From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============7601844739018104230==" MIME-Version: 1.0 From: Pierre-Louis Bossart To: kbuild-all@lists.01.org Subject: Re: sound/soc/intel/boards/hda_dsp_common.c:76: undefined reference to `snd_hda_codec_build_controls' Date: Thu, 06 Feb 2020 12:01:24 -0600 Message-ID: <9d8d130f-47d5-03a5-484e-b0865c6de6f8@linux.intel.com> In-Reply-To: List-Id: --===============7601844739018104230== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable >>> =C2=A0=C2=A0=C2=A0 ld: sound/soc/intel/boards/hda_dsp_common.o: in func= tion = >>> `hda_dsp_hdmi_build_controls': >>>>> sound/soc/intel/boards/hda_dsp_common.c:76: undefined reference to = >>>>> `snd_hda_codec_build_controls' >> >> Looks like the revert select enforcing the built-in of SOF while the >> legacy HDA is a module.=C2=A0 It doesn't look so trivial to fix... > = > SOF in this case is build as a module, but the machine driver isn't. > = > It seems like the SND_SOC_INTEL_SKL_HDA_DSP_GENERIC_MACH option is = > different from others machine drivers. All others can only be either M = > or not selected, but here we have a case where the selection can be M or = y. Suggested fix is here, running it through our CI tests for now: https://github.com/thesofproject/linux/pull/1768 This should be caught by kbuild tests as well. --===============7601844739018104230==--