* 回复：Re: Xenomai latency test not good
@ 2019-09-09 10:14 鲁克文
2019-09-09 11:19 ` Jan Kiszka
0 siblings, 1 reply; 2+ messages in thread
From: 鲁克文 @ 2019-09-09 10:14 UTC (permalink / raw)
To: Jan Kiszka, xenomai
Dear Jan Kiszka:
I re-enable ACPI_PROCESSOR in the kernel, but also the latency test seems no good,attached file is the new kernel config file.
/*********************************************************************************************/RTT| 00:01:04 (periodic user-mode task, 100 us period, priority 99)RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worstRTD| 0.072| 0.785| 19.712| 2| 0| -0.177| 131.284RTD| 0.096| 2.280| 15.701| 2| 0| -0.177| 131.284RTD| 0.059| 1.573| 12.868| 2| 0| -0.177| 131.284RTD| 0.058| 2.512| 25.545| 2| 0| -0.177| 131.284RTD| 0.286| 1.928| 153.611| 3| 0| -0.177| 153.611RTD| 0.285| 2.834| 17.032| 3| 0| -0.177| 153.611RTD| 0.128| 2.104| 26.467| 3| 0| -0.177| 153.611RTD| 0.140| 1.105| 14.978| 3| 0| -0.177| 153.611RTD| 0.163| 0.950| 12.708| 3| 0| -0.177| 153.611RTD| 0.199| 1.012| 15.109| 3| 0| -0.177| 153.611RTD| 0.042| 1.007| 10.511| 3| 0| -0.177| 153.611RTD| 0.102| 0.763| 125.684| 4| 0| -0.177| 153.611RTD| 0.138| 0.810| 12.226| 4| 0| -0.177| 153.611RTD| 0.125| 1.312| 13.858| 4| 0| -0.177| 153.611RTD| 0.256| 0.854| 67.039| 4| 0| -0.177| 153.611RTD| 0.280| 0.841| 14.205| 4| 0| -0.177| 153.611RTD| 0.099| 0.631| 76.185| 4| 0| -0.177| 153.611RTD| 0.230| 0.953| 16.041| 4| 0| -0.177| 153.611RTD| 0.086| 0.985| 65.416| 4| 0| -0.177| 153.611RTD| 0.110| 0.856| 18.357| 4| 0| -0.177| 153.611/************************************************************************************/
--------------------------------
Best regards
Kevin
----- 原始邮件 -----
发件人：Jan Kiszka <jan.kiszka@siemens.com>
收件人：hikwlu@sina.com, xenomai <xenomai@xenomai.org>
主题：Re: Xenomai latency test not good
日期：2019年09月09日 17点46分
On 09.09.19 11:34, 鲁克文 via Xenomai wrote:
> Dear Sir:
> I have a Intel Atom J1900, using Ubuntu 16.04, Kernel version:4.9.24, patch with xenomai 3.0.5, using ipipe-core-4.9.24-x86-2.patch.
> When I start to run the latency test, I get this result(it is not good):/******************************************************/RTT| 00:00:43 (periodic user-mode task, 100 us period, priority 99)RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worstRTD| 0.192| 0.621| 10.228| 5| 0| 0.070| 130.271RTD| 0.312| 0.512| 116.038| 6| 0| 0.070| 130.271RTD| 0.311| 0.513| 6.818| 6| 0| 0.070| 130.271RTD| 0.311| 0.609| 13.132| 6| 0| 0.070| 130.271RTD| 0.178| 0.643| 10.743| 6| 0| 0.070| 130.271RTD| 0.309| 0.482| 4.416| 6| 0| 0.070| 130.271RTD| 0.057| 0.549| 10.501| 6| 0| 0.057| 130.271RTD| 0.297| 0.643| 16.239| 6| 0| 0.057| 130.271RTD| 0.164| 0.683| 9.876| 6| 0| 0.057| 130.271RTD| 0.308| 0.410| 2.888| 6| 0| 0.057| 130.271RTD| 0.139| 0.474| 169.443| 7| 0| 0.057| 169.443RTD| 0.318| 0.400| 2.924| 7| 0| 0.057| 169.443RTD| 0.318| 0.442| 3.559| 7| 0| 0.057| 169.443RTD| 0.137| 0.434| 3.379| 7| 0| 0.057| 169.443RTD| 0.316| 0.465| 125.119| 8| 0| 0.057| 169.443RTD| 0.316| 0.410| 5.106| 8| 0| 0.057| 169.443RTD| 0.231| 0.477| 3.953| 8| 0| 0.057| 169.443RTD| 0.303| 0.549| 5.225| 8| 0| 0.057| 169.443RTD| 0.279| 0.443| 6.689| 8| 0| 0.057| 169.443RTD| 0.218| 0.428| 6.521| 8| 0| 0.057| 169.443RTD| 0.109| 0.468| 122.246| 9| 0| 0.057| 169.443/*****************************************************************************/
Your email client mangled the linefeeds.
