Revert/pr4620

[plbossart]

let's double-check if the bad results in https://sof-ci.01.org/linuxpr/PR4620/build6903/devicetest/index.html are indeed created by PR 4620.

@ujfalusi please double-check the results when you have a chance

[ujfalusi]

@plbossart, I think the issue is coming from the fact that we do not have video support for MTL, thus no HDMI, thus with the split registering of the dais we will not register the ones for HDMI and that will break the tplg which expects that those are there.

The analog codec will register the things for HDMI and even if there is no HDMI audio support available the SW can probe.

[ujfalusi]

@plbossart, confirmed that this is the case. If you boot with nomodeset for example then the card will not form, because
(on a TGL HDA - upx-i11)

cat /proc/asound/cards
cat /sys/kernel/debug/asoc/dais
cat /sys/kernel/debug/asoc/components

diff normal.txt nomodeset.txt :

2,3d1
<  0 [sofhdadsp      ]: sof-hda-dsp - sof-hda-dsp
<                       AAEON-UPX_TGL01-V1.0
23,31d20
< HDMI1
< HDMI2
< HDMI3
< HDA Analog
< DeepBuffer
< intel-hdmi-hifi1
< intel-hdmi-hifi2
< intel-hdmi-hifi3
< intel-hdmi-hifi4
40d28
< ehdaudio0D2

and in kernel log:

[   66.868211] skl_hda_dsp_generic skl_hda_dsp_generic: ASoC: Failed to add route iDisp3 Tx -> hifi3(*)
[   66.868220] skl_hda_dsp_generic skl_hda_dsp_generic: ASoC: Failed to add route iDisp2 Tx -> hifi2(*)
[   66.868226] skl_hda_dsp_generic skl_hda_dsp_generic: ASoC: Failed to add route iDisp1 Tx -> hifi1(*)
[   66.868267] skl_hda_dsp_generic skl_hda_dsp_generic: snd_soc_bind_card: snd_soc_dapm_add_routes failed: -19

Sorry for missing this 'detail'

[ujfalusi]

Sorry, I forgot to switch to the rebase+apply and it created a merge commit.