You are not logged in.
Pages: 1
Device: ASUS TUF Gaming A15 FA506IHRB_FA506IHRZ
Kernel Version: 6.9.7-arch1-1
MT7921 802.11ax PCI Express Wireless Network Adapter wifi abruptly stops working with the kernel log
mt7921e 0000:03:00.0: Timeout for driver own
mt7921e 0000:03:00.0: driver own failed
After rebooting the wifi device disappears for 1-5 boots (sometimes more) from lshw. There are no clear steps to reproduce this and it randomly happens.
Link to kernel log: https://pastebin.com/BNYcTnmY
Offline
Please don't post random excerpts.
https://bbs.archlinux.org/viewtopic.php?id=292150 was down to windows fast-start.
Offline
Both fast startup and hibernation are off.
Offline
There are no clear steps to reproduce this and it randomly happens.
Please don't post random excerpts.
Esp. since the behavior is "random" the context will matter if you want an informed comment - "-p err" will only say "yeah, happened"
I can tell you that you've
Jun 30 14:11:05 frisk-laptop kernel: nvme0n1: Read(0x2) @ LBA 654504104, 64 blocks, Unrecovered Read Error (sct 0x2 / sc 0x81) MORE
Jun 30 14:11:05 frisk-laptop kernel: critical medium error, dev nvme0n1, sector 654504104 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
an issue w/ your nvme and
Jun 30 14:11:10 frisk-laptop kernel: ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed
What's most likely the usb hub on one of your GPUs and
Jun 30 14:11:23 frisk-laptop kernel: warning: `panel-4-wavelan' uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211
you're using xfce.
None of that is inherently relevant to your wifi issues.
Offline
Please don't post random excerpts.
I'm sorry, I'm not sure what you meant by this but if it's about posting logs with warning priority here are the logs at a higher priority level: https://0x0.st/Xa2p.txt
Offline
Please post your complete system journal for the boot:
sudo journalctl -b | curl -F 'file=@-' 0x0.st
Offline
Yes, I've uploaded the full log for that boot at: https://0x0.st/Xa2p.txt
Offline
The NIC shows up, scans and connects effortlessly and pretty much exactly 7 minutes later stops responding - smell like powersaving.
Try to add "mt7921e.disable_aspm=1 pcie_aspm=off" to the https://wiki.archlinux.org/title/Kernel_parameters
Offline
The issue still persists although it did work for longer. Here are the logs from that boot: https://0x0.st/XaLC.txt
Offline
To test whether this is power-saving related at all, try to throw the NIC a life-line and keep "ping google.com" running in the background.
Also, in case some tool sets that
iw dev wlp3s0 get power_save
iw dev wlp3s0 set power_save off
Offline
iw dev wlp3s0 get power_save
This outputs "on" on boot, is this expected even with the mt7921e.disable_aspm=1 pcie_aspm=off kernel parameters?
The output was still off after the driver failing and being set manually with that command.
To test whether this is power-saving related at all, try to throw the NIC a life-line and keep "ping google.com" running in the background.
The driver still failed even with the ping command running in the background.
Also, if I don't shutdown soon after the driver failing the kernel seems to also crash (the caps lock light starts blinking and everything freezes and I'm not able to switch into any other tty either). I tried testing if the LTS kernel fixed anything but it doesn't seem like it. The journal log (by using -b -1) doesn't contain any information regarding the crash, the last message is the driver own failed.
Last edited by Frisk17 (2024-07-03 10:52:39)
Offline
Did you get anywhere with this? My system has been very stable since I built it 6 months ago, up until this morning when I woke up to this error and the wifi wouldn't connect.
Then today it happened mid-session and completely locked the whole PC up.
This is my log:
My mobo is an Asus ROG Strix B650E-I
I don't think I'm doing any sleep stuff, it's a desktop and I didn't set it up as such
I do have this in sway that turns the output off:
exec swayidle -w \
timeout 300 'swaylock -f -c 000000' \
timeout 600 'swaymsg "output * power off"' resume 'swaymsg "output * power on"'
Offline
If it's a regression, but not the kernel, the cause might be https://archlinux.org/packages/core/any/linux-firmware/ which was updated ~3 weeks ago.
I didn't check whether the relevant file changed but you could compare the FW version against the one loaded in an older journal and/or simply try to downgrade linux-firmware
Offline
I found this message on one of the boots where the wifi device is not detected, although I didn't find messages like this on other boots where the device is not detected
Jul 08 12:25:08 frisk-laptop kernel: mt7921e 0000:03:00.0: Unable to change power state from D3cold to D0, device inaccessible
Jul 08 12:25:08 frisk-laptop kernel: mt7921e 0000:03:00.0: disabling ASPM L0s L1
Here are the full logs http://0x0.st/XBIS.txt
Offline
No mt7921e message should show up if the device doesn't even show up because the module alias won't get triggered.
Did you try to downgrade linux-firmware?
Offline
Did you try to downgrade linux-firmware?
Yes, I did downgrade linux firmware to linux-firmware 20240510.b9d2bf23-1 but the issue still persisted.
Offline
Pages: 1