00:58 <Ashy> are any of you running a kernelci cluster at all?

00:58 <HdkR> I'm not at least

00:59 <Ashy> it's my plan to eventually get into kernel dev for arm boards but i think my first step will be to set up a small kernelci cluster tracking the relevant repos

00:59 <Ashy> i want to run a fully open source platform at some point

01:00 <Lyude> Ashy: HdkR was planning on doing some ci stuff I thought

01:00 <Ashy> probably gonna buy a raptor power9 system next year sometime but the whole world carries arm machines in their pockets already so that's clearly the long term future

01:00 <Lyude> also man i hope i'm almost finished fixing all of these driver bugs so I can actually figure out what's going on with mali_kbase :(

01:05 <HdkR> Does kernelci just test kernel boot?

01:07 <Lyude> HdkR: it is meant to test most things, but what it actually tests at the moment I'm not sure

01:08 <Lyude> the goal is though to have a ci system that's just available to anyone who needs it for any driver for anything basicalluy

01:08 <Lyude> *basically

01:10 <HdkR> Makes sense

01:11 <Lyude> yesss

01:11 <Lyude> finally no kasan warnings on my vim2

01:11 <Lyude> now hopefully the filesystem drivers will stop randomly crashing

01:12 <HdkR> Always a nice result :)

01:13 <Lyude> today has quite literally been a "why is my panfrost system randomly crashing in f2fs? oh, of course, it's the wifi driver and video driver"

01:14 <Ashy> i think it mostly just tests boot but yeah supposedly you can plug in whatever tests you want

01:15 <Ashy> i'd like to have some phoronix benchmarks set up to get boots plus performance regressions etc

02:55 <chewitt> @Lyude is https://patchwork.kernel.org/patch/10696569/ related to https://github.com/chewitt/LibreELEC.tv/blob/amlogic/projects/Amlogic/patches/linux/linux-9999-emmc-freq.patch ??

02:55 <chewitt> I'm reading the backlog discussion with @narmstrong but didn't quite follow it

12:11 <belgin> hello

13:51 <lvrp16> chewitt: jerome of baylibre is working on emmc again, i gave him the wrong schematic of le potato and he is re-working the tuning code again

13:55 <NeuroScr> hrm now to find one of my 2 USB to TTL adapters

14:19 <daniels> Lyude: awesome :) lmk if i can help any more

14:19 <daniels> Ashy: kci is far more difficult to set up than you'd think - to the point where it's easier to just get the stuff you want to do upstream

14:20 <daniels> Ashy: most kci labs (including collabora's) use lava though, so you could just reuse the lava definitions to run and test new kernels; those are pretty simple and standard

17:19 <NeuroScr> found one but I don’t want to unwire it, I’ll keep looking

18:20 <Lyude> chewitt: no-that's another issue I found while trying to get lockdep working (it'd turn lockdep off whenever I hit it)

18:21 <Lyude> want to see if I can get lockdep working for the mali_kbase debugging, so I can get resetting the GPU to work

21:24 <Ashy> nanopc-t4 looks pretty damn nice