[Unstable Update] 2021-10-23 - Kernels, Nvidia, Haskell, Kodi, Thunderbird, PHP, Wine

Is fixed upstream already:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/gpu/drm/ttm/ttm_bo_util.c?h=v5.10.77&id=c21b4002214c1c7e7b627b9b53375612f7aab6db

2 Likes

Well the fix introduced a regression so they reverted it at Arch… I’ve you look at stable-queue.git you will find more reverts reported by Fedora and Arch.

https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/tree/queue-5.10

Other kernel series might be affected too…

1 Like

Yup, have checked that so next Kernel Series should be fine.

Update to kernel 5.15.1-1 has rendered Virtualbox unable to launch a guest OS. sudo vboxreload did not fix the issue either. (That command threw up another error)It was the message that the drivers were not installed or installed incorrectly etc… It is working fine with 5.10 kernel. For some reason the update ( pacman via TTY) did not pick up the linux515-virtualbox-host-modules update. I had to update manually.

The following is from pamac history:

[2021-11-06T16:18:31-0400] [ALPM] installed linux515-virtualbox-host-modules (6.1.28-3)
[2021-11-06T16:06:22-0400] [ALPM] upgraded linux515-headers (5.15.0-2 -> 5.15.1-1)
[2021-11-06T16:06:20-0400] [ALPM] upgraded linux515 (5.15.0-2 -> 5.15.1-1)
[2021-11-06T16:06:12-0400] [ALPM] upgraded linux510-virtualbox-host-modules (6.1.28-3 -> 6.1.28-4)
[2021-11-06T16:06:12-0400] [ALPM] upgraded linux510-headers (5.10.77-1 -> 5.10.78-1)
[2021-11-06T16:06:09-0400] [ALPM] upgraded linux510 (5.10.77-1 -> 5.10.78-1)
[2021-11-06T16:06:07-0400] [ALPM] upgraded foomatic-db (3:20211009-1 -> 3:20211106-1)
[2021-11-06T16:06:06-0400] [ALPM] upgraded fmt (8.0.1-1 -> 8.0.1-2)
[2021-11-06T16:06:06-0400] [ALPM] upgraded ffmpeg (2:4.4-6 -> 2:4.4.1-1)
sudo vboxreload                                                          
[sudo] password for jrichard: 
Unloading modules: 
Loading modules: modprobe: ERROR: could not insert 'vboxnetadp': Exec format error
modprobe: ERROR: could not insert 'vboxnetflt': Exec format error
modprobe: ERROR: could not insert 'vboxdrv': Exec format error

Screenshot_2021-11-06_19-57-59

edit: FIXED with update to 6.128-4 and sudo vboxreload.
Thank you for the quick fix.

Magic number mismatch again. Between kernel 5.15.1-1 and linux515-nvidia-470xx 470.82.00-2.
Looking at Branch Compare, a new revision of Nvidia package hasn’t been built? Mirror list refresh didn’t bring it up, at least.

Edit: More precisely kernel: nvidia: version magic '5.15.0-2-MANJARO SMP preempt mod_unload ' should be '5.15.1-1-MANJARO SMP preempt mod_unload '


Fixed! Thanks, @oberon !

System: Gnome unstable / B550 Board / MSI RX5500 8GB / Ryzen 7 3700X /
(only with:) Kernel 5.15.1-1

Warning in systemd.log:

.....
kernel:  ata3.00: Read log page 0x08 failed, Emask 0x1
kernel:  ata3.00: Read log page 0x08 failed, Emask 0x1
.....

Solution taken from:
https://askubuntu.com/questions/1060245/what-do-these-ata-error-messages-mean
==> add to: GRUB_CMDLINE_LINUX_DEFAULT new entry: “libata.force=noncq”

EDIT: please look for:

2-EDIT: problem has gone with kernl-version 5.15.2-2

2 Likes

There’s an nvidia-fake-powerd package in the AUR now, built from those instructions.
I think it should go without saying, but use at your own discretion! Used with 495xx drivers.

