+1<div>A blog format where we can also comment and document our experiences/tweaks.<span></span><br><br>On Saturday, March 5, 2016, David Lang <<a href="mailto:david@lang.hm">david@lang.hm</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">A blog format for hardware testing would be a good idea.<br>
<br>
David Lang<br>
<br>
On Sat, 5 Mar 2016, Dave Taht wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Date: Sat, 5 Mar 2016 12:23:36 -0800<br>
From: Dave Taht <<a>dave.taht@gmail.com</a>><br>
To: moeller0 <<a>moeller0@gmx.de</a>><br>
Cc: "<a>cerowrt-devel@lists.bufferbloat.net</a>"<br>
<<a>cerowrt-devel@lists.bufferbloat.net</a>><br>
Subject: [Cerowrt-devel] odroid C1+ status<br>
<br>
wow, thx for all the suggestions on alternate x86 router hardware... I<br>
will read more later.<br>
<br>
Would using a blog format for things like the following work better<br>
for people? I could more easily revise, including graphics, etc,<br>
etc... could try to hit on our hot buttons (upgradability, bloat,<br>
reliability, kernel versions, manufacturer support) with some sort of<br>
grading system...<br>
<br>
<a href="http://the-edge.taht.net/post/odroid_c1_plus/" target="_blank">http://the-edge.taht.net/post/odroid_c1_plus/</a> in this case<br>
<br>
...<br>
<br>
I got the odroid C1+ to work better. (either a cable or power supply<br>
issue, I swapped both). On output it peaks at about 416Mbits with 26%<br>
of cpu being spent in a softirq interrupt. On input I can get it to<br>
gbit, with 220% of cpu in use.<br>
<br>
The rrul tests were pretty normal, aside from the apparent 400mbit<br>
upload limit causing contention on rx/tx (at the moment I have no good<br>
place to put these test results since snapon is now behind a firewall.<br>
I'd like to get more organized about how we store and index these<br>
results also)<br>
<br>
There is no BQL support in the odroid driver for it, and it ships with<br>
linux 3.10.80. At least its a LTS version.... I am totally unfamiliar<br>
with the odroid ecosystem but maybe there is active kernel dev on it<br>
somewhere?<br>
<br>
(The pi 2, on the other hand, is kernel 4.1.17-v7 AND only has a<br>
100mbit phy, so it is hard to complain about only getting 400mbit from<br>
the odroid c1+, but, dang it, a much later kernel would be nice in the<br>
odroid)<br>
<br>
My goal in life, generally, is to have a set of boxes with known<br>
characteristics to drive tests with, that are reliable enough to setup<br>
once and ignore.<br>
<br>
A) this time around, I definitely wanted variety, particularly in tcp<br>
implementations, kernel versions, ethernet and wifi chips - as it<br>
seemed like drawing conclusions from "perfect" drivers like the e1000e<br>
all the time was a bad idea. We have a very repeatable testbed in<br>
karlstad, already - I'm interested in what random sort of traffic can<br>
exist on a home network that messes life up.<br>
<br>
One of the things I noticed while using kodi is that the box announces<br>
2k of multicast ipv4 packets every 30 seconds or so on the upnp<br>
port... AND over 4k of multicast ipv6 packets, if ipv6 is enabled.<br>
<br>
B) Need to be able to drive 802.11ac as hard as possible with as many<br>
stations as possible.<br>
<br>
C) needs to be low power and quiet (cheap is good too!)<br>
<br>
Has anyone tried the banana pi? That's what comcast is using in their tests....<br>
_______________________________________________<br>
Cerowrt-devel mailing list<br>
<a>Cerowrt-devel@lists.bufferbloat.net</a><br>
<a href="https://lists.bufferbloat.net/listinfo/cerowrt-devel" target="_blank">https://lists.bufferbloat.net/listinfo/cerowrt-devel</a><br>
</blockquote>
_______________________________________________<br>
Cerowrt-devel mailing list<br>
<a>Cerowrt-devel@lists.bufferbloat.net</a><br>
<a href="https://lists.bufferbloat.net/listinfo/cerowrt-devel" target="_blank">https://lists.bufferbloat.net/listinfo/cerowrt-devel</a><br>
</blockquote></div>