<?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>eIQ Machine Learning Softwareのトピック[ArmNN on IMX6 (CORTEXA9)]: CpuAcc causes segmentation faults when Armnn compiled in STATIC</title>
    <link>https://community.nxp.com/t5/eIQ-Machine-Learning-Software/ArmNN-on-IMX6-CORTEXA9-CpuAcc-causes-segmentation-faults-when/m-p/1326190#M400</link>
    <description>&lt;P&gt;Hello ArmNN/IMX experts,&lt;/P&gt;&lt;P&gt;I have been able to build from sources (outside eIQ context) ArmNN Static librairies for an iMx6 target.&lt;/P&gt;&lt;P&gt;I have built a static version of the libs because I have been disapointed by the performances of the Shared versions that I also built, and was hoping an improvement as some internet litterature mention ...&lt;/P&gt;&lt;P&gt;Using the SHARED ArmNN libs (*.so) I'm able to run an app which executes inference of a TFLite model on CpuRef (CPU-only) and CpuAcc (using NEON).&lt;/P&gt;&lt;P&gt;Using the STATIC ArmNN libs (*.a), the CPURef is working but not the CpuAcc which generates a segmentation fault at runtime.&lt;/P&gt;&lt;P&gt;I have tried, in the gcc command line the "whole-archive" option, in order to include all the libs, but this is not sufficient and provide a huge binary size.&lt;/P&gt;&lt;P&gt;Since the ArmNN framework does not generates any specific warnings at runtime (even if logging is activated) I am stucked.&lt;/P&gt;&lt;P&gt;## Does someone have an idea ?&lt;/P&gt;&lt;P&gt;For information, I have measured an improvement of ~5% by using the CpuRef with STATIC ArmNN built compared with SHARED (default build option in tutorials).&lt;/P&gt;&lt;P&gt;Kind regards,&lt;/P&gt;&lt;P&gt;Nicolas&lt;/P&gt;</description>
    <pubDate>Thu, 19 Aug 2021 09:57:58 GMT</pubDate>
    <dc:creator>nvassal</dc:creator>
    <dc:date>2021-08-19T09:57:58Z</dc:date>
    <item>
      <title>[ArmNN on IMX6 (CORTEXA9)]: CpuAcc causes segmentation faults when Armnn compiled in STATIC</title>
      <link>https://community.nxp.com/t5/eIQ-Machine-Learning-Software/ArmNN-on-IMX6-CORTEXA9-CpuAcc-causes-segmentation-faults-when/m-p/1326190#M400</link>
      <description>&lt;P&gt;Hello ArmNN/IMX experts,&lt;/P&gt;&lt;P&gt;I have been able to build from sources (outside eIQ context) ArmNN Static librairies for an iMx6 target.&lt;/P&gt;&lt;P&gt;I have built a static version of the libs because I have been disapointed by the performances of the Shared versions that I also built, and was hoping an improvement as some internet litterature mention ...&lt;/P&gt;&lt;P&gt;Using the SHARED ArmNN libs (*.so) I'm able to run an app which executes inference of a TFLite model on CpuRef (CPU-only) and CpuAcc (using NEON).&lt;/P&gt;&lt;P&gt;Using the STATIC ArmNN libs (*.a), the CPURef is working but not the CpuAcc which generates a segmentation fault at runtime.&lt;/P&gt;&lt;P&gt;I have tried, in the gcc command line the "whole-archive" option, in order to include all the libs, but this is not sufficient and provide a huge binary size.&lt;/P&gt;&lt;P&gt;Since the ArmNN framework does not generates any specific warnings at runtime (even if logging is activated) I am stucked.&lt;/P&gt;&lt;P&gt;## Does someone have an idea ?&lt;/P&gt;&lt;P&gt;For information, I have measured an improvement of ~5% by using the CpuRef with STATIC ArmNN built compared with SHARED (default build option in tutorials).&lt;/P&gt;&lt;P&gt;Kind regards,&lt;/P&gt;&lt;P&gt;Nicolas&lt;/P&gt;</description>
      <pubDate>Thu, 19 Aug 2021 09:57:58 GMT</pubDate>
      <guid>https://community.nxp.com/t5/eIQ-Machine-Learning-Software/ArmNN-on-IMX6-CORTEXA9-CpuAcc-causes-segmentation-faults-when/m-p/1326190#M400</guid>
      <dc:creator>nvassal</dc:creator>
      <dc:date>2021-08-19T09:57:58Z</dc:date>
    </item>
  </channel>
</rss>

