Message-ID: <1361197982.8555.24.camel@jlt4.sipsolutions.net> (sfid-20130218_153311_039970_DF720F40)
Subject: Re: [RFC] design discussion: Collecting information for (non-peer)
 stations
From: Johannes Berg <johannes@sipsolutions.net>
To: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de>
Cc: linux-wireless@vger.kernel.org,
	Thomas Pedersen <thomas@cozybit.com>, antonio@open-mesh.com,
	marek@open-mesh.com,
	Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de>
Date: Mon, 18 Feb 2013 15:33:02 +0100
In-Reply-To: <1361197831.8555.23.camel@jlt4.sipsolutions.net> (sfid-20130218_153041_336693_8960F747)
References: <20130215171938.GA4140@pandem0nium>
	 <1361197831.8555.23.camel@jlt4.sipsolutions.net>
	 (sfid-20130218_153041_336693_8960F747)
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Sender: linux-wireless-owner@vger.kernel.org

On Mon, 2013-02-18 at 15:30 +0100, Johannes Berg wrote:
> On Fri, 2013-02-15 at 18:19 +0100, Simon Wunderlich wrote:
> 
> > Commands we would like to propose are:
> >  * start collecting - this feature should not run by default to avoid bloating memory for users who
> >    don't even need this
> >  * stop collecting
> >  * read - dumps the data for all stations
> >  * read + reset - dump the data and reset information for all stations. This should also clean up stations,
> >    at least those which are not connected to the BSS, to not bloat the station table.
> > 
> > I guess the right position to implement this is mac80211 receive path. Our intended platform
> > is ath9k/ath5k, but that feature should work with any mac80211 driver. We don't care if sta_info
> > structs are allocated or custom structures are used, as long as we can receive a list of stations
> > which includes peer and non-peer stations, along with their statistics.
> > 
> > We are looking forward to your thoughts. :)
> 
> I would argue that since most of the sta_info struct is used for
> operational stuff, you shouldn't use it, but have a separate struct and
> maybe embed that separate struct in sta_info for its statistics.
> 
> I'd also not use the existing nl80211 station APIs since this could be
> an optional feature for many things, and it will likely break existing
> expectations, e.g. that all stations listed by "iw wlan0 station dump"
> are clients connected to an AP interface.
> 
> It could be argued that this API then should also not even include the
> connected stations when listing ones, i.e. explicitly be non-connected
> stations.

Or maybe use the APIs, but require including a special attribute in the
dump/get request message in order to dump/get the/a non-connected
station(s), and only include those attributes that are relevant.

johannes