<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    yesterday i shortly tested this with tplink 841 and LEDE FW, these
    smaller embedded Router System <br>
    (as mentioned some emails before)<br>
    <br>
    and we got 30 Mbit Through in on direction and 11-14 the other. <br>
    (which doesnt make so much sense to me - but i hadnt time to debug
    this, or find out more)<br>
    <br>
    so that as short notice ;)   - iperf3 Test (tcp - with/out reverse
    and - server/client changed)<br>
    <br>
    <pre><code class="lang-auto hljs coffeescript">[ ID] Interval           Transfer     Bandwidth
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.04</span>  sec  <span class="hljs-number">0.00</span> Bytes  <span class="hljs-number">0.00</span> bits/sec                  sender
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.04</span>  sec  <span class="hljs-number">13.9</span> MBytes  <span class="hljs-number">11.6</span> Mbits/sec                  receiver

[ ID] Interval           Transfer     Bandwidth       Retr
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.06</span>  sec  <span class="hljs-number">38.7</span> MBytes  <span class="hljs-number">32.3</span> Mbits/sec   <span class="hljs-number">50</span>             sender
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.06</span>  sec  <span class="hljs-number">0.00</span> Bytes  <span class="hljs-number">0.00</span> bits/sec                  receiver

[ ID] Interval           Transfer     Bandwidth
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.03</span>  sec  <span class="hljs-number">0.00</span> Bytes  <span class="hljs-number">0.00</span> bits/sec                  sender
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.03</span>  sec  <span class="hljs-number">14.2</span> MBytes  <span class="hljs-number">11.8</span> Mbits/sec                  receiver

[ ID] Interval           Transfer     Bandwidth       Retr
[  <span class="hljs-number">4</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.00</span>  sec  <span class="hljs-number">38.3</span> MBytes  <span class="hljs-number">32.1</span> Mbits/sec   <span class="hljs-number">40</span>             sender
[  <span class="hljs-number">4</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.00</span>  sec  <span class="hljs-number">37.9</span> MBytes  <span class="hljs-number">31.8</span> Mbits/sec                  receiver

[ ID] Interval           Transfer     Bandwidth
[  <span class="hljs-number">4</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">4.00</span>   sec  <span class="hljs-number">0.00</span> Bytes  <span class="hljs-number">0.00</span> bits/sec                  sender
[  <span class="hljs-number">4</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">4.00</span>   sec  <span class="hljs-number">6.78</span> MBytes  <span class="hljs-number">14.2</span> Mbits/sec                  receiver

[ ID] Interval           Transfer     Bandwidth       Retr
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.04</span>  sec  <span class="hljs-number">38.7</span> MBytes  <span class="hljs-number">32.4</span> Mbits/sec   <span class="hljs-number">46</span>             sender
[  <span class="hljs-number">5</span>]   <span class="hljs-number">0.00</span>-<span class="hljs-number">10.04</span>  sec  <span class="hljs-number">0.00</span> Bytes  <span class="hljs-number">0.00</span> bits/sec                  receiver</code></pre>
    <br>
<a class="moz-txt-link-freetext" href="https://forum.freifunk.net/t/lede-test-wireguard-und-blanko-durchsatz-tp841nv11-1-und-bug/13163/11">https://forum.freifunk.net/t/lede-test-wireguard-und-blanko-durchsatz-tp841nv11-1-und-bug/13163/11</a><br>
    <br>
    On 01.10.2016 01:34, Jason A. Donenfeld wrote:<br>
    <blockquote type="cite">Hello,<br>
      <br>
      A new experimental snapshot, `experimental-0.0.20161001`, has been
      tagged in<br>
      the git repository.<br>
      <br>
      Please note that this snapshot is, like the rest of the project at
      this point<br>
      in time, experimental, and does not consitute a real release that
      would be<br>
      considered secure and bug-free. However, if you'd like to test
      this snapshot<br>
      out, there are a few relevent changes.<br>
      <br>
      == Changes ==<br>
      <br>
        * poly1305: optimize unaligned access<br>
        <br>
        This is a very appreciated fix from René van Dorst, adjusting
      the arithmetic<br>
        in Poly1305 to work fast on platforms with slow unaligned
      access, such as<br>
        MIPS. According to his calculation, this gives a 50% improvement
      on small MIPS<br>
        boxes.<br>
        <br>
        * hashtables: use rdrand() instead of counter<br>
        <br>
        Rather than incrementing a counter, we instead use rdrand, which
      gives us an<br>
        extremely fast source of random numbers. We're still running
      this through<br>
        siphash with a secret, so a backdoored rdrand implementation
      won't be a<br>
        problem.<br>
        <br>
        * examples: add nat-hole-punching<br>
        <br>
       
