<?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Re: IMX8M plus: NNRT segmentation fault</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209901#M167568</link>
    <description>&lt;P&gt;Did you read and know what you are talking bout ?&lt;/P&gt;&lt;P&gt;I was using prebuilt tensorflow lite benchmark tool for Linux ARM 64 bits.&lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;</description>
    <pubDate>Mon, 11 Jan 2021 22:48:36 GMT</pubDate>
    <dc:creator>keithmok</dc:creator>
    <dc:date>2021-01-11T22:48:36Z</dc:date>
    <item>
      <title>IMX8M plus: NNRT segmentation fault</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209285#M167503</link>
      <description>&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;&lt;SPAN class="Apple-converted-space"&gt;Segmentation fault on /usr/lib/libnnrt.so.1.1.9 when running the attached tflite model using tensorflow lite on IMX8M plus using NPU..&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;Procedure : Use latest NXP release for IMX8M plus&lt;BR /&gt;Download Google tensorflow prebuilt benchmark tools for linux_aarch64 from&amp;nbsp;&lt;A href="https://www.tensorflow.org/lite/performance/measurement" target="_blank" rel="noopener"&gt;https://www.tensorflow.org/lite/performance/measurement&lt;/A&gt;.&lt;BR /&gt;&lt;BR /&gt;&lt;/SPAN&gt;Run /home/root/linux_aarch64_benchmark_model --graph=ai_benchmark/memnet_quant.tflite --use_nnapi=true&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;GDB shows&amp;nbsp; it segmentation fault in /usr/lib/libnnrt.so.1.1.9 .&lt;BR /&gt;&lt;BR /&gt;[Thread debugging using libthread_db enabled]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Using host libthread_db library "/lib/libthread_db.so.1".&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;STARTING!&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Log parameter values verbosely: [0]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Graph: [ai_benchmark/memnet_quant.tflite]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Use NNAPI: [1]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;NNAPI accelerators available: [vsi-npu]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Loaded model ai_benchmark/memnet_quant.tflite&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;INFO: Created TensorFlow Lite delegate for NNAPI.&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Explicitly applied NNAPI delegate, and the model graph will be completely executed by the delegate.&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;The input model file size (MB): 0.656504&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Initialized session in 87.803ms.&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Running benchmark for at least 1 iterations and at least 0.5 seconds but terminate if exceeding 150 seconds.&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[New Thread 0xffffe5b7c1c0 (LWP 3343)]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[New Thread 0xffffe537b1c0 (LWP 3344)]&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p2"&gt;&amp;nbsp;&lt;/P&gt;&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Thread 1 "linux_aarch64_b" received signal SIGSEGV, Segmentation fault.&lt;/SPAN&gt;&lt;/P&gt;&lt;P class="p3"&gt;&lt;SPAN class="s2"&gt;0x0000fffff7ac89bc&lt;/SPAN&gt;&lt;SPAN class="s3"&gt; in &lt;/SPAN&gt;&lt;SPAN class="s1"&gt;nnrt::op::Operand::echo(unsigned int) const&lt;/SPAN&gt;&lt;SPAN class="s3"&gt; () from /usr/lib/libnnrt.so.1.1.9&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 11 Jan 2021 23:15:11 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209285#M167503</guid>
      <dc:creator>keithmok</dc:creator>
      <dc:date>2021-01-11T23:15:11Z</dc:date>
    </item>
    <item>
      <title>Re: IMX8M plus: NNRT segmentation fault</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209898#M167567</link>
      <description>&lt;P&gt;I think you should use the tool for ARM.&lt;/P&gt;</description>
      <pubDate>Mon, 11 Jan 2021 22:36:28 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209898#M167567</guid>
      <dc:creator>jimmychan</dc:creator>
      <dc:date>2021-01-11T22:36:28Z</dc:date>
    </item>
    <item>
      <title>Re: IMX8M plus: NNRT segmentation fault</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209901#M167568</link>
      <description>&lt;P&gt;Did you read and know what you are talking bout ?&lt;/P&gt;&lt;P&gt;I was using prebuilt tensorflow lite benchmark tool for Linux ARM 64 bits.&lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 11 Jan 2021 22:48:36 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209901#M167568</guid>
      <dc:creator>keithmok</dc:creator>
      <dc:date>2021-01-11T22:48:36Z</dc:date>
    </item>
    <item>
      <title>Re: IMX8M plus: NNRT segmentation fault</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209902#M167569</link>
      <description>&lt;P&gt;"&lt;SPAN class="s1"&gt;&lt;SPAN class="Apple-converted-space"&gt;benchmark tools for linux_x86-64&lt;/SPAN&gt;&lt;/SPAN&gt;" in your question. So, I was thinking you are using the tool not for ARM.&lt;/P&gt;</description>
      <pubDate>Mon, 11 Jan 2021 22:53:19 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209902#M167569</guid>
      <dc:creator>jimmychan</dc:creator>
      <dc:date>2021-01-11T22:53:19Z</dc:date>
    </item>
    <item>
      <title>Re: IMX8M plus: NNRT segmentation fault</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209904#M167570</link>
      <description>&lt;P&gt;That's a typ (I corrected that), but pls read the full log and command I post:&lt;BR /&gt;Run /home/root/&lt;STRONG&gt;linux_aarch64_benchmark_model&lt;BR /&gt;&lt;/STRONG&gt;aarch64 is clearly a&amp;nbsp; linux arm 64 bits binary.&lt;/P&gt;</description>
      <pubDate>Mon, 11 Jan 2021 23:17:18 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/IMX8M-plus-NNRT-segmentation-fault/m-p/1209904#M167570</guid>
      <dc:creator>keithmok</dc:creator>
      <dc:date>2021-01-11T23:17:18Z</dc:date>
    </item>
  </channel>
</rss>

