From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from dedo.coelho.fi ([88.198.205.34]:48878 "EHLO dedo.coelho.fi" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1753077AbaJ1Ncl (ORCPT ); Tue, 28 Oct 2014 09:32:41 -0400 Message-ID: <1414503152.27833.17.camel@coelho.fi> (sfid-20141028_143252_454131_78822A14) From: Luca Coelho To: johannes@sipsolutions.net Cc: j@w1.fi, linux-wireless@vger.kernel.org Date: Tue, 28 Oct 2014 15:32:32 +0200 In-Reply-To: <1414495985-26937-1-git-send-email-luca@coelho.fi> References: <1414495985-26937-1-git-send-email-luca@coelho.fi> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Subject: Re: [PATCH 0/2] mac80211: fix HT+ to HT- (and vice-versa) CSA Sender: linux-wireless-owner@vger.kernel.org List-ID: (fixed Johannes' email address) Both these bugs go far back, so IMHO they should go to 3.8 and stable. 1/2 goes back to 3.13 2/2 goes back to 2.6.30 (though the affected code changed somewhat since then) -- Luca. On Tue, 2014-10-28 at 13:33 +0200, Luca Coelho wrote: > From: Luciano Coelho > > These two patches fix a bug that Jouni found when running his hostap > hwsim test scripts (ap_ht40_csa). The failure was caused by two > different bugs which are fixed in this patchset. > > The first bug was that we were not using the secondary channel offset > IE of beacons for the switch, but the offset of the current channel, > which cause us to use the same offset. > > The second bug was that we were trying to switch at the exact time of > the switch on the station side, which, because of the processing time, > was causing us to be too late. > > Luciano Coelho (2): > mac80211: use secondary channel offset IE also beacons during CSA > mac80211: schedule the actual switch of the station before CSA count 0 > > net/mac80211/ibss.c | 2 +- > net/mac80211/ieee80211_i.h | 3 +-- > net/mac80211/mesh.c | 2 +- > net/mac80211/mlme.c | 5 +++-- > net/mac80211/spectmgmt.c | 18 ++++++------------ > 5 files changed, 12 insertions(+), 18 deletions(-) >