From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp73.iad3a.emailsrvr.com (smtp73.iad3a.emailsrvr.com [173.203.187.73]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 96AD03B2CE for ; Sat, 5 Mar 2016 17:19:32 -0500 (EST) Received: from smtp18.relay.iad3a.emailsrvr.com (localhost.localdomain [127.0.0.1]) by smtp18.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id 4987B2802F1; Sat, 5 Mar 2016 17:19:32 -0500 (EST) Received: from app12.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by smtp18.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id 3DB3C2802D6; Sat, 5 Mar 2016 17:19:32 -0500 (EST) X-Sender-Id: dpreed@reed.com Received: from app12.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by 0.0.0.0:25 (trex/5.5.4); Sat, 05 Mar 2016 17:19:32 -0500 Received: from reed.com (localhost [127.0.0.1]) by app12.wa-webapps.iad3a (Postfix) with ESMTP id 2FA3EC0067; Sat, 5 Mar 2016 17:19:32 -0500 (EST) Received: by apps.rackspace.com (Authenticated sender: dpreed@reed.com, from: dpreed@reed.com) with HTTP; Sat, 5 Mar 2016 17:19:32 -0500 (EST) Date: Sat, 5 Mar 2016 17:19:32 -0500 (EST) From: dpreed@reed.com To: "Dave Taht" Cc: "moeller0" , "cerowrt-devel@lists.bufferbloat.net" MIME-Version: 1.0 Content-Type: text/plain;charset=UTF-8 Content-Transfer-Encoding: quoted-printable Importance: Normal X-Priority: 3 (Normal) X-Type: plain In-Reply-To: References: X-Auth-ID: dpreed@reed.com Message-ID: <1457216372.19224891@apps.rackspace.com> X-Mailer: webmail/12.2.3-RC Subject: Re: [Cerowrt-devel] odroid C1+ status X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 05 Mar 2016 22:19:32 -0000 I have a Banana Pi, but I can't imagine why it would be useful as a router.= =0AWhy would Comcast even bother? A Raspberry Pi 3 would be better, and fa= r more available. (though I think the slow Ethernet port and low end WiFi o= n the Raspberry Pi 3 would make it sort of marginal, it's certainly quite f= ine for a low-end OpenWRT machine if you want to live at 50 Mb/sec)=0A=0A= =0A=0AOn Saturday, March 5, 2016 3:23pm, "Dave Taht" = said:=0A=0A> wow, thx for all the suggestions on alternate x86 router hardw= are... I=0A> will read more later.=0A> =0A> Would using a blog format for t= hings like the following work better=0A> for people? I could more easily re= vise, including graphics, etc,=0A> etc... could try to hit on our hot butto= ns (upgradability, bloat,=0A> reliability, kernel versions, manufacturer su= pport) with some sort of=0A> grading system...=0A> =0A> http://the-edge.tah= t.net/post/odroid_c1_plus/ in this case=0A> =0A> ...=0A> =0A> I got the odr= oid C1+ to work better. (either a cable or power supply=0A> issue, I swappe= d both). On output it peaks at about 416Mbits with 26%=0A> of cpu being spe= nt in a softirq interrupt. On input I can get it to=0A> gbit, with 220% of= cpu in use.=0A> =0A> The rrul tests were pretty normal, aside from the app= arent 400mbit=0A> upload limit causing contention on rx/tx (at the moment I= have no good=0A> place to put these test results since snapon is now behin= d a firewall.=0A> I'd like to get more organized about how we store and ind= ex these=0A> results also)=0A> =0A> There is no BQL support in the odroid d= river for it, and it ships with=0A> linux 3.10.80. At least its a LTS versi= on.... I am totally unfamiliar=0A> with the odroid ecosystem but maybe ther= e is active kernel dev on it=0A> somewhere?=0A> =0A> (The pi 2, on the othe= r hand, is kernel 4.1.17-v7 AND only has a=0A> 100mbit phy, so it is hard t= o complain about only getting 400mbit from=0A> the odroid c1+, but, dang it= , a much later kernel would be nice in the=0A> odroid)=0A> =0A> My goal in = life, generally, is to have a set of boxes with known=0A> characteristics t= o drive tests with, that are reliable enough to setup=0A> once and ignore.= =0A> =0A> A) this time around, I definitely wanted variety, particularly in= tcp=0A> implementations, kernel versions, ethernet and wifi chips - as it= =0A> seemed like drawing conclusions from "perfect" drivers like the e1000e= =0A> all the time was a bad idea. We have a very repeatable testbed in=0A> = karlstad, already - I'm interested in what random sort of traffic can=0A> e= xist on a home network that messes life up.=0A> =0A> One of the things I no= ticed while using kodi is that the box announces=0A> 2k of multicast ipv4 p= ackets every 30 seconds or so on the upnp=0A> port... AND over 4k of multic= ast ipv6 packets, if ipv6 is enabled.=0A> =0A> B) Need to be able to drive = 802.11ac as hard as possible with as many=0A> stations as possible.=0A> =0A= > C) needs to be low power and quiet (cheap is good too!)=0A> =0A> Has anyo= ne tried the banana pi? That's what comcast is using in their tests....=0A>= _______________________________________________=0A> Cerowrt-devel mailing = list=0A> Cerowrt-devel@lists.bufferbloat.net=0A> https://lists.bufferbloat.= net/listinfo/cerowrt-devel=0A> =0A