02:22 <steev> what kind of crashes, what kernels

06:44 <pstef> weird, I took wip/sc8280xp-v6.8-rc7 from jhovold/linux, added config from Fedora Rawhide and "make -sj7 all" complains about pahole and suggests to disable CONFIG_DEBUG_INFO_BTF. I can do that, but how did compilation work for Fedora? I must be missing something that is not in any of the installation instructions

07:17 <jhovold> steev: ok, don't think there was anything ASPM related in 6.7.9, perhaps random changes in timing

07:19 <jhovold> adamcstephens: yeah, there are definitely still issues with the external display (e.g don't try disconnecting while suspended) but I don't think that is new

09:54 <_[m]123> <steev> "what kind of crashes, what..." <- system freeze, resets - kernel 6.8rc7

10:26 <jhovold> _[m]123: can you try disabling efivars and see if that helps? Just disable CONFIG_QCOM_QSEECOM

10:26 <jhovold> would be interesting to log any efivars accesses during boot of your distro as well

10:27 <jhovold> e.g. by slapping a printk in each of the qcom_efivar_ops

10:29 <jhovold> you and konrad both run debian/ubuntu, may or may not be relevant

10:29 <jhovold> konradybcio: ^

10:36 <jhovold> _[m]123: also is this with an external display, and are you using wayland or x?

10:48 <_[m]123> wayland, had 2 crashes, one in the night with lid closed ? and one just this morning without external screen on

10:48 <_[m]123> though it seems like the samsung screen somehow still communicates with the laptop even when it's turn off - very normal 🙂

11:11 <_[m]123> it says "source enabled" in the tv logs for reason why it's still on - weird

11:13 <_[m]123> > disable Anynet+ (HDMI-CEC).

11:13 <_[m]123> this should fix

14:38 <steev> hrmpf

14:38 <steev> same wifi router/modem, moved it to the new place. instead of the mdsu bit insufficient, i now get spammed with htc rx: insufficient length

14:40 <_[m]123> <_[m]123> "> disable Anynet+ (HDMI-CEC)...." <- > <@nino:matrix.libre.brussels> > disable Anynet+ (HDMI-CEC).... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/AOGaCpCgaqthYyokGqcSXkgp>)

14:40 <_[m]123> * it didn't grrr crap software

14:40 <_[m]123> I only see kde related stuff in logs mm

14:52 <jhovold> Here's an updated wip branch for the X13s:

14:52 <jhovold> - https://github.com/jhovold/linux/tree/wip/sc8280xp-v6.8

14:52 <jhovold> Changes include:

14:52 <jhovold> - fix serial console regression in 6.4 (CRD)

14:52 <jhovold> - disable pcie aspm l0s on all sc8280xp and derivative platforms

14:53 <jhovold> I also spent some more time tracking down the resets when disconnecting (or powering off) an external display. It looks like the HPD notify revert in rc7 indeed also fixed that regression even if the underlying bug is still there. Some more details here:

14:53 <jhovold> - https://lore.kernel.org/lkml/Ze8Ke_M2xHyPYCu-@hovoldconsulting.com/

16:37 <pstef> jhovold: hey, thanks for the config, it looks pretty lean and I can see the compilation time greatly reduced. Just to confirm, the way I'm supposed to use it is this? make -sj7 johan_defconfig all; sudo make modules_install install dtbs_install; sudo dracut --kver ...; sudo reboot

17:07 <lollaritshe[m]> when i try to install linux-x13s i get an error "Errors occured, no packages were upgraded"

17:07 <lollaritshe[m]> im currently chrooting from the USB

17:09 <lollaritshe[m]> apparently haveing Linux-aarch64 and linux-firmware installed was the error

17:09 <pstef> jhovold: just to report some success, I executed that sequence of commands and at first it didn't boot, but then I noticed you had removed btrfs from the kernel, and I put it back in - now I can get to the desktop, thanks!

17:14 <adamcstephens> jhovold: haha, oops i just tried to disconnect while suspended on rc7 :)

17:17 <_[m]123> * ```... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/NPVdJpyPlxFLjQCNZUMIIMYg>)

17:17 <_[m]123> * ````... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/VMucfnOvePJJEcqMRjFhGeNK>)

17:17 <_[m]123> * ````... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/VHMeQEqruckBmylWcyWfxYTh>)

17:20 <pstef> _[m]123: yes, according to a recent commit

17:21 <pstef> _[m]123: sorry, when I said "yes" I meant both to =yy

17:21 <pstef> single y

17:24 <lollaritshe[m]> Even with the linux-x13s kernel i dont get the option to boot it

17:25 <lollaritshe[m]> did i miss something?

