Currently we notify a voltage change whenever we exit set_voltage(),
even if the change failed for some reason (eg, a constraints issue).
This shouldn't cause any substantial ill effects but is wasteful as
listeners get notified on noops. Fix this by moving the notification
into _do_set_voltage() and only notifying if we don't return an error.
Reported-by: Saravana Kannan <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
---
drivers/regulator/core.c | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index a0579f0..3d72cc8 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -1667,6 +1667,10 @@ static int _regulator_do_set_voltage(struct regulator_dev *rdev,
ret = -EINVAL;
}
+ if (ret == 0)
+ _notifier_call_chain(rdev, REGULATOR_EVENT_VOLTAGE_CHANGE,
+ NULL);
+
trace_regulator_set_voltage_complete(rdev_get_name(rdev), selector);
return ret;
@@ -1718,7 +1722,6 @@ int regulator_set_voltage(struct regulator *regulator, int min_uV, int max_uV)
ret = _regulator_do_set_voltage(rdev, min_uV, max_uV);
out:
- _notifier_call_chain(rdev, REGULATOR_EVENT_VOLTAGE_CHANGE, NULL);
mutex_unlock(&rdev->mutex);
return ret;
}
--
1.7.1
On Thu, 2010-12-16 at 13:59 +0000, Mark Brown wrote:
> Currently we notify a voltage change whenever we exit set_voltage(),
> even if the change failed for some reason (eg, a constraints issue).
> This shouldn't cause any substantial ill effects but is wasteful as
> listeners get notified on noops. Fix this by moving the notification
> into _do_set_voltage() and only notifying if we don't return an error.
>
> Reported-by: Saravana Kannan <[email protected]>
> Signed-off-by: Mark Brown <[email protected]>
> ---
> drivers/regulator/core.c | 5 ++++-
> 1 files changed, 4 insertions(+), 1 deletions(-)
>
Applied.
Thanks
Liam
--
Freelance Developer, SlimLogic Ltd
ASoC and Voltage Regulator Maintainer.
http://www.slimlogic.co.uk