04:22 <Jasper[m]> <steev> "Jasper: right, but what does..." <- Nothing hahaha

11:30 <erebion[m]> Noch seriously looking to buy an X13s with 32 GB RAM somewhere as my X270 just died.

11:51 <erebion[m]> *Now

13:31 <_[m]1> who is using x13s as his main laptop? I've bought a macbook air m1 1st gen meanwhile to bridge the gap until I get it stable

13:33 <_[m]1> <clover[m]> "new kernel overwrites old kernel" <- is this wanted behaviour?

13:33 <_[m]1> @jasper how many times u reboot?

13:35 <jhovold> _[m]1: i am, for exampl, and without any issues so whatever problems you're having are likely due to user space and/or configuration issues

13:38 <jhovold> (unless there are any problems with whatever patches steev puts on top of my branches, I wouldn't know)

13:41 <jhovold> _[m]1: what's your kernel command line for a start?

13:42 <jhovold> I happened to see that clover[m] suggests running with only efi=noruntime, which is not enough and could cause trouble if you try to use full disk encryption, for example

14:34 <_[m]1> I added cryptsetup unlock with uuid

14:46 <_[m]1> ```

14:46 <_[m]1> GRUB_CMDLINE_LINUX=""

14:46 <_[m]1> GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"

14:46 <_[m]1> weird

14:48 <robclark> I'm using x13s as main laptop.. but using systemd-boot, not grub.. because at the time I set it up fedora's grub was unhappy about x13s.. I think that is fixed now but since it is my daily driver I've not had a chance to re-install

14:49 <danielt> Ditto... been my daily driver since I got it. I run with FDE and grub (and a fair bit of legacy on the kernel command line that I've never bothered to tidy up).

15:02 <steev> not just using windows on it smh

15:04 <leezu> robclark: Do you know if SC7180 supports DRM PRIME frames (ChooseV4L2PixelFormat)? Asking based on this Firefox bug https://bugzilla.mozilla.org/show_bug.cgi?id=1852765#c3

15:05 <javierm> leezu: isn't the problem in the v4l2 m2m driver?

15:05 <javierm> if I understand the report correctly

15:06 <javierm> that the v4l2 driver is not exporting dma-bufs

15:08 <leezu> Possible, yes. I've just verified that ffmpeg 6 alone is able to use the hw decode correctly: SDL_VIDEODRIVER=wayland ffplay -codec:v h264_v4l2m2m ~/video.mp4 Do you know if ffmpeg supports/uses non-DRM-PRIME decode?

15:14 <clover[m]> <_[m]1> "is this wanted behaviour?" <- yes, if you want to keep it you can rename the kernel first with the mv command. or you can use downgrade from the AUR to rollback to older kernels

15:28 <jhovold> clover[m]: you should update your wiki entry to say that clk_ignore_unused pd_ignore_unused are still needed

15:28 <jhovold> may be a good idea to include iommu.passthrough=0 iommu.strict=0 as well

15:28 <clover[m]> ok

15:28 <jhovold> without the former, things may potentially break left and right

15:29 <jhovold> clover[m]: i also just noticed that srinik's updated alsa-ucm-conf pull request is broken and results in too low speaker volume again

15:30 <jhovold> I had mine patched locally and only noticed today when alarm moved to 1.2.10 so I had to rebuild myself

15:31 <jhovold> i noticed some people saying that had regressed here but didn't make the connection then

15:31 <clover[m]> if you know how to fix it let me know. i tried doing what he had in his other repo (sourcehut?) to adjust default playback volume and it didn't work

15:32 <jhovold> https://github.com/alsa-project/alsa-ucm-conf/pull/335

15:32 <jhovold> Just take the latest branch and add back these two lines:

15:32 <jhovold> https://github.com/Srinivas-Kandagatla/alsa-ucm-conf/commit/39420b0

15:32 <clover[m]> ok just apply that patch?

15:32 <jhovold> fixed the issue here

15:33 <jhovold> No, you need to create a patch first I guess

15:33 <jhovold> I just patched by corresponding file locally and reported it

15:33 <robclark> leezu: no idea what a "non-DRM-PRIME" dmabuf is.. I guess v4l2 exported dma-buf? Either direction should work (drm exporting, v4l2 importing.. or visa versa)

15:34 <robclark> there should be no need for cpu copy in either case

15:34 <jhovold> clover[m]: I think srinik will push an updated branch in a not too distant future so you could wait for that too

15:36 <clover[m]> sounds good, will take a look when i get some free time

15:37 <jhovold> and you should need that "unfix" device number patch you carry:

15:37 <jhovold> https://github.com/ironrobin/x13s-alarm/blob/trunk/alsa-ucm-conf-git-x13s/unfix-device-numbers.patch

15:37 <jhovold> should *not*

15:38 <jhovold> with the latest topology file, that would not work

15:38 <clover[m]> yeah i can probably safely dump that. thanks

15:45 <leezu> robclark: Ack. Thank you. I'll mention this on the bug report. By the way, I noticed ffplay reports "[h264_v4l2m2m @ 0x7018011e40] capture: driver decode errorB f=0/1" for a few frames when playing video with v4l2m2m. Do you know what this means? I wonder if such error may be triggering the Dmabuf Got non-DRM-PRIME frame from FFmpeg V4L2 error and thus triggering Firefox to

15:45 <leezu> fallback to software decode

15:50 <robclark> leezu: no idea.. but I'm probably also the wrong one to ask

16:15 <_[m]1> <jhovold> "clover: you should update your..." <- I put those though

16:15 <_[m]1> <jhovold> "may be a good idea to include..." <- not this

16:15 <jhovold> the iommu bits should only give your better usb throughput

17:23 <clover[m]> steev: i noticed that if i am in a meeting using my bluetooth headset, and then i leave the meeting, disconnect my headset, and switch my audio sink back to internal speakers, no sound plays. not sure if this is a bluetooth bug or a sound bug

18:10 <steev> clover[m]: hm... i am not sure there

20:25 <clover[m]> with the latest brave and vs code updates that released today i was able to remove llvm15 everything is working perfectly now

20:26 <_[m]1> but had to build sway yourself right

20:26 <clover[m]> nope

20:27 <clover[m]> i get sway from extra repo just like everyone else that uses arch

21:12 <clover[m]> jhovold ok im carrying alsa-ucm-config with the patch you suggested. i don't notice much difference however, maybe volume is slightly higher but i feel like it was better before. arch people can try it by doing a pacman update

21:14 <clover[m]> music sounds better actually!

22:21 <Jasper[m]> <_[m]1> "@jasper how many times u reboot?" <- About 8 times if I'm unlucky

22:21 <Jasper[m]> otherwise 2-4

22:22 <Jasper[m]> I'll check the suggestions posted before, I think I use the correct kernel options

22:22 <clover[m]> i also told eholzbach from github to add the those kernel options, fingers crossed.

22:23 <_[m]1> Jasper[m]: yeah same

22:23 <_[m]1> I got more today even

23:38 <steev> danielt said he does fde and no issues so maybe figure out whats different between y'all?