Return-path: Received: from packetmixer.de ([79.140.42.25]:57338 "EHLO mail.mail.packetmixer.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753277Ab3KVMIQ (ORCPT ); Fri, 22 Nov 2013 07:08:16 -0500 From: Simon Wunderlich To: Karl Beldan Subject: Re: [PATCH v3] mac80211_hwsim: claim CSA support for AP Date: Fri, 22 Nov 2013 13:08:11 +0100 Cc: Johannes Berg , "linux-wireless" References: <1385111186-19551-1-git-send-email-karl.beldan@gmail.com> <20131122104527.GD8527@magnum.frso.rivierawaves.com> In-Reply-To: <20131122104527.GD8527@magnum.frso.rivierawaves.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="us-ascii" Message-Id: <201311221308.11994.sw@simonwunderlich.de> (sfid-20131122_130819_737655_4CEAE04D) Sender: linux-wireless-owner@vger.kernel.org List-ID: Hello Karl, > > Hmm, I thought the CSA code would make ieee80211_beacon_get_tim return > NULL (or do something alike) after the last ieee80211_beacon_get_tim > returned a beacon with a null CSA count until the config is done - but > it seems it doesn't - in that case this change would be race prone. > Did I miss something ? No, you have to do the check yourself (as you appearently did). In ath9k I'm checking if the CSA finished before scheduling the next beacon. Can you do the same for hwsim? Where do you see the race? Simon P.S.: Please use my new e-mail address, the old one will go out of service by end of the year.