<?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 Truncated data with dual ISP capturing in i.MX Processors</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1388934#M184666</link>
    <description>&lt;P&gt;Hi Community,&lt;/P&gt;&lt;P&gt;We are testing the capturing of dual camera sensors behind ISP blocks in iMX8M Plus with &lt;STRONG&gt;video_test&lt;/STRONG&gt; utility (included in &lt;STRONG&gt;isp-imx&lt;/STRONG&gt; package).&lt;/P&gt;&lt;P&gt;Our sensors' resolution is 1440x1080 with RAW10 Bayer output and our custom board disposes 2GB of RAM.&amp;nbsp;&lt;/P&gt;&lt;P&gt;We verified each pipeline (sensorX-&amp;gt;MIPI-CSI2x-&amp;gt;ISPx-&amp;gt;/dev/videoX) with test pattern and each works well, &lt;STRONG&gt;video_test&lt;/STRONG&gt; could capture valid patterns.&lt;/P&gt;&lt;P&gt;We then launched dual capturing with &lt;STRONG&gt;video_test&lt;/STRONG&gt; in separate folder (to avoid that rawdump files with same name being overridden) and observed that :&lt;/P&gt;&lt;P&gt;1. If we launch both captures at nearly &lt;STRONG&gt;same&lt;/STRONG&gt; time, there's only one that is able to dump the files out, other does NOT,&lt;/P&gt;&lt;P&gt;2. If we launch 1st capture then wait until it really captures (with rawdump_1440x1080_xy.pgm), then launch the 2nd capture (also with rawdump_1440x1080_xy.pgm in different folder) then 1st capture data starts to be truncated/shifted from 10 bit to 8 bit :&lt;/P&gt;&lt;P&gt;Valid data of 2nd capture :&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="valid-data(1).png" style="width: 859px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/165556i36D8C36DF8FE481F/image-size/large?v=v2&amp;amp;px=999" role="button" title="valid-data(1).png" alt="valid-data(1).png" /&gt;&lt;/span&gt;&lt;BR /&gt;&lt;BR /&gt;Invalid data of 1st capture since the moment 2nd capture is launched (00 columns should &amp;lt;= 03, not always 00):&lt;BR /&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Invalid-data(1).png" style="width: 868px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/165557i2263C41EFC15D1BA/image-size/large?v=v2&amp;amp;px=999" role="button" title="Invalid-data(1).png" alt="Invalid-data(1).png" /&gt;&lt;/span&gt;&lt;BR /&gt;Therefore, I would like to know if there's any limitation of ISPs and/or of &lt;STRONG&gt;video_test&lt;/STRONG&gt; utility with dual capturing regarding our resolution toward the size of DDR in our custom board and/or the default CMA memory in the system, please ?&lt;/P&gt;&lt;P&gt;We are using BSP-5.4.70-2.3.2.&lt;/P&gt;&lt;P&gt;Thanks in advance and best regards,&lt;BR /&gt;Khang&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Fri, 17 Dec 2021 17:21:59 GMT</pubDate>
    <dc:creator>khang_letruong</dc:creator>
    <dc:date>2021-12-17T17:21:59Z</dc:date>
    <item>
      <title>Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1388934#M184666</link>
      <description>&lt;P&gt;Hi Community,&lt;/P&gt;&lt;P&gt;We are testing the capturing of dual camera sensors behind ISP blocks in iMX8M Plus with &lt;STRONG&gt;video_test&lt;/STRONG&gt; utility (included in &lt;STRONG&gt;isp-imx&lt;/STRONG&gt; package).&lt;/P&gt;&lt;P&gt;Our sensors' resolution is 1440x1080 with RAW10 Bayer output and our custom board disposes 2GB of RAM.&amp;nbsp;&lt;/P&gt;&lt;P&gt;We verified each pipeline (sensorX-&amp;gt;MIPI-CSI2x-&amp;gt;ISPx-&amp;gt;/dev/videoX) with test pattern and each works well, &lt;STRONG&gt;video_test&lt;/STRONG&gt; could capture valid patterns.