CPU architecture wrong in terms of cores and threads

BollerwagenPicard · 2024-05-31 18:48:07

I would like to apologize at the beginning if I am explaining the problem in a strange way. Unfortunately, I am not a native English speaker and the problem is confusing me a lot right now.

I wanted to use Virtualbox today and I noticed that I can't assign all my CPU cores to a VM. Since I haven't changed anything in the BIOS or system, I was very confused at first and wanted to check the hardware virtualization functions to be on the safe side. For this I use the command `lscpu` out of habit, I noticed a strange output for my CPU:

CPU(s):                   24
  On-line CPU(s) list:    0-23
Vendor ID:                GenuineIntel
  Model name:             Intel(R) Xeon(R) CPU E5-2640 0 @ 2.50GHz
    CPU family:           6
    Model:                45
    Thread(s) per core:   12
    Core(s) per socket:   1
    Socket(s):            2
[...]
NUMA:                     
  NUMA node(s):           2
  NUMA node0 CPU(s):      0-5,12-17
  NUMA node1 CPU(s):      6-11,18-23

Using the command `journalctl -b` I was able to find many of these following errors in the boot process...

Mai 31 20:26:03 archlinux kernel: BUG: arch topology borken
Mai 31 20:26:03 archlinux kernel:      the SMT domain not a subset of the CLS domain

Therefore I started my system with the kernel parameter `mds=full,nosmt` as a test, which changed the output (unfortunately I did not save it) but did not lead to the correct CPU layout of my system.

I currently running with Kernel: Linux [Hostname] 6.9.2-arch1-1 #1 SMP PREEMPT_DYNAMIC Sun, 26 May 2024 01:30:29 +0000 x86_64 GNU/Linux
microcode: 0x71a

Unfortunately, I currently have no idea how to solve the problem and am looking for help and ideas. Many thanks in advance

loqs · 2024-05-31 18:51:39

https://gitlab.archlinux.org/archlinux/ … /issues/57? Please post the full output of `dmesg`.

gromit · 2024-05-31 18:56:17

Which CPU does your system have? And could you please post a full dmesg?

BollerwagenPicard · 2024-05-31 19:08:10

Hi

the log is hosted at: https://paste.3server.de/?ae37ded4be46a … i4iY2VEm2E

My system has two Intel Xeon CPU E5-26400, it is an old server changed to a desktop.

gromit · 2024-05-31 19:23:57

Is this a new problem? i.e. does it occur when you downgrade your kernel to 6.8.X or when you switch to the linux-lts package?

gromit · 2024-05-31 19:26:12

Also does this problem occur inside of the VM or on the host?

A quick search gave me for example https://www.suse.com/zh-cn/support/kb/doc/?id=000018616

BollerwagenPicard · 2024-05-31 19:26:40

So it's a relatively new problem and it used to work, but unfortunately I don't know exactly when the problem started.

I don't have the confidence to downgrade or switch to LTS.

loqs · 2024-05-31 19:50:39

If the system journal has any boots from before the update to linux 6.9 you could check if the same error message is present.

BollerwagenPicard · 2024-05-31 19:55:55

gromit wrote:

Also does this problem occur inside of the VM or on the host?

It is the host

BollerwagenPicard · 2024-05-31 20:10:57

loqs wrote:

If the system journal has any boots from before the update to linux 6.9 you could check if the same error message is present.

Found the log of 6.8.9-arch1-2... and I do not see the error in the old logs

loqs · 2024-05-31 21:37:44

It would appear to be an issue introduced in linux 6.9 but there is not much that can be done if you are not willing to test kernels other than the latest stable release.

BollerwagenPicard · 2024-06-01 09:51:09

loqs wrote:

It would appear to be an issue introduced in linux 6.9 but there is not much that can be done if you are not willing to test kernels other than the latest stable release.

that's not quite what I said...

I just said I didn't have the confidence to do it... if you help me or lead me to a verfitziert good guide you can also do such tests

gromit · 2024-06-01 13:43:03

So how we usually do that is that for every debugging step involved you get a kernel image to install which you test on your machine, see https://gitlab.archlinux.org/archlinux/ … ote_187653 for examples

But yeah the linked issue loqs linked in https://bbs.archlinux.org/viewtopic.php … 8#p2174898 sounds related so maybe applying the fix for that already solves your problem

A prebuilt version is linked here: https://gitlab.archlinux.org/archlinux/ … ote_189079

