Return-path: Received: from mail.candelatech.com ([208.74.158.172]:43837 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751530Ab1JSSsS (ORCPT ); Wed, 19 Oct 2011 14:48:18 -0400 Message-ID: <4E9F1B6B.1080401@candelatech.com> (sfid-20111019_204821_659778_187B05B5) Date: Wed, 19 Oct 2011 11:48:11 -0700 From: Ben Greear MIME-Version: 1.0 To: Johannes Berg CC: Stanislaw Gruszka , linux-wireless@vger.kernel.org, Jouni Malinen , reinette chatre , Eliad Peller Subject: Re: [RFC] mac80211: properly go back to operational channel? References: <1317121970-3638-1-git-send-email-sgruszka@redhat.com> <1317123289.4082.12.camel@jlt3.sipsolutions.net> <20110929150018.GA4554@redhat.com> <1317740824.6741.22.camel@jlt3.sipsolutions.net> <20111005111507.GA2184@redhat.com> <1317813762.4839.7.camel@jlt3.sipsolutions.net> <20111018141922.GA6204@redhat.com> <1318948251.3958.29.camel@jlt3.sipsolutions.net> In-Reply-To: <1318948251.3958.29.camel@jlt3.sipsolutions.net> Content-Type: text/plain; charset=UTF-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 10/18/2011 07:30 AM, Johannes Berg wrote: > On Tue, 2011-10-18 at 16:19 +0200, Stanislaw Gruszka wrote: >> For local->tmp_channel == NULL and local->scan_channel == NULL >> ieee80211_cfg_on_oper_channel() will return false if >> local->oper_channel != local->hw.conf.channel, >> hece we do not properly go back to oper_channel from tmp_channel. > > Huh, good catch. > >> Does patch have sense? > > Let's see what Ben says. It seems a bit like the > ieee80211_cfg_on_oper_channel() test there should be inverted instead of > removed? > >> Could it fixes problems we are talking >> in this thread? > > Yes, could be related, also some issue Jouni and Reinette have been > seeing with P2P might be related as well. Please see the patch I just posted. I think it fixes the issue (which was originally reported by Eliad Peller , who also posted a patch). I re-worked his suggested patch to what I believe is a more proper patch. Also, Eliad had posted another patch that seems to be good as is (which I also just re-posted). Obviously, my tests cases are not catching all of these issues, so please review the patches and/or test them out to see if they fix the problem. Thanks, Ben > > johannes > -- Ben Greear Candela Technologies Inc http://www.candelatech.com