&lt;/P&gt;&lt;P&gt;We then launched dual capturing with &lt;STRONG&gt;video_test&lt;/STRONG&gt; in separate folder (to avoid that rawdump files with same name being overridden) and observed that :&lt;/P&gt;&lt;P&gt;1. If we launch both captures at nearly &lt;STRONG&gt;same&lt;/STRONG&gt; time, there's only one that is able to dump the files out, other does NOT,&lt;/P&gt;&lt;P&gt;2. If we launch 1st capture then wait until it really captures (with rawdump_1440x1080_xy.pgm), then launch the 2nd capture (also with rawdump_1440x1080_xy.pgm in different folder) then 1st capture data starts to be truncated/shifted from 10 bit to 8 bit :&lt;/P&gt;&lt;P&gt;Valid data of 2nd capture :&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="valid-data(1).png" style="width: 859px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/165556i36D8C36DF8FE481F/image-size/large?v=v2&amp;amp;px=999" role="button" title="valid-data(1).png" alt="valid-data(1).png" /&gt;&lt;/span&gt;&lt;BR /&gt;&lt;BR /&gt;Invalid data of 1st capture since the moment 2nd capture is launched (00 columns should &amp;lt;= 03, not always 00):&lt;BR /&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Invalid-data(1).png" style="width: 868px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/165557i2263C41EFC15D1BA/image-size/large?v=v2&amp;amp;px=999" role="button" title="Invalid-data(1).png" alt="Invalid-data(1).png" /&gt;&lt;/span&gt;&lt;BR /&gt;Therefore, I would like to know if there's any limitation of ISPs and/or of &lt;STRONG&gt;video_test&lt;/STRONG&gt; utility with dual capturing regarding our resolution toward the size of DDR in our custom board and/or the default CMA memory in the system, please ?&lt;/P&gt;&lt;P&gt;We are using BSP-5.4.70-2.3.2.&lt;/P&gt;&lt;P&gt;Thanks in advance and best regards,&lt;BR /&gt;Khang&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Fri, 17 Dec 2021 17:21:59 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1388934#M184666</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2021-12-17T17:21:59Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1391893#M185006</link>
      <description>&lt;P&gt;Hi again,&lt;/P&gt;&lt;P&gt;To update the situation, I did 2 tests with 2 physical displays :&lt;/P&gt;&lt;P&gt;1. &lt;SPAN&gt;In the 1st test, I launched &lt;EM&gt;gst-launch-1.0 -v v4l2src device=/dev/video0 ! waylandsink&lt;/EM&gt; for &lt;STRONG&gt;cam0&lt;/STRONG&gt; in smaller display first then &lt;EM&gt;gst-launch-1.0 -v v4l2src device=/dev/video1 ! waylandsink&lt;/EM&gt; for &lt;STRONG&gt;cam1&lt;/STRONG&gt; in bigger display later : &lt;A href="https://drive.google.com/file/d/1MaGugWX7-DRBvFiecD-BHZTlmMmd6C42/view?usp=sharing" target="_blank" rel="noopener nofollow noreferrer"&gt;https://drive.google.com/file/d/1MaGugWX7-DRBvFiecD-BHZTlmMmd6C42/view?usp=sharing&lt;/A&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;2. In the 2nd test, I changed the camera streaming's order by launching &lt;EM&gt;gst-launch-1.0 -v v4l2src device=/dev/video1 ! waylandsink&lt;/EM&gt; for &lt;STRONG&gt;cam1&lt;/STRONG&gt; in bigger display first then&lt;EM&gt; gst-launch-1.0 -v v4l2src device=/dev/video0 ! waylandsink&lt;/EM&gt; for &lt;STRONG&gt;cam0&lt;/STRONG&gt; in smaller display later : &lt;A href="https://drive.google.com/file/d/1ITv7wQPzavTaCC9Lphod_oBugAWLEQp3/view?usp=sharing" target="_blank" rel="noopener nofollow noreferrer"&gt;https://drive.google.com/file/d/1ITv7wQPzavTaCC9Lphod_oBugAWLEQp3/view?usp=sharing&lt;/A&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;My observation is that the later GST launching/streaming impacts the former one which was initially good (i.e. in the videos the later makes the former turn &lt;STRONG&gt;blue&lt;/STRONG&gt; or &lt;STRONG&gt;darker&lt;/STRONG&gt; visually) and this seems to be an issue of ISP.&lt;/P&gt;&lt;P&gt;&lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/39586"&gt;@joanxie&lt;/a&gt;, &lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/34846"&gt;@Bio_TICFSL&lt;/a&gt;&amp;nbsp; : Would this be a known issue and if YES, has there's been any solution in more recent release(s), please ?&lt;/P&gt;&lt;P&gt;Best regards,&lt;BR /&gt;Khang.&lt;/P&gt;</description>
