sound/soc/codecs/Kconfig | 1 + 1 file changed, 1 insertion(+)
From: Arnd Bergmann <arnd@arndb.de>
Selecting SND_SOC_SDCA_HID only works if HID is enabled:
WARNING: unmet direct dependencies detected for SND_SOC_SDCA_HID
Depends on [n]: SOUND [=y] && SND [=y] && SND_SOC [=y] && SND_SOC_SDCA [=y] && (HID [=n]=y [=y] || HID [=n]=SND_SOC_SDCA [=y])
Selected by [y]:
- SND_SOC_RT766_SDCA_SDW [=y] && SOUND [=y] && SND [=y] && SND_SOC [=y] && SOUNDWIRE [=y] && SND_SOC_SDCA [=y]
Add this an a dependency to SND_SOC_RT766_SDCA_SDW as well.
Fixes: b50ecf5873df ("ASoC: rt766: add RT766/RT767 SDCA driver")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
sound/soc/codecs/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index d3730c4da51b..35879cf79472 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -1992,6 +1992,7 @@ config SND_SOC_RT766_SDCA_SDW
tristate "Realtek RT766 SDCA Codec - SDW"
depends on SOUNDWIRE
depends on SND_SOC_SDCA
+ depends on HID
select SND_SOC_SDCA_HID
select SND_SOC_SDCA_IRQ
select REGMAP_SOUNDWIRE
--
2.53.0
On Tue, Sep 15, 2026 at 11:16:34PM +0200, Arnd Bergmann wrote: > From: Arnd Bergmann <arnd@arndb.de> > Selecting SND_SOC_SDCA_HID only works if HID is enabled: > --- a/sound/soc/codecs/Kconfig > +++ b/sound/soc/codecs/Kconfig > @@ -1992,6 +1992,7 @@ config SND_SOC_RT766_SDCA_SDW > tristate "Realtek RT766 SDCA Codec - SDW" > depends on SOUNDWIRE > depends on SND_SOC_SDCA > + depends on HID > select SND_SOC_SDCA_HID > select SND_SOC_SDCA_IRQ > select REGMAP_SOUNDWIRE Should that be HID=y || HID=SND_SOC_SDCA like for SND_SOC_SDCA_HID and SND_SOC_SDCA_CLASS? The dependencies on _CLASS should stop the problematic cases from actually occurring but I'm not sure if it stops the warning.
On Wed, Sep 16, 2026, at 17:10, Mark Brown wrote:
> On Tue, Sep 15, 2026 at 11:16:34PM +0200, Arnd Bergmann wrote:
>> From: Arnd Bergmann <arnd@arndb.de>
>
>> Selecting SND_SOC_SDCA_HID only works if HID is enabled:
>
>> --- a/sound/soc/codecs/Kconfig
>> +++ b/sound/soc/codecs/Kconfig
>> @@ -1992,6 +1992,7 @@ config SND_SOC_RT766_SDCA_SDW
>> tristate "Realtek RT766 SDCA Codec - SDW"
>> depends on SOUNDWIRE
>> depends on SND_SOC_SDCA
>> + depends on HID
>> select SND_SOC_SDCA_HID
>> select SND_SOC_SDCA_IRQ
>> select REGMAP_SOUNDWIRE
>
> Should that be HID=y || HID=SND_SOC_SDCA like for SND_SOC_SDCA_HID and
> SND_SOC_SDCA_CLASS? The dependencies on _CLASS should stop the
> problematic cases from actually occurring but I'm not sure if it stops
> the warning.
Yes. Just as I was trying to write up my presentation slides for next
week trying to explain how nobody ever gets this right, you have reminded
me that I can't do it either ;-)
How about this version?
diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index 181ed1b8a6dc..f88381523ea3 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -1991,10 +1991,7 @@ config SND_SOC_RT715_SDCA_SDW
config SND_SOC_RT766_SDCA_SDW
tristate "Realtek RT766 SDCA Codec - SDW"
depends on SOUNDWIRE
- depends on SND_SOC_SDCA
- depends on HID
- select SND_SOC_SDCA_HID
- select SND_SOC_SDCA_IRQ
+ depends on SND_SOC_SDCA_CLASS
select REGMAP_SOUNDWIRE
select REGMAP_SOUNDWIRE_MBQ
I think this is much clearer than the original version or my
first patch, as it avoids the fragile 'select'. If you like this
one, I'll send a v2 after this passes some more randconfig testing,
or you can apply that directly.
