Re: [PATCH 1/2] ASoC: SOF: Intel: select SND_SOC_SDW_UTILS=y from SND_SOC_SOF_HDA_GENERIC=y

[Mark Brown <broonie@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 1203 bytes --]

On Thu, Jun 11, 2026 at 08:22:21PM +0200, Arnd Bergmann wrote:
> On Thu, Jun 11, 2026, at 16:20, Mark Brown wrote:

> > Can we end up with SND_SOC_SDW_UTILS=y but SOUNDWIRE=m?  Though I think
> > that's a potential issue anyway...

> I don't see that combination in any randconfig builds and haven't
> managed to produce it in menuconfig, but I can't swear that it's
> impossible either.

I think

  SND_SOC_SOF_SKYLAKE=y
  SND_SOC_SOF_LUNARLAKE=m
  SOUNDWIRE=m

Skylake doesn't have Soundwire but does select the Intel stuff.  I
didn't test though.

> I do feel that the entire set of dependencies around
> SOUNDWIRE, HDA and the Intel/AMD drivers is too fragile,
> so it would be good if someone could manage to simplify this
> in some form, but the best I can do at the moment it to
> address regressions when they happen.

Yes, this stuff is all miserable and I'm not convinced the granularity
is useful.  Kconfig is pretty unhelpful for this sort of library code.

> With the patches that I have sent, plus the "ASoC: amd: acp: Fix
> linker error with SDCA quirks" one from Syed Saba Kareem that
> I'm testing at the moment, I get no link failures in randconfig
> kernels.

It does fix some stuff.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]