From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from c60.cesmail.net ([216.154.195.49]:1087 "EHLO c60.cesmail.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753215AbZHFUj2 (ORCPT ); Thu, 6 Aug 2009 16:39:28 -0400 Subject: Re: [PATCH] cfg80211: fix disassociation warning due to misuse of wdev->current_bss From: Pavel Roskin To: Johannes Berg Cc: linux-wireless@vger.kernel.org, John Linville In-Reply-To: <1249564158.4852.24.camel@johannes.local> References: <20090806085242.23637.82992.stgit@ct.roinet.com> <1249549574.4852.17.camel@johannes.local> <1249563308.15613.9.camel@ct> <1249564158.4852.24.camel@johannes.local> Content-Type: text/plain Date: Thu, 06 Aug 2009 16:39:24 -0400 Message-Id: <1249591164.19169.41.camel@mj> Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thu, 2009-08-06 at 15:09 +0200, Johannes Berg wrote: > On Thu, 2009-08-06 at 08:55 -0400, Pavel Roskin wrote: > > > Now that the check is fixed, it turn out there is a related problem in > > the same function. If the device is in the managed mode and disconnects > > from an AP due to a low signal, > > Who decides to disconnect? I was unscrewing the antenna from the AP, and the AP wasn't visible in the scan results without an antenna, so I assume that the station was deciding to disconnect on its own. > This isn't making any sense to me. Once you're connected, > wdev->auth_bsses[i] shouldn't be set any more. I see that you have posted several patches for the issue. I'll test them on the same hardware tomorrow to avoid introducing extra variables. -- Regards, Pavel Roskin