Return-path: Received: from cp-out9.libero.it ([212.52.84.109]:58640 "EHLO cp-out9.libero.it" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754076AbZGWVuD convert rfc822-to-8bit (ORCPT ); Thu, 23 Jul 2009 17:50:03 -0400 From: Fabio Rossi To: "Luis R. Rodriguez" Subject: Re: WARNING: at net/mac80211/tx.c:561 Date: Thu, 23 Jul 2009 23:47:43 +0200 Cc: "Luis R. Rodriguez" , linux-wireless@vger.kernel.org References: <200907230051.57201.rossi.f@inwind.it> <43e72e890907231119s5e992c91r2225f6ea77543dc8@mail.gmail.com> <43e72e890907231123w7267c022yc33f4aad63e3a3fd@mail.gmail.com> In-Reply-To: <43e72e890907231123w7267c022yc33f4aad63e3a3fd@mail.gmail.com> MIME-Version: 1.0 Message-Id: <200907232347.44096.rossi.f@inwind.it> Content-Type: text/plain; charset="utf-8" Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thursday 23 July 2009, Luis R. Rodriguez wrote: > >> Please try this patch: > >> > >> From: Luis R. Rodriguez > >> Subject: [PATCH] mac80211: do not trigger beacon work if scanning > >> > >> We were issues probe requests to the associated AP on the wrong > >> band. This leads to finding not bitrate. We should not be doing > >> this, so prevent the timer from stuffing beacon loss work on > >> the mac80211 workqueue. > >> > >> Signed-off-by: Luis R. Rodriguez > >> --- > >>  net/mac80211/mlme.c |    2 +- > >>  1 files changed, 1 insertions(+), 1 deletions(-) > >> > >> diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c > >> index 523c0d9..e90992e 100644 > >> --- a/net/mac80211/mlme.c > >> +++ b/net/mac80211/mlme.c > >> @@ -2187,7 +2187,7 @@ static void ieee80211_sta_bcn_mon_timer(unsigned > >> long data) (struct ieee80211_sub_if_data *) data; > >>        struct ieee80211_local *local = sdata->local; > >> > >> -       if (local->quiescing) > >> +       if (local->quiescing || local->sw_scanning || > >> local->hw_scanning) return; > > > > Heh no wait, this is very wrong. > > Never mind, please do try it. If we are scanning we should not be > sending probe requests through a timer. That clarifies what I mean and > I should add that to the commit log. But please do try it. Patch applied, now the warning is disappeared! Fabio