From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: [net-next-2.6 PATCH v4 1/2] net: implement mechanism for HW based QOS Date: Tue, 04 Jan 2011 07:32:19 -0500 Message-ID: <1294144339.4927.79.camel@mojatatu> References: <20110104030553.13187.69135.stgit@jf-dev1-dcblab> Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: davem@davemloft.net, jarkao2@gmail.com, shemminger@vyatta.com, tgraf@infradead.org, eric.dumazet@gmail.com, bhutchings@solarflare.com, nhorman@tuxdriver.com, netdev@vger.kernel.org To: John Fastabend Return-path: Received: from mail-vw0-f46.google.com ([209.85.212.46]:32869 "EHLO mail-vw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750895Ab1ADMcR (ORCPT ); Tue, 4 Jan 2011 07:32:17 -0500 Received: by vws16 with SMTP id 16so6038579vws.19 for ; Tue, 04 Jan 2011 04:32:17 -0800 (PST) In-Reply-To: <20110104030553.13187.69135.stgit@jf-dev1-dcblab> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, 2011-01-03 at 19:05 -0800, John Fastabend wrote: [..] > To steer the skb we mask out the lower 4 bits of the priority > and allow the hardware to configure upto 15 distinct classes > of traffic. This is expected to be sufficient for most applications > at any rate it is more then the 8021Q spec designates and is > equal to the number of prio bands currently implemented in > the default qdisc. This seems very hardware specific. i.e may be true for 8021q (or like 8021q) type hardware and therefore your qdisc. There are people with hardware that has thousands of hardware flow queues (or at least give impressions as such). Maybe a naming convention prefixed with 8021Q would be nicer or moving it into your qdisc infrastructure and export it so drivers needing it can. cheers, jamal