On Thu, 9 Jun 2022 11:32:00 +0300
Claudiu Beznea <claudiu.beznea@xxxxxxxxxxxxx> wrote:
> When buffers are enabled conversion may start asynchronously thus
> allowing changes on actual hardware could lead to bad behavior. Thus
> do not allow changing oversampling ratio and sample frequency when
> buffers are enabled.
Less than desirable behavior perhaps, but broken? I don't see this
as a fix from what you have mentioned - though I'm not against it.
(just drop the fixes tag)
It is an ABI change, but unlikely to be one any sane code hits.
>
> Fixes: 5e1a1da0f8c9 ("iio: adc: at91-sama5d2_adc: add hw trigger and buffer support")
> Signed-off-by: Claudiu Beznea <claudiu.beznea@xxxxxxxxxxxxx>
> ---
> drivers/iio/adc/at91-sama5d2_adc.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/iio/adc/at91-sama5d2_adc.c b/drivers/iio/adc/at91-sama5d2_adc.c
> index a672a520cdc0..b76328da0cb2 100644
> --- a/drivers/iio/adc/at91-sama5d2_adc.c
> +++ b/drivers/iio/adc/at91-sama5d2_adc.c
> @@ -1644,6 +1644,9 @@ static int at91_adc_write_raw(struct iio_dev *indio_dev,
> {
> struct at91_adc_state *st = iio_priv(indio_dev);
>
> + if (iio_buffer_enabled(indio_dev))
> + return -EBUSY;
This is racy as nothing stops buffers being enabled after this point.
Use the iio_device_claim_direct_mode() and release for this as they
protect against the race.
> +
> switch (mask) {
> case IIO_CHAN_INFO_OVERSAMPLING_RATIO:
> if ((val != AT91_OSR_1SAMPLES) && (val != AT91_OSR_4SAMPLES) &&
