On Thu, 20 Oct 2022 12:02:53 +0300
Cosmin Tanislav <demonsingur@xxxxxxxxx> wrote:
> From: Cosmin Tanislav <cosmin.tanislav@xxxxxxxxxx>
>
> Currently, every time the device wakes up from sleep, the
> iio_chan array is reallocated, leaking the previous one
> until the device is removed (basically never).
>
> Move the allocation to the probe function to avoid this.
>
> Fixes: f110f3188e56 ("iio: temperature: Add support for LTC2983")
> Signed-off-by: Cosmin Tanislav <cosmin.tanislav@xxxxxxxxxx>
Note I already have this queued as a fix and I've sent the pull request to Greg this
morning. For future versions of this series, maybe just put a note in the cover
letter to cover the dependency.
Jonathan
> ---
> drivers/iio/temperature/ltc2983.c | 13 ++++++-------
> 1 file changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/iio/temperature/ltc2983.c b/drivers/iio/temperature/ltc2983.c
> index b652d2b39bcf..a60ccf183687 100644
> --- a/drivers/iio/temperature/ltc2983.c
> +++ b/drivers/iio/temperature/ltc2983.c
> @@ -1385,13 +1385,6 @@ static int ltc2983_setup(struct ltc2983_data *st, bool assign_iio)
> return ret;
> }
>
> - st->iio_chan = devm_kzalloc(&st->spi->dev,
> - st->iio_channels * sizeof(*st->iio_chan),
> - GFP_KERNEL);
> -
> - if (!st->iio_chan)
> - return -ENOMEM;
> -
> ret = regmap_update_bits(st->regmap, LTC2983_GLOBAL_CONFIG_REG,
> LTC2983_NOTCH_FREQ_MASK,
> LTC2983_NOTCH_FREQ(st->filter_notch_freq));
> @@ -1514,6 +1507,12 @@ static int ltc2983_probe(struct spi_device *spi)
> gpiod_set_value_cansleep(gpio, 0);
> }
>
> + st->iio_chan = devm_kzalloc(&spi->dev,
> + st->iio_channels * sizeof(*st->iio_chan),
> + GFP_KERNEL);
> + if (!st->iio_chan)
> + return -ENOMEM;
> +
> ret = ltc2983_setup(st, true);
> if (ret)
> return ret;
