From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from s3.sipsolutions.net ([5.9.151.49]:32923 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752514AbbA0Lnc (ORCPT ); Tue, 27 Jan 2015 06:43:32 -0500 Message-ID: <1422359002.1890.65.camel@sipsolutions.net> (sfid-20150127_124337_047766_CF2F8687) Subject: Re: [RFCv4 3/3] mac80211: add VHT support for IBSS From: Johannes Berg To: Janusz Dziedzic Cc: linux-wireless@vger.kernel.org Date: Tue, 27 Jan 2015 12:43:22 +0100 In-Reply-To: (sfid-20150127_123951_413522_6EE70539) References: <1422348243-4928-1-git-send-email-janusz.dziedzic@tieto.com> <1422348243-4928-3-git-send-email-janusz.dziedzic@tieto.com> <1422354271.1890.62.camel@sipsolutions.net> (sfid-20150127_123951_413522_6EE70539) Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, 2015-01-27 at 12:39 +0100, Janusz Dziedzic wrote: > On 27 January 2015 at 11:24, Johannes Berg wrote: > > On Tue, 2015-01-27 at 09:44 +0100, Janusz Dziedzic wrote: > > > >> +u8 *ieee80211_ie_build_vht_oper(u8 *pos, struct ieee80211_sta_vht_cap *vht_cap, > >> + const struct cfg80211_chan_def *chandef) > > > >> + /* 1 stream, MCS0-7 as a min Basic VHT MCS rates */ > >> + vht_oper->basic_mcs_set = cpu_to_le16(0xfffc); > > > > Unless I'm mistaken in my reading of the spec, this will make any > > well-behaved client (i.e. not mac80211) not join this network since it > > supports VHT MCSes. > > > > We seem to do the same for HT: > > > > /* It seems that Basic MCS set and Supported MCS set > > are identical for the first 10 bytes */ > > memset(&ht_oper->basic_set, 0, 16); > > memcpy(&ht_oper->basic_set, &ht_cap->mcs, 10); > > > > but I'm not convinced it's right. It probably only works because nobody > > ever tested against a well-behaved non-HT client? Or perhaps there isn't > > even one? > > > for HT (spec, 20.3.5) - MCS0 .. MSC7 are mandatory - but this is for > all stations (not sure about ibss) > for VHT (22.5, table 22-30) - MCS0...MCS7 are mandatory for nss=1 (20,40,80) > Seems such set is secure? > > If I understand correctly case you describe with non-HT client: > non-HT client should not understand HT ies, and HT-only client should > not understand VHT ies? Yeah that's a good point, I guess older clients won't even look at this, and we don't include the HT or VHT marker in the basic rate set IE. johannes