Return-path: Received: from mail.atheros.com ([12.36.123.2]:39464 "EHLO mail.atheros.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753563AbYH0XbI (ORCPT ); Wed, 27 Aug 2008 19:31:08 -0400 Date: Wed, 27 Aug 2008 16:31:03 -0700 From: "Luis R. Rodriguez" To: Tomas Winkler CC: Michael Buesch , "John W. Linville" , "davem@davemloft.net" , "linux-wireless@vger.kernel.org" , "netdev@vger.kernel.org" , "linux-kernel@vger.kernel.org" , Bruno Randolf Subject: Re: pull request: wireless-2.6 2008-08-26 Message-ID: <20080827233103.GE5927@tesla> (sfid-20080828_013113_832234_D0471CBB) References: <20080827013009.GA15781@tuxdriver.com> <1ba2fa240808271226i95a7789k481a68b09dc60164@mail.gmail.com> <200808272225.10557.mb@bu3sch.de> <1ba2fa240808271611v382631ecn2a24e2816562d434@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" In-Reply-To: <1ba2fa240808271611v382631ecn2a24e2816562d434@mail.gmail.com> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wed, Aug 27, 2008 at 04:11:15PM -0700, Tomas Winkler wrote: > if (beacon_timestamp > rx_timestamp) > merge > > The patch from Assaf just disable reporting RX timestamp to mac and > thus disabling merging which gives incorrect spec behavior but smooth > traffic. We *should follow the spec*. > Actually we've checked few cards including broadcom and various > windows NICs What are windows NICs? > and non of them implements this correctly so this WA is > probably the solution. Disagreed! If there is hardware which is not capable of handling this we should simply have a HW flag which specifies this to handle this as a work around (WA). Just because some cards are not capable it doesn't mean it should impose that on the rest. > Other solution would be to mark leader with highest TSF and not > reconnecting to the same station again and again. > > Last solution would be to remove this merging all together but then > I'm not sure if Bruno added this code just implement the spec or > really tested it with any hardware. I'm not sure if any vendor > implements PS in IBSS so this merging is probably not important > anyway. Absolutely not! IBSS merge is per spec, otherwise you don't really have a real IBSS. Luis