> For the Intel X86 SMI thing,here is my system setting(could not be disabled):/**********************************************************************************/root@googol:/home/googol# dmesg |grep -i xenomai[ 0.000000] Kernel command line: BOOT_IMAGE=/vmlinuz-4.9.24 root=UUID=ba9a8ea7-08a9-4772-9744-da7b70a9360f ro quiet splash vt.handoff=7 xenomai.smi=enabled[ 1.694624] [Xenomai] scheduling class idle registered.[ 1.694626] [Xenomai] scheduling class rt registered.[ 1.694670] [Xenomai] SMI-enabled chipset found[ 1.694688] [Xenomai] SMI workaround failed![ 1.694731] I-pipe: head domain Xenomai registered.[ 1.695859] [Xenomai] Cobalt v3.0.5 (Sisyphus's Boulder) /**************************************************************************************/
> The attached file is my kernel config file.According to the testing result, it seems that the system has large latency, so it is not good when doing realtime control.Could you help me to improve its realtime behavior, or guide me how to improve it?
You may have to re-enable ACPI_PROCESSOR, even if Xenomai warns about this. I've
seen cases where disabling it triggered the firmware to throttle the CPU and
cause such high latencies.
Jan
--
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux
-------------- next part --------------
A non-text attachment was scrubbed...
Name: .config
Type: application/xml
Size: 200369 bytes
Desc: not available
URL: <http://xenomai.org/pipermail/xenomai/attachments/20190909/df057a8c/attachment.wsdl>
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: 回复：Re: Xenomai latency test not good
2019-09-09 10:14 回复：Re: Xenomai latency test not good 鲁克文
@ 2019-09-09 11:19 ` Jan Kiszka
0 siblings, 0 replies; 2+ messages in thread
From: Jan Kiszka @ 2019-09-09 11:19 UTC (permalink / raw)
To: hikwlu, xenomai
On 09.09.19 12:14, 鲁克文 wrote:
> Dear Jan Kiszka:
>
> I re-enable ACPI_PROCESSOR in the kernel, but also the latency test seems no
> good,attached file is the new kernel config file.
>
> /*********************************************************************************************/
> RTT| 00:01:04 (periodic user-mode task, 100 us period, priority 99)
> RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
> RTD| 0.072| 0.785| 19.712| 2| 0| -0.177| 131.284
> RTD| 0.096| 2.280| 15.701| 2| 0| -0.177| 131.284
> RTD| 0.059| 1.573| 12.868| 2| 0| -0.177| 131.284
> RTD| 0.058| 2.512| 25.545| 2| 0| -0.177| 131.284
> RTD| 0.286| 1.928| 153.611| 3| 0| -0.177| 153.611
> RTD| 0.285| 2.834| 17.032| 3| 0| -0.177| 153.611
> RTD| 0.128| 2.104| 26.467| 3| 0| -0.177| 153.611
> RTD| 0.140| 1.105| 14.978| 3| 0| -0.177| 153.611
> RTD| 0.163| 0.950| 12.708| 3| 0| -0.177| 153.611
> RTD| 0.199| 1.012| 15.109| 3| 0| -0.177| 153.611
> RTD| 0.042| 1.007| 10.511| 3| 0| -0.177| 153.611
> RTD| 0.102| 0.763| 125.684| 4| 0| -0.177| 153.611
> RTD| 0.138| 0.810| 12.226| 4| 0| -0.177| 153.611
> RTD| 0.125| 1.312| 13.858| 4| 0| -0.177| 153.611
> RTD| 0.256| 0.854| 67.039| 4| 0| -0.177| 153.611
> RTD| 0.280| 0.841| 14.205| 4| 0| -0.177| 153.611
> RTD| 0.099| 0.631| 76.185| 4| 0| -0.177| 153.611
> RTD| 0.230| 0.953| 16.041| 4| 0| -0.177| 153.611
> RTD| 0.086| 0.985| 65.416| 4| 0| -0.177| 153.611
> RTD| 0.110| 0.856| 18.357| 4| 0| -0.177| 153.611
> /************************************************************************************/
>
Yeah, these numbers are still off. Likely, we see some power management effects.
They might be triggered by Linux, and then some kernel switch can do the magic
(check e.g. CONFIG_ACPI_PROCESSOR_CSTATE). You may also try "idle=poll" as
kernel parameter. Or the issue is caused by the firmware (bios). Check power
settings there. In the worst case, the behaviour is SMI induced and can't be
disabled. Then only a different firmware can help - or a different hardware.
Jan
--
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-09-09 11:19 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-09-09 10:14 回复：Re: Xenomai latency test not good 鲁克文
2019-09-09 11:19 ` Jan Kiszka
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.