05:20 <bamse> steev: presume you mean on the flex 5g...and no i haven't figured out why

05:20 <bamse> steev: but it's not just the trackpad, both bt and usb mouse shows the same behavior

05:21 <bamse> steev: and i think i'm seeing the same issue when i'm running out of ram...plus i see sustained 110+MB/s read from ufs

17:23 <macc24> robclark: o.

17:24 <macc24> * i can't type o/

17:30 <macc24> so about #84 on cadmium, it looks like fact that strongbad technically is baseboard of coachz doesn't really matter, since "baseboard" for cadmium is close to "broadest group of devices that make sense to refer to" and "board" being "device different from others physically"

18:52 <robclark> macc24: sure, I wasn't sure how closely you wanted to keep $BASEBOARD matching what it is for CrOS builds.. I'm not really sure it matters for cadmium images

18:52 <macc24> robclark: now literally only architecture and bootloader matters for cadmium images B-)

18:53 <macc24> i can insteall cadmium onto a krane, then lazor, then kevin with no re-imaging the installation medium B-)

18:53 <robclark> yeah, like I mentioned, I think it mostly just matters for sensors that trigger tablet mode vs desktop mode UI, and that sort of thing

18:53 <macc24> yea thanks for the comment

18:53 <robclark> np

20:48 <Erisa> bamse: Flex 5G is called Yoga 5G outside of the US :)

20:48 <Erisa> oh i see the message just said 5G without context, i get you now

20:51 <Erisa> steev: I tried with almost full battery and plugged in, first time it threw some UFS errored and rebooted and the second time its doing CPU lockup again. I guess I should try a newer kernel?

20:52 <Erisa> I was going to mess with kernel stuff afterr installing but might as well try it now since the install isnt working

20:52 <steev> Erisa: possibly - you could try rebuilding the iso on a debian virtual machine and using the branch from bamse's github i linked above

20:56 <Erisa> sure i'll have a look, cheers

23:02 <bamse> Erisa: ohh, i didn't know those are the same

23:18 <steev> surprised they're available in europe since they're EOL in US

23:18 <bamse> steev: the flex 5g is EOL already?

23:18 <steev> yeah

23:18 <bamse> amazing

23:18 <steev> https://www.lenovo.com/us/en/p/laptops/ideapad/ideapad-flex-series/lenovo-flex-5g-14q8cx05/88lgf801419

23:21 <steev> sadly, they're still extremely expensive on ebay

23:21 <steev> bamse: since you're around... what's up with patchwork? no patches since the 13th

23:22 <bamse> wow, and opening that link screwed up my machine

23:22 <steev> oof

23:22 <bamse> the mouse pointer moves a little every few seconds

23:23 <steev> it should just show [DISCONTINUED] on the page and there's an overlay that says "this product is no longer available, perhaps these would interest you" and shows 3 other machines

23:23 <bamse> will have to wait until the oom killer kicks in and solves my problem

23:23 <bamse> yeah, i got that...and i guess chromium decided to use a lot of ram to tell me

23:24 <steev> is that... on your flex 5g?

23:24 <steev> because the c630 has no problems

23:24 <bamse> right, i've not made another attempt at the external monitor thing on my yoga

23:25 <bamse> but as you say, we don't have the same problem there

23:25 <steev> true, we just have other problems :P

23:26 <steev> i really should look into the audio stuff. i can't correct the popping sound, but i'd really like to get sound coming out of both speakers

23:26 <steev> i was also thinking of doing something like

23:26 <steev> https://linux-meson.com/

23:26 <bamse> i spoke with Srini about it, he has it working properly and i'm told that there's some interaction with the ucm maintainer(s)

23:27 <steev> oh, nice

23:27 <steev> i don't expect much going on with the holidays

23:28 <bamse> i rebased my c630 tree yesterday and noticed that i haven't send out the backlight dts changes :(

23:28 <bamse> but those 2 included i'm down to 5 patches ontop of v5.16-rc

23:29 <bamse> some-battery, battery dts, backlight dts, panel dts and defconfig

23:29 <bamse> although i think it's only battery and rfcomm that i actually need the defconfig for now

23:29 <bamse> but even if we're close, something linux-meson like would be nice

23:30 <bamse> hard to do the "mainlining progress" in some sensible way though, given that the relevant patches are scattered all over the place

23:31 <steev> i'm a bit higher in the patches count but those are mostly niceities

23:31 <steev> Your branch is ahead of 'origin/master' by 58 commits.

23:31 <steev> that's torvalds, but it's stuff like

23:32 <bamse> right, the 5 patches is for c630

23:32 <steev> the cpu freq stuff

23:32 <steev> srini's patches for the audio

23:32 <bamse> the 5g is currently at 34

23:32 <steev> attempts at the rcg2 stuff

23:32 <steev> and that dtpm thing

23:35 <steev> https://github.com/steev/linux/commits/c630-5.16.0-rc6

23:35 <steev> i'm kinda dumb though and i dunno what clocks should actually have the clk_ops_enable_parent on

23:36 <steev> and venus still does it, i should probably ask lumag about it since he was the last one to touch those clocks

23:36 <steev> bamse: oh, i did notice one regression from 5.15 to 5.16

23:36 <steev> the webcam can't do 720p with 5.16

23:37 <steev> it stutters like crazy and the console just says the system is too slow

23:37 <bamse> i don't think CLK_OPS_PARENT_ENABLE is the solution to our problem...at least not in the general case

23:37 <bamse> sounds like a performance regression to me

23:37 <steev> the gcc stuff is based on a patch i saw earlier on the list for the 8250 i think

23:38 <steev> but the one in dispcc does fix the rcg2 messages

23:38 <steev> the ones*

23:38 <bamse> yeah, i haven't figured out the exact life cycle for the clocks parented by the dsi phy

23:39 <steev> i'm assuming the venus one is shut down already by the time the module loads (i don't iirc, have the venus firmware in the initrd)

23:39 <bamse> looking at the callstack i get on my c630 i wonder if we're cutting the power to the phy...and then trying to reparent the rcg, before finally re-enabling the pll in the phy again - which luckily seems to unlock the wedged rcg

23:41 <steev> maybe

23:44 <steev> fwiw, i'm on vacation, but not going anywhere now, so if stuff needs testing, i'm all for it

23:44 <steev> even if i wanted to do work... my boss broke my jenkins server so... i gotta wait on him to fix it

23:45 <bamse> i'm trying to finish up the reworked dp/usb-c integration...then i'll probably be able to take a look

23:54 <Dylanger> Does anyone in here have virtio driving writing experience + is available for contract hire?

23:54 <Dylanger> (Sorry if this isn't allowed)

23:55 <Dylanger> * isn't allowed/asking etc)