[Unstable Update] November 2023 Edition

firefox developer edition is behaving strange after kernel rebuilding.

One could suspect that hardware acceleration is in play - I am not completely sure why it is hanging - actually completely freezing the system - a switch to console (TTY)
takes 5s to complete - running a pkill firefox brings system back from the ‘dead’

[master U:1 ?:4] $ uname -r
6.7.0-rc1-next-20231114-1-next-git-01346-g5ba73bec5e7b-dirty
[master U:1 ?:4] $ inxi -SCGm
System:
  Host: tiger
    Kernel: 6.7.0-rc1-next-20231114-1-next-git-01346-g5ba73bec5e7b-dirty
    arch: x86_64 bits: 64 Desktop: KDE Plasma v: 5.27.9 Distro: Manjaro Linux
Memory:
  System RAM: total: 64 GiB note: est. available: 62.63 GiB
    used: 4.35 GiB (6.9%)
  RAM Report: permissions: Unable to run dmidecode. Root privileges
    required.
CPU:
  Info: 12-core model: AMD Ryzen Threadripper PRO 5945WX s bits: 64
    type: MT MCP cache: L2: 6 MiB
  Speed (MHz): avg: 1893 min/max: 1800/7015 cores: 1: 1800 2: 1800 3: 1800
    4: 1777 5: 1800 6: 1800 7: 1800 8: 1800 9: 1800 10: 1800 11: 1800 12: 1800
    13: 1800 14: 1800 15: 1800 16: 1800 17: 1800 18: 1800 19: 1800 20: 1777
    21: 1800 22: 4100 23: 1800 24: 1800
Graphics:
  Device-1: AMD Navi 31 [Radeon RX 7900 XT/7900 XTX] driver: amdgpu v: kernel
  Display: wayland server: X.org v: 1.21.1.9 with: Xwayland v: 23.2.2
    compositor: kwin_wayland driver: X: loaded: modesetting dri: radeonsi
    gpu: amdgpu resolution: 5120x1440~120Hz
  API: EGL v: 1.5 drivers: radeonsi,swrast
    platforms: wayland,x11,surfaceless,device
  API: OpenGL v: 4.6 compat-v: 4.5 vendor: amd mesa v: 23.1.9-manjaro1.1
    renderer: AMD Radeon RX 7900 XTX (gfx1100 LLVM 16.0.6 DRM 3.56
    6.7.0-rc1-next-20231114-1-next-git-01346-g5ba73bec5e7b-dirty)

Using backup browser Vivaldi to write this.

Normal firefox seems to be less unaffected … although hanging can be observed for a few seconds - system becomes unresponsive - no mouse movement etc. - the returns to normal after a few seconds.

No problems here but everything here is so different I doubt it will help…
I have been using 6.7 since the middle of the merge window, all Manjaro patches are commented out for my build
Firefox Dev 120.0b9 (64-bit) is the Mozilla build

[hjw@tux ~]$ inxi -SCGm
System:
  Host: tux Kernel: 6.7.0-1-TUX arch: x86_64 bits: 64 Desktop: Xfce v: 4.18.1
    Distro: Manjaro Linux
Memory:
  System RAM: total: 16 GiB available: 15.51 GiB used: 2.01 GiB (13.0%)
  RAM Report: permissions: Unable to run dmidecode. Root privileges
    required.
CPU:
  Info: quad core model: Intel Core i5-6500 bits: 64 type: MCP cache:
    L2: 1024 KiB
  Speed (MHz): avg: 800 min/max: 800/3600 cores: 1: 800 2: 800 3: 800 4: 800
Graphics:
  Device-1: Intel HD Graphics 530 driver: i915 v: kernel
  Display: x11 server: X.Org v: 21.1.9 driver: X: loaded: modesetting
    dri: iris gpu: i915 s-res: 1920x1080 resolution: 1920x1080
  API: OpenGL Message: Unable to show GL data. glxinfo is missing.

Whoooops! Thank you. I was a little tired when I pushed that PKGBUILD last night, I guess …
Fixed now and currently rebuilding.

After a rest and returning to my computer - I find the freeze and hanging to be more widespread.

Vivaldi is hanging so I am thinking GPU driver - all kernel based on my system.