      <pubDate>Sat, 25 Dec 2021 07:23:34 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1391893#M185006</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2021-12-25T07:23:34Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392254#M185066</link>
      <description>&lt;P&gt;&lt;SPAN&gt;ISP is limited to 2 camera with 1080p 60 using MIPI CSI to ISP, but your camera resolution is 1440x1080, so this is fine, as I known, we tested dual balser dtb already, how about output two camera to one display? any different? the two ISP uses the same dewarp, so you need check this too&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 28 Dec 2021 13:12:13 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392254#M185066</guid>
      <dc:creator>joanxie</dc:creator>
      <dc:date>2021-12-28T13:12:13Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392274#M185070</link>
      <description>&lt;P&gt;Hi &lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/39586"&gt;@joanxie&lt;/a&gt;,&lt;/P&gt;&lt;P&gt;The same phenomenon (difference in brightness/color) also happens on same display :&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="dual-cam-single-display.jpg" style="width: 999px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/166295i0C3DE0C89B45EFFA/image-size/large?v=v2&amp;amp;px=999" role="button" title="dual-cam-single-display.jpg" alt="dual-cam-single-display.jpg" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;The 2 instances of our&amp;nbsp; ISP driver share everything from XML to .json.&lt;/P&gt;&lt;P&gt;We also tested on more recent BSP (5.10.52-2.1.0) release and observed the same thing : Lately launched streaming is brighter and makes the formerly launched streaming darker. More illustration is in following links&lt;/P&gt;&lt;P&gt;1. cam0(/dev/video2) before cam1 (/dev/video3) : &lt;A href="https://drive.google.com/file/d/15bAhZcY6-YArsLZY2KlLv9BvKxarGGaw/view?usp=sharing" target="_blank" rel="nofollow noopener noreferrer"&gt;https://drive.google.com/file/d/15bAhZcY6-YArsLZY2KlLv9BvKxarGGaw/view?usp=sharing&lt;/A&gt;&lt;/P&gt;&lt;P&gt;2. cam1 (/dev/video3) before cam0 (/dev/video2) : &lt;A href="https://drive.google.com/file/d/1mluJeyFnvp8SJozEfaMsajTUi3dg6rcD/view?usp=sharing" target="_blank" rel="nofollow noopener noreferrer"&gt;https://drive.google.com/file/d/1mluJeyFnvp8SJozEfaMsajTUi3dg6rcD/view?usp=sharing&lt;/A&gt;&lt;/P&gt;&lt;P&gt;We also output same test pattern from both sensors to avoid the dependency of external condition (light, lens).&lt;/P&gt;&lt;P&gt;Best regards,&lt;/P&gt;&lt;P&gt;Khang.&lt;/P&gt;</description>
      <pubDate>Mon, 27 Dec 2021 12:54:45 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392274#M185070</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2021-12-27T12:54:45Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392292#M185072</link>
      <description>&lt;P&gt;Hi again &lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/39586"&gt;@joanxie&lt;/a&gt;,&amp;nbsp;&lt;/P&gt;&lt;P&gt;I might also have a small doubt on the behavior of gstreamer command we are using :&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;gst-launch-1.0 -v v4l2src device=/dev/videoX ! "video/x-raw,format=YUY2,width=1440,height=1080, framerate=15/1" ! queue ! imxvideoconvert_g2d ! waylandsink&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;From attached files, the difference of debug messages when calling next gst-launch on second camera (&lt;STRONG&gt;/dev/video3&lt;/STRONG&gt; mapped to &lt;STRONG&gt;imx296 2-0036&lt;/STRONG&gt;) following the previous gst-launch on first camera (&lt;STRONG&gt;/dev/video2&lt;/STRONG&gt; mapped to &lt;STRONG&gt;imx296 1-001a&lt;/STRONG&gt;) is below :&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;[  114.381215] imx296 1-001a: enter imx296_query_capability
[  114.381402] imx296 2-0036: enter imx296_query_capability
[  114.394979] enter isp_mi_stop
