[SOLVED] Question about meaning of various journalctl warnings

mesaprotector · 2024-03-20 00:22:38

Apologies if this should be in Newbie Corner. My laptop boots fine. I just have been looking into system logs for a while now and after doing a lot of research, there are still a small handful of anomalies I can't figure out. I'm curious and am enjoying learning more.

#1

kernel: wmi_bus wmbi_bus=PNP0C14:00: WQ00 data block query control method not found

#2

kernel: i8042: PNP: PS/2 appears to have AUX port disabled, if this is incorrect please boot with i8042.nopnp

These device errors always occur soon after switching root. #2 seems to be referring to my laptop's built-in keyboard, which works fine - any references I find to this error online involve non-working keyboards. I could add i8042.pnp to my kernel parameters but... since it works, why?

#3

kernel: platform regulatory.0: Direct firmware load for regulatory.db failed with error -2

#4

kernel: nvidia-modeset: WARNING: GPU:0: Unable to read EDID for display device DP-2

These don't occur together, but both are presumably Nvidia-related. For #3, I don't have a regulatory.db file in the expected location and I'm not sure I need one. For #4, DP-2 is not actually the label for my laptop's built-in monitor (that would be eDP-1) so... what is this? Looking at the output of xrandr --verbose doesn't really help. DP-2 is listed as disconnected, and has no information other than the generic stuff listed under all of DP-1 through 4.

The final weirdness is that journalctl repeats the entirety of the log up until it switches root. As in, when reading the log, lines 1-1000 will be identical to lines 1001-2000 or similar, including timestamps. Then after systemd kicks in it just goes like normal.

Last edited by mesaprotector (2024-03-21 19:39:58)

seth · 2024-03-20 09:06:17

#1 DELL? https://bugs.launchpad.net/ubuntu/+sour … ug/1890467 - you can typically ignore this and it'll only (at best) go away w/ a firmware update.
#2 is a general warning, i8042 tells what it found and that it might be wrong and what to do in case.
#3 https://wiki.archlinux.org/title/Networ … ory_domain
#4 Sure it's not DFP-2 (nvidia's internal output addressing)? Otherwise DP is generally "DisplayPort". => Xorg log?

The final weirdness is that journalctl repeats

systemd-boot?

mesaprotector · 2024-03-20 09:39:56

seth wrote:

#1 DELL? https://bugs.launchpad.net/ubuntu/+sour … ug/1890467 - you can typically ignore this and it'll only (at best) go away w/ a firmware update.
#2 is a general warning, i8042 tells what it found and that it might be wrong and what to do in case.
#3 https://wiki.archlinux.org/title/Networ … ory_domain
#4 Sure it's not DFP-2 (nvidia's internal output addressing)? Otherwise DP is generally "DisplayPort". => Xorg log?
The final weirdness is that journalctl repeats
systemd-boot?

I swear, if Mr. and Mrs. Arch don't pay seth, they really should. I think he's about 1/3 of this forum's tech support.

#4: It's definitely DP-2; I also supposedly have DP-1 through DP-4, as well as DP-1-0 through DP-1-2, despite not having a DisplayPort connector on my laptop.

And not quite, I use a unified kernel image without an additional bootloader, but I do use systemd init rather than busybox.

seth · 2024-03-20 13:34:26

I do use systemd init rather than busybox.

The cause.

Do you have an xorg log and can you

for OUT in /sys/class/drm/card*; do echo $OUT; edid-decode $OUT/edid; echo "================="; done

?
You'll need https://aur.archlinux.org/packages/edid-decode-git

mesaprotector · 2024-03-20 17:19:40

seth wrote:

I do use systemd init rather than busybox.
The cause.
Do you have an xorg log and can you
for OUT in /sys/class/drm/card*; do echo $OUT; edid-decode $OUT/edid; echo "================="; done
?
You'll need https://aur.archlinux.org/packages/edid-decode-git

Huh. I wonder why in the world systemd behaves like that with journals.

