ret;
>> - mutex_lock(&st->lock);
> hence this lot can be shared with the above.
To be sure of what I've understood correctly: in the end you prefer to have
a patch with the point you suggested rather then the initial patch?
Thank you,
Claudiu Beznea
>
>> + mutex_lock(&st->lock);
>>
>> - ret = at91_adc_read_pressure(st, chan->channel,
>> - &tmp_val);
>> + if (fn) {
>> + ret = fn(st, chan->channel, &tmp_val);
>> *val = tmp_val;
>> ret = at91_adc_adjust_val_osr(st, val);
>> mutex_unlock(&st->lock);
>> @@ -1616,11 +1606,6 @@ static int at91_adc_read_info_raw(struct iio_dev *indio_dev,
>>
>> /* in this case we have a voltage channel */
>>
>> - ret = iio_device_claim_direct_mode(indio_dev);
>> - if (ret)
>> - return ret;
>> - mutex_lock(&st->lock);
>> -
>> st->chan = chan;
>>
>> at91_adc_cor(st, chan);
>