I have rebuilt kernel to

 $ uname -r
6.7.0-rc1-next-20231115-1-next-git-01493-gbc962b35b139-dirty

But I had to use Xorg where I usually use plasma-wayland-session - to be able to open a brower to do followup

EDIT: switching between wayland and xorg is ATM the difference between a working system or a hung system - so I am thinking

 $ grep /var/log/pacman.log -e wayland
[2023-10-03T13:08:36+0200] [ALPM] removed kwayland (5.110.0-1)
[2023-10-03T13:08:41+0200] [ALPM] upgraded qt5-wayland (5.15.10+kde+r57-1 -> 5.15.10+kde+r62-1)
[2023-10-03T13:08:42+0200] [ALPM] installed kwayland5 (5.110.0-1)
[2023-10-03T13:08:43+0200] [ALPM] upgraded kwayland-integration (5.27.8-1 -> 5.27.8-2)
[2023-10-06T08:18:50+0200] [PACMAN] Running 'pacman -Syu plasma-wayland-session'
[2023-10-06T08:18:52+0200] [ALPM] installed xorg-xwayland (23.2.1-1)
[2023-10-06T08:18:52+0200] [ALPM] installed plasma-wayland-session (5.27.8-2)
[2023-10-09T13:20:52+0200] [ALPM] upgraded qt5-wayland (5.15.10+kde+r62-1 -> 5.15.11+kde+r59-1)
[2023-10-10T07:39:54+0200] [ALPM] removed wayland-utils (1.2.0-1)
[2023-10-18T06:07:51+0200] [ALPM] installed qt6-wayland (6.6.0-1)
[2023-10-22T07:25:13+0200] [ALPM] upgraded kwayland5 (5.110.0-1 -> 5.111.0-1)
[2023-10-25T11:03:14+0200] [ALPM] upgraded qt5-wayland (5.15.11+kde+r59-1 -> 5.15.11+kde+r60-1)
[2023-10-25T11:03:16+0200] [ALPM] upgraded kwayland-integration (5.27.8-2 -> 5.27.9-1)
[2023-10-25T11:03:16+0200] [ALPM] upgraded plasma-wayland-session (5.27.8-2 -> 5.27.9-1)
[2023-10-26T10:28:49+0200] [ALPM] upgraded xorg-xwayland (23.2.1-1 -> 23.2.2-1)
[2023-10-26T10:28:49+0200] [ALPM] upgraded plasma-wayland-session (5.27.9-1 -> 5.27.9.1-1)
[2023-11-09T01:04:41+0100] [ALPM] upgraded plasma-wayland-session (5.27.9.1-1 -> 5.27.9.1-2)
[2023-11-14T10:17:37+0100] [ALPM] upgraded kwayland5 (5.111.0-1 -> 5.112.0-1)
[2023-11-14T10:17:39+0100] [ALPM] installed lib32-wayland (1.22.0-1)

And since the freeze started Nov. 14 - it appears to be kwayland5 optionally also lib32-wayland

I also installed 6.7.0-rc1 for a little rc-testing and found that firefox needed seconds to start (normally it starts without any noticeable delay). Also running vlc had some issues. Running journalctl I found lots of

kernel: i915 0000:00:02.0: [drm] *ERROR* Failed to send PXP TEE message
kernel: i915 0000:00:02.0: [drm] *ERROR* Failed to send tee msg for inv-stream-key-15, ret=[-62]

messages, which have not been there using 6.6.1. Being no linux kernel expert I then switched back to 6.6.1 and everything is o.k. again.

PS: I am running newest Manjaro KDE Unstable with X11.

I don’t see any usefulness in posting issues with release candidate kernels in this forum/thread.
If you think there are issues: please use the proper channels so kernel developers can be made aware and possibly fix some.

2 Likes

i’ve got similar problems with wayland but i’m on stable. i could minimize it by installing

libva-vdpau-driver-wayland

but i still suffer from a taskbar that freezes and get’s unresponsive. (time is frozen, taskbar-apps can’t be opened).

I will try - and see if it helps - tomorrow - it is strange though - as my system has been extremely stable for a very long time now.

question: is your taskbar freezing when it’s happening ? mine does including the time that is displayed.

Also interesting because this is the vdpau backend for vaapi (wayland variant)… which, in the current repos, with cards that support it, has been replaced (conflicts with) the newer libva-nvidia-driver.
(NVDEC backend for vaapi)

When upgrading “mkinitcpio”:

==> WARNING: Possibly missing ‘/bin/sh’ for script: /usr/bin/mount.glusterfs

I downgraded, back to 36-6 before reboot, didn’t know what could cause.