[  114.458285] imx296 2-0036: enter imx296_query_capability
[  114.458328] imx296 1-001a: enter imx296_query_capability
[  114.467034] imx296 2-0036: enter imx296_query_capability
[  114.485855] imx296 2-0036: enter imx296_query_supports
[  114.486714] imx296 2-0036: enter imx296_query_capability
[  114.486860] imx296 1-001a: enter imx296_query_capability
[  114.571087] imx296 2-0036: enter imx296_query_supports
[  114.571110] imx296 2-0036: enter imx296_get_clk
[  114.571120] imx296 2-0036: enter imx296_set_sensor_mode
[  114.571127] imx296 2-0036: enter imx296_get_sensor_mode
[  114.571134] imx296 2-0036: enter imx296_query_supports
[  114.571258] enter isp_mi_stop
[  114.576791] imx296 2-0036: enter imx296_set_fmt
[  114.576801] imx296 2-0036: I2C writing value 80 at address 0x3a00
[  114.577437] imx296 2-0036: I2C writing value f0 at address 0x3005
[  114.578038] imx296 2-0036: I2C writing value 7f at address 0x350b
[  114.578679] imx296 2-0036: I2C writing value 00 at address 0x300d
[  114.579281] imx296 2-0036: I2C writing value 58 at address 0x400e
[  114.579881] imx296 2-0036: I2C writing value 4c at address 0x3014
[  114.580473] imx296 2-0036: I2C writing value 04 at address 0x3015
[  114.581058] imx296 2-0036: I2C writing value 1c at address 0x4014
[  114.581642] imx296 2-0036: I2C writing value 77 at address 0x3516
[  114.582224] imx296 2-0036: I2C writing value 00 at address 0x321a
[  114.582808] imx296 2-0036: I2C writing value 02 at address 0x3226
[  114.583397] imx296 2-0036: I2C writing value f5 at address 0x3832
[  114.583986] imx296 2-0036: I2C writing value 00 at address 0x3833
[  114.584611] imx296 2-0036: I2C writing value 72 at address 0x3541
[  114.585267] imx296 2-0036: I2C writing value 2a at address 0x4041
[  114.586252] imx296 2-0036: I2C writing value a3 at address 0x3d48
[  114.587537] imx296 2-0036: I2C writing value 00 at address 0x3d49
[  114.588195] imx296 2-0036: I2C writing value 85 at address 0x3d4a
[  114.588865] imx296 2-0036: I2C writing value 00 at address 0x3d4b
[  114.589507] imx296 2-0036: I2C writing value 01 at address 0x3256
[  114.590155] imx296 2-0036: I2C writing value a3 at address 0x3758
[  114.590837] imx296 2-0036: I2C writing value 00 at address 0x3759
[  114.591503] imx296 2-0036: I2C writing value 85 at address 0x375a
[  114.592152] imx296 2-0036: I2C writing value 00 at address 0x375b
[  114.592795] imx296 2-0036: I2C writing value 00 at address 0x3165
[  114.593437] imx296 2-0036: I2C writing value 10 at address 0x3169
[  114.594045] imx296 2-0036: I2C writing value 02 at address 0x316a
[  114.594669] imx296 2-0036: I2C writing value 00 at address 0x4174
[  114.595288] imx296 2-0036: I2C writing value 08 at address 0x3079
[  114.595913] imx296 2-0036: I2C writing value 04 at address 0x3090
[  114.596544] imx296 2-0036: I2C writing value 04 at address 0x3094
[  114.597142] imx296 2-0036: I2C writing value 04 at address 0x3098
[  114.597729] imx296 2-0036: I2C writing value 04 at address 0x309e
[  114.598327] imx296 2-0036: I2C writing value 04 at address 0x30a0
[  114.598929] imx296 2-0036: I2C writing value 3c at address 0x30a1
[  114.599453] imx296 2-0036: I2C writing value f6 at address 0x38a2
[  114.600046] imx296 2-0036: I2C writing value 00 at address 0x38a3
[  114.600632] imx296 2-0036: I2C writing value 06 at address 0x40a2
[  114.601222] imx296 2-0036: I2C writing value 5f at address 0x30a4
[  114.601807] imx296 2-0036: I2C writing value 91 at address 0x30a8
[  114.602389] imx296 2-0036: I2C writing value 28 at address 0x30ac
[  114.604075] imx296 2-0036: I2C writing value 09 at address 0x30af
[  114.604671] imx296 2-0036: I2C writing value f6 at address 0x40c1
[  114.605280] imx296 2-0036: I2C writing value 0f at address 0x40c7
[  114.605904] imx296 2-0036: I2C writing value f3 at address 0x31c8
[  114.606742] imx296 2-0036: I2C writing value f4 at address 0x31d0
