* Debugging PREEMPT_RT spikes on kernelci
@ 2026-11-16 11:11 Sebastian Andrzej Siewior
2026-11-16 11:28 ` Denys Fedoryshchenko
0 siblings, 1 reply; 3+ messages in thread
From: Sebastian Andrzej Siewior @ 2026-11-16 11:11 UTC (permalink / raw)
To: kernelci; +Cc: John Ogness
Hi,
I have been looking at the results of meson-g12b-a311d-khadas-vim3 on
defconfig+preempt_rt. The run for v7.4-rc3 is available at
https://dashboard.kernelci.org/test/maestro%3A6afa52b25b429d354e5ef864
which run about 10h ago. The reported cyclictest results were
| T: 0 ( 482) P:98 I:1000 C: 441036 Min: 0 Act: 1258 Avg: 274 Max: 9556
| T: 1 ( 489) P:98 I:1500 C: 360000 Min: 5 Act: 9 Avg: 17 Max: 85
and are a bit high. I opened a PR against kernelci-pipeline to change a
few config settings and hoped for the best. This PR was merged FRI 6th
NOV and according to the artifacts of this run, it is still the "old"
config.
I checked a few other trees and boards and none of them is using the
newer config. So either I did something wrong or the git repo was not
yet updated.
What would be the delay one can expect until the git tree is updated?
Sebastian
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Debugging PREEMPT_RT spikes on kernelci
2026-11-16 11:11 Debugging PREEMPT_RT spikes on kernelci Sebastian Andrzej Siewior
@ 2026-11-16 11:28 ` Denys Fedoryshchenko
2026-11-19 8:09 ` Sebastian Andrzej Siewior
0 siblings, 1 reply; 3+ messages in thread
From: Denys Fedoryshchenko @ 2026-11-16 11:28 UTC (permalink / raw)
To: Sebastian Andrzej Siewior; +Cc: kernelci, John Ogness
Hello,
I will do deploy production right now.
Usually changes are deployed weekly, unless there is breaking changes that might need more needing.
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: "kernelci"<kernelci@lists.linux.dev>
Cc: "John Ogness"<john.ogness@linutronix.de>
Date: Mon, 16 Nov 2026 13:11:35 +0200
Subject: Debugging PREEMPT_RT spikes on kernelci
> Hi,
>
> I have been looking at the results of meson-g12b-a311d-khadas-vim3 on
> defconfig+preempt_rt. The run for v7.4-rc3 is available at
> https://dashboard.kernelci.org/test/maestro%3A6afa52b25b429d354e5ef864
>
> which run about 10h ago. The reported cyclictest results were
> | T: 0 ( 482) P:98 I:1000 C: 441036 Min: 0 Act: 1258 Avg: 274 Max: 9556
> | T: 1 ( 489) P:98 I:1500 C: 360000 Min: 5 Act: 9 Avg: 17 Max: 85
>
> and are a bit high. I opened a PR against kernelci-pipeline to change a
> few config settings and hoped for the best. This PR was merged FRI 6th
> NOV and according to the artifacts of this run, it is still the "old"
> config.
> I checked a few other trees and boards and none of them is using the
> newer config. So either I did something wrong or the git repo was not
> yet updated.
> What would be the delay one can expect until the git tree is updated?
>
> Sebastian
>
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Debugging PREEMPT_RT spikes on kernelci
2026-11-16 11:28 ` Denys Fedoryshchenko
@ 2026-11-19 8:09 ` Sebastian Andrzej Siewior
0 siblings, 0 replies; 3+ messages in thread
From: Sebastian Andrzej Siewior @ 2026-11-19 8:09 UTC (permalink / raw)
To: Denys Fedoryshchenko; +Cc: kernelci, John Ogness
On 2026-11-16 13:28:26 [+0200], Denys Fedoryshchenko wrote:
> Hello,
Hi,
> I will do deploy production right now.
Thank you.
> Usually changes are deployed weekly, unless there is breaking changes
> that might need more needing.
I see. Thank you for the information.
Judging from yesterday's run
https://dashboard.kernelci.org/test/maestro%3A6afe07fc99c74e1a99855381
the config bits are merged and cyclictest reports
| T: 0 ( 484) P:98 I:1000 C: 540000 Min: 4 Act: 4 Avg: 13 Max: 98
| T: 1 ( 491) P:98 I:1500 C: 360000 Min: 4 Act: 7 Avg: 14 Max: 92
which looks good.
Sebastian
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-11-19 8:09 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-16 11:11 Debugging PREEMPT_RT spikes on kernelci Sebastian Andrzej Siewior
2026-11-16 11:28 ` Denys Fedoryshchenko
2026-11-19 8:09 ` Sebastian Andrzej Siewior
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox