Kernel update?

[nickb333]

Are there any plans to update the kernel package on armv7h? Currently I have linux-armv7 5.11.2-1.

Code: Select all

 $ uname -a
Linux alarm-pcd3-hd 5.11.2-1-ARCH #1 SMP PREEMPT Sun Feb 28 22:03:25 UTC 2021 armv7l GNU/Linux


Thanks.

[rautesamtr]

If i remember correctly the problem is that the maintainer doesn't have time to update the armv7h kernel. The package probably needs a new maintainer or co-maintainer since the last update was 9 months ago.

[Mettacrawler]

linux-armv7-5.16.1-1 was released as stable. It works great for rme.

[graysky]

linux-armv7-5.16.1-1 was released as stable. It works great for rme.

Yep: https://github.com/archlinuxarm/PKGBUIL ... ac4f4aec69

[karog]

This morning I upgraded my odroid XU4 from linux-armv7-5.11.2-1 to linux-armv7-5.16.1-1.

It boots ok from a USB attached SSD with /root. But I also have a USB 3.0 WD My Passport which normally mounts on boot via fstab but not since upgrade. ls /dev/sd* shows the WD as does lsusb. But if I do a blkid, it takes more than 4 minutes but then finally completes. After that, I can mount -a and everything is fine and subsequent blkid is fast.

However, if I reboot, the scenario repeats.

The following is from dmesg showing the WD being recognized:

Code: Select all

[    7.010441] scsi 1:0:0:0: Direct-Access     WD       My Passport 0748 1022 PQ: 0 ANSI: 6
[    7.017545] sd 1:0:0:0: Attached scsi generic sg1 type 0
[    7.017778] sd 1:0:0:0: [sdb] 1953458176 512-byte logical blocks: (1.00 TB/931 GiB)
[    7.023606] scsi 1:0:0:1: Enclosure         WD       SES Device       1022 PQ: 0 ANSI: 6
[    7.030355] sd 1:0:0:0: [sdb] Write Protect is off
[    7.039879] scsi 1:0:0:1: Attached scsi generic sg2 type 13
[    7.042806] sd 1:0:0:0: [sdb] Mode Sense: 47 00 10 08
[    7.043270] sd 1:0:0:0: [sdb] No Caching mode page found
[    7.053638] sd 1:0:0:0: [sdb] Assuming drive cache: write through


[csoroiu]

Also, on my odroidxu, after update, the network interface is not getting an ip address, but seems up.
I'm still trying to figure out whats wrong.

Only the below packages got updated before it broke. (extract from pacman log)

Code: Select all

[2022-01-18T17:36:37+0200] [ALPM] upgraded expat (2.4.2-1 -> 2.4.3-1)
[2022-01-18T17:36:38+0200] [ALPM] upgraded libxml2 (2.9.12-6 -> 2.9.12-7)
[2022-01-18T17:37:05+0200] [ALPM] upgraded linux-armv7 (5.11.2-1 -> 5.16.1-1)
[2022-01-18T17:37:05+0200] [ALPM] upgraded netctl (1.26-1 -> 1.27-1)
[2022-01-18T17:37:06+0200] [ALPM] upgraded runc (1.0.3-1 -> 1.1.0-1)
[2022-01-18T17:37:13+0200] [ALPM] upgraded vim-runtime (8.2.3890-1 -> 8.2.4106-1)
[2022-01-18T17:37:14+0200] [ALPM] upgraded vim (8.2.3890-1 -> 8.2.4106-1)


[Mettacrawler]

karog: But if I do a blkid, it takes more than 4 minutes but then finally completes. After that, I can mount -a and everything is fine and subsequent blkid is fast.

When blkid runs for the first time, it scans all devices and creates a cache (/run/blkid/blkid.tab). That's why subsequent calls are faster, they don't need to scan anymore, they just read the cache.

[autumnontape]

My Hisense Chromebook C11 doesn't move past a black screen, with backlight on, on boot since the kernel update. I have to hold the power button to turn it off. It's the same after re-flashing the old vmlinux.kpart, with the only difference being that the backlight turns off... I don't know if that sort of brokenness is expected when the flashed vmlinux.kpart doesn't match the installed linux-armv7.

I'd like to look at/provide logs, but I can't figure out how to view the logs stored on the MicroSD card from another (x86_64) device. There's nothing interesting in /var/log, at least nothing that's legible when treated as a plain text file. So I'd appreciate it if someone could help me with that, it'd make a good first step... :p

[autumnontape]

Update: I figured out how to view the systemd journal, but I'm only seeing the logs from when I tried booting with the old vmlinux.kpart. I'm not great at interpreting these, but it looks like it failed loading kernel modules, which isn't surprising.

With the new kernel, it looks like it doesn't even get far enough to start writing to the journal.

[rautesamtr]

Sorry for hijacking. I just wanted to say a really big "Thank You!" to Kevin Mihelich for making time to upgrade the v7 kernel.