Return-path: Received: from mail.deathmatch.net ([72.66.92.28]:1746 "EHLO mail.deathmatch.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751901AbZHaMyq (ORCPT ); Mon, 31 Aug 2009 08:54:46 -0400 Date: Mon, 31 Aug 2009 08:53:12 -0400 From: Bob Copeland To: Nick Kossifidis Cc: "John W. Linville" , "Luis R. Rodriguez" , proski@gnu.org, ath5k-devel@lists.ath5k.org, linux-wireless@vger.kernel.org, ic.felix@gmail.com, Felix Fietkau Subject: Re: Ath5k and proprietary extensions Message-ID: <20090831125312.GB11006@hash.localnet> References: <40f31dec0908282151t245912f0ye79684d4a519f3c9@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <40f31dec0908282151t245912f0ye79684d4a519f3c9@mail.gmail.com> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Sat, Aug 29, 2009 at 07:51:29AM +0300, Nick Kossifidis wrote: > Most code is there for half/quarter and (static) turbo operation, we > just have to deal with pll programming, clock settings etc. Question > is how do we use it, NL80211_CMD_TESTMODE is an option i guess and in > case no one else thinks that we could use a channel width API (or > extend what we have) for setting 5/10/40MHz width through cfg80211, we > can just stick to NL80211_CMD_TESTMODE. Well, we will have a channel width API, I think; several people have expresed an interest in doing so. Your summary is useful, and I really don't have any qualms about keeping the non-standard stuff as long as someone has plans to write the code to support it. I personally have no interest in using those features, but of course I only speak for me. If we do remove turbo code for now, we can leave the initvals, as they are such a pain to verify. But I'm not sure half a page of text is worth the churn. -- Bob Copeland %% www.bobcopeland.com