ChanServ changed the topic of #linux-cros-arm to:
alpernebbi has quit [Ping timeout: 480 seconds]
alpernebbi has joined #linux-cros-arm
hexdump01 has joined #linux-cros-arm
hexdump0815 has quit [Ping timeout: 480 seconds]
dcz_ has joined #linux-cros-arm
dcz_ has quit [Ping timeout: 480 seconds]
<jenneron[m]> janrinze: try adding this to dojo dts https://dpaste.com/DJS9GMUQM
<janrinze> jenneron[m]: will add to the dts, also tested pmOS on snow. works, audio needs a little persuasion using alsamixer. My speakers were already toasted well so can't say anyting on speaker output ;-) headphones work as expected, mesa is a little old thus not allowing much testing on gl-es/openGL. youtube... well.. audio works :D video is (of course) quite stuttery etc.
<jenneron[m]> janrinze: what persuation exactly?
<jenneron[m]> also, chromium should work with gles2
<jenneron[m]> mesa is not old, but it doesn't support gles3 on mali v4
<jenneron[m]> and probably never will
<janrinze> jenneron[m]: with alsamixer we need to select the sound card explicitly, also unmute the headphones and related DAC. Can't seem to persuade it to have microphone input.. Once alsamixer has enabled the audio, all is well. alsaucm reload cannot find the soundcard.
<jenneron[m]> janrinze: try symlnking to snow max98090 configuration
<jenneron[m]> you can find the path with strace alsaucm reload
<janrinze> the dts change did not help for dojo. still the same error. ETDM2_OUT_BE: codec dai not found.
<jenneron[m]> janrinze: do i correctly understand that you don't have chrome os installation anymore?
<janrinze> jenneron[m]: that is correct
<janrinze> jenneron[m]: sof-mt8195.tplg does not exist, also i don't seem to be able to find it on the internet. sof-mt8195.ri is available..
<jenneron[m]> janrinze: this is why i asked about chrome os
<jenneron[m]> there is this file
<janrinze> jenneron[m]: Wish I knew this before changing nvme partitions.
<jenneron[m]> it still should be somewhere in chromiumos git repos
<janrinze> jenneron[m]: I'll live without sound for now. When someone finds the proper files I hope to get a copy ;-)
<jenneron[m]> i will take a look later
<jenneron[m]> need to solve device-tree first
<janrinze> jenneron[m]: yes, that's fine. I have taken a look on where these are required. it appears that the name of the tplg is supposed to be different. also i found.. tada.. a dojo directory with these firmwares.
<janrinze> jenneron[m]: i
<jenneron[m]> where?
<janrinze> on my laptop.. apparently I have forgotten how I got them there.
<janrinze> May have copied them from chromeos previously already.
<janrinze> jenneron[m]: the firmwares for dojo are mediatek/sof/dojo/sof-mt8195.ri and mediatek/sof-tplg/dojo/sof-mt8195-mt6359-max98390-rt5682.tplg
<jenneron[m]> i don't see it in the package
<jenneron[m]> only intel directory
<janrinze> usr/lib/firmware/intel/sof-tplg/sof-mt8195-mt6359-max98390-rt5682.tplg
<janrinze> yes, seems like someone messed up the directory structure.
<janrinze> this clearly does not belong there.
<jenneron[m]> it is also not in linux-firmware
<jenneron[m]> janrinze: we have these 2 files https://pkgs.alpinelinux.org/contents?file=sof-mt8195-*.tplg&path=&name=linux-firmware-mediatek&branch=edge&arch=aarch64
<jenneron[m]> the problem is that they are for rt1019
<jenneron[m]> so it will probably not work for you
<jenneron[m]> maybe you can upload your firmware files somewhere and document it on pmOS wiki?
dcz_ has joined #linux-cros-arm
<jenneron[m]> janrinze: does firmware load fine?
<jenneron[m]> when you place it where kernel expects
<janrinze> I did the birany include in the kernel.
<janrinze> *binary
<jenneron[m]> meh
<jenneron[m]> but does it load?
<janrinze> it does, but i have also changed the name in the source code.
<jenneron[m]> janrinze: i came up with this https://pastebin.com/raw/K55rbrSJ
<jenneron[m]> i'm not sure it is correct
<jenneron[m]> janrinze: you would need to patch device-tree for this to work
<janrinze> jenneron[m]: I also tried what you did, remove the node.
<jenneron[m]> i think
<jenneron[m]> janrinze: and what happened?
<janrinze> jenneron[m]: actually the dojo has mt8195-mt6359-max98390-rt5682 set
<jenneron[m]> does audio work?
