Return-path: Received: from mail-pw0-f46.google.com ([209.85.160.46]:35733 "EHLO mail-pw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932667Ab0JDVHn (ORCPT ); Mon, 4 Oct 2010 17:07:43 -0400 Received: by pwj5 with SMTP id 5so1231475pwj.19 for ; Mon, 04 Oct 2010 14:07:43 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20101004203900.GA14524@jm.kir.nu> References: <1285965233-11097-1-git-send-email-lrodriguez@atheros.com> <1286198845.3620.37.camel@jlt3.sipsolutions.net> <20101004164149.GE2105@tux> <1286210566.3620.42.camel@jlt3.sipsolutions.net> <20101004172504.GL2105@tux> <1286213426.3620.43.camel@jlt3.sipsolutions.net> <20101004173507.GN2105@tux> <20101004203900.GA14524@jm.kir.nu> From: "Luis R. Rodriguez" Date: Mon, 4 Oct 2010 14:07:22 -0700 Message-ID: Subject: Re: [PATCH v2 0/3] mac80211: fix rate_control_send_low warnings for delbas To: Jouni Malinen Cc: Johannes Berg , Luis Rodriguez , "linville@tuxdriver.com" , "linux-wireless@vger.kernel.org" , "stable@kernel.org" , Paul Stewart , Amod Bodas , Vasanth Thiagarajan Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Mon, Oct 4, 2010 at 1:39 PM, Jouni Malinen wrote: > On Mon, Oct 04, 2010 at 10:35:07AM -0700, Luis R. Rodriguez wrote: >> I was wondering if actively disassociating might help with a smoother >> transition. I was under this why we were doing this in the first place. >> I frankly do not know, but if it does not help then I do agree with >> your patch replacement. > > In many cases, it may end up harming more than helping.. At minimum, it > takes some time to transmit the frame (and do the channel changes, if > needed). Furthermore, this makes it more difficult for centrally managed > networks to optimize roaming since we would be disassociating and > associating as a new association instead of doing proper re-association. > In such networks, the APs (or well, likely some sort of central manager) > takes care of clearing the old association when the reassociation is > being processed. In addition, this could potentially tunnel some frames > through the new AP or at least make sure that bridge tables gets > updated. > Jouni, just to be clear so you are fine with dropping explicitly the tear down of the BA agreement to the old AP? Luis