Return-path: Received: from bear.ext.ti.com ([192.94.94.41]:44675 "EHLO bear.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754152Ab0KKTrh (ORCPT ); Thu, 11 Nov 2010 14:47:37 -0500 Message-ID: <4CDC4851.8080005@ti.com> Date: Thu, 11 Nov 2010 21:47:29 +0200 From: Gery Kahn MIME-Version: 1.0 To: Juuso Oikarinen CC: "Coelho Luciano (Nokia-MS/Helsinki)" , "linux-wireless@vger.kernel.org" Subject: Re: [PATCH 2/2] Revert "wl1271: Change supported channel order for a more optimal scan" References: <1289384840-28305-1-git-send-email-juuso.oikarinen@nokia.com> <1289384840-28305-2-git-send-email-juuso.oikarinen@nokia.com> <4CDA7F7A.4000705@ti.com> <1289388086.2522.72.camel@wimaxnb.nmp.nokia.com> <4CDA83BD.2000306@ti.com> <1289389378.2522.78.camel@wimaxnb.nmp.nokia.com> In-Reply-To: <1289389378.2522.78.camel@wimaxnb.nmp.nokia.com> Content-Type: text/plain; charset=UTF-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 11/10/2010 1:42 PM, Juuso Oikarinen wrote: > On Wed, 2010-11-10 at 12:36 +0100, ext Gery Kahn wrote: >> On 11/10/2010 01:21 PM, Juuso Oikarinen wrote: >>> On Wed, 2010-11-10 at 12:18 +0100, ext Gery Kahn wrote: >>>> On 11/10/2010 12:27 PM, juuso.oikarinen@nokia.com wrote: >>>>> From: Juuso Oikarinen >>>>> >>>>> This reverts commit fa21c7a9e4be439e217fe72edbd39b643b643791. >>>>> >>>>> The reverted patch caused more harm than benefit. >>>>> >>>>> Signed-off-by: Juuso Oikarinen >>>>> >>>> Just couple words what harm it was? >>> The harm is that the supported channels are reported to user-space in >>> the order they are registered in the driver. The result of that is quite >>> messy in userspace PoV. >>> >>> -Juuso >>> >> Thank you for answer. >> >> The reason of doing such channel sorting was channel overlapping. >> >> Let's say for 11a (bandwidth 20MHz) it is not a problem. >> While 11g, preferred channel sorting would be 1,6,11,... >> >> Gery > Yes, this is why we had this reordering in the first place. > > One benefit of sorting channels would be to get faster results for the > whole band, as there are responses from the overlapping channels too. > Another benefit is to have max band coverage with minimum prob reqs. The overlapping answers have collisions, which discard all packets and if timeout scan - no responce from AP or even silence in channel. > Currently the the scanning code however waits for all channels to be > scanned before sending results to userspace, so this benefit is not > utilized, and removing this patch makes no difference in that sense. > > To take advantage of channel reordering, some other changes are needed > too. And the reordering needs to be performed in the mac80211/cfg80211 > level (I actually earlier submitted a RFC patch doing just this.) > > -Juuso > >> -- >> 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 >