<?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 BayTrail-I E3845 MIPI CSI2 interface receiving latency in Embedded Intel Atom® Processors</title>
    <link>https://community.intel.com/t5/Embedded-Intel-Atom-Processors/BayTrail-I-E3845-MIPI-CSI2-interface-receiving-latency/m-p/215369#M1063</link>
    <description>&lt;P&gt;Hi, &lt;/P&gt;&lt;P&gt;I'm Choi. &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I use E3845's MIPI interface for camera interface.&lt;/P&gt;&lt;P&gt;I measured the receiving latency of E3845 MIPI Interface.&lt;/P&gt;&lt;P&gt;The measure step is as follows,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;1. With FPGA via D-PHY IC, I send image frame(640 x 480 x 16bit, with 2 lane D-Phy)&lt;/P&gt;&lt;P&gt;2. I set a GPIO pin high at FPGA side just after completion of each image frame transmission.&lt;/P&gt;&lt;P&gt;    (The GPIO pin return to low state after tens of micro seconds.)&lt;/P&gt;&lt;P&gt;3. CPU's ISP receives the frame successfully and saves the frame on DDR memory.&lt;/P&gt;&lt;P&gt;4. CPU set a GPIO pin high, (One of CPU's GPIO pin)&lt;/P&gt;&lt;P&gt;5. I check the interval FPGA GPIO pin's rising point to CPU GPIO pin's rising point.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;The received frame was perfect. There was no Checksum error or sequence Number error.&lt;/P&gt;&lt;P&gt;And with viewer program, the received image look same as the source image.&lt;/P&gt;&lt;P&gt;D-PHY IC has just small size buffer.&lt;/P&gt;&lt;P&gt;CPU's application was developed with WindRiver Linux 7.x.&lt;/P&gt;&lt;P&gt;I know WR Linux didn't support pre-emptive RT for MIPI ISP.&lt;/P&gt;&lt;P&gt;So, we raised the task's priority to the highest level as high as possible to reduce the latency.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Now, my question is this,&lt;/P&gt;&lt;P&gt;I expected the the latency is below 1 msec, &lt;/P&gt;&lt;P&gt;But, the latency was over 3 msec.&lt;/P&gt;&lt;P&gt;The latency is disappointingly long.&lt;/P&gt;&lt;P&gt;Why the latency is so long ?&lt;/P&gt;&lt;P&gt;Does the ISP use PCIe bus internally ?&lt;/P&gt;&lt;P&gt;Or there are 2 step buffering ?&lt;/P&gt;&lt;P&gt;Isn't there any know-how to reduce the latency dramatically under 1 msec ?&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Regard,&lt;/P&gt;&lt;P&gt;Choi&lt;/P&gt;</description>
    <pubDate>Fri, 07 Apr 2017 01:07:04 GMT</pubDate>
    <dc:creator>SChoi8</dc:creator>
    <dc:date>2017-04-07T01:07:04Z</dc:date>
    <item>
      <title>BayTrail-I E3845 MIPI CSI2 interface receiving latency</title>
      <link>https://community.intel.com/t5/Embedded-Intel-Atom-Processors/BayTrail-I-E3845-MIPI-CSI2-interface-receiving-latency/m-p/215369#M1063</link>
      <description>&lt;P&gt;Hi, &lt;/P&gt;&lt;P&gt;I'm Choi. &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I use E3845's MIPI interface for camera interface.&lt;/P&gt;&lt;P&gt;I measured the receiving latency of E3845 MIPI Interface.&lt;/P&gt;&lt;P&gt;The measure step is as follows,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;1. With FPGA via D-PHY IC, I send image frame(640 x 480 x 16bit, with 2 lane D-Phy)&lt;/P&gt;&lt;P&gt;2. I set a GPIO pin high at FPGA side just after completion of each image frame transmission.&lt;/P&gt;&lt;P&gt;    (The GPIO pin return to low state after tens of micro seconds.)&lt;/P&gt;&lt;P&gt;3. CPU's ISP receives the frame successfully and saves the frame on DDR memory.&lt;/P&gt;&lt;P&gt;4. CPU set a GPIO pin high, (One of CPU's GPIO pin)&lt;/P&gt;&lt;P&gt;5. I check the interval FPGA GPIO pin's rising point to CPU GPIO pin's rising point.