<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>i.MX Processorsのトピックimx6q realtime performance</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/imx6q-realtime-performance/m-p/442301#M67989</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hello,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I am trying to achieve better real time performance on imx6qsabresd board. The idea is to run high priority real time task on a shielded core.&lt;/P&gt;&lt;P&gt;My test are based on cyclictest where I measure the maximum latency. The measured values are between 28 - 40ns.&lt;/P&gt;&lt;P&gt;My kernel version with preempt-rt patch:&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="jive_macro_code _jivemacro_uid_14338575244176534 jive_text_macro" data-renderedposition="134_8_1232_16" jivemacro_uid="_14338575244176534"&gt;&lt;P&gt;Linux qemu 3.14.23-rt20-PD13.2.3 #24 SMP PREEMPT RT Tue Jun 9 16:03:30 EEST 2015 armv7l GNU/Linux&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Cyclictest result&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="_jivemacro_uid_14338575359194323 jive_macro_code jive_text_macro" data-renderedposition="192_8_1232_48" jivemacro_uid="_14338575359194323"&gt;&lt;P&gt;root@qemu:~ taskset -c 2 cyclictest -t 1 -p 80 -n -m -q -l 100000&lt;/P&gt;&lt;P&gt;# /dev/cpu_dma_latency set to 0us&lt;/P&gt;&lt;P&gt;T: 0 ( 1447) P:80 I:1000 C: 100000 Min: 9 Act: 16 Avg: 16 Max: 28&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Going deeper I have found that the latency came from linux scheduler.&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="jive_macro_code jive_text_macro _jivemacro_uid_14338590205559401" data-renderedposition="282_8_1232_80" jivemacro_uid="_14338590205559401"&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 1us+: 0:120:R + [002] 1511: 94:R sleep&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 5us+: 0&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 8us+: sched_wakeup: comm=sleep pid=1511 prio=94 success=1 target_cpu=002&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2d...3.. 28us+: __schedule&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2d...3.. 31us : 0:120:R ==&amp;gt; [002] 1511: 94:R sleep&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Is it possible this delay to come from i.MX6 context switching?&lt;/P&gt;&lt;P&gt;Does anyone have already done similar tests with i.MX6 processors and if yes what are the results? Is it possible to achieve better maximum latency with i.MX6Q processor?&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Tue, 09 Jun 2015 14:18:28 GMT</pubDate>
    <dc:creator>ivanstoyanov</dc:creator>
    <dc:date>2015-06-09T14:18:28Z</dc:date>
    <item>
      <title>imx6q realtime performance</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/imx6q-realtime-performance/m-p/442301#M67989</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hello,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I am trying to achieve better real time performance on imx6qsabresd board. The idea is to run high priority real time task on a shielded core.&lt;/P&gt;&lt;P&gt;My test are based on cyclictest where I measure the maximum latency. The measured values are between 28 - 40ns.&lt;/P&gt;&lt;P&gt;My kernel version with preempt-rt patch:&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="jive_macro_code _jivemacro_uid_14338575244176534 jive_text_macro" data-renderedposition="134_8_1232_16" jivemacro_uid="_14338575244176534"&gt;&lt;P&gt;Linux qemu 3.14.23-rt20-PD13.2.3 #24 SMP PREEMPT RT Tue Jun 9 16:03:30 EEST 2015 armv7l GNU/Linux&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Cyclictest result&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="_jivemacro_uid_14338575359194323 jive_macro_code jive_text_macro" data-renderedposition="192_8_1232_48" jivemacro_uid="_14338575359194323"&gt;&lt;P&gt;root@qemu:~ taskset -c 2 cyclictest -t 1 -p 80 -n -m -q -l 100000&lt;/P&gt;&lt;P&gt;# /dev/cpu_dma_latency set to 0us&lt;/P&gt;&lt;P&gt;T: 0 ( 1447) P:80 I:1000 C: 100000 Min: 9 Act: 16 Avg: 16 Max: 28&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Going deeper I have found that the latency came from linux scheduler.&lt;/P&gt;&lt;PRE __default_attr="plain" __jive_macro_name="code" class="jive_macro_code jive_text_macro _jivemacro_uid_14338590205559401" data-renderedposition="282_8_1232_80" jivemacro_uid="_14338590205559401"&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 1us+: 0:120:R + [002] 1511: 94:R sleep&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 5us+: 0&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2dn.h4.. 8us+: sched_wakeup: comm=sleep pid=1511 prio=94 success=1 target_cpu=002&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2d...3.. 28us+: __schedule&lt;/P&gt;&lt;P&gt;&amp;lt;idle&amp;gt;-0 2d...3.. 31us : 0:120:R ==&amp;gt; [002] 1511: 94:R sleep&lt;/P&gt;&lt;/PRE&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Is it possible this delay to come from i.MX6 context switching?&lt;/P&gt;&lt;P&gt;Does anyone have already done similar tests with i.MX6 processors and if yes what are the results? Is it possible to achieve better maximum latency with i.MX6Q processor?&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Tue, 09 Jun 2015 14:18:28 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/imx6q-realtime-performance/m-p/442301#M67989</guid>
      <dc:creator>ivanstoyanov</dc:creator>
      <dc:date>2015-06-09T14:18:28Z</dc:date>
    </item>
    <item>
      <title>Re: imx6q realtime performance</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/imx6q-realtime-performance/m-p/442302#M67990</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hi, &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Unfortunately I have not been able to find any helpful data for you.&lt;/P&gt;&lt;P&gt;I believe that you may get some information from where you downloaded that Linux Version.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Perhaps this can help.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;A href="https://community.nxp.com/message/305302"&gt;i.MX6: Has someone tried PREEMPT_RT patch&lt;/A&gt; &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Best Regards,&lt;/P&gt;&lt;P&gt;Alejandro&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Mon, 15 Jun 2015 23:51:29 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/imx6q-realtime-performance/m-p/442302#M67990</guid>
      <dc:creator>alejandrolozan1</dc:creator>
      <dc:date>2015-06-15T23:51:29Z</dc:date>
    </item>
  </channel>
</rss>

