ASoC: ops: Log unknown controls in snd_soc_limit_volume()
Mark Brown
maintainerPATCH 1/1 ASoC: ops: Log unknown controls in snd_soc_limit_volume()
When we fail to look up the control name in snd_soc_limit_volume() we don't log anything, the error code isn't particularly descriptive and checking the return value of the function at all is a bit erratic among the callers. Since there is no reason why anyone should ever be attempting to limit the volume of a nonexistant control add a log message in the core to improve usability.
--- sound/soc/soc-ops.c | 3 +++ 1 file changed, 3 insertions(+)
sound/soc/soc-ops.c +3 -0
@@ -503,6 +503,9 @@ int snd_soc_limit_volume(struct snd_soc_card *card, const char *name, int max) mc->platform_max = max; ret =snd_soc_clip_to_platform_max (kctl); } + }else { +dev_warn (card->dev,"Volume limit for unknown control '%s'\n" , + name); }return ret;
---
base-commit: 028ef9c96e96197026887c0f092424679298aae8
change-id: 20260421-asoc-limit-volume-error-ca28188dc7c9
Best regards,
Signature · 1 line
Mark Brown <[email protected]>
Quoted 7 lines · Mark Brown
When we fail to look up the control name in
snd_soc_limit_volume()we don't log anything, the error code isn't particularly descriptive and checking the return value of the function at all is a bit erratic among the callers. Since there is no reason why anyone should ever be attempting to limit the volume of a nonexistant control add a log message in the core to improve usability.
Can confirm this works as expected for the case we discussed previously. With this I get:
[ 19.585906] snd-x1e80100 sound: Volume limit for unknown control 'WSA_RX0 Digital Volume' [ 19.585909] snd-x1e80100 sound: Volume limit for unknown control
'WSA_RX1 Digital Volume'
One thing I'd be worried about though is the number of false positives this might generate.
The x1e80100 patch we discussed previously handles a 2-speaker and a 4-speaker configuration
by unconditionally adding limits for all of them. Previously the missing ones didn't really matter with
this they also print warnings.
Quoted 29 lines
Signed-off-by Mark Brown <[email protected]>--- sound/soc/soc-ops.c | 3 +++ 1 file changed, 3 insertions(+)sound/soc/soc-ops.c +3
-1@@ -503,6 +503,9 @@ int snd_soc_limit_volume(struct snd_soc_card *card, const char *name, int max) mc->platform_max = max; ret =snd_soc_clip_to_platform_max (kctl); } + }else { +dev_warn (card->dev,"Volume limit for unknown control '%s'\n" , + name); }return ret; ---base-commit:
028ef9c96e96197026887c0f092424679298aae8
change-id: 20260421-asoc-limit-volume-error-ca28188dc7c9Best regards,
--
Mark Brown <[email protected]>
Mark Brown
maintainerQuoted 7 lines · Tobias Heider
One thing I'd be worried about though is the number of false positives this might generate.
The x1e80100 patch we discussed previously handles a 2-speaker and a 4-speaker configuration
by unconditionally adding limits for all of them. Previously the missing ones didn't really matter with
this they also print warnings.
Right. Ideally the card would know what it was registering but perhaps that's more trouble than it's worth. I'll split the difference and lower the print to a dev_dbg().