Nvidia GT730 Geforce not longer working after update (unstable) 07.11.2021.
Error: “Driver doesnot support your GT730”
SOLUTION:
in GRUB press “E” and add to bootoptions: 3 nomodeset
Boot (F10) => login prompt => login and type:

sudo mhwd -r pci $(mhwd -li |grep 'video-' | awk '{print $1}')
sudo mhwd -a pci free 0300
reboot…
(Nvidia-driver replaced by opensource-driver)

1 Like

Or install the new Nvidia driver package made for the occasion (The previous driver)?

mhwd -l

After updates of btrfs-progs-5.15-1, harfbuzz-3.1.1-1. harfbuzz-icu-3.1.1-1, and lib32-harfbuzz-3.1.1-1
==> Building image from preset: /etc/mkinitcpio.d/linux514.preset: 'default'
  -> -k /boot/vmlinuz-5.14-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.14-x86_64.img
==> ERROR: '/lib/modules/5.14.17-1-MANJARO' is not a valid kernel module directory
==> Building image from preset: /etc/mkinitcpio.d/linux514.preset: 'fallback'
  -> -k /boot/vmlinuz-5.14-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.14-x86_64-fallback.img -S autodetect
==> ERROR: '/lib/modules/5.14.17-1-MANJARO' is not a valid kernel module directory
==> Building image from preset: /etc/mkinitcpio.d/linux515.preset: 'default'
  -> -k /boot/vmlinuz-5.15-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.15-x86_64.img
==> Starting build: 5.15.1-1-MANJARO
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [autodetect]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [mdadm_udev]
Custom /etc/mdadm.conf file will be used in initramfs for assembling arrays.
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: /boot/initramfs-5.15-x86_64.img
==> Image generation successful
==> Building image from preset: /etc/mkinitcpio.d/linux515.preset: 'fallback'
  -> -k /boot/vmlinuz-5.15-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.15-x86_64-fallback.img -S autodetect
==> Starting build: 5.15.1-1-MANJARO
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [mdadm_udev]
Custom /etc/mdadm.conf file will be used in initramfs for assembling arrays.
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: /boot/initramfs-5.15-x86_64-fallback.img
==> Image generation successful
error: command failed to execute correctly

Edit: Something had nuked /usr/lib/modules/5.14.17-1-MANJARO entirely… :eyes:
I uninstalled and reinstalled linux514, and the folder and its contents are back.

All set 5.10.78 is fine on ThinkPad T60

I’m getting a strange update issue that sounds like it would be system-breaking, as of just now. I declined, of course. What should I do? Here’s what I saw in my terminal:

[alex@alex-pc ~]$ sudo pacman -Syu
:: Synchronizing package databases...
 core is up to date
 extra is up to date
 community is up to date
 multilib is up to date
:: Starting full system upgrade...
resolving dependencies...
looking for conflicting packages...
:: xf86-input-elographics and xorg-server are in conflict. Remove xorg-server? [y/N] N
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: xf86-input-elographics and xorg-server are in conflict (xorg-server<21.1.1)
[alex@alex-pc ~]$
2 Likes

Refresh your mirrors and/or wait until your mirror syncs, you need the xorg-server 21.1.1-2 packages first. Sorry, they came in slightly out of order. :grimacing:

:warning: Neither uninstall nor downgrade anything. :warning:

1 Like

Yup, waiting worked. Thanks for the info! I hope nobody accepted that prompt while it was active.

After the last update (Plasma 5.23.3) Octopi fails to start with the following error:

octopi: error while loading shared libraries: libqtermwidget5.so.0: cannot open shared object file: No such file or directory

UPD: sudo ln -s /lib/libqtermwidget5.so.1.0.0 /lib/libqtermwidget5.so.0 worked for now. I know it’s a workaround but I don’t care much as I use Octopi only for ez browsing of packages anyway.

2 Likes

Install it from the AUR for now, apparently no one here is interesting in updating it. :man_shrugging:

Gnome 41 finally came.
I did the update in TTY, rebooted normally.

Smooth and seamless.

Latest xorg-server update to 21.1.1-2 can cause some “fun” with scaling on plasma. I had to force the old default of 96dpi to reduce huge icons and fonts to proper size again. Plasma scaling was set to 100% at all times.

There’s also this thread in the arch forums.

1 Like

Here nvidia-390xx driver is not starting after xorg-server update.

Apparently the change in xorg-server will be reverted in 21.1.2 next week:

Arch itself might be opting to revert in it’s package sooner:
https://bugs.archlinux.org/task/72661