I meant the phosh screen, which is locked by default, we need to swipe up and enter the pin to see the home screen. Just now it rebooted again. I will check journalctl now.
1 Like
journalctl ālist-boots showed 02:36:39 as the boot time and running journalctl āsince 02:36:38 shows
Feb 21 02:36:38 pureos systemd-logind[714]: The system will suspend now!
So it seems system is unable to suspend unless login screen is unlocked at least once.
and below
Feb 21 02:36:39 pureos systemd[1] Starting systemd-suspend.service - System Suspendā¦
This is the last message before it shows
ā Boot 3026bad1c3b74ā¦ā¦ ā
as next boot starts
Thatās very unlikely to be related.
You ignored my question:
Wifi is disabled and suspend is enabled. It just happened now even after I unlocked. So that is unrelated. It is related to suspend. The wifi device itself is disabled (greyed out) in phosh drop down - I cannot turn it on. All 3 kill switches are on position. For now removing this kernel.
Please figure out whatās wrong with your WiFi then, as this is a likely culprit.
WiFi works fine in the older kernel. On 6.12 it is completely missing - in phosh drop down as well as in settings.
So please figure out whatās wrong with it. I donāt have psychic powers, I canāt see the logs or check stuff that you donāt provide me with. You havenāt even told me which card your phone has installed.
I managed to reproduce this on a phone with Redpine card. This entire conversation could be avoided if the initial report included kernel logs, as itās clear from the dmesg and the huge screaming WARN trace in there whatās going on. I may even have a fix already, but itās still compiling 
3 Likes
I installled it, donāt know what to test, it doest have any difference, but I know it is upgraded. Ha
1 Like
You could use testing steps from here: Making sure you're not a bot!
Not everything will apply to L5 PureOS.
1 Like
Iām using kernel 6.12 (until now only for 3 few hours, but I will continue using it). No problems found so far in my up to date crimson system. If I find an issue I will report it. Thank you for the good work.
Thanks. I was checking the wifi card details and saw you already reproduced and fixed it. I was not sure if it was related to WiFi when I reported initially. As I donāt usually look at kernel logs often, I was not sure where to start.
Kernel 6.12.74pureos1 is now available in crimson-updates-proposed, which aside of coming with fresh updates from both upstream and downstream also fixes two regressions from 6.6:
- Redpine driver failing to initialize network interface (which could lead to kernel panics on suspend)
- CPU frequency scaling being broken on phones from the Dogwood batch
If no further regressions are found Iāll likely make it the default on Crimson next week.
4 Likes
Thanks, now the wifi is initialized correctly (it connects to the wifi without issues). But when suspending manually from phosh power drop down, it does not wake up from suspend after pressing power button.
last boot was at 00:48 (when I long pressed power button), journaclt āsince 00:47:25 showed a red line in the log
00:47:26 pureos kernel: watchdog0: watchdog did not stop!
But this may be an irrelevant line.
Now second time pressing suspend actually did a reboot.
and logs before this reboot is Aw- ful. I can't understand a word an the tears came to.
This log shows a regular, graceful reboot initiated at 00:59:48.
When on battery power, suspend seems to work. But when it is charging and pressing suspend, it cannot wake up with pressing power button. This seems to be an issue with one charger - charger of Lenovo T14s laptop, when connecting a USB C cable to laptop, suspend works fine.
Charger spec,
20.0 V 3.25A 65.0W / 15.0V 3.0A 9.0V 2.0A 10.0 W.
I have noticed it does not auto suspend well when using this charger before (the screen keeps waking up by itself after going blank).
Any other way to get relevant logs? For now, I can avoid suspend with this specific charger.
The phone will wake up when the charger is renegotiating the PD contract. See https://source.puri.sm/Librem5/fw/firmware-tps6598x-nonfree#debugging
This would be unrelated to this topic though, as the kernel doesnāt really influence that process.
reboot system boot 6.12.0-1-librem5 Tue Feb 24 02:03 still running
reboot system boot 6.12.0-1-librem5 Tue Feb 24 01:49 still running
reboot system boot 6.12.0-1-librem5 Tue Feb 24 01:48 - 01:49 (00:01)
reboot system boot 6.12.0-1-librem5 Tue Feb 24 01:39 - 01:49 (00:09)
reboot system boot 6.12.0-1-librem5 Tue Feb 24 01:00 - 01:49 (00:48)
reboot system boot 6.12.0-1-librem5 Tue Feb 24 00:50 - 00:59 (00:09)
reboot system boot 6.12.0-1-librem5 Tue Feb 24 00:48 - 00:59 (00:11)
reboot system boot 6.6.0-1-librem5 Tue Feb 24 00:28 - 00:47 (00:18)
reboot system boot 6.6.0-1-librem5 Mon Feb 23 01:48 - 00:28 (22:39)
reboot system boot 6.6.0-1-librem5 Sat Feb 21 03:09 - 00:28 (2+21:18)
This shows the reboot times. Iām capturing logs from 01:48 - 01:49 (00:01) as I think that was the likely crash time.
See if this has something useful "Yes, that's just like you. Getting rid of a champion foot-and-mouth.
Do you need logs from any other time?
Same here - thereās no crash logged, just a regular reboot that has been initiated at 01:49:16 and finished successfully at 01:49:22.
last reports crashed sessions like this:
reboot system boot 6.6.0-1-librem5 Sun Feb 22 20:17 - 21:14 (00:57)
purism pts/0 fe80::1958:d140: Sun Feb 22 20:13 - crash (00:03)
purism tty7 Sun Feb 22 20:13 - crash (00:04)
purism tty7 Sun Feb 22 20:13 - 20:13 (00:00)
reboot system boot 6.6.0-1-librem5 Sun Feb 22 20:13 - 21:14 (01:01)
May be I will just avoid this specific charger. I also tried with another charger (Xiomi poco F5 67W charger) which can suspend fine when charging. So I can confirm it is specific to this charger. Laptop and Furilabs FLX1 charges fine with that charger.