Considering that’s just a warning, I wouldn’t be too worried if I were you…

Kernel5.4 fails to build

...
(1/1) reinstalling mkinitcpio                                                                                      
...
==> Building image from preset: /etc/mkinitcpio.d/linux54.preset: 'default'
==> Using configuration file: '/etc/mkinitcpio.conf'
  -> -k /boot/vmlinuz-5.4-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.4-x86_64.img
==> Starting build: '5.4.260-1-MANJARO'
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [autodetect]
  -> Running build hook: [modconf]
  -> Running build hook: [kms]
  -> Running build hook: [keyboard]
  -> Running build hook: [keymap]
  -> Running build hook: [consolefont]
==> WARNING: consolefont: no font found in configuration
  -> Running build hook: [block]
  -> Running build hook: [filesystems]
  -> Running build hook: [resume]
  -> Running build hook: [fsck]
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: '/boot/initramfs-5.4-x86_64.img'
==> WARNING: errors were encountered during the build. The image may not be complete.
==> Building image from preset: /etc/mkinitcpio.d/linux54.preset: 'fallback'
==> Using configuration file: '/etc/mkinitcpio.conf'
  -> -k /boot/vmlinuz-5.4-x86_64 -c /etc/mkinitcpio.conf -g /boot/initramfs-5.4-x86_64-fallback.img -S autodetect
==> Starting build: '5.4.260-1-MANJARO'
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [modconf]
  -> Running build hook: [kms]
==> WARNING: Possibly missing firmware for module: 'ast'
  -> Running build hook: [keyboard]
  -> Running build hook: [keymap]
  -> Running build hook: [consolefont]
==> WARNING: consolefont: no font found in configuration
  -> Running build hook: [block]
==> WARNING: Possibly missing firmware for module: 'qed'
==> WARNING: Possibly missing firmware for module: 'bfa'
==> WARNING: Possibly missing firmware for module: 'qla1280'
==> WARNING: Possibly missing firmware for module: 'aic94xx'
==> WARNING: Possibly missing firmware for module: 'qla2xxx'
==> WARNING: Possibly missing firmware for module: 'wd719x'
  -> Running build hook: [filesystems]
  -> Running build hook: [resume]
  -> Running build hook: [fsck]
==> Generating module dependencies
==> Creating gzip-compressed initcpio image: '/boot/initramfs-5.4-x86_64-fallback.img'
==> WARNING: errors were encountered during the build. The image may not be complete.
...
error: command failed to execute correctly

Is it ok to update? No problem?

I have decided not to test libva-vdpau-driver-wayland as it is a custom script and I want to test upfront issues with the latest next kernel code using Manjaro defaults.

Everything stalls.

It is only doing so with Linux 6.7 kernel - no stalling with 6.6 - Wayland works as reliable as before.

Only when I boot into 6.7 …

Uhh - this is unstable branch - if you need to ask if it is safe - you are not in the right place :slight_smile:

To address your specific consern - it is just a warning message - on systemd based systems /bin is symlinked to /usr/bin so sh will be available …

1 Like

I think it’s fine. I also get WARNINGS IIRC and I have no problems.

The BEST would be if you made a backup, do the update and test. If any problems, then you can always roll back. If you don’t have a backup to roll back to, you can always downgrade and restore from a chroot environment.

1 Like

Do you have /bin/sh in your /etc/shells?

Indeed. See the changelog:

  • The error message when binaries are not found will now be more
    helpful and include the name of the binary.

I had this in the end of mkinitcpio run as a part of the update:

