07:51 <wens> lvrp16: you mean like mt8195?

17:55 <lvrp16> wens: well any mediatek chip. given mediatek's wifi portfolio, they seem to be taking rockchip's space seriously.

18:32 <lvrp16> you have a76 going up against a78 and x2s

20:49 <hexdump0815> is anyone here running mainline linux on veyron chromebooks and really using them? i'm currently rotating through several of the systems i build images for and try to dogfood daily drive them for a few weeks each and on veyron i reproducably get a hanging system

20:51 <hexdump0815> i can reproduce it on my mighty and my minnie, so its not a hardware issue - i'm using xorg xfce and it might be related to using firefox as well maybe

20:52 <hexdump0815> it manifests in the system simply locking up, i.e. screen freezes and i can only reboot - system is bullseye, kernels tested are v5.19.2, v5.19.1, v5.18.1 and v5.15.22 - all show this

20:52 <hexdump0815> i tested with the default debian bullseye mesa and with a self built v22.1.1 mesa

20:53 <hexdump0815> i tried to isolate the source of the problem by blacklisting panfrost, the sound, the wifi and the bt modules - problem still happens after a while of use

20:55 <hexdump0815> the easiest way to reproduce it so far is: watch a youtube video in firefox, suspend the system while doing so, resume ... it usually freezes a while later when accessing anything in the menubar ... latest afte rthe second suspend/resume it will freeze for sure a while later

20:57 <hexdump0815> i'm still trying to figure out what might be the problem - just wanted to ask if this sounds familiar to anyone here or if you have seen this is something similar yourself already on veyron of if it is working reliable in such a scenario for you?

20:59 <hexdump0815> i think i also got it wihout suspending, but then it might take longer until it appears - just using terminals in the same xfce session everything seems to be fine so far and otherwise the system is very stable: i can easily make -j 4 kernels all the time without any problem

21:03 <lvrp16> hexdump0815: I have an roc-rk3399-pc that runs 24/7 as a build server, it hasn't experienced any hangs yet and it has been running for 5.19-rc8 since that was released

21:04 <lvrp16> try lowering the temperature in u-boot device tree to see if it's the hardware opps

21:05 <hexdump0815> lvrp16: veyron is rk3288 - rk3399 seems to be fine, was not able to reproduce the problem on a bob chromebook with rk3399 with the same setup (aarch64 instead of armv7l there of course then)

21:06 <lvrp16> hexdump0815: sorry got my codenames confused

21:06 <hexdump0815> lvrp16: np - i do not expect it to be temperature related as there is no real load involved - system is more idle than busy

21:13 <mps> hexdump0815: not sure if I'm of the help but mainline works fine on gru-kevin chromebook

21:16 <urja> mps, again that is rk3399

21:16 <urja> my veyron speedy is running ...

21:16 <urja> 5.15.61

21:17 <hexdump0815> urja: are you using it regularly with firefox on xorg?

21:17 <urja> yes, but i dont suspend them

21:18 <urja> and i cant say them to be completely stable, the graphics gets glitchy over time (and there might be an occasional hang), but they do run for days

21:19 <urja> well... only one tbh

21:19 <urja> but suspend is one of the things i gave up on testing at some point (mostly because it always broke the broadcom wifi...)

21:20 <urja> (only one in that i only have one of my veyrons running a recent kernel, the other one has an old sorta hacked chromeos lol)

21:22 <hexdump0815> urja: thanks for the info