Return-path: Received: from nbd.name ([46.4.11.11]:42665 "EHLO nbd.name" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751531Ab3JELS5 (ORCPT ); Sat, 5 Oct 2013 07:18:57 -0400 Message-ID: <524FF599.5040706@openwrt.org> (sfid-20131005_131915_663385_ECB61BD5) Date: Sat, 05 Oct 2013 13:18:49 +0200 From: Felix Fietkau MIME-Version: 1.0 To: Johannes Berg CC: linux-wireless@vger.kernel.org Subject: Re: [PATCH 2/2] mac80211: implement support for configuring antenna gain References: <1378472763-36062-1-git-send-email-nbd@openwrt.org> <1378472763-36062-2-git-send-email-nbd@openwrt.org> <1380625512.14430.19.camel@jlt4.sipsolutions.net> <524AAE97.5020708@openwrt.org> <1380628244.14430.37.camel@jlt4.sipsolutions.net> In-Reply-To: <1380628244.14430.37.camel@jlt4.sipsolutions.net> Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 2013-10-01 1:50 PM, Johannes Berg wrote: > On Tue, 2013-10-01 at 13:14 +0200, Felix Fietkau wrote: >> On 2013-10-01 1:05 PM, Johannes Berg wrote: >> > On Fri, 2013-09-06 at 15:06 +0200, Felix Fietkau wrote: >> >> Report the maximum allowable extra antenna gain to the driver to allow >> >> it to reduce the tx power even further based on internal data >> > >> > I don't quite understand the maximum thing here - what's a user to do >> > who has an antenna that goes over? Is that then intended to not be >> > supported? That seems odd. A very high gain antenna might just result in >> > signal distortions, but what's the reason for limiting it this way? >> Very high gain antennas are useful for long distance links. >> The signal is not distorted, but focused directionally, which can easily >> make it exceed regulatory EIRP limits, unless tx power is reduced >> appropriately. > > Sure. > >> If the user explicitly configures the gain of the directional antenna >> using this patch, mac80211 will reduce the maximum allowed tx power >> setting to stay within the legal limit. > > I understand. I don't understand the pieces about "max_antenna_gain". Right now ath9k has an antenna gain value in the EEPROM, and it compares it against the channel max_antenna_gain value. Let's assume we have configured the tx power to the maximum value, the regdb allows 3 dB antenna gain, and the ath9k EEPROM contains an antenna gain of 3 dB as well. If we now add another 3 dB of user-configured antenna gain, it first starts tapping into the regulatory-allowed antenna gain before reducing tx power in mac80211. The driver needs to know about this, so I put the calculated maximum antenna gain into the hw conf as well. - Felix