From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.mediaspot.net (mail.mediaspot.net [109.73.92.181]) by huchra.bufferbloat.net (Postfix) with ESMTP id 666DE21F1C8 for ; Mon, 6 May 2013 13:15:38 -0700 (PDT) X-Spam-Status: No, hits=4.1 required=6.5 tests=AWL: 0.115,BAYES_99: 4.07,TOTAL_SCORE: 4.185,autolearn=no X-Spam-Level: **** X-Footer: bWVkaWFzcG90Lm5ldA== Received: from exchange.mediaspot.net ([192.168.0.22]) by mail.mediaspot.net (using TLSv1/SSLv3 with cipher AES128-SHA (128 bits)) for codel@lists.bufferbloat.net; Mon, 6 May 2013 22:15:35 +0200 Received: from EXCHANGE.mediaspot.local ([fe80::d853:1561:649a:ab52]) by EXCHANGE.mediaspot.local ([fe80::d853:1561:649a:ab52%6]) with mapi id 14.03.0123.003; Mon, 6 May 2013 22:15:36 +0200 From: Alessandro Bolletta To: "codel@lists.bufferbloat.net" Thread-Topic: Wifi improvements on debloating Thread-Index: Ac5KlnfyYLM3qDJGRgShXKGokcrqCA== Date: Mon, 6 May 2013 20:15:34 +0000 Message-ID: Accept-Language: it-IT, en-US Content-Language: it-IT X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.0.90] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [Codel] Wifi improvements on debloating X-BeenThere: codel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: CoDel AQM discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 May 2013 20:15:39 -0000 Hi everybody, I saw that you released a new version of cerowrt based on linux kernel 3.7.= 5 and, as I can see, cerowrt includes many fixes for debloating also at dri= ver level, right? Since we forked a OpenWRT svn based on kernel 3.7.5 for our communitary mes= h network, we are thinking to port your debloating improvements made over c= erowrt on our forked OS for our growing-up mesh network. Do you think that = the effort made on your fixes is too high to made a port of these features,= so it would be better to fork directly cerowrt, or do you think we shouldn= 't encounter big troubles in porting the features that we need? Moreover, have you made improvements in wifi driver's queuing to fight buff= erbloat at driver level?=20 We are planning to use atk9k-based wireless card on our devices and we plan= to build high-gain links between nodes. Don't you think that, if nodes are= connected through high throughput links (as 100Mbit of real UDP flow link = speed, for example), the driver-level queuing, as 802.11n packet aggregatio= n feature, shouldn't be a source of bufferbloat? Can I ask you if you ever = encountered configurations like these and what are your suggestions about t= hat? Thanks so much, greetings from Italy! -- Alessandro Bolletta