Re: [PATCH v8 5/5] pwm: core: Set enable state properly on failed call to enable

[Thierry Reding]

On Tue, May 26, 2015 at 01:08:20PM -0700, Jonathan Richardson wrote:
> The pwm_enable function didn't clear the enabled bit if a call to a
> clients enable function returned an error. The result was that the state
> of the pwm core was wrong. Clearing the bit when enable returns an error
> ensures the state is properly set.
> 
> Tested-by: Jonathan Richardson <jonathar@xxxxxxxxxxxx>
> Reviewed-by: Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx>
> Signed-off-by: Jonathan Richardson <jonathar@xxxxxxxxxxxx>
> ---
>  drivers/pwm/core.c |   16 +++++++++++++---
>  1 file changed, 13 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c
> index 224645f..18f5ac4 100644
> --- a/drivers/pwm/core.c
> +++ b/drivers/pwm/core.c
> @@ -477,10 +477,20 @@ EXPORT_SYMBOL_GPL(pwm_set_polarity);
>   */
>  int pwm_enable(struct pwm_device *pwm)
>  {
> -	if (pwm && !test_and_set_bit(PWMF_ENABLED, &pwm->flags))
> -		return pwm->chip->ops->enable(pwm->chip, pwm);
> +	int err;
> +
> +	if (!pwm)
> +		return -EINVAL;
> +
> +	if (!test_and_set_bit(PWMF_ENABLED, &pwm->flags)) {
> +		err = pwm->chip->ops->enable(pwm->chip, pwm);
> +		if (err) {
> +			clear_bit(PWMF_ENABLED, &pwm->flags);
> +			return err;
> +		}
> +	}

I think with this new pattern we're now actually going to need a lock to
make sure pwm->flags doesn't change between the test_and_set_bit() and
clear_bit() calls.

Thierry

Attachment: pgpNpBGcuITeb.pgp
Description: PGP signature