[  114.607388] imx296 2-0036: I2C writing value 00 at address 0x40c8
[  114.608037] imx296 2-0036: I2C writing value 00 at address 0x30df
[  114.608677] imx296 2-0036: I2C writing value 03 at address 0x3300
[  114.609402] imx296 2-0036: I2C writing value 08 at address 0x3310
[  114.610020] imx296 2-0036: I2C writing value 00 at address 0x3311
[  114.610619] imx296 2-0036: I2C writing value 00 at address 0x3312
[  114.611273] imx296 2-0036: I2C writing value 00 at address 0x3313
[  114.611939] imx296 2-0036: I2C writing value a0 at address 0x3314
[  114.612556] imx296 2-0036: I2C writing value 05 at address 0x3315
[  114.613160] imx296 2-0036: I2C writing value 40 at address 0x3316
[  114.613761] imx296 2-0036: I2C writing value 04 at address 0x3317
[  114.614362] imx296 2-0036: I2C writing value 00 at address 0x3204
[  114.614992] imx296 2-0036: I2C writing value 00 at address 0x3205
[  114.615602] imx296 2-0036: I2C writing value 08 at address 0x3212
[  114.616202] imx296 2-0036: I2C writing value 3c at address 0x3254
[  114.616801] imx296 2-0036: I2C writing value 00 at address 0x3255
[  114.617597] imx296 2-0036: I2C writing value 80 at address 0x3089
[  114.618200] imx296 2-0036: I2C writing value 0b at address 0x308a
[  114.618800] imx296 2-0036: I2C writing value 80 at address 0x308b
[  114.620092] imx296 2-0036: I2C writing value 08 at address 0x308c
[  114.620698] imx296 2-0036: I2C writing value 74 at address 0x418c
[  114.621302] imx296 2-0036: enter imx296_set_test_pattern
[  114.621306] imx296 2-0036: imx296_set_test_pattern:9 
[  114.621329] imx296 2-0036: enter imx296_set_fps( fps/1024=15 )
[  114.621333] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.621924] imx296 2-0036: I2C writing value 77 at address 0x308d
[  114.622509] imx296 2-0036: I2C writing value 0f at address 0x308e
[  114.623249] imx296 2-0036: I2C writing value 00 at address 0x308f
[  114.623857] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.642551] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.643161] imx296 2-0036: I2C writing value 94 at address 0x3010
[  114.643762] imx296 2-0036: I2C writing value 11 at address 0x3011
[  114.644360] imx296 2-0036: I2C writing value 00 at address 0x3012
[  114.644959] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.645566] imx296 2-0036: enter imx296_get_sensor_mode
[  114.645870] imx296 2-0036: enter imx296_set_exp( exp=4490 )
[  114.645877] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.646481] imx296 2-0036: I2C writing value 8a at address 0x308d
[  114.647103] imx296 2-0036: I2C writing value 11 at address 0x308e
[  114.647715] imx296 2-0036: I2C writing value 00 at address 0x308f
[  114.648316] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.648931] imx296 2-0036: gain = 228
[  114.648937] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.649620] imx296 2-0036: I2C writing value e4 at address 0x3204
[  114.650291] imx296 2-0036: I2C writing value 00 at address 0x3205
[  114.650907] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.652411] enter isp_s_comp
[  114.655357] enter isp_s_comp
[  114.658240] enter isp_s_comp
[  114.674969] enter wdr3_hw_init
[  114.678068] wdr3 res: 1440 1080 
[  114.682009] enter isp_mi_start
[  114.686187] imx296 2-0036: enter imx296_s_stream
[  114.715375] imx296 2-0036: TPG_REG 0x3238 : 0x17
[  114.720027] imx296 2-0036: I2C writing value 17 at address 0x3238
[  114.720626] imx296 2-0036: TPG_REG 0x3239 : 0x00
[  114.725273] imx296 2-0036: I2C writing value 00 at address 0x3239
[  114.725878] imx296 2-0036: TPG_REG 0x323a : 0x00
[  114.730538] imx296 2-0036: I2C writing value 00 at address 0x323a
[  114.731138] imx296 2-0036: TPG_REG 0x323c : 0x00
[  114.735773] imx296 2-0036: I2C writing value 00 at address 0x323c
[  114.736367] imx296 2-0036: TPG_REG 0x323d : 0x00
[  114.741017] imx296 2-0036: I2C writing value 00 at address 0x323d
