<?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IMX8M Plus GPU DRAM contention</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-Plus-GPU-DRAM-contention/m-p/2401351#M246246</link>
    <description>&lt;P&gt;We are running Yocto Linux on an IMX8M Plus&lt;/P&gt;&lt;P&gt;What we have noticed is that during any GPU activity, we get huge latency spikes for memory access from the CPU (1-3ms spikes). This is a huge issue, as we are using XDP for networking.&lt;/P&gt;&lt;P&gt;We have tried tuning&amp;nbsp;IMX8MP_ICM_A53, IMX8MP_ICM_GPU3D and&amp;nbsp;IMX8MP_ICM_GPU2D, setting them to 7, 2, 2 for QoS control. This made the latency spikes less frequent, but they are still there. The only option we found was doing absolutely no GPU activity or removing the GPU from the device tree, however this is not acceptable for our use-case. Are there some other setting that could be tuned to alleviate this behavior? We cannot have the GPU locking the DRAM for more than a 50-100 μs at most&lt;/P&gt;</description>
    <pubDate>Mon, 03 Aug 2026 08:04:18 GMT</pubDate>
    <dc:creator>richardlovgren</dc:creator>
    <dc:date>2026-08-03T08:04:18Z</dc:date>
    <item>
      <title>IMX8M Plus GPU DRAM contention</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-Plus-GPU-DRAM-contention/m-p/2401351#M246246</link>
      <description>&lt;P&gt;We are running Yocto Linux on an IMX8M Plus&lt;/P&gt;&lt;P&gt;What we have noticed is that during any GPU activity, we get huge latency spikes for memory access from the CPU (1-3ms spikes). This is a huge issue, as we are using XDP for networking.&lt;/P&gt;&lt;P&gt;We have tried tuning&amp;nbsp;IMX8MP_ICM_A53, IMX8MP_ICM_GPU3D and&amp;nbsp;IMX8MP_ICM_GPU2D, setting them to 7, 2, 2 for QoS control. This made the latency spikes less frequent, but they are still there. The only option we found was doing absolutely no GPU activity or removing the GPU from the device tree, however this is not acceptable for our use-case. Are there some other setting that could be tuned to alleviate this behavior? We cannot have the GPU locking the DRAM for more than a 50-100 μs at most&lt;/P&gt;</description>
      <pubDate>Mon, 03 Aug 2026 08:04:18 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-Plus-GPU-DRAM-contention/m-p/2401351#M246246</guid>
      <dc:creator>richardlovgren</dc:creator>
      <dc:date>2026-08-03T08:04:18Z</dc:date>
    </item>
    <item>
      <title>Re: IMX8M Plus GPU DRAM contention</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-Plus-GPU-DRAM-contention/m-p/2401923#M246272</link>
      <description>&lt;P&gt;Hi&amp;nbsp;&lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/265262"&gt;@richardlovgren&lt;/a&gt;,&lt;BR /&gt;Thank you for contacting NXP Support!&lt;/P&gt;
&lt;P&gt;Could you please tell me which BSP version you are using?&lt;/P&gt;
&lt;P&gt;Does this issue reproduce on an EVK as well, or is it only occurring on your custom hardware?&lt;/P&gt;
&lt;P&gt;Could you also provide the log files, your device tree, and any other relevant information so that I can investigate and diagnose the issue on my side?&lt;/P&gt;
&lt;P&gt;Additionally, please share the exact steps to reproduce the problem, so I can try to replicate it in my setup.&lt;/P&gt;
&lt;P&gt;Best regards,&lt;BR /&gt;Chavira&lt;/P&gt;</description>
      <pubDate>Tue, 04 Aug 2026 17:33:51 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-Plus-GPU-DRAM-contention/m-p/2401923#M246272</guid>
      <dc:creator>Chavira</dc:creator>
      <dc:date>2026-08-04T17:33:51Z</dc:date>
    </item>
  </channel>
</rss>

