* [PATCH] iio: dac: ad5446: return unshifted values in read_raw
@ 2016-08-30 6:27 Martin Hundebøll
2016-09-03 16:21 ` Jonathan Cameron
0 siblings, 1 reply; 2+ messages in thread
From: Martin Hundebøll @ 2016-08-30 6:27 UTC (permalink / raw)
To: linux-iio; +Cc: Martin Hundebøll, Jonathan Cameron
The Analog Devices DAC chips supported by the ad5456 driver ignore
between 0 and 6 of the least significant bits, and thus the written raw
value is shifted accordingly before being cached and written.
Since the shifted value is cached, it is returned to the user in
ad5446_read_raw(), which might be confusing to the user (at least it was
for me).
Instead store the unshifted value and do the shifting when needed (i.e.
when writing data and disabling powerdown mode).
Signed-off-by: Martin Hundebøll <mnhu@prevas.dk>
---
drivers/iio/dac/ad5446.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/iio/dac/ad5446.c b/drivers/iio/dac/ad5446.c
index 46bb62a..8d172d7 100644
--- a/drivers/iio/dac/ad5446.c
+++ b/drivers/iio/dac/ad5446.c
@@ -118,7 +118,7 @@ static ssize_t ad5446_write_dac_powerdown(struct iio_dev *indio_dev,
shift = chan->scan_type.realbits + chan->scan_type.shift;
val = st->pwr_down_mode << shift;
} else {
- val = st->cached_val;
+ val = st->cached_val << chan->scan_type.shift;
}
ret = st->chip_info->write(st, val);
@@ -195,9 +195,9 @@ static int ad5446_write_raw(struct iio_dev *indio_dev,
if (val >= (1 << chan->scan_type.realbits) || val < 0)
return -EINVAL;
- val <<= chan->scan_type.shift;
mutex_lock(&indio_dev->mlock);
st->cached_val = val;
+ val <<= chan->scan_type.shift;
if (!st->pwr_down)
ret = st->chip_info->write(st, val);
mutex_unlock(&indio_dev->mlock);
--
2.9.3
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] iio: dac: ad5446: return unshifted values in read_raw
2016-08-30 6:27 [PATCH] iio: dac: ad5446: return unshifted values in read_raw Martin Hundebøll
@ 2016-09-03 16:21 ` Jonathan Cameron
0 siblings, 0 replies; 2+ messages in thread
From: Jonathan Cameron @ 2016-09-03 16:21 UTC (permalink / raw)
To: Martin Hundebøll, linux-iio, Lars-Peter Clausen
On 30/08/16 07:27, Martin Hundebøll wrote:
> The Analog Devices DAC chips supported by the ad5456 driver ignore
> between 0 and 6 of the least significant bits, and thus the written raw
> value is shifted accordingly before being cached and written.
>
> Since the shifted value is cached, it is returned to the user in
> ad5446_read_raw(), which might be confusing to the user (at least it was
> for me).
>
> Instead store the unshifted value and do the shifting when needed (i.e.
> when writing data and disabling powerdown mode).
>
> Signed-off-by: Martin Hundebøll <mnhu@prevas.dk>
Lars?
Looks right to me, could you just sanity check this.
Stable material I think.
Jonathan
> ---
> drivers/iio/dac/ad5446.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/iio/dac/ad5446.c b/drivers/iio/dac/ad5446.c
> index 46bb62a..8d172d7 100644
> --- a/drivers/iio/dac/ad5446.c
> +++ b/drivers/iio/dac/ad5446.c
> @@ -118,7 +118,7 @@ static ssize_t ad5446_write_dac_powerdown(struct iio_dev *indio_dev,
> shift = chan->scan_type.realbits + chan->scan_type.shift;
> val = st->pwr_down_mode << shift;
> } else {
> - val = st->cached_val;
> + val = st->cached_val << chan->scan_type.shift;
> }
>
> ret = st->chip_info->write(st, val);
> @@ -195,9 +195,9 @@ static int ad5446_write_raw(struct iio_dev *indio_dev,
> if (val >= (1 << chan->scan_type.realbits) || val < 0)
> return -EINVAL;
>
> - val <<= chan->scan_type.shift;
> mutex_lock(&indio_dev->mlock);
> st->cached_val = val;
> + val <<= chan->scan_type.shift;
> if (!st->pwr_down)
> ret = st->chip_info->write(st, val);
> mutex_unlock(&indio_dev->mlock);
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2016-09-03 16:22 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-08-30 6:27 [PATCH] iio: dac: ad5446: return unshifted values in read_raw Martin Hundebøll
2016-09-03 16:21 ` Jonathan Cameron
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).