ChanServ changed the topic of #msm8937-mainline to: Boot Linux on your MSM8917/37/40 and QM215 mobile! | GitHub: https://github.com/msm89x7-mainline | Logs: https://oftc.irclog.whitequark.org/msm8937-mainline | Bridged to #msm8937-mainline:kde.org on Matrix
<M0xCAFEBABE[m]> is it just something wrong on my side or wifi and audio has really stopped working on 6.15?
<M0xCAFEBABE[m]> msm89x7/6.15 branch
<Biswa96[m]> those are working in xiaomi-land in pmos linux 6.15-r0.
<barni2000[m]> <M0xCAFEBABE[m]> "is it just something wrong on my..." <- i have never tried wifi audio
<barni2000[m]> * audio should be on your side, wifi can be stopped working because of msm-firmware-loader changes
<barni2000[m]> santoni?
<M0xCAFEBABE[m]> iirc it was working on 6.15-rcX though
<M0xCAFEBABE[m]> ugg and ugglite
<barni2000[m]> both are working for me
<barni2000[m]> but i can recheck
<barni2000[m]> audio is working
<barni2000[m]> wifi also
<M0xCAFEBABE[m]> okay..
<M0xCAFEBABE[m]> barni2000[m]: are you on the same branch as I mentioned?
<barni2000[m]> yes
<M0xCAFEBABE[m]> <barni2000[m]> "i have never tried wifi audio" <- Looks like msm-firmware-loader didn't get any change recently
<barni2000[m]> there is an 2-3 months old bug
<barni2000[m]> 4
<barni2000[m]> 0xCAFEBABE: what do you know about VoiceMMode and CSVoice? should we support both?
<barni2000[m]> maybe i should try fw update on riva for VoiceMMode
<barni2000[m]> maybe it should work with oreo fw
<M0xCAFEBABE[m]> barni2000[m]: Android 6.0 (some early 7.0 too?) firmware supports only CSVoice, newer firmwares supports both (iirc)
<M0xCAFEBABE[m]> and sdm439 supports only VoiceMMode iirc
<barni2000[m]> SDM439 is does not matter until i have not write an dsi driver for it
<barni2000[m]> i can try define CSVoice
<barni2000[m]> * define CSVoice for test
<M0xCAFEBABE[m]> tho can we somehow implement a fallback logic...
<M0xCAFEBABE[m]> like, if VoiceMMode is unavailable, then fallback to CSVoice
<M0xCAFEBABE[m]> or, define that on dts
<M0xCAFEBABE[m]> has the same lines on both ugg and ugglite
<M0xCAFEBABE[m]> ooh wait I haven't updated kernel config yet, lemme do
<M0xCAFEBABE[m]> * ~~ooh wait, * lemme do~~ doesn't help
<M0xCAFEBABE[m]> * <del>ooh wait, * lemme do doesn't help</del>
<M0xCAFEBABE[m]> * <del>ooh wait, * lemme do</del> doesn't help
<M0xCAFEBABE[m]> oh well, audio not working is because of it's failing to set HPHL/HPHR (manually setting via command works tho)
<M0xCAFEBABE[m]> but still, only just the kernel has been changed
<barni2000[m]> <M0xCAFEBABE[m]> "```..." <- i don't have these lines
<barni2000[m]> i have a feeling something is missing from your config
<M0xCAFEBABE[m]> barni2000[m]: what, it had worked fine on 6.15.0-rc3 (or rc4?)
<M0xCAFEBABE[m]> btw, did you change anything on your side from -rc4 to -r0
<barni2000[m]> maybe
<M0xCAFEBABE[m]> barni2000[m]: what...
<barni2000[m]> i dont remember
<M0xCAFEBABE[m]> do u still have the HEAD of 6.15-rc3?
<barni2000[m]> no
<barni2000[m]> maybe in reflog
<barni2000[m]> can you send me a hash?
