02:30 <steev> i really hate audio. like really really hate it

02:36 <steev> enable valid mismatch

02:36 <steev> bamse: https://paste.debian.net/1217712/ any idea what might cause this? trying to play audio and all i'm getting is 2 quick pops and nothing, dmesg shows issue(s) with soundwire? but i haven't seen anything that touches that in a long time. tree is https://github.com/steev/linux/commits/linux-5.15.y I'm worried the issue comes from whatever this pipewire shit is that is supposed to replace pulse, but even aplay throws some swr read

03:12 <bamse> steev: i've not paid any attention to audio lately :/

03:42 <bamse> steev: you should have a patch in your inbox for it the SW_LID issue

03:43 <bamse> steev: on my machine i run evtest, select gpio-keys...it says SW_LID is 1 and if i close/open the lid a few times it doesn't report the changes properly...also /proc/interrupt reports 1 interrupt per lid-close, none per lid-open

03:43 <bamse> steev: with the patch evtest reports close/open and /proc/interrupts increments as expected

03:43 <bamse> and the machine resumes immediately when i open the lid, not the 4-5 seconds delay that i had previously

14:49 <steev> bamse: oh spiffy, that will make gwolf happy, i know he suspends a lot

15:26 <steev> bamse: okay audio is userland side, i need to look at wtf they did. this seems to have to do with pipewire and they moved from some pipewire-media-session to wireplumber and now all i get any time i try to play anything is two pops. i checked on 5.14 where i know 100% audio was working

16:04 <bamse> vkoul: did we test pipewire again on any of the qcom platforms?

16:05 <bamse> steev: having touchpad beyond the first suspend is really nice :)

16:05 <vkoul> bamse: srini had tried it and ran into issues, but that was much earlier..

16:05 <vkoul> We should try it again

16:05 <bamse> vkoul: forgot to ask, did you get your flex 5g up and running?

16:06 <vkoul> bamse: yes my flex is up and running but not the audio yet...

16:12 <steev> fwiw, audio worked with pipewire 0.3.38 it breaks with 0.3.39

16:15 <steev> i just... can't go back to 0.3.38 because that wants to uninstall all of the gnome stuff *sigh*

16:35 <steev> bamse: can confirm, have not lost touchpad at all here

16:35 <bamse> what worries me is that it seems to not have been failing 100% before...and i don't know why

16:37 <steev> same reason why it takes 3 reboots to make audio work when it fails... it's quirky ;)

16:38 <steev> well, prior to these pipewire changes

16:48 <gwolf> FWIW I never had any issues with suspending! Track pad always works right

16:48 <gwolf> Audio is less predictable though

16:48 <gwolf> (boots)

16:48 <gwolf> Blue screen... I almost never reboot, but it might happen on half of all reboots

16:49 <steev> well this new pipewire crap in testing/unstable is... garbage

16:50 <steev> i really wish bm16ton would just join irc instead of the emails to me every so often :(

16:51 <gwolf> Yes, I have noticed pipe wire... And been to lazy to try to understand how audio is now supposed to work

16:51 <bamse> gwolf: per my testing, i can't explain why your trackpad worked...

16:51 <gwolf> bamse: well, I got a lucky firmware subrelease?

16:51 <bamse> gwolf: i never get an interrupt for the falling edge of the lid gpio, with or without actually suspending the system...

16:52 <steev> i can't remember how to tell which edition of the c630 one has

16:52 <steev> i know that mine and gwolfs have the different display so it could be that his has the "no issues ever" with it

16:52 <bamse> steev: shouldn't matter, if i just poll the value of the gpio it goes high when i close the lid and low when i open the lid, but i wouldn't get the interrupts from the gpio controller

16:52 <steev> ah

16:53 <bamse> with the wakeirq quirk i get both interrupts

16:55 <bamse> anyway, it seems to work now :)

16:55 <bamse> and people seem hapy with the patch

16:56 <steev> i definitely am :D

16:56 <steev> i also tested it across 5.12, 5.13, 5.14 and 5.15