17:25 <lollaritshe[m]> i used archinstall for the rest of the installation

17:29 <vadikas[m]> lollaritshe[m]: have you see ironrobin's instructions? https://github.com/ironrobin/archiso-x13s/wiki/EndeavourOS-Tutorial

17:29 <lollaritshe[m]> i was on the arch specific one

17:29 <vadikas[m]> is it arch-specific one

17:29 <lollaritshe[m]> maybe this is a bit more detailed 😅

17:29 <lollaritshe[m]> https://github.com/ironrobin/x13s-alarm

17:29 <lollaritshe[m]> im on this one

17:30 <lollaritshe[m]> is it possible that systemdboot isnt supported?

17:31 <vadikas[m]> I don't know. But those instructions worked fine, and you can skip the custom os install to stay with generif arch/arm

17:33 <_[m]123> <pstef> "single y" <- jhovold asked just before to test with this disabled I think

17:34 <pstef> ah, sorry, I didn't notice that

17:34 <_[m]123> my question really was, should I disable the second var too

18:09 <lollaritshe[m]> <vadikas[m]> "I don't know. But those instruct..." <- will try

18:49 <steev> lollaritshe[m]: fwiw, those docs are clover[m]

18:53 <clover[m]> <lollaritshe[m]> "is it possible that systemdboot..." <- systemd-boot should work, you just have to use an uncompressed kernel IIRC

18:54 <lollaritshe[m]> how do i do that? sorry new to the aarch64 thing

18:54 <lollaritshe[m]> at least i can destroy my girlfriends windows on her x86 machine right now

18:55 <clover[m]> compressed kernel: vmlinuz-linux

18:55 <clover[m]> uncompressed kernel: vmlinux

19:01 <pstef> jhovold: on my x13s systemd-udevd is busy. sysctl status shows that it's in a loop of trying to "echo 4-0010 > /sys/bus/i2c/drivers/i2c_hid_of/bind" but if I run that I get "No such device or address". Indeed, I don't have the touchscreen. I've tried to disable it first by setting status="disabled" in the dts, then removing it. I do make dtbs_install, but that shows no change.

19:10 <clover[m]> if you just watch arch you can still follow the endeavourOS tutorial, lollar. at step 14 instead of installing endeavourOS you can just leave it and install whatever desktop packages you want manually

19:12 <vadikas[m]> clover BTW, your repository has mesa version older than in official repos, so patched mesa don't install being superseeded by official build

19:13 <vadikas[m]> I manually updated it, can make a PR if you want

19:29 <_[m]123> jhovold I compiled latest and turned those 2 vars off - we'll see, but I don't know how to put that print tho

20:01 <clover[m]> vadikas: i pushed Mesa 24.0.2 to my repo, can you check to see if done correctly?

20:06 <vadikas[m]> <clover[m]> "vadikas: i pushed Mesa 24.0.2 to..." <- pkgrel=0 , but should be >2 IMO

20:07 <vadikas[m]> > <@clover:ironrobin.net> vadikas: i pushed Mesa 24.0.2 to my repo, can you check to see if done correctly?

20:07 <vadikas[m]> * pkgrel=0 , but should be >2 IMO -- in stock repos mesa is 23.0.2-2 now

20:07 <vadikas[m]> I was thinking -- may be 690? :-)

20:08 <jglathe_x13s> X13s is on 6.8.0 now

20:09 <clover[m]> steev: will you push 6.8?

20:12 <steev> soon yeah

20:12 <clover[m]> vadikas, pkgrel has more to do with the packaging than the software version for example if a patch was added you bump the pkgrel

20:13 <vadikas[m]> yep, but they're compared directly, so larger the number the more precedence package has

20:14 <clover[m]> Yeah, but if you want my packages instead of upstream (which I recommend) you can just put the priority higher in the pacman.conf

20:15 <vadikas[m]> ok, make sense!

20:40 <steev> clover[m]: fwiw, I’m still waiting on stable to post the linux-6.8.y branch, kinda odd that it hasn’t yet

20:41 <clover[m]> Got it, ty for the update

23:13 <_[m]123> https://paste.debian.net/1310372/

23:13 <_[m]123> is this something relevant ?