BollerwagenPicard · 2024-06-01 14:34:19

Hi
I get a lot of errors by installing the 6.9.2.arch1-1.5

(4/6) Updating linux initcpios...
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'default'
==> Using default configuration file: '/etc/mkinitcpio.conf'
  -> -k /boot/vmlinuz-linux -g /boot/initramfs-linux.img
==> Starting build: '6.9.2-arch1-1.5'
  -> Running build hook: [base]
  -> Running build hook: [systemd]
  -> Running build hook: [autodetect]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [sd-vconsole]
  -> Running build hook: [btrfs]
==> ERROR: module not found: 'nvidia'
==> ERROR: module not found: 'nvidia_modeset'
==> ERROR: module not found: 'nvidia_uvm'
==> ERROR: module not found: 'nvidia_drm'
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: '/boot/initramfs-linux.img'
==> WARNING: errors were encountered during the build. The image may not be complete.
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'fallback'
==> Using default configuration file: '/etc/mkinitcpio.conf'
  -> -k /boot/vmlinuz-linux -g /boot/initramfs-linux-fallback.img -S autodetect
==> Starting build: '6.9.2-arch1-1.5'
  -> Running build hook: [base]
  -> Running build hook: [systemd]
  -> Running build hook: [modconf]
  -> Running build hook: [block]
==> WARNING: Possibly missing firmware for module: 'aic94xx'
==> WARNING: Possibly missing firmware for module: 'bfa'
==> WARNING: Possibly missing firmware for module: 'qed'
==> WARNING: Possibly missing firmware for module: 'qla1280'
==> WARNING: Possibly missing firmware for module: 'qla2xxx'
==> WARNING: Possibly missing firmware for module: 'wd719x'
  -> Running build hook: [filesystems]
  -> Running build hook: [keyboard]
  -> Running build hook: [sd-vconsole]
  -> Running build hook: [btrfs]
==> ERROR: module not found: 'nvidia'
==> ERROR: module not found: 'nvidia_modeset'
==> ERROR: module not found: 'nvidia_uvm'
==> ERROR: module not found: 'nvidia_drm'
==> Generating module dependencies
==> Creating zstd-compressed initcpio image: '/boot/initramfs-linux-fallback.img'
==> WARNING: errors were encountered during the build. The image may not be complete.
Fehler: Befehl konnte nicht korrekt ausgeführt werden

But I will try 6.9.3.arch1-1 as well and hope for the best

loqs · 2024-06-01 14:44:19

For 6.9.2.arch1-1.5 install both linux-6.9.2.arch1-1.5-x86_64.pkg.tar.zst and linux-headers-6.9.2.arch1-1.5-x86_64.pkg.tar.zst and replace the nvidia package with nvidia-dkms if the system is not already using nvidia-dkms then DKMS will build modules for that kernel.

BollerwagenPicard · 2024-06-01 15:02:45

6.9.3.arch1-1 did not fix it...

linux-6.9.2.arch1-1.5-x86_64.pkg.tar.zst and linux-headers-6.9.2.arch1-1.5-x86_64.pkg.tar.zst with nvidia-dkms did fix CPU problems and adds working GUI thank you loqs...

But I have a new problem in this combination...

I know it's hardly part of the original problem, but anyone have any idea how I can get cdemu to work?

loqs · 2024-06-01 17:45:55

Did you replace the vhba package with vhba-dkms? You need to change all packages that provide kernel modules to their -dkms version.

BollerwagenPicard · 2024-06-01 17:50:51

Thank you... I did not

do I need to change them back if it is fixed in the ArchLinux kernel or can i keep it on dkms?

loqs · 2024-06-01 17:53:58

You can keep them on the -dkms versions or change them back if you want after the issue is fixed in the linux package.

Arch Linux

#1 2024-05-31 18:48:07

CPU architecture wrong in terms of cores and threads

#2 2024-05-31 18:51:39

Re: CPU architecture wrong in terms of cores and threads

#3 2024-05-31 18:56:17

Re: CPU architecture wrong in terms of cores and threads

#4 2024-05-31 19:08:10

Re: CPU architecture wrong in terms of cores and threads

#5 2024-05-31 19:23:57

Re: CPU architecture wrong in terms of cores and threads

#6 2024-05-31 19:26:12

Re: CPU architecture wrong in terms of cores and threads

#7 2024-05-31 19:26:40

Re: CPU architecture wrong in terms of cores and threads

#8 2024-05-31 19:50:39

Re: CPU architecture wrong in terms of cores and threads

#9 2024-05-31 19:55:55

Re: CPU architecture wrong in terms of cores and threads

#10 2024-05-31 20:10:57

Re: CPU architecture wrong in terms of cores and threads

#11 2024-05-31 21:37:44

Re: CPU architecture wrong in terms of cores and threads

#12 2024-06-01 09:51:09

Re: CPU architecture wrong in terms of cores and threads

#13 2024-06-01 13:43:03

Re: CPU architecture wrong in terms of cores and threads

#14 2024-06-01 14:34:19

Re: CPU architecture wrong in terms of cores and threads

#15 2024-06-01 14:44:19

Re: CPU architecture wrong in terms of cores and threads

#16 2024-06-01 15:02:45

Re: CPU architecture wrong in terms of cores and threads

#17 2024-06-01 17:45:55

Re: CPU architecture wrong in terms of cores and threads

#18 2024-06-01 17:50:51

Re: CPU architecture wrong in terms of cores and threads

#19 2024-06-01 17:53:58

Re: CPU architecture wrong in terms of cores and threads

Board footer