On 28.08.2020 08:48, Krzysztof Kozlowski wrote:
diff --git a/sound/soc/codecs/wm8994.c b/sound/soc/codecs/wm8994.c index b3ba053..fc9ea19 100644 --- a/sound/soc/codecs/wm8994.c +++ b/sound/soc/codecs/wm8994.c @@ -3514,6 +3514,8 @@ int wm8994_mic_detect(struct snd_soc_component *component, struct snd_soc_jack * return -EINVAL; }
- pm_runtime_get_sync(component->dev);
The driver enables PM runtime unconditionally so you should probably handle the error code here. I know that driver does not do it in other cases but it should not be a reason to multiple this pattern... unless it really does not matter as there are no runtime PM ops?
The regmap is provided by the parent MFD device (drivers/mfd/wm8994-core.c) and that is where those runtime PM calls get propagated, we could possibly get en error if there is something wrong with resuming the parent device.
If you don't mind I would prefer to omit the return value tests in that fix patch. Existing callers of the wm89*_mic_detect() functions are ignoring the return value anyway. Probably the checks could be added in a separate patch.