Return-path: Received: from smtp1.irobot.com ([206.83.81.187]:24034 "EHLO smtp1.irobot.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932504AbZKXRRX (ORCPT ); Tue, 24 Nov 2009 12:17:23 -0500 Message-ID: <4B0C1527.60305@irobot.com> Date: Tue, 24 Nov 2009 09:17:27 -0800 From: Adam Wozniak MIME-Version: 1.0 To: Johannes Berg CC: Derek Smithies , Christian Lamparter , linux-wireless@vger.kernel.org, Felix Fietkau Subject: Re: [PATCH 2/2] mac80211: minstrel try all rates References: <4AF0D54D.4090303@irobot.com> <4AFC655A.5020706@irobot.com> <200911122103.27455.chunkeey@googlemail.com> <4AFC8E4F.5090307@irobot.com> <4AFC8F0D.5020700@irobot.com> <1258097352.3899.75.camel@johannes.local> <4AFDDF19.8060202@irobot.com> <1258191039.6167.40.camel@johannes.local> <4B018B26.2070008@irobot.com> <1258392453.32159.28.camel@johannes.local> <4B0192A6.9050808@irobot.com> <4B01D4A5.7060204@irobot.com> <4B01E298.3030602@irobot.com> <1258443484.1375.8.camel@johannes.local> <4B02DFE3.9040602@irobot.com> <4B0AEEE3.2050808@irobot.com> <1259018866.7094.189.camel@johannes.local> <4B0B2F6E.7050907@irobot.com> <1259025067.7094.208.camel@johannes.local> In-Reply-To: <1259025067.7094.208.camel@johannes.local> Content-Type: text/plain; charset=UTF-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: Johannes Berg wrote: > I really don't like this patch anyhow though, it's at best a hackish > workaround around an issue that should just be fixed instead. > > Will take another look at the other patch tomorrow. > I concur 2/2 is a hack. I'm not an 802.11 ad-hoc expert, and I haven't been allocated time to pursue the issue any more deeply than this. It is still not clear to me how the list of usable rates is supposed to be determined when the IBSS is composed of stations with heterogeneous rate capabilities, beaconing is stochastically distributed, and not all nodes can hear all beacons, much less what is wrong with the way the current code is doing it. I do not believe 1/2 was a hack. It really was possible to get to the end of that function and use that variable without it being assigned the correct value. Also, it is important to reinitialize the rate control layer when the list of usable rates changes. I would certainly be in favor of a better patch that resolves all the issues. --Adam Wozniak