      <a class="moz-txt-link-freetext" href="https://lists.zx2c4.com/pipermail/wireguard/2016-August/000372.html">https://lists.zx2c4.com/pipermail/wireguard/2016-August/000372.html</a><br>
       
<a class="moz-txt-link-freetext" href="https://git.zx2c4.com/WireGuard/tree/contrib/examples/nat-hole-punching/README">https://git.zx2c4.com/WireGuard/tree/contrib/examples/nat-hole-punching/README</a><br>
        <br>
        * examples: add key extractor<br>
        <br>
       
      <a class="moz-txt-link-freetext" href="https://lists.zx2c4.com/pipermail/wireguard/2016-August/000373.html">https://lists.zx2c4.com/pipermail/wireguard/2016-August/000373.html</a><br>
       
<a class="moz-txt-link-freetext" href="https://git.zx2c4.com/WireGuard/tree/contrib/examples/extract-keys/README">https://git.zx2c4.com/WireGuard/tree/contrib/examples/extract-keys/README</a><br>
        <br>
        * tools: allow multiple AllowedIPs invocations<br>
        <br>
        Multiple AllowedIPs= lines can now be specified, which could
      improve<br>
        readability of the config files.<br>
        <br>
        * send: properly encapsulate ECN<br>
        <br>
        Thanks to the guidance of Dave Taht, we now support ECN.<br>
        <br>
        * Rework headers and includes<br>
        * compat: Isolate more functions<br>
        <br>
        In anticipation of upstreaming WireGuard, we've now moved most
      of our<br>
        version-specific #ifdefs to compat.h, where we use horrible
      macro tricks to<br>
        redefine functions for old versions. This allows us to keep the
      actual code as<br>
        clean as possible. When we merge to mainline, compat.h will be
      deleted<br>
        wholesale.<br>
        <br>
        * tests: test jumbo frames with more transfer<br>
        * tests: add crypto-RP filter test<br>
        * qemu: enhancements<br>
        <br>
        With this an numerous other commits, we've further expanded the
      test suite.<br>
      <br>
      As always, the source is available at
      <a class="moz-txt-link-freetext" href="https://git.zx2c4.com/WireGuard/">https://git.zx2c4.com/WireGuard/</a> and<br>
      information about the project is available at
      <a class="moz-txt-link-freetext" href="https://www.wireguard.io/">https://www.wireguard.io/</a> .<br>
      <br>
      This snapshot is available in tarball form here:<br>
       
<a class="moz-txt-link-freetext" href="https://git.zx2c4.com/WireGuard/snapshot/WireGuard-experimental-0.0.20161001.tar.xz">https://git.zx2c4.com/WireGuard/snapshot/WireGuard-experimental-0.0.20161001.tar.xz</a><br>
        SHA256:
      ac3abb7b940716ac12b96a2cb3f7666598cbefd26f19c268f627dc47cd113ac8<br>
      <br>
      If you're a snapshot package maintainer, please bump your package
      version. If<br>
      you're a user, the WireGuard team welcomes any and all feedback on
      this latest<br>
      snapshot.<br>
      <br>
      Thank you,<br>
      Jason Donenfeld<br>
      <br>
      <br>
    </blockquote>
    <span style="white-space: pre;">> _______________________________________________
> WireGuard mailing list
> <a class="moz-txt-link-abbreviated" href="mailto:WireGuard@lists.zx2c4.com">WireGuard@lists.zx2c4.com</a>
> <a class="moz-txt-link-freetext" href="http://lists.zx2c4.com/mailman/listinfo/wireguard">http://lists.zx2c4.com/mailman/listinfo/wireguard</a></span><br>
    <br>
    -- <br>
    make the world nicer, please use PGP encryption<br>
    <br>
  </body>
</html>