Return-path: Received: from mail.deathmatch.net ([72.66.92.28]:3879 "EHLO mail.deathmatch.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935370AbZJOTsv (ORCPT ); Thu, 15 Oct 2009 15:48:51 -0400 Date: Thu, 15 Oct 2009 15:45:30 -0400 From: Bob Copeland To: Nick Kossifidis Cc: linville@tuxdriver.com, linux-wireless@vger.kernel.org, ath5k-devel@lists.ath5k.org, jirislaby@gmail.com, lrodriguez@atheros.com Subject: Re: [PATCH] ath5k: use noise calibration from madwifi hal Message-ID: <20091015194530.GA32509@hash.localnet> References: <1255544190-30020-1-git-send-email-me@bobcopeland.com> <40f31dec0910142357w6e395640uf9690498d0c3854e@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <40f31dec0910142357w6e395640uf9690498d0c3854e@mail.gmail.com> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thu, Oct 15, 2009 at 09:57:45AM +0300, Nick Kossifidis wrote: > a) Please leave some comments to let people know that when nf > calibration runs antennas are detached (it measures the noise produced > from the card itself), that if we enable nf calibration before agc > calibration it always fails + the link to Atheros patent. Sure, I'll add that back. I read the patent but I'm not sure it applies 100%, it also talks at the end about keeping the rx antennas connected and measuring the noise in the environment instead of just the thermal noise. I guess one way to know for sure is to hit it with noise and look at the resulting values. > b) What about thresh62 ? Do we also need to update/rewrite this value > (we got it from EEPROM and write it on the register during reset) ? Yeah, I looked at that, madwifi/legacy-hal also set it (just the EEPROM value for the threshold bits) before doing the false detect backoff stuff. > Acked-by: Nick Kossifidis Thanks for the review! -- Bob Copeland %% www.bobcopeland.com