Summary
...
...
( 4/10) Updating linux initcpios...
...
...
==> Unified kernel image generation successful
==> Running post hooks
  -> Running post hook: [uki-sbsign]
Signing Unsigned original image
==> Post processing done
==> Building image from preset: /etc/mkinitcpio.d/linux66.preset: 'default'
==> Using default configuration file: '/etc/mkinitcpio.conf'
  -> -k /boot/vmlinuz-6.6-x86_64 -U /efi/EFI/Linux/manjaro-6.6-x86_64.efi -g /boot/initramfs-6.6-x86_64.img --cmdline=/etc/cmdline.d/default.conf --splash=/sys/firmware/acpi/bgrt/image --microcode /boot/intel-ucode.img
==> Starting build: '6.6.1-1-MANJARO'
  -> Running build hook: [base]
  -> Running build hook: [udev]
  -> Running build hook: [keyboard]
==> WARNING: Possibly missing firmware for module: 'xhci_pci'
  -> Running build hook: [keymap]
  -> Running build hook: [consolefont]
  -> Running build hook: [autodetect]
  -> Running build hook: [plymouth]
  -> Running build hook: [modconf]
  -> Running build hook: [kms]
  -> Running build hook: [block]
  -> Running build hook: [plymouth-tpm2-totp]
  -> Running build hook: [tpm2]
  -> Running build hook: [plymouth-encrypt]
  -> Running build hook: [lvm2]
  -> Running build hook: [resume]
  -> Running build hook: [filesystems]
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: '/boot/initramfs-6.6-x86_64.img'
==> Image generation successful
==> Creating unified kernel image: '/efi/EFI/Linux/manjaro-6.6-x86_64.efi'
  -> Using UEFI stub: '/usr/lib/systemd/boot/efi/linuxx64.efi.stub'
  -> Using os-release file: '/etc/os-release'
  -> Using cmdline file: '/etc/cmdline.d/default.conf'
  -> Using splash image: '/sys/firmware/acpi/bgrt/image'
  -> Using kernel image: '/boot/vmlinuz-6.6-x86_64'
  -> Using microcode image: '/boot/intel-ucode.img'
==> Unified kernel image generation successful
==> Running post hooks
  -> Running post hook: [uki-sbsign]
Signing Unsigned original image
==> Post processing done
==> Building image from preset: /etc/mkinitcpio.d/linux66.preset: 'systemd'
==> Using configuration file: '/etc/mkinitcpio.systemd.conf'
  -> -k /boot/vmlinuz-6.6-x86_64 -c /etc/mkinitcpio.systemd.conf -U /efi/EFI/Linux/manjaro-6.6-x86_64-systemd.efi -g /boot/initramfs-6.6-x86_64-systemd.img --cmdline=/etc/cmdline.d/systemd.conf --splash=/sys/firmware/acpi/bgrt/image --microcode /boot/intel-ucode.img
==> Starting build: '6.6.1-1-MANJARO'
  -> Running build hook: [base]
  -> Running build hook: [systemd]
  -> Running build hook: [keyboard]
==> WARNING: Possibly missing firmware for module: 'xhci_pci'
  -> Running build hook: [autodetect]
  -> Running build hook: [sd-vconsole]
  -> Running build hook: [sd-plymouth]
  -> Running build hook: [modconf]
  -> Running build hook: [kms]
  -> Running build hook: [block]
  -> Running build hook: [sd-plymouth-tpm2-totp]
  -> Running build hook: [sd-encrypt]
  -> Running build hook: [lvm2]
  -> Running build hook: [filesystems]
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: '/boot/initramfs-6.6-x86_64-systemd.img'
==> Image generation successful
==> Creating unified kernel image: '/efi/EFI/Linux/manjaro-6.6-x86_64-systemd.efi'
  -> Using UEFI stub: '/usr/lib/systemd/boot/efi/linuxx64.efi.stub'
  -> Using os-release file: '/etc/os-release'
  -> Using cmdline file: '/etc/cmdline.d/systemd.conf'
  -> Using splash image: '/sys/firmware/acpi/bgrt/image'
  -> Using kernel image: '/boot/vmlinuz-6.6-x86_64'
  -> Using microcode image: '/boot/intel-ucode.img'
==> Unified kernel image generation successful
==> Running post hooks
  -> Running post hook: [uki-sbsign]
Signing Unsigned original image
==> Post processing done
ошибка: не удалось корректно выполнить команду
( 5/10) Refreshing PackageKit...
( 6/10) Reloading system bus configuration...
...
...

Which means something like: “error: failed to execute command correctly”.
This was shown only after the last kernel image generated by mkinicpio (6.6). But a manual run of mkinitcpio finished with no failure and previously built efi’s looked normal anyway.

yes, I do have.

Ok, I found, I used for months this hooks in my initcpio.conf, with no problem:

HOOKS=(systemd autodetect modconf block keyboard sd-vconsole filesystems)

Now, with mkinitcpio-37-1 I needed to add “base” at the begining of the list of hooks, to get rid of the warning.

1 Like

I don’t need to add base and as far as i know systemd doesn’t need base?
Also putting base in front of systemd stops it being a systemd led mkinitcpio systemd must be first.