Arnd
On Wed, Sep 16, 2026 at 05:36:02PM +0200, Arnd Bergmann wrote: > On Wed, Sep 16, 2026, at 17:10, Mark Brown wrote: > > On Tue, Sep 15, 2026 at 11:16:34PM +0200, Arnd Bergmann wrote: > >> depends on SOUNDWIRE > >> depends on SND_SOC_SDCA > >> + depends on HID > > Should that be HID=y || HID=SND_SOC_SDCA like for SND_SOC_SDCA_HID and > > SND_SOC_SDCA_CLASS? The dependencies on _CLASS should stop the > > problematic cases from actually occurring but I'm not sure if it stops > > the warning. > Yes. Just as I was trying to write up my presentation slides for next > week trying to explain how nobody ever gets this right, you have reminded > me that I can't do it either ;-) > How about this version? ... > I think this is much clearer than the original version or my > first patch, as it avoids the fragile 'select'. If you like this > one, I'll send a v2 after this passes some more randconfig testing, > or you can apply that directly. Looks good to me but I'm not entirely confident in my Kconfig special case knowledge, I'll wait for your new version after testing.
On Wed, Sep 16, 2026 at 05:36:02PM +0200, Arnd Bergmann wrote: > On Wed, Sep 16, 2026, at 17:10, Mark Brown wrote: > > On Tue, Sep 15, 2026 at 11:16:34PM +0200, Arnd Bergmann wrote: > >> Selecting SND_SOC_SDCA_HID only works if HID is enabled: > > > >> --- a/sound/soc/codecs/Kconfig > >> +++ b/sound/soc/codecs/Kconfig > >> @@ -1992,6 +1992,7 @@ config SND_SOC_RT766_SDCA_SDW > >> tristate "Realtek RT766 SDCA Codec - SDW" > >> depends on SOUNDWIRE > >> depends on SND_SOC_SDCA > >> + depends on HID > >> select SND_SOC_SDCA_HID > >> select SND_SOC_SDCA_IRQ > >> select REGMAP_SOUNDWIRE > > > > Should that be HID=y || HID=SND_SOC_SDCA like for SND_SOC_SDCA_HID and > > SND_SOC_SDCA_CLASS? The dependencies on _CLASS should stop the > > problematic cases from actually occurring but I'm not sure if it stops > > the warning. > > Yes. Just as I was trying to write up my presentation slides for next > week trying to explain how nobody ever gets this right, you have reminded > me that I can't do it either ;-) > > How about this version? > > diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig > index 181ed1b8a6dc..f88381523ea3 100644 > --- a/sound/soc/codecs/Kconfig > +++ b/sound/soc/codecs/Kconfig > @@ -1991,10 +1991,7 @@ config SND_SOC_RT715_SDCA_SDW > config SND_SOC_RT766_SDCA_SDW > tristate "Realtek RT766 SDCA Codec - SDW" > depends on SOUNDWIRE > - depends on SND_SOC_SDCA > - depends on HID > - select SND_SOC_SDCA_HID > - select SND_SOC_SDCA_IRQ > + depends on SND_SOC_SDCA_CLASS > select REGMAP_SOUNDWIRE > select REGMAP_SOUNDWIRE_MBQ > > I think this is much clearer than the original version or my > first patch, as it avoids the fragile 'select'. If you like this > one, I'll send a v2 after this passes some more randconfig testing, > or you can apply that directly. This seems a little odd though, as the driver doesn't actually depend on SND_SOC_SDCA_CLASS. That builds in the actual class driver, which this driver doesn't use, it only uses functions from the SDCA library. THanks, Charles
On Wed, Sep 16, 2026 at 04:10:08PM +0100, Mark Brown wrote: > On Tue, Sep 15, 2026 at 11:16:34PM +0200, Arnd Bergmann wrote: ... > > config SND_SOC_RT766_SDCA_SDW > > tristate "Realtek RT766 SDCA Codec - SDW" > > depends on SOUNDWIRE > > depends on SND_SOC_SDCA > > + depends on HID > > select SND_SOC_SDCA_HID > > select SND_SOC_SDCA_IRQ > > select REGMAP_SOUNDWIRE > > Should that be HID=y || HID=SND_SOC_SDCA like for SND_SOC_SDCA_HID and > SND_SOC_SDCA_CLASS? The dependencies on _CLASS should stop the > problematic cases from actually occurring but I'm not sure if it stops > the warning. (Based just on this comment without looking into details) there is also 'depends on FOO if BAR' syntax that might be useful. -- With Best Regards, Andy Shevchenko
© 2016 - 2026 Red Hat, Inc.