[  114.741614] imx296 2-0036: TPG_REG 0x323e : 0x00
[  114.746632] imx296 2-0036: I2C writing value 00 at address 0x323e
[  114.747248] imx296 2-0036: TPG_REG 0x323f : 0x00
[  114.751906] imx296 2-0036: I2C writing value 00 at address 0x323f
[  114.752513] imx296 2-0036: TPG_REG 0x3240 : 0x01
[  114.757210] imx296 2-0036: I2C writing value 01 at address 0x3240
[  114.757825] imx296 2-0036: TPG_REG 0x3241 : 0x01
[  114.762489] imx296 2-0036: I2C writing value 01 at address 0x3241
[  114.763101] imx296 2-0036: TPG_REG 0x3244 : 0x00
[  114.767778] imx296 2-0036: I2C writing value 00 at address 0x3244
[  114.768391] imx296 2-0036: TPG_REG 0x3245 : 0x00
[  114.773075] imx296 2-0036: I2C writing value 00 at address 0x3245
[  114.773690] imx296 2-0036: TPG_REG 0x3246 : 0x00
[  114.778393] imx296 2-0036: I2C writing value 00 at address 0x3246
[  114.778990] imx296 2-0036: TPG_REG 0x3247 : 0x00
[  114.783639] imx296 2-0036: I2C writing value 00 at address 0x3247
[  114.784282] imx296 2-0036: TPG_REG 0x3249 : 0x4a
[  114.789011] imx296 2-0036: I2C writing value 4a at address 0x3249
[  114.789680] imx296 2-0036: TPG_REG 0x3254 : 0x00
[  114.794352] imx296 2-0036: I2C writing value 00 at address 0x3254
[  114.795000] imx296 2-0036: TPG_REG 0x3255 : 0x00
[  114.799648] imx296 2-0036: I2C writing value 00 at address 0x3255
[  114.800249] imx296 2-0036: TPG_REG 0x3022 : 0xf0
[  114.804887] imx296 2-0036: I2C writing value f0 at address 0x3022
[  114.805520] imx296 2-0036: imx296_set_ctrl : ctrl = {.id = 0x00980913, .val = 0x000000c8}
[  114.805530] imx296 2-0036: gain = 200
[  114.805534] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.806161] imx296 2-0036: I2C writing value c8 at address 0x3204
[  114.807005] imx296 2-0036: I2C writing value 00 at address 0x3205
[  114.807635] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.808264] imx296 2-0036: imx296_set_ctrl : ctrl = {.id = 0x00980911, .val = 0x000003e8}
[  114.808273] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.808900] imx296 2-0036: I2C writing value e8 at address 0x308d
[  114.809510] imx296 2-0036: I2C writing value 03 at address 0x308e
[  114.810134] imx296 2-0036: I2C writing value 00 at address 0x308f
[  114.811092] imx296 2-0036: I2C writing value 00 at address 0x3008
[  114.811715] imx296 2-0036: I2C writing value 00 at address 0x3000
[  114.850559] imx296 2-0036: I2C writing value 00 at address 0x300a
[  114.884800] imx296 2-0036: gain = 130
[  114.884810] imx296 2-0036: I2C writing value 01 at address 0x3008
[  114.885380] imx296 2-0036: I2C writing value 82 at address 0x3204
[  114.885977] imx296 2-0036: I2C writing value 00 at address 0x3205
[  114.886769] imx296 2-0036: I2C writing value 00 at address 0x3008
[  115.078787] imx296 2-0036: gain = 73
[  115.078798] imx296 2-0036: I2C writing value 01 at address 0x3008
[  115.079483] imx296 2-0036: I2C writing value 49 at address 0x3204
[  115.080156] imx296 2-0036: I2C writing value 00 at address 0x3205
[  115.080839] imx296 2-0036: I2C writing value 00 at address 0x3008
[  115.285077] imx296 2-0036: gain = 41
[  115.285090] imx296 2-0036: I2C writing value 01 at address 0x3008
[  115.285710] imx296 2-0036: I2C writing value 29 at address 0x3204
[  115.286327] imx296 2-0036: I2C writing value 00 at address 0x3205
[  115.286914] imx296 2-0036: I2C writing value 00 at address 0x3008
[  115.472066] imx296 2-0036: gain = 23
[  115.472085] imx296 2-0036: I2C writing value 01 at address 0x3008
[  115.472806] imx296 2-0036: I2C writing value 17 at address 0x3204
[  115.473491] imx296 2-0036: I2C writing value 00 at address 0x3205
[  115.474159] imx296 2-0036: I2C writing value 00 at address 0x3008
[  115.671961] imx296 2-0036: gain = 12
[  115.671991] imx296 2-0036: I2C writing value 01 at address 0x3008
[  115.672642] imx296 2-0036: I2C writing value 0c at address 0x3204
