[SOLVED] x86/split lock detection log entries after reboot

[Raid007]

Hello everyone,

I hope you're doing well. I've recently encountered an issue regarding the x86/split lock detection log entries after a system reboot and unfortunately couldn't find a suitable solution through my research. Hence, I'm reaching out here to seek some assistance.

I've noticed that after a reboot, the system generates approximately 25-30 log entries with varying process IDs (PIDs) and CPUs. The entry typically looks like this:

Mar 01 13:58:07 pve1 kernel: x86/split lock detection: #AC: CPU 2/KVM/2056 took a split_lock trap at address: 0x7ef1d050

Interestingly, these entries are frequent initially but tend to diminish after about 8 hours, and then they cease until the next reboot.

I'd like to mention that all virtual machines are running smoothly without any issues, and overall system performance is unaffected. These log entries are the only anomaly I've observed.

My question is whether these entries signify a potential issue or if it's normal for them to occur after a reboot. If they indicate a problem, I'd appreciate any insights on how to proceed with troubleshooting. I am using Proxmox 8.1.4 with the enterprise rep.

Thank you all in advance for your assistance.

Best regards,
Raid007

 

[Raid007]

I've continued monitoring, and the errors persist. Can anyone confirm if they also see these errors in their logging?
Or has anyone an idea if I should be worried about this warnings? Maybe they are pretty normal after rebooting a system?

System is running like a Porsche but I am still a little worried...

Thanks for any assistance.

Best,
Raid007

 

[cheiss]

Hi,

this is a basically a feature of the hardware and the kernel: Bus lock detection and handling
This normally does and should not have any impact on the stability of the system.
Probably a newer kernel was loaded after the reboot which enabled these warnings by default.

It basically detects situations which can impact performance (in the <1% range) - but if they vanish anyway after a certain amount of time, I personally would not worry much about it. Especially if you cannot notice any performance impact, as you said.

You can also turn these warnings off on the kernel command line if you really want, as explained in the linked documentation.

 

[Raid007]

Thank you for your response. I really appreciate it.
I won't worry too much as long as everything runs smoothly!

Have a great day!

 

[cheiss]

Great!

Please just mark the thread as SOLVED, by editing the first post - there should be a dropdown for that near the title field. This helps others to find this thread more easily in the future!