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;The received frame was perfect. There was no Checksum error or sequence Number error.&lt;/P&gt;&lt;P&gt;And with viewer program, the received image look same as the source image.&lt;/P&gt;&lt;P&gt;D-PHY IC has just small size buffer.&lt;/P&gt;&lt;P&gt;CPU's application was developed with WindRiver Linux 7.x.&lt;/P&gt;&lt;P&gt;I know WR Linux didn't support pre-emptive RT for MIPI ISP.&lt;/P&gt;&lt;P&gt;So, we raised the task's priority to the highest level as high as possible to reduce the latency.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Now, my question is this,&lt;/P&gt;&lt;P&gt;I expected the the latency is below 1 msec, &lt;/P&gt;&lt;P&gt;But, the latency was over 3 msec.&lt;/P&gt;&lt;P&gt;The latency is disappointingly long.&lt;/P&gt;&lt;P&gt;Why the latency is so long ?&lt;/P&gt;&lt;P&gt;Does the ISP use PCIe bus internally ?&lt;/P&gt;&lt;P&gt;Or there are 2 step buffering ?&lt;/P&gt;&lt;P&gt;Isn't there any know-how to reduce the latency dramatically under 1 msec ?&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Regard,&lt;/P&gt;&lt;P&gt;Choi&lt;/P&gt;</description>
      <pubDate>Fri, 07 Apr 2017 01:07:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Embedded-Intel-Atom-Processors/BayTrail-I-E3845-MIPI-CSI2-interface-receiving-latency/m-p/215369#M1063</guid>
      <dc:creator>SChoi8</dc:creator>
      <dc:date>2017-04-07T01:07:04Z</dc:date>
    </item>
    <item>
      <title>Re: BayTrail-I E3845 MIPI CSI2 interface receiving latency</title>
      <link>https://community.intel.com/t5/Embedded-Intel-Atom-Processors/BayTrail-I-E3845-MIPI-CSI2-interface-receiving-latency/m-p/215370#M1064</link>
      <description>&lt;P&gt;Hello, Sung-hyuk_Choi:&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Thank you for contacting Intel Embedded Community.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;In order to better understand this situation, we would like to address the following questions:&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Could please try to reproduce this situation using any of following the Operating System (OS) and let us know the results?&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Wind River VxWorks*&lt;/P&gt;&lt;P&gt;Microsoft Windows* 8&lt;/P&gt;&lt;P&gt;Windows Embedded Standard 8 (non-connected standby)&lt;/P&gt;&lt;P&gt;Microsoft Windows 7&lt;/P&gt;&lt;P&gt;Windows Embedded Standard 7&lt;/P&gt;&lt;P&gt;Linux* Tizen (select in-vehicle infotainment (IVI) customers only)&lt;/P&gt;&lt;P&gt;Linux based on Yocto Project*&lt;/P&gt;&lt;P&gt;Linux based on Fedora*&lt;/P&gt;&lt;P&gt;Microsoft Embedded Compact 7 and 2013&lt;/P&gt;&lt;P&gt;Android* (JB MR2 4.3)&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Could you please clarify if it is a third party design or your design? In case that it is a third party one, please give us all the information related to it. On the other hand, in case that it is your design, could you please let us know in a detailed way how did you implement it? &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Please let us know this information to have a better idea of this situation. &lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Best regards,&lt;/P&gt;&lt;P&gt;Carlos_A.&lt;/P&gt;</description>
      <pubDate>Fri, 07 Apr 2017 15:01:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Embedded-Intel-Atom-Processors/BayTrail-I-E3845-MIPI-CSI2-interface-receiving-latency/m-p/215370#M1064</guid>
      <dc:creator>CarlosAM_INTEL</dc:creator>
      <dc:date>2017-04-07T15:01:53Z</dc:date>
    </item>
  </channel>
</rss>

