Return-path: Received: from mail-qy0-f174.google.com ([209.85.216.174]:49857 "EHLO mail-qy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753239Ab1LLOiE convert rfc822-to-8bit (ORCPT ); Mon, 12 Dec 2011 09:38:04 -0500 Received: by qcqz2 with SMTP id z2so3548872qcq.19 for ; Mon, 12 Dec 2011 06:38:03 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <1323680785.3442.9.camel@jlt3.sipsolutions.net> References: <20111209223131.71B17206F0@glenhelen.mtv.corp.google.com> <1323680785.3442.9.camel@jlt3.sipsolutions.net> Date: Mon, 12 Dec 2011 06:38:02 -0800 Message-ID: (sfid-20111212_153821_913786_588FB004) Subject: Re: [PATCH] cfg80211: Return beacon loss count in station From: Paul Stewart To: Johannes Berg Cc: linux-wireless@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Mon, Dec 12, 2011 at 1:06 AM, Johannes Berg wrote: > On Fri, 2011-12-09 at 11:01 -0800, Paul Stewart wrote: >> If station info contains a beacon loss count, return >> it to userspace. > > >> + * @NL80211_STA_INFO_BEACON_LOSS: count of times beacon loss was detected (u32) > > Hmm, this is very vague. Are you sure it does what you think it does? > Some drivers like wl12xx might not trigger the beacon loss function for > every beacon, instead aggregating things. I implemented this in iwlwifi > once and it would only tell mac80211 roughly every 20 beacons about > missed ones... > > This should at least be documented, but I fear it makes the value rather > useless in the general case. I actually differ on that. I completely agree that this is not a count of how many beacons were lost, but it does represent an important statistic worth measuring, which is how often the system enters a state where it is trying to recover from beacon loss. You can't get this information anywhere else (unless you turn on MAC80211_VERBOSE_DEBUG and go trawling through the logs) and my aim is to collect swaths of this information to get qualitative information about the occurrence count of this situation for a large population of users. It seems like the cleanest way to get at it, regardless of the somewhat device-specific interpretation (which one can always account for). > johannes > > > -- > To unsubscribe from this list: send the line "unsubscribe linux-wireless" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at ?http://vger.kernel.org/majordomo-info.html