<?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 Low MMC Throughput on i.MX28 in i.MX Processors</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642970#M98149</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I have a custom i.MX28 board with a SDIO wifi chip attached. Using iperf I noticed a much lower data throughput on Linux mainline 4.7 (5Mbit/sec) compared to the legacy vendor Kernel 2.6.35 (20Mbit/sec).&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I think, the difference is that legacy Linux implements a polling mode, you can find it by the reference&amp;nbsp;&lt;SPAN style="color: #333333; font-size: 16.6667px; background-color: #ffffff;"&gt;&lt;A href="http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_2.6.35_1.1.0&amp;amp;id=c105f3ef1d461aaeedbc6361941096b6684cc812"&gt;ENGR0015175&lt;/A&gt;&amp;nbsp;&lt;SPAN style="font-size: 15px;"&gt;which is missing in the mainline kernel.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="background-color: #ffffff; color: #333333; font-size: 15px;"&gt;I know that mainline Linux is not officially supported, however the vendor kernel did not receive any updates for several years!&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="background-color: #ffffff; color: #333333; font-size: 15px;"&gt;Please be so kind and add this missing feature to mainline Linux&amp;nbsp;:smileyhappy:.&lt;/SPAN&gt;&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Tue, 18 Oct 2016 06:01:49 GMT</pubDate>
    <dc:creator>jörg_krause</dc:creator>
    <dc:date>2016-10-18T06:01:49Z</dc:date>
    <item>
      <title>Low MMC Throughput on i.MX28</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642970#M98149</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I have a custom i.MX28 board with a SDIO wifi chip attached. Using iperf I noticed a much lower data throughput on Linux mainline 4.7 (5Mbit/sec) compared to the legacy vendor Kernel 2.6.35 (20Mbit/sec).&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I think, the difference is that legacy Linux implements a polling mode, you can find it by the reference&amp;nbsp;&lt;SPAN style="color: #333333; font-size: 16.6667px; background-color: #ffffff;"&gt;&lt;A href="http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_2.6.35_1.1.0&amp;amp;id=c105f3ef1d461aaeedbc6361941096b6684cc812"&gt;ENGR0015175&lt;/A&gt;&amp;nbsp;&lt;SPAN style="font-size: 15px;"&gt;which is missing in the mainline kernel.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="background-color: #ffffff; color: #333333; font-size: 15px;"&gt;I know that mainline Linux is not officially supported, however the vendor kernel did not receive any updates for several years!&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="background-color: #ffffff; color: #333333; font-size: 15px;"&gt;Please be so kind and add this missing feature to mainline Linux&amp;nbsp;:smileyhappy:.&lt;/SPAN&gt;&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Tue, 18 Oct 2016 06:01:49 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642970#M98149</guid>
      <dc:creator>jörg_krause</dc:creator>
      <dc:date>2016-10-18T06:01:49Z</dc:date>
    </item>
    <item>
      <title>Re: Low MMC Throughput on i.MX28</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642971#M98150</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hello Jörg Krause,&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;I would suggest looking at the kernels available for the i.MX28 as part of the Community BSP (link below). This BSP is maintained by the community and it is part of the Yocto Project so it gets updated with some regularity. Please note that this BSP is not officially supported by NXP but you may raise questions in the Community.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;A href="http://freescale.github.io/"&gt;http://freescale.github.io/&lt;/A&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Regards,&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Tue, 18 Oct 2016 16:36:18 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642971#M98150</guid>
      <dc:creator>gusarambula</dc:creator>
      <dc:date>2016-10-18T16:36:18Z</dc:date>
    </item>
    <item>
      <title>Re: Low MMC Throughput on i.MX28</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642972#M98151</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Many thanks for the link to the community BSP. Unfortunately, this BSP does not have the optimization, too.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Sadly, it looks like NXP leaves the maintenance for the i.MX28 completely to the community.&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Fri, 28 Oct 2016 15:18:37 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642972#M98151</guid>
      <dc:creator>jörg_krause</dc:creator>
      <dc:date>2016-10-28T15:18:37Z</dc:date>
    </item>
    <item>
      <title>Re: Low MMC Throughput on i.MX28</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642973#M98152</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;I got the PIO mode working for the mxs-mmc driver. For this I ported the PIO code from the vendor kernel and removed the usage of the DMA&amp;nbsp;engine entirely.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Testing network bandwidth with iperf, I get about ~10Mbit/sec with PIO&amp;nbsp;mode compared to ~6.5Mbit/sec with DMA mode for UDP and about&amp;nbsp;~6.5Mbit/sec compared to ~4.5Mbit/sec with DMA mode for TCP.&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Wed, 02 Nov 2016 09:29:09 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/Low-MMC-Throughput-on-i-MX28/m-p/642973#M98152</guid>
      <dc:creator>jörgkrause</dc:creator>
      <dc:date>2016-11-02T09:29:09Z</dc:date>
    </item>
  </channel>
</rss>

