Return-path: Received: from mail-vw0-f46.google.com ([209.85.212.46]:55790 "EHLO mail-vw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753442Ab1B1SuV convert rfc822-to-8bit (ORCPT ); Mon, 28 Feb 2011 13:50:21 -0500 Received: by vws12 with SMTP id 12so3380724vws.19 for ; Mon, 28 Feb 2011 10:50:20 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <20110227104723.GA8482@jm.kir.nu> References: <1298683960-20835-1-git-send-email-thomas@cozybit.com> <1298683960-20835-3-git-send-email-thomas@cozybit.com> <20110227104723.GA8482@jm.kir.nu> From: Javier Cardona Date: Mon, 28 Feb 2011 10:44:32 -0800 Message-ID: Subject: Re: [PATCH 2/5] mac80211: Let user space receive and send mesh auth/deauth frames To: Jouni Malinen Cc: Thomas Pedersen , linux-wireless@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: Hi Jouni, On Sun, Feb 27, 2011 at 2:47 AM, Jouni Malinen wrote: > On Fri, Feb 25, 2011 at 05:32:37PM -0800, Thomas Pedersen wrote: >> From: Javier Cardona >> @@ -529,7 +529,9 @@ ieee80211_default_mgmt_stypes[NUM_NL80211_IFTYPES] = { >> ? ? ? [NL80211_IFTYPE_MESH_POINT] = { >> + ? ? ? ? ? ? .rx = BIT(IEEE80211_STYPE_ACTION >> 4) | >> + ? ? ? ? ? ? ? ? ? ? BIT(IEEE80211_STYPE_AUTH >> 4) | >> + ? ? ? ? ? ? ? ? ? ? BIT(IEEE80211_STYPE_DEAUTH >> 4), > > This seems to be pointing towards processing the full Authentication > frame in user space. However, some other patches in this series seemed > to be extending the NL80211_CMD_AUTHENTICATE mechanism to allow SAE to > be implemented on top of it. Which direction is being taken here? Or are > both mechanisms going to be used depending on iftype? The patches were sent to the list before proper review. The changes to NL80211_CMD_AUTHENTICATE were left over from an earlier implementation variant and will be removed in the final submission. > For non-mesh cases, I would probably prefer NL80211_CMD_AUTHENTICATE > extension for the station vif to continue with the style we started for > FT. The AP vif is already handling the full Authentication frame in user > space, so that is fine to for SAE, too. Ah, thanks for the suggestion. As I mentioned on a previous e-mail, our current goal is to implement SAE for mesh interfaces. But will definitely follow your advice if we ever attempt to implement station mode support. Cheers, Javier -- Javier Cardona cozybit Inc. http://www.cozybit.com