* runaway latency detection
@ 2016-08-20 17:03 T. Smith
2016-08-20 17:54 ` Nikolay Borisov
0 siblings, 1 reply; 4+ messages in thread
From: T. Smith @ 2016-08-20 17:03 UTC (permalink / raw)
To: Daniel Bristot de Oliveira, Sebastian Andrzej Siewior,
Jon Masters, Carsten Emde, Peter Zijlstra, Thomas Gleixner,
Andrew Morton, Ingo Molnar, Clark Williams, Steven Rostedt,
Daniel Wagner
Cc: linux-kernel
Trying to isolate PREEMPT_RT (full preemption) runaway latency conditions seen when using the cyclictest on the 4.1.8 SMP PREEMPT_RT kernel. Seen average latency of 3uS, but runaway or outlier max latency seems indeterminate. Noted 72uS with one run of the cyclictest for example. Is this a known issue, the cause understood, and is there a patch for this?
The goal is to isolate causes of indeterminism when using the PREEMPT_RT kernel configuration with full preemption and to characterize latency and jitter using ftrace, any recommendations?
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: runaway latency detection
2016-08-20 17:03 runaway latency detection T. Smith
@ 2016-08-20 17:54 ` Nikolay Borisov
[not found] ` <CAChUvXP8Zy4uz95NWFEtF9oU6rHjPZ0Uz5qPMqY1n3i2eBeaSw@mail.gmail.com>
0 siblings, 1 reply; 4+ messages in thread
From: Nikolay Borisov @ 2016-08-20 17:54 UTC (permalink / raw)
To: T. Smith, Daniel Bristot de Oliveira, Sebastian Andrzej Siewior,
Jon Masters, Carsten Emde, Peter Zijlstra, Thomas Gleixner,
Andrew Morton, Ingo Molnar, Clark Williams, Steven Rostedt,
Daniel Wagner
Cc: linux-kernel
On 20.08.2016 20:03, T. Smith wrote:
> The goal is to isolate causes of indeterminism when using the PREEMPT_RT kernel configuration with full preemption and to characterize latency and jitter using ftrace, any recommendations?
What hardware is it ? If it's x86 it's entirely possible you are seeing
SMI. Perhaps you should look into disabling those if you are on x86?
^ permalink raw reply [flat|nested] 4+ messages in thread
* runaway latency detection
@ 2016-08-22 2:22 T. Smith
0 siblings, 0 replies; 4+ messages in thread
From: T. Smith @ 2016-08-22 2:22 UTC (permalink / raw)
To: linux-kernel, linux-scsi, irqbalance
Trying to isolate PREEMPT_RT (full preemption) runaway latency conditions seen when using the cyclictest on the 4.1.8 SMP PREEMPT_RT kernel. Seen average latency of 3uS, but runaway or outlier max latency random spikes seem indeterminate. Noted 72uS with one run of the cyclictest for example. Is this a known issue, the cause understood, and is there a patch for this? This is on ARM8 Cortex-A53 four core. Running the cyclictest on an isolated core 2 or 3 shows same behavior as do other platforms. No Trusted OS configured.
The goal is to isolate causes of indeterminism when using the PREEMPT_RT kernel configuration with full preemption and to characterize latency and jitter using ftrace, any recommendations on catching random latency spikes using ftrace or profiling when running the cyclictest?
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2016-08-22 2:22 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-08-20 17:03 runaway latency detection T. Smith
2016-08-20 17:54 ` Nikolay Borisov
[not found] ` <CAChUvXP8Zy4uz95NWFEtF9oU6rHjPZ0Uz5qPMqY1n3i2eBeaSw@mail.gmail.com>
2016-08-21 3:45 ` Steven Rostedt
2016-08-22 2:22 T. Smith
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).