<M0xCAFEBABE[m]> nope, i don't have that also
<barni2000[m]> 1337misom: can you please fix your Signed of by? You should put your full name there
<M0xCAFEBABE[m]> also, now ugg's screen dies without any indication on dmesg when it's booted for 2~3 minutes
<M0xCAFEBABE[m]> ugglite's is still fine
<barni2000[m]> what have you changed in the kernel
<barni2000[m]> my ugg is working fine
<M0xCAFEBABE[m]> nothing
<barni2000[m]> i only have annyoing backlight bug but thats all
<M0xCAFEBABE[m]> barni2000[m]: oh yeah, backlight became not changeable on my ugglite
<barni2000[m]> i cannot change the brigthness
<barni2000[m]> it is working like a sinus
<barni2000[m]> s/sinus/sine/
<barni2000[m]> ∿
<barni2000[m]> do you have sc_ili9881c in ugglite?
<M0xCAFEBABE[m]> nope
<M0xCAFEBABE[m]> my ugglite has hx otm1901a
<M0xCAFEBABE[m]> and ugg has the tm one
<barni2000[m]> so hx variant also have issues :(
<barni2000[m]> you cannot change the brigthness or it is alternating
<barni2000[m]> maybe i should check what they changed in mdss driver at downstream
<M0xCAFEBABE[m]> xiaomi only changed for sc_ili9881c
<M0xCAFEBABE[m]> otm1901a doesn't have any change
<barni2000[m]> tm variant is working for me
<barni2000[m]> so you have find some change
<M0xCAFEBABE[m]> barni2000[m]: the special thing for ili9881c is it uses a special dcs command to set brightness
<barni2000[m]> i have set the regulators
<barni2000[m]> M0xCAFEBABE[m]: i had a feeling it will be something like that
<M0xCAFEBABE[m]> <barni2000[m]> "maybe in reflog" <- could u try finding it please...
<M0xCAFEBABE[m]> all of those were so fine on rc3
<M0xCAFEBABE[m]> once we found it we could do a diff and see what could've broke stuff
<barni2000[m]> it is between 6.14 and 6.15
<M0xCAFEBABE[m]> M0xCAFEBABE[m]: u could find the closest commit hash faster by trying to match the date you rebased onto newer version
<barni2000[m]> M0xCAFEBABE[m]: i only changed what you have asked me
<barni2000[m]> oh, there is a patch upstream for wifi
<barni2000[m]> maybe i have cherry-picked
<barni2000[m]> i have not cherry-picked
<barni2000[m]> 6.15 contains this fix
<M0xCAFEBABE[m]> barni2000[m]: yeah, when this commit is included the dmesg lines I posted above appears
<barni2000[m]> i don't understand why because i don't have issues with 6.15.0
<M0xCAFEBABE[m]> btw, did anything of ours got upstreamed during 6.15-rc3 to 6.15?
<M0xCAFEBABE[m]> i'm trying to rebase onto msm8916 6.15-rc3 now
<barni2000[m]> M0xCAFEBABE[m]: some commits maybe but not much, i have stucked with 8937 upstreaming
<barni2000[m]> maybe 6.15 was the first version what contains 8917
<barni2000[m]> i dont remember
<barni2000[m]> no, 6.14
<barni2000[m]> 6.15 contains small commits from 8937 submission
<M0xCAFEBABE[m]> do u have a list of what you've tried to submit so far?
<barni2000[m]> yes on the mail list
<barni2000[m]> now there are 4 commits left gcc gcc schema 8937 and xiaomi-land
<barni2000[m]> it seems to me non panel binded changes
<M0xCAFEBABE[m]> <barni2000[m]> "audio is working" <- just figured that out, it's because of HPHL/HPHR should've passed in the value "Switch" rather than "1"
<M0xCAFEBABE[m]> well, you used "1" in alsa-ucm-conf, and I followed it when making my own config...