From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from ra.tuxdriver.com ([70.61.120.52]:4740 "EHLO ra.tuxdriver.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933971AbYBTUde (ORCPT ); Wed, 20 Feb 2008 15:33:34 -0500 Date: Wed, 20 Feb 2008 15:25:18 -0500 From: "John W. Linville" To: Johannes Berg Cc: bruno randolf , "Luis R. Rodriguez" , Michael Renzmann , linux-wireless Subject: Re: when does ath5k filter frames? Message-ID: <20080220202518.GD3825@tuxdriver.com> (sfid-20080220_203340_020826_46DA6E91) References: <1203514108.17534.41.camel@johannes.berg> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1203514108.17534.41.camel@johannes.berg> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wed, Feb 20, 2008 at 02:28:28PM +0100, Johannes Berg wrote: > Hi, > > In mac80211 we have a whole bunch of cryptic code to handle "filtered" > frames. The only driver that does that seems to be ath5k, and it doesn't > really seem to use this since it is intimately tied to > IEEE80211_TXCTL_CLEAR_DST_MASK which it doesn't check... > > Any ideas? Should we kill all this code and not use whatever hardware > powersave helpers atheros hardware might have? > > On the other hand, Broadcom also has powersave helpers in > hardware/firmware, maybe we should generalise to be able to use them as > well? Do we understand either of these (i.e. Broadcom or Atheros) helpers well enough to utilize them? If so, it sounds like a nice feature to support. Hardware/firmware may be able to do nifty things to save power. If we don't really know how those helpers work, it may be better just to remove the unused infrastructure. Is my opinion useful? :-) John -- John W. Linville linville@tuxdriver.com