Return-path: Received: from mail-wi0-f176.google.com ([209.85.212.176]:39242 "EHLO mail-wi0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751828AbaEWJQp convert rfc822-to-8bit (ORCPT ); Fri, 23 May 2014 05:16:45 -0400 Received: by mail-wi0-f176.google.com with SMTP id n15so505562wiw.9 for ; Fri, 23 May 2014 02:16:43 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1400835498.4358.9.camel@jlt4.sipsolutions.net> References: <1400769938.4174.29.camel@jlt4.sipsolutions.net> <1400812703-15685-1-git-send-email-luca@coelho.fi> <1400835498.4358.9.camel@jlt4.sipsolutions.net> Date: Fri, 23 May 2014 11:16:43 +0200 Message-ID: (sfid-20140523_111659_991325_0848AA8C) Subject: Re: [PATCH] mac80211: add a single-transaction driver op to switch contexts From: Michal Kazior To: Johannes Berg Cc: Luca Coelho , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 23 May 2014 10:58, Johannes Berg wrote: > On Fri, 2014-05-23 at 10:01 +0200, Michal Kazior wrote: > >> So.. which patch are we going forward with? Luca's or mine? Either way >> is fine with me as long as we reach a conclusion :-) > > Looks like both are missing some things? Luca's doesn't have tracing, > and supports just a single interface (which you'll not like :) ) but > personally I liked the struct API a bit better than multiple double > pointer arrays you had. Well, the 'single interface' problem isn't even a thing. Luca just happens to rework ieee80211_vif_use_reserved_ context() in one go. My patch simply adds the new op since the old/single-vif channel csa code gets thrown out later anyway. I suppose we can just split Luca's patch and drop the vif_use_reserved_context() part (as well as IEEE80211_HW_CHANGE_RUNNING_CHANCTX flag removal) as it will get overwritten by my patches later anyway? Michał