<?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>topic imx93 dm-crypt crash with RT-patch in i.MX Processors</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2004344#M231360</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;We are trying to use dm-crypt with OP-TEE based encryption as described in imx Linux User's Guide chapter 10.5.5. We are using BSP 6.6.36 as a base. This seems to be working fine,&amp;nbsp; however we also need RT functionality in Linux so we have applied the RT patch manually.&lt;/P&gt;&lt;P&gt;When running a dm-crypt I/O test with RT patch + CONFIG_PREEMPT_RT=y in the kernel, we get strange behavior with random processes crashing, ssh sessions being terminated etc. This makes us suspect some memory corruption occuring in the kernel.&lt;/P&gt;&lt;P&gt;Command used for testing I/O and which triggers the error:&lt;/P&gt;&lt;P&gt;fio --filename=/mnt/encrypted/fio.bin --size=128MB --rw=rw --bs=8k --ioengine=sync --iodepth=1 --runtime=120 --numjobs=1 --time_based --group_reporting --name=throughput-test-job --eta-newline=1&lt;/P&gt;&lt;P&gt;Usually fio segfaults in addition to other errors occuring or not. It can take a while to trigger the errors.&lt;/P&gt;&lt;P&gt;We have also tried the recently released:&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;lf-6.6.52 with manual RT patch&lt;/LI&gt;&lt;LI&gt;real-time-edge-linux Real-Time-Edge-v3.0-202412 which already has the RT patch&lt;/LI&gt;&lt;/UL&gt;&lt;P&gt;All kernels give the same result. We have not been able to reproduce the error when using CONFIG_PREEMPT=y instead of CONFIG_PREEMPT_RT=y.&lt;/P&gt;&lt;P&gt;We have also tried non OP-TEE based encryption with dm-crypt (capi:cbc-aes-ce-plain algorithm) as a comparision and not gotten any errors with RT preemption. However, we need the hidden key support.&lt;/P&gt;&lt;P&gt;Thankful for any help.&lt;/P&gt;</description>
    <pubDate>Fri, 29 Nov 2024 12:52:24 GMT</pubDate>
    <dc:creator>electro1</dc:creator>
    <dc:date>2024-11-29T12:52:24Z</dc:date>
    <item>
      <title>imx93 dm-crypt crash with RT-patch</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2004344#M231360</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;We are trying to use dm-crypt with OP-TEE based encryption as described in imx Linux User's Guide chapter 10.5.5. We are using BSP 6.6.36 as a base. This seems to be working fine,&amp;nbsp; however we also need RT functionality in Linux so we have applied the RT patch manually.&lt;/P&gt;&lt;P&gt;When running a dm-crypt I/O test with RT patch + CONFIG_PREEMPT_RT=y in the kernel, we get strange behavior with random processes crashing, ssh sessions being terminated etc. This makes us suspect some memory corruption occuring in the kernel.&lt;/P&gt;&lt;P&gt;Command used for testing I/O and which triggers the error:&lt;/P&gt;&lt;P&gt;fio --filename=/mnt/encrypted/fio.bin --size=128MB --rw=rw --bs=8k --ioengine=sync --iodepth=1 --runtime=120 --numjobs=1 --time_based --group_reporting --name=throughput-test-job --eta-newline=1&lt;/P&gt;&lt;P&gt;Usually fio segfaults in addition to other errors occuring or not. It can take a while to trigger the errors.&lt;/P&gt;&lt;P&gt;We have also tried the recently released:&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;lf-6.6.52 with manual RT patch&lt;/LI&gt;&lt;LI&gt;real-time-edge-linux Real-Time-Edge-v3.0-202412 which already has the RT patch&lt;/LI&gt;&lt;/UL&gt;&lt;P&gt;All kernels give the same result. We have not been able to reproduce the error when using CONFIG_PREEMPT=y instead of CONFIG_PREEMPT_RT=y.&lt;/P&gt;&lt;P&gt;We have also tried non OP-TEE based encryption with dm-crypt (capi:cbc-aes-ce-plain algorithm) as a comparision and not gotten any errors with RT preemption. However, we need the hidden key support.&lt;/P&gt;&lt;P&gt;Thankful for any help.&lt;/P&gt;</description>
      <pubDate>Fri, 29 Nov 2024 12:52:24 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2004344#M231360</guid>
      <dc:creator>electro1</dc:creator>
      <dc:date>2024-11-29T12:52:24Z</dc:date>
    </item>
    <item>
      <title>Re: imx93 dm-crypt crash with RT-patch</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2005765#M231461</link>
      <description>&lt;P&gt;Hi,&amp;nbsp;&lt;/P&gt;
&lt;P&gt;We're working on the issue, please wait.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Regards&lt;/P&gt;
&lt;P&gt;Harvey&lt;/P&gt;</description>
      <pubDate>Tue, 03 Dec 2024 07:10:49 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2005765#M231461</guid>
      <dc:creator>Harvey021</dc:creator>
      <dc:date>2024-12-03T07:10:49Z</dc:date>
    </item>
    <item>
      <title>Re: imx93 dm-crypt crash with RT-patch</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2011578#M231831</link>
      <description>&lt;P&gt;hi&amp;nbsp;&lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/241121"&gt;@electro1&lt;/a&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Will reply back to you in other ticket created by you -&amp;nbsp;&lt;A href="https://community.nxp.com/t5/i-MX-Processors/imx93-cbc-aes-tee-immediate-crash/m-p/2005146/highlight/false#M231418" target="_blank"&gt;imx93 cbc-aes-tee immediate crash - NXP Community&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Regards&lt;/P&gt;
&lt;P&gt;Harvey&lt;/P&gt;</description>
      <pubDate>Wed, 11 Dec 2024 05:24:51 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/imx93-dm-crypt-crash-with-RT-patch/m-p/2011578#M231831</guid>
      <dc:creator>Harvey021</dc:creator>
      <dc:date>2024-12-11T05:24:51Z</dc:date>
    </item>
  </channel>
</rss>