[  115.673262] imx296 2-0036: I2C writing value 00 at address 0x3205
[  115.673935] imx296 2-0036: I2C writing value 00 at address 0x3008
[  115.882370] imx296 2-0036: gain = 6
[  115.882381] imx296 2-0036: I2C writing value 01 at address 0x3008
[  115.884094] imx296 2-0036: I2C writing value 06 at address 0x3204
[  115.884824] imx296 2-0036: I2C writing value 00 at address 0x3205
[  115.885430] imx296 2-0036: I2C writing value 00 at address 0x3008
[  116.082000] imx296 2-0036: gain = 3
[  116.082013] imx296 2-0036: I2C writing value 01 at address 0x3008
[  116.083649] imx296 2-0036: I2C writing value 03 at address 0x3204
[  116.084262] imx296 2-0036: I2C writing value 00 at address 0x3205
[  116.084920] imx296 2-0036: I2C writing value 00 at address 0x3008
[  116.482268] imx296 2-0036: gain = 1
[  116.482281] imx296 2-0036: I2C writing value 01 at address 0x3008
[  116.483943] imx296 2-0036: I2C writing value 01 at address 0x3204
[  116.484550] imx296 2-0036: I2C writing value 00 at address 0x3205
[  116.485234] imx296 2-0036: I2C writing value 00 at address 0x3008
[  116.690356] imx296 2-0036: gain = 0
[  116.690370] imx296 2-0036: I2C writing value 01 at address 0x3008
[  116.692233] imx296 2-0036: I2C writing value 00 at address 0x3204
[  116.692835] imx296 2-0036: I2C writing value 00 at address 0x3205
[  116.693418] imx296 2-0036: I2C writing value 00 at address 0x3008&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;From the very beginning of second gst-launching of /dev/video3 (i.e. 2-0036) , it seems that /dev/video2 (i.e. 1-001a) is also queried :&lt;/P&gt;&lt;PRE&gt;[  114.381215] imx296 1-001a: enter imx296_query_capability
...
[  114.458328] imx296 1-001a: enter imx296_query_capability
...
[  114.486860] imx296 1-001a: enter imx296_query_capability&lt;/PRE&gt;&lt;P&gt;Then in the meantime there appears to be 2 &lt;EM&gt;isp_mi_stop&lt;/EM&gt; calls :&lt;/P&gt;&lt;PRE&gt;[  114.394979] enter isp_mi_stop
...
[  114.571258] enter isp_mi_stop&lt;/PRE&gt;&lt;P&gt;But only one &lt;EM&gt;isp_mi_start&lt;/EM&gt; call later :&lt;/P&gt;&lt;PRE&gt;[  114.652411] enter isp_s_comp
[  114.655357] enter isp_s_comp
[  114.658240] enter isp_s_comp
[  114.674969] enter wdr3_hw_init
[  114.678068] wdr3 res: 1440 1080 
[  114.682009] enter isp_mi_start&lt;/PRE&gt;&lt;P&gt;I'm not sure which&amp;nbsp;&lt;EM&gt;isp_mi_stop&lt;/EM&gt; or&amp;nbsp;&lt;EM&gt;isp_mi_start&lt;/EM&gt; belong to which ISP controller but it does not seem logical that&amp;nbsp; there are 2 stop and 1 start. Possibly one ISP is in stopping state. What do you think?&lt;/P&gt;&lt;P&gt;Best regards,&lt;BR /&gt;K.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 27 Dec 2021 14:18:05 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392292#M185072</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2021-12-27T14:18:05Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392635#M185111</link>
      <description>&lt;P&gt;pls see the private message&lt;/P&gt;</description>
      <pubDate>Tue, 28 Dec 2021 09:56:06 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392635#M185111</guid>
      <dc:creator>joanxie</dc:creator>
      <dc:date>2021-12-28T09:56:06Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392669#M185116</link>
      <description>&lt;P&gt;Hi &lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/39586"&gt;@joanxie&lt;/a&gt; ,&lt;/P&gt;&lt;P&gt;Thanks for your kind support. Really appreciate.&lt;/P&gt;&lt;P&gt;Best regards,&lt;BR /&gt;K.&lt;/P&gt;</description>
      <pubDate>Tue, 28 Dec 2021 12:05:25 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1392669#M185116</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2021-12-28T12:05:25Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1713891#M211734</link>