<janrinze> jenneron[m]: I added ETDM2_OUT_BE before I fixed the firmware. still fiddling with it. Let's try your dts fix first.
<janrinze> no audio yet.
<janrinze> jenneron[m]: your last pastebin should go in which file?
<jenneron[m]> janrinze: dojo dts
<janrinze> jenneron[m]: replacing the current &sound or add it to it?
<jenneron[m]> huh
<jenneron[m]> let me look
<jenneron[m]> ah
<janrinze> jenneron[m]: I'm on 6.8.0-rc6
<jenneron[m]> nevermind
<jenneron[m]> it's annoying
<janrinze> it is ;-)
<jenneron[m]> my patch is probably incorrect
<jenneron[m]> let's wait for angelo's response
<janrinze> Okay. I'm also curious why snd-soc-max98390 not gets loaded.
<jenneron[m]> janrinze: check `dmesg | grep max98390`
<weirdtreething[m]> <jenneron[m]> "it still should be somewhere..." <- nah
<weirdtreething[m]> it would be in a recovery image
<jenneron[m]> i think janrinze found it
<janrinze> jenneron[m]: as long as they are in the correct directory the kernel will load them
dcz_ has quit [Ping timeout: 480 seconds]
<janrinze> Nothing I can fix here for audio , it seems. we have found the firmware, know that the source code is not selecting the right firmware (needs a patch) and know that the dts is incorrectly setup for sound. As a side note: lsmod and dmesg uploaded as dojo_dmesg.txt and dojo_lsmod.txt
<weirdtreething[m]> nice
KREYREN_oftc has quit [Remote host closed the connection]
KREYREN_oftc has joined #linux-cros-arm
<jenneron[m]> yeah, but your patch is incorrect
<jenneron[m]> it breaks tomato
<janrinze> jenneron[m]: how come? what should be the path? Oh, there is a tomato dir too
<janrinze> jenneron[m]: ah, yes, the patch. That was a quick and dirty hack.
<janrinze> jenneron[m]: The code needs to be split up there in some way to handle firmwares and configs.
<jenneron[m]> yep
KREYREN_oftc has quit [Remote host closed the connection]
KREYREN_oftc has joined #linux-cros-arm
<jenneron[m]> i think
<jenneron[m]> do you have downstream kernel as well? can you send dmesg from there?
<janrinze> jenneron[m]: which downstream kernel?
<jenneron[m]> janrinze: well the downstream kernel
<jenneron[m]> which comes from chrome os
<jenneron[m]> iirc you managed to get it working with debian
<janrinze> I think that's 5.10.199 but I have 5.10.209. Close to the same kernel. Will check the config and rebuild.
<weirdtreething[m]> is the 5.10.209 kernel build from google's source though
<janrinze> jenneron[m]: with 5.10.209 no audio either.
<jenneron[m]> janrinze: i'm interested in dmesg
<janrinze> jenneron[m]: wrong ABI in sof firmware..
<jenneron[m]> janrinze: this is not what i'm interested in
<jenneron[m]> do you have the link to the source of this kernel?
<janrinze> jenneron[m]: the colors in X are all messed up with this kernel.
<janrinze> branch "chromeos-5.10"
<jenneron[m]> what tag though?
<janrinze> ^^
<janrinze> commit 2b27d0c16348f8c7d577647dd85372148443a05a (grafted, HEAD -> chromeos-5.10, origin/chromeos-5.10)
<jenneron[m]> nah not that one
<jenneron[m]> thanks
<janrinze> jenneron[m]: perhaps I should get the 5.10.199 tag to get it same as the chromeos?
<jenneron[m]> there might be a problem with max98390 driver in mainline
<ellyq> i wonder if we even have proper driver for that in mainline aside from asahi's speakersafetyd
<janrinze> ellyq: there is snd-soc-max98090 is mainline.
<jenneron[m]> janrinze: it is max98390
<janrinze> jenneron[m]: yes, sorry for the bad info, it is snd-soc-max98390
<ellyq> dialog semiconductor trying to not confuse people with their naming scheme for 5 minutes challenge
<ellyq> s/dialog/analog
<weirdtreething[m]> <ellyq> "i wonder if we even have..." <- cometlake chromebooks have max98390
<weirdtreething[m]> maybe even some alderlake
KREYREN_oftc has quit [Remote host closed the connection]
KREYREN_oftc has joined #linux-cros-arm
KREYREN_oftc has quit [Remote host closed the connection]
KREYREN_oftc has joined #linux-cros-arm
<ellyq> weirdtreething[m]: but that's SOF, MT8173 didn't use SOF
<weirdtreething[m]> <ellyq> "weirdtreething: but that's SOF..." <- huh what, when did I mention mt8173