* big rt latencies compared to non rt
@ 2021-05-11 16:00 Angelo Dureghello
2021-05-11 19:50 ` Jonathan Schwender
2021-05-11 20:14 ` John Ogness
0 siblings, 2 replies; 4+ messages in thread
From: Angelo Dureghello @ 2021-05-11 16:00 UTC (permalink / raw)
To: linux-rt-users
Hi all,
i am running some RT tests on imx8qxp.
Kernel is a imx (nxp) modified kernel, but with
rt patch (5.4.70) that at least applies properly.
Linux version 5.4.70-rt40-00120-g3e7dde1168e5-dirty
I am running some histograms by
cyclictest --duration=180 -t=1 -m -Sp99 \
-i2000 -h3000 -q > histogram
Average is near 16uS but there are some outside
values near 40000, as
000009 000000 000000 000000 000000
000010 000001 000000 000002 000000
000011 000002 000000 000000 000000
000012 000023 000006 000008 000004
000013 001565 001504 002010 001367
000014 000425 001959 000649 000700
000015 001447 001917 001129 001415
000016 034084 026118 040328 032334
000017 040537 015000 033773 041092 <-----
000018 005771 009092 007999 007512
000019 002709 023378 001790 003117
000020 000167 006238 000109 000136
000021 000135 002369 000099 000093
000022 000039 000622 000025 000041
000023 000153 000123 000026 000023
000024 000986 000178 000564 000500
000025 001872 000292 001291 001515
000026 000049 000269 000112 000081
000027 000010 000561 000011 000014
000028 000003 000296 000016 000009
000029 000005 000035 000014 000002
000030 000001 000008 000006 000000
000031 000001 000008 000006 000001
000032 000011 000002 000004 000000
Rest of the histogram is mostly 00000 values.
As far as you may know, is there something i am
missing for this cpu ?
Thanks a lot,
Regards,
angelo
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: big rt latencies compared to non rt
2021-05-11 16:00 big rt latencies compared to non rt Angelo Dureghello
@ 2021-05-11 19:50 ` Jonathan Schwender
2021-05-11 20:14 ` John Ogness
1 sibling, 0 replies; 4+ messages in thread
From: Jonathan Schwender @ 2021-05-11 19:50 UTC (permalink / raw)
To: Angelo Dureghello, linux-rt-users
Hi Angelo,
>
> Average is near 16uS but there are some outside
> values near 40000, as
You are reading the table wrong. The leftmost column is the microsecond
bucket. The other columns give the amount of times a latency was inside
that microsecond bucket for each measurement thread.
> 000017 040537 015000 033773 041092 <-----
For example, in this line CPU 0 had 40537 latencies in the 17 us bucket,
while CPU 1 had 15000 measured latencies of 17 us (and so forth).
You should also check the line at the End `# Max Latencies: xxxxx`,
which will give you the maximum latency, and the line below which shows
you how often the histogram overflowed.
> Thanks a lot,
>
> Regards,
> angelo
>
Hope I could help,
Jonathan
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: big rt latencies compared to non rt
2021-05-11 16:00 big rt latencies compared to non rt Angelo Dureghello
2021-05-11 19:50 ` Jonathan Schwender
@ 2021-05-11 20:14 ` John Ogness
2021-05-11 20:19 ` Angelo Dureghello
1 sibling, 1 reply; 4+ messages in thread
From: John Ogness @ 2021-05-11 20:14 UTC (permalink / raw)
To: Angelo Dureghello, linux-rt-users
Hi Angelo,
On 2021-05-11, Angelo Dureghello <angelo@kernel-space.org> wrote:
> cyclictest --duration=180 -t=1 -m -Sp99 \
> -i2000 -h3000 -q > histogram
>
> Average is near 16uS but there are some outside
> values near 40000, as
The first column is the latency in microseconds. The other columns
(2nd-5th) is the count of times that latency was seen for a CPU.
> 000009 000000 000000 000000 000000
> 000010 000001 000000 000002 000000
> 000011 000002 000000 000000 000000
> 000012 000023 000006 000008 000004
> 000013 001565 001504 002010 001367
> 000014 000425 001959 000649 000700
> 000015 001447 001917 001129 001415
> 000016 034084 026118 040328 032334
> 000017 040537 015000 033773 041092 <-----
CPU3 saw a latency of 17us 41092 times.
John Ogness
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: big rt latencies compared to non rt
2021-05-11 20:14 ` John Ogness
@ 2021-05-11 20:19 ` Angelo Dureghello
0 siblings, 0 replies; 4+ messages in thread
From: Angelo Dureghello @ 2021-05-11 20:19 UTC (permalink / raw)
To: John Ogness, linux-rt-users, schwenderjonathan
Hi John and Jonathan,
thanks a lot for the help.
On 11/05/21 10:14 PM, John Ogness wrote:
> Hi Angelo,
>
> On 2021-05-11, Angelo Dureghello <angelo@kernel-space.org> wrote:
>> cyclictest --duration=180 -t=1 -m -Sp99 \
>> -i2000 -h3000 -q > histogram
>>
>> Average is near 16uS but there are some outside
>> values near 40000, as
>
> The first column is the latency in microseconds. The other columns
> (2nd-5th) is the count of times that latency was seen for a CPU.
>
>> 000009 000000 000000 000000 000000
>> 000010 000001 000000 000002 000000
>> 000011 000002 000000 000000 000000
>> 000012 000023 000006 000008 000004
>> 000013 001565 001504 002010 001367
>> 000014 000425 001959 000649 000700
>> 000015 001447 001917 001129 001415
>> 000016 034084 026118 040328 032334
>> 000017 040537 015000 033773 041092 <-----
>
> CPU3 saw a latency of 17us 41092 times.
>
> John Ogness
>
--
Angelo Dureghello
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-05-11 20:26 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-05-11 16:00 big rt latencies compared to non rt Angelo Dureghello
2021-05-11 19:50 ` Jonathan Schwender
2021-05-11 20:14 ` John Ogness
2021-05-11 20:19 ` Angelo Dureghello
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.