From owner-freebsd-wireless@FreeBSD.ORG Tue Aug 9 12:34:08 2011 Return-Path: Delivered-To: freebsd-wireless@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 31F061065673; Tue, 9 Aug 2011 12:34:08 +0000 (UTC) (envelope-from lev@serebryakov.spb.ru) Received: from onlyone.friendlyhosting.spb.ru (onlyone.friendlyhosting.spb.ru [IPv6:2a01:4f8:131:60a2::2]) by mx1.freebsd.org (Postfix) with ESMTP id C1E808FC16; Tue, 9 Aug 2011 12:34:07 +0000 (UTC) Received: from lion.home.serebryakov.spb.ru (unknown [IPv6:2001:470:923f:1:40d5:9d99:be75:3272]) (Authenticated sender: lev@serebryakov.spb.ru) by onlyone.friendlyhosting.spb.ru (Postfix) with ESMTPA id EC9F84AC1C; Tue, 9 Aug 2011 16:34:05 +0400 (MSD) Date: Tue, 9 Aug 2011 16:34:02 +0400 From: Lev Serebryakov X-Priority: 3 (Normal) Message-ID: <845369356.20110809163402@serebryakov.spb.ru> To: Bernhard Schmidt In-Reply-To: <201108091332.16269.bschmidt@freebsd.org> References: <1151152206.20110808220720@serebryakov.spb.ru> <281311475.20110809142033@serebryakov.spb.ru> <201108091332.16269.bschmidt@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: freebsd-wireless@freebsd.org Subject: Re: Bad host-vased AP performace -- and qouple of questions about FreeBSD WiFi stack tuning X-BeenThere: freebsd-wireless@freebsd.org X-Mailman-Version: 2.1.5 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: Tue, 09 Aug 2011 12:34:08 -0000 Hello, Bernhard. You wrote 9 =D0=B0=D0=B2=D0=B3=D1=83=D1=81=D1=82=D0=B0 2011 =D0=B3., 15:32:= 15: > On Tuesday, August 09, 2011 12:20:33 Lev Serebryakov wrote: >> Hello, Adrian. >> You wrote 9 =D0=B0=D0=B2=D0=B3=D1=83=D1=81=D1=82=D0=B0 2011 =D0=B3., 14:= 14:19: >>=20 >> > txpower 15 is actually 15dBm. Ie, try "ifconfig wlan0 txpower 1.5" =20 >> Bernhard Schmidt says, that iwn(4) uses half-dBm units and "txpower >> 30" on iwn(4) means 15dBm physically. > The ifconfig txpower command gets its argument in 0.5 dBm steps. That > means if you want to set 15dBm you have to use 30, as 30 * 0.5 =3D 15. > This is because someone might want to set eg 7.5 dBm but we still > want use an integer to store that value. It is in contradiction to what Adrian wrote, that "ifconfig" takes TRUE dBm as argument, and that "15.5" is allowed. Even more, it is in contradiction with ifconfig source code (ifieee80211.c:978 on 8-STABLE) and with practice: gateway# ifconfig wlan0 wlan0: flags=3D8843 metric 0 mtu 15= 00 txpower 30 scanvalid 60 protmode CTS wme burst dtimperiod 1 -dfs gateway# ifconfig wlan0 txpower 20.5 gateway# ifconfig wlan0 wlan0: flags=3D8843 metric 0 mtu 15= 00 txpower 20.5 scanvalid 60 protmode CTS wme burst dtimperiod 1 -dfs gateway# ifconfig wlan0 txpower 30 gateway# ifconfig wlan0 wlan0: flags=3D8843 metric 0 mtu 15= 00 txpower 30 scanvalid 60 protmode CTS wme burst dtimperiod 1 -dfs gateway# ifconfig wlan0 txpower 20.6 ifconfig: invalid tx power (must be .5 dBm units) gateway# So, at COMMAND LINE it seems, that "txpower" is TRUE dBm, 1 unit of txpower =3D=3D 1dBm. Yes, INTERNAL representation is integer with "1 unit =3D 0.5 dBm", but command line and output txpower seems to be true dBms > The 30 you are seeing is probably coming somewhere else, I assume > it is the regulatory limit allowed by that channel. Try playing > with different regdomains/countries. 30 is a max in regdomains.xml --=20 // Black Lion AKA Lev Serebryakov