From owner-freebsd-wireless@FreeBSD.ORG Thu Nov 29 21:53:51 2012 Return-Path: Delivered-To: freebsd-wireless@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 5BD8DC8F; Thu, 29 Nov 2012 21:53:51 +0000 (UTC) (envelope-from nakal@web.de) Received: from mout.web.de (mout.web.de [212.227.15.4]) by mx1.freebsd.org (Postfix) with ESMTP id DE9E88FC16; Thu, 29 Nov 2012 21:53:50 +0000 (UTC) Received: from zelda.sugioarto.com ([217.236.14.77]) by smtp.web.de (mrweb003) with ESMTPSA (Nemesis) id 0LmLac-1T4JKx0kmg-00aGPT; Thu, 29 Nov 2012 22:53:49 +0100 Date: Thu, 29 Nov 2012 22:53:47 +0100 From: Martin To: Adrian Chadd Subject: Re: kern/149516: [ath] ath(4) hostap with fake MAC/BSSID results in station dropping packets when associated Message-ID: <20121129225347.757b1fc8@zelda.sugioarto.com> In-Reply-To: References: <201211291940.qATJe1Jf033574@freefall.freebsd.org> <20121129214512.4909fa8f@zelda.sugioarto.com> <20121129223901.686290ce@zelda.sugioarto.com> X-Mailer: Claws Mail 3.9.0 (GTK+ 2.24.13; x86_64-unknown-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Provags-ID: V02:K0:w3jjdnwGozC1h2BFNZcl77PuP4Cf6DcJ/cj8LDA7VoB y//BAlc2J1k1DIYADPvH3pLwU7RsbdaQJdfiKR9J9ix4O191R+ rGiZq08elPq9JIUosVvWTLH8vlPlumJ0suG4YNAEVBlWkku40h gmF70byu88iGNiJeT/R55NyVSu+RowNKXz9o3MnwUFr2gGzBJa 9vpE05yLDvpPBJtXpoEmA== Cc: freebsd-wireless@freebsd.org X-BeenThere: freebsd-wireless@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: "Discussions of 802.11 stack, tools device driver development." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 29 Nov 2012 21:53:51 -0000 Am Thu, 29 Nov 2012 13:43:04 -0800 schrieb Adrian Chadd : > There was someone who posted on -wireless a while ago about how to > "correctly" program different mac addresses for different VAPs. The > whole setup path is a little un-intuitive and needs some love. > > So take a look over the -wireless archives for multi-VAP and separate > mac addresses. :-) > > Aha. It's actually having the LSB set in the first octet. Not the MSB. > > So 0x80 is fine. 0x90 is fine. 0x09 is not. > > http://en.wikipedia.org/wiki/Multicast_address > > The hardware treats that address as special, so don't ever configure a > MAC with that multicast bit set. Or things won't work out well. > > > > Adrian When I understand you correctly, this only explains why 0x0d does not work. It still does not explain why 0x00 and 0xd0 do not work. I also wrote that the lowest bit needs to be "0" in my first PR (duh! it has been too long ago!). [1] As I said, everything was OK on FreeBSD 7.x and broke somewhere in 8.1-RELEASE. Please also notice that the problem affects ath(4) on the driver layer (I suppose) not generally on IEEE 802.11 layer, because I also tested with rum(4) and I could use it properly. [1] http://lists.freebsd.org/pipermail/freebsd-stable/2010-July/058059.html -- Martin