Return-path: Received: from mail-iy0-f174.google.com ([209.85.210.174]:46534 "EHLO mail-iy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750960Ab1KYSkV convert rfc822-to-8bit (ORCPT ); Fri, 25 Nov 2011 13:40:21 -0500 Received: by iage36 with SMTP id e36so4896257iag.19 for ; Fri, 25 Nov 2011 10:40:20 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: <1321886416-12597-1-git-send-email-ivan.bezyazychnyy@gmail.com> Date: Fri, 25 Nov 2011 22:40:20 +0400 Message-ID: (sfid-20111125_194025_241521_A8AB149D) Subject: Re: [RFC v2 0/8] mesh power saving implementation: indication From: Ivan Bezyazychnyy To: Javier Cardona Cc: linux-wireless@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 25 November 2011 05:40, Javier Cardona wrote: > Hi Ivan, > > Good work! ?Comments inline. > > On Mon, Nov 21, 2011 at 6:40 AM, Ivan Bezyazychnyy > wrote: >> We have updated our patches and also we would like to add some words >> what our work is about. >> >> Based on the latest IEEE 802.11s standard the power saving mechanism >> can be divided into the following implementation phases: >> * indication of mesh STAs' power modes in frames; >> * a buffering and identification sending frames in mesh STA; >> * Mesh Awake Window implementation; >> * Peer service periods. > > My recommendation would be to implement this bottom up. ?That is, > start by implementing the Extensible synchronization framework > (Section 13.12.2, 802.11mb D10.0), as power save will not work without > syncronization. ?After that, continue as... > > 13.13.8.4 Operation in light sleep mode for a mesh peering > 13.13.4 TIM transmissions in an MBSS (and buffering) > 13.13.6 Mesh awake window > 13.13.9 Mesh peer service periods > > Only after this is implemented I would advertise this capability in > beacons and make it configurable via userspace (i.e. what you've done > in this patchset). ?That approach would also make it easier to test; > only after you get the TSF right you should attempt to implement the > awake windows and so on. > > Does that make sense to you? > Thanks. Definitely that makes sense, but we have to better understand the first part about synchronization. We will study it. Best regards, Ivan