From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ie0-f174.google.com (mail-ie0-f174.google.com [209.85.223.174]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 5AADB21F0A2 for ; Sun, 9 Dec 2012 09:18:35 -0800 (PST) Received: by mail-ie0-f174.google.com with SMTP id c11so5912680ieb.19 for ; Sun, 09 Dec 2012 09:18:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=sblcSbzE/ZmD0VVRsW8YRjOc7UtgLAFBFtaSlBglttY=; b=ieiu+gmnUQ8Y3u/+pp6E66PQArh/ZGPvlX70ipbIgNMO7Uo0yUVUP/iaimHvnhXMeS 8AJD5gl4YtjcRhuj2TvzIm732jKBRENPo/PUPrg6jZzBNN70cFu87CDwoPZUB6nGn18L /2ckKSa+Jnqdpw8k2ErkXRpoOxqmhBhKQXmCFkYsNXh+NxoaHG+y0QZH4Stj5XqDugyG Tw00tdP92sPcdB12MPm/QYiNs6OCciMxpptglwhGzRWLpPpyXGhFTp5P9IIPIhrtkDgN xrNZt87GZ7dDjLu+qoJWUiCJsxUZHIFRqpd/Ryz/5UBXuFPP0LXnHFXuc4bVq6hzL4d7 OLtA== MIME-Version: 1.0 Received: by 10.50.157.162 with SMTP id wn2mr4469231igb.27.1355073514756; Sun, 09 Dec 2012 09:18:34 -0800 (PST) Received: by 10.64.135.39 with HTTP; Sun, 9 Dec 2012 09:18:34 -0800 (PST) Date: Sun, 9 Dec 2012 18:18:34 +0100 Message-ID: From: Dave Taht To: cerowrt-devel@lists.bufferbloat.net Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: [Cerowrt-devel] the iwl driver sucks X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 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: Sun, 09 Dec 2012 17:18:35 -0000 I am happy to report that ad-hoc mode (used by cero in mesh networking mode) still works in 3.6.9-5. On the other hand the iwl driver in linux 3.6.8 is nearly unusable in paris, generically, on my laptop. I have no idea why, I just want to get an ath9k chip into it so I don't have to deal with it.... (it takes 2 minutes or more to send an email via google, IF it succeeds. I'm now 10 minutes into downloading richard's preso with only 320k downloaded so far) I did some rrul experiments on the adhoc interfaces today that were very interesting. One result was that in good conditions I'd get 100Mbit total, bidirectionally, with rrul, with ath9k set to 128 packets, and could see tcp global synchronization with pfifo fast, and fq_codel had very little visible effect. uploads were at about 82Mbit and down 24. ~30ms median RTT latency. I also saw a really neat one where BE and BK traffic underwent priority inversion for a while. With ath9k set to 2, and fq_codel I'd get about 16/13 and ~10ms median RTT latency. So... far less throughput, but a vastly better up/down ratio - and far less latency. under bad conditions (router in the microwave for shielding), oh... my... god... at 128 ath9k buffers... I don't want to talk about it. --=20 Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.= html