      <description>&lt;P&gt;&lt;SPAN&gt;Hi,&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;"My observation is that the later GST launching/streaming impacts the former one which was initially good (i.e. in the videos the later makes the former turn&amp;nbsp;&lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;blue&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt;&amp;nbsp;or&amp;nbsp;&lt;/SPAN&gt;&lt;STRONG&gt;&lt;SPAN&gt;darker&lt;/SPAN&gt;&lt;/STRONG&gt;&lt;SPAN&gt;&amp;nbsp;visually) and this seems to be an issue of ISP."&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Did you find a solution to this problem? I have the same problem with a dual ar1335 camera setup on the DART-MX8M-PLUS device. Running an OS build with Yocto and using layers from isp-imx and isp-vvcam.&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Wed, 30 Aug 2023 08:50:06 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1713891#M211734</guid>
      <dc:creator>martinty</dc:creator>
      <dc:date>2023-08-30T08:50:06Z</dc:date>
    </item>
    <item>
      <title>Re: Truncated data with dual ISP capturing</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1724226#M212684</link>
      <description>&lt;P&gt;Hi &lt;a href="https://community.nxp.com/t5/user/viewprofilepage/user-id/222221"&gt;@martinty&lt;/a&gt; ,&lt;/P&gt;&lt;P&gt;As far as I remember, the proper patch or solution for the issue has just been integrated since the BSP release lf-5.15.32-.2.0.0 :&amp;nbsp;&lt;/P&gt;&lt;PRE&gt;static void mipi_csis_imx8mp_phy_reset(struct csi_state *state)
{
	int ret = 0;
	struct v4l2_subdev *sen_sd;

	struct v4l2_subdev_mbus_code_enum code = {
		.which = V4L2_SUBDEV_FORMAT_ACTIVE,
	};

	mipi_csis_imx8mn_phy_reset(state);
	sen_sd = csis_get_remote_subdev(state, __func__);
	if (!sen_sd)
		goto csi_phy_initial_cfg;

	ret = v4l2_subdev_call(sen_sd, pad, enum_mbus_code, NULL, &amp;amp;code);
	if (ret &amp;lt; 0 &amp;amp;&amp;amp; ret != -ENOIOCTLCMD) {
		v4l2_err(&amp;amp;state-&amp;gt;sd, "enum_mbus_code error !!!\n");
		return;
	} else if (ret == -ENOIOCTLCMD)
		goto csi_phy_initial_cfg;

	/* temporary place */
	if (state-&amp;gt;mix_gpr) {
		if ((code.code == MEDIA_BUS_FMT_SRGGB8_1X8) ||
				(code.code == MEDIA_BUS_FMT_SGRBG8_1X8) ||
				(code.code == MEDIA_BUS_FMT_SGBRG8_1X8) ||
				(code.code == MEDIA_BUS_FMT_SBGGR8_1X8)) {
			mipi_csis_imx8mp_dewarp_ctl_data_type(state,
					 ISP_DEWARP_CTRL_DATA_TYPE_RAW8);
			v4l2_dbg(1, debug, &amp;amp;state-&amp;gt;sd,
					"%s: bus fmt is 8 bit!\n", __func__);
		} else if ((code.code == MEDIA_BUS_FMT_SRGGB10_1X10) ||
				(code.code == MEDIA_BUS_FMT_SGRBG10_1X10) ||
				(code.code == MEDIA_BUS_FMT_SGBRG10_1X10) ||
				(code.code == MEDIA_BUS_FMT_SBGGR10_1X10)) {
			mipi_csis_imx8mp_dewarp_ctl_data_type(state,
					ISP_DEWARP_CTRL_DATA_TYPE_RAW10);
			v4l2_dbg(1, debug, &amp;amp;state-&amp;gt;sd,
					"%s: bus fmt is 10 bit !\n", __func__);
		} else {
			mipi_csis_imx8mp_dewarp_ctl_data_type(state,
					ISP_DEWARP_CTRL_DATA_TYPE_RAW12);
			v4l2_dbg(1, debug, &amp;amp;state-&amp;gt;sd,
					"%s: bus fmt is 12 bit !\n", __func__);
		}
		goto write_regmap;
	}

csi_phy_initial_cfg:
	mipi_csis_imx8mp_dewarp_ctl_data_type(state,
			ISP_DEWARP_CTRL_DATA_TYPE_RAW12);
	v4l2_dbg(1, debug, &amp;amp;state-&amp;gt;sd,
			"%s: bus fmt is 12 bit !\n", __func__);

write_regmap:
	state-&amp;gt;val |= ISP_DEWARP_CTRL_ID_MODE(ISP_DEWARP_CTRL_ID_MODE_012);
	mipi_csis_imx8mp_dewarp_ctl_left_just_mode(state);
	regmap_update_bits(state-&amp;gt;mix_gpr, ISP_DEWARP_CTRL,
			state-&amp;gt;val, state-&amp;gt;val);

	return;
}&lt;/PRE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Best Regards,&lt;/P&gt;&lt;P&gt;Khang&lt;/P&gt;</description>
      <pubDate>Sun, 17 Sep 2023 04:31:21 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Truncated-data-with-dual-ISP-capturing/m-p/1724226#M212684</guid>
      <dc:creator>khang_letruong</dc:creator>
      <dc:date>2023-09-17T04:31:21Z</dc:date>
    </item>
  </channel>
</rss>

