From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from ozlabs.org ([203.10.76.45]:56698 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753979AbYESGfQ (ORCPT ); Mon, 19 May 2008 02:35:16 -0400 From: Rusty Russell To: David Miller Subject: Re: [PATCH] mac80211: rewrite fragmentation code Date: Mon, 19 May 2008 13:08:13 +1000 Cc: herbert@gondor.apana.org.au, mb@bu3sch.de, johannes@sipsolutions.net, linux-wireless@vger.kernel.org, netdev@vger.kernel.org, ron.rindjunsky@intel.com, tomasw@gmail.com, ivdoorn@gmail.com, peter.p.waskiewicz.jr@intel.com References: <200805161201.48835.rusty@rustcorp.com.au> <200805162032.48469.rusty@rustcorp.com.au> <20080516.124039.253626477.davem@davemloft.net> In-Reply-To: <20080516.124039.253626477.davem@davemloft.net> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Message-Id: <200805191308.14548.rusty@rustcorp.com.au> (sfid-20080519_083532_241435_045DEA96) Sender: linux-wireless-owner@vger.kernel.org List-ID: On Saturday 17 May 2008 05:40:39 David Miller wrote: > My description describes how I want the mid-layer queue to disappear > entirely. Queueing would be done by the driver only. Ok, the more I ponder this, the more I like it. It has a very nice side benefit for virtio_net: we can xmit a whole bunch of packets before notifying the host. Real NICs might gain similarly. The bit I can't see is what to do about qdisc if the driver manages its own queue(s). Leave the qdisc as currently in place and have the driver call dev_dequeue_skb() (or some wrapper) directly? Modulo locking issues, that should be a fairly simple change. Thanks, Rusty.