lkml.org 
[lkml]   [2023]   [Jan]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: rseq CPU ID not correct on 6.0 kernels for pinned threads
From
On 2023-01-12 11:33, Florian Weimer wrote:
> * Mathieu Desnoyers:
>
>> As you also point out, it can also be caused by some other task
>> modifying the affinity of your task concurrently. You could print
>> the result of sched_getaffinity on error to get a better idea of
>> the expected vs actual mask.
>>
>> Lastly, it could be caused by CPU hotplug which would set all bits
>> in the affinity mask as a fallback. As you mention it should not be
>> the cause there.
>>
>> Can you share your kernel configuration ?
>
> Attached.
>
> cpupower frequency-info says:
>
> analyzing CPU 0:
> driver: intel_cpufreq
> CPUs which run at the same hardware frequency: 0
> CPUs which need to have their frequency coordinated by software: 0
> maximum transition latency: 20.0 us
> hardware limits: 800 MHz - 4.60 GHz
> available cpufreq governors: conservative ondemand userspace powersave performance schedutil
> current policy: frequency should be within 800 MHz and 4.60 GHz.
> The governor "schedutil" may decide which speed to use
> within this range.
> current CPU frequency: Unable to call hardware
> current CPU frequency: 3.20 GHz (asserted by call to kernel)
> boost state support:
> Supported: yes
> Active: yes
>
> And I have: kernel.sched_energy_aware = 1
>
>> Is this on a physical machine or in a virtual machine ?
>
> I think it happened on both.
>
> I added additional error reporting to the test (running on kernel
> 6.0.18-300.fc37.x86_64), and it seems that there is something that is
> mucking with affinity masks:
>
> info: Detected CPU set size (in bits): 64
> info: Maximum test CPU: 19
> error: Pinned thread 17 ran on impossible cpu 7
> info: getcpu reported CPU 7, node 0
> info: CPU affinity mask: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
> error: Pinned thread 3 ran on impossible cpu 13
> info: getcpu reported CPU 13, node 0
> info: CPU affinity mask: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
> info: Main thread ran on 2 CPU(s) of 20 available CPU(s)
> info: Other threads ran on 20 CPU(s)
>
> For each of these threads, the affinity mask should be a singleton set.
> Now I need to find out if there is a process that changes affinity
> settings.

If it's not cpu hotunplug, then perhaps something like systemd modifies
the AllowedCPUs of your cpuset concurrently ?

Thanks,

Mathieu

--
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com

\
 
 \ /
  Last update: 2023-03-26 23:38    [W:0.069 / U:0.176 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site