08:35 <z3ntu> Anyone have a suggestion how battery temperature ADC (

08:37 <z3ntu> Anyone have a suggestion how battery temperature ADC (ADC5_BAT_THERM_*K_PU) can be handled because some batteries have e.g. a 10kΩ NTC resistor in the battery, so the available values of 30K, 100K and 400K all don't give correct readings for the battery temperature?

08:38 <z3ntu> This also seems to be an unsolved problem on downstream msm-* kernels, I've seen ODMs just patch e.g. the 30K lookup table in the vadc driver but obviously that's not a solution that can work upstream

09:38 <marc|gonzalez> hey I got a pretty informative kernel error message :)

09:38 <marc|gonzalez> MultiMedia1: ASoC: no backend DAIs enabled for MultiMedia1, possibly missing ALSA mixer-based routing or UCM profile

09:38 <marc|gonzalez> msm-snd-msm8998 sound: YO snd_pcm_prepare: 1999 ret=-22

12:20 <marc|gonzalez> lumag: konradybcio: krzk: In the following DT node, https://paste.debian.net/1330482/ doesn't sound-dai = <&q6asmdai MSM_FRONTEND_DAI_MULTIMEDIA1>; define a backend DAI for MultiMedia1 ?

12:22 <marc|gonzalez> Looking more closely at dpcm_be_connect()

13:02 <krzk> marc|gonzalez: no, it's not backend but dai on CPU side.

13:02 <marc|gonzalez> ah

13:03 <marc|gonzalez> I'm enabling more debugging to get a better picture

13:03 <krzk> marc|gonzalez: "no backend DAIs enabled for MultiMedia1" is not an error. It only means given frontend is not yet configured when card is brought up, e.g. because of missing ALSA UC<

13:03 <krzk> ALSA UCM

13:04 <marc|gonzalez> There's something missing in the DT, I imagine?

13:04 <krzk> For example if you manually (without ALSA UCM) configure mixers when system and audio comes up, this error is kind of expected

13:04 <krzk> No, it is not related to DT, nothing is missing, at least in case of recent platforms. I don't know that much older designs.

13:05 <marc|gonzalez> "recent platforms" meaning recent kernels?

13:05 <krzk> marc|gonzalez: no, recent hardware, so Audioreach based. We always talk here about most recent kernel.

13:06 <marc|gonzalez> understood. So I'm discussing not so recent HW, 2018 msm8998, which is already similar to previous msm8996 probably 2016

13:18 <marc|gonzalez> https://paste.debian.net/1330486/ is the kernel debug log I get when I run tinyplay equinox.wav from userspace

13:18 <marc|gonzalez> I think I somehow need to specify the routing to HDMI Playback

13:57 <marc|gonzalez> krzk: is https://www.kernel.org/doc/html/latest/sound/soc/dpcm.html the right doc to peruse?

14:27 <krzk> marc|gonzalez: yes, at least part of it.

14:28 <krzk> you almost always need to configure proper mixers and some audio-routing. The dmesg mentioned above looks like exactly that - missing mixers.

14:28 <krzk> marc|gonzalez: and the log is quite explicit: "possibly missing ALSA mixer-based routing or UCM profile"

14:29 <marc|gonzalez> thanks, is mixer config done in the machine driver?

14:29 <krzk> no, in userspace

14:29 <marc|gonzalez> roger that, thanks again

14:29 <krzk> just like UCM. Basically UCM is mixers, just nicely packaged for OS

14:31 <krzk> marc|gonzalez: these might be a bit useful for you: https://github.com/krzk/tools/tree/master/tests-var/qcom but do not treat them literally, because mixers are heavily codec/platform drivers dependend and you have something much, much older.

14:31 <marc|gonzalez> interesting, thanks

14:31 <krzk> RB3 is the oldest there and is actually pre-Audioreach. ALSA UCM repo might have some more stuff for older boards, dunno, never needed

14:31 <marc|gonzalez> I can't load a Use Case Manager using the tinyalsa tools?

14:33 <krzk> hm, not sure, but anyway UCM is just "helper" and you can achieve the same with amixer/tinyalsa.

14:36 <marc|gonzalez> tinymix contents returns Number of controls: 1029

14:36 <marc|gonzalez> including "Playback Channel Map" which sounds relevant