All of the EDIDs are empty except for eDP-1. I did run edid-decode but it didn't show anything exciting. Here's my /sys/class/drm directory (card0 is Nvidia, card1 is Intel):

[asuka@Marojejy drm]$ ls
card0       card0-eDP-2     card1       card1-DP-2  card1-DP-4   renderD128  version
card0-DP-5  card0-HDMI-A-1  card1-DP-1  card1-DP-3  card1-eDP-1  renderD129

[asuka@Marojejy ~]$ cat /var/log/Xorg.0.log | grep \ DP\-
[    25.629] (II) modeset(0): Output DP-1 has no monitor section
[    25.629] (II) modeset(0): Output DP-2 has no monitor section
[    25.629] (II) modeset(0): Output DP-3 has no monitor section
[    25.629] (II) modeset(0): Output DP-4 has no monitor section
[    25.634] (II) modeset(0): EDID for output DP-1
[    25.634] (II) modeset(0): EDID for output DP-2
[    25.634] (II) modeset(0): EDID for output DP-3
[    25.634] (II) modeset(0): EDID for output DP-4
[    25.634] (II) modeset(0): Output DP-1 disconnected
[    25.634] (II) modeset(0): Output DP-2 disconnected
[    25.634] (II) modeset(0): Output DP-3 disconnected
[    25.634] (II) modeset(0): Output DP-4 disconnected

I've found people online who have similar despite no physical port. I'm guessing that this just means I could connect up to four external monitors to my laptop, and if I did they'd show up here. So the error, and it listing DP-2 specifically every boot, is confusing but doesn't really mean anything.

seth · 2024-03-20 17:34:37

nvidia will rather not use the modeseting driver…
If you want a better comment, I'd have to see the complete journal and xorg log for what's going on.

Also it's not very likely that the internal panel is wired to the nvidia chip and you're running on the modesetting driver what makes it extremely likely that those are actually forwarded via prime.

Just ftr: you do not expect anything to be attached to the display port output?

mesaprotector · 2024-03-21 19:39:36

It's the other way around - the HDMI is connected to nvidia (card0) and the panel is connected to intel (card1).

I'm not sure what the "modesetting driver" means... I tried booting without

nvidia_drm.modeset=1

and even with

nomodeset

and the warnings are exactly the same. (And my laptop predictably gets capped to 60 Hz if I do that.) Unless this is about the manually built nvidia-dkms driver.

I think I'm going to mark this thread as solved, because I'm satisfied that none of the warnings I mentioned are going to cause any problems.

Last edited by mesaprotector (2024-03-21 20:10:58)

seth · 2024-03-21 20:54:24

The modeseting driver is the default Xorg driver that operates on top of the KMS - instead of the ddx drivers like eg. xf86-video-intel (don't install that!)
"nomodeset" will prevent any of this from working and you're getting the vesa driver - with the glorious hardware features of the mid-nineties.

If you've issues w/ your setup, just open a new thread and post xorg log and system journal there.

Arch Linux

#1 2024-03-20 00:22:38

[SOLVED] Question about meaning of various journalctl warnings

#2 2024-03-20 09:06:17

Re: [SOLVED] Question about meaning of various journalctl warnings

#3 2024-03-20 09:39:56

Re: [SOLVED] Question about meaning of various journalctl warnings

#4 2024-03-20 13:34:26

Re: [SOLVED] Question about meaning of various journalctl warnings

#5 2024-03-20 17:19:40

Re: [SOLVED] Question about meaning of various journalctl warnings

#6 2024-03-20 17:34:37

Re: [SOLVED] Question about meaning of various journalctl warnings

#7 2024-03-21 19:39:36

Re: [SOLVED] Question about meaning of various journalctl warnings

#8 2024-03-21 20:54:24

Re: [SOLVED] Question about meaning of various journalctl warnings

Board footer