Received: with ECARTIS (v1.0.0; list netdev); Tue, 01 Apr 2003 13:23:46 -0800 (PST)
Received: from caduceus.fm.intel.com (fmr02.intel.com [192.55.52.25])
	by oss.sgi.com (8.12.8/8.12.5) with SMTP id h31LMWoJ032287
	for <netdev@oss.sgi.com>; Tue, 1 Apr 2003 13:23:13 -0800
Received: from talaria.fm.intel.com (talaria.fm.intel.com [10.1.192.39])
	by caduceus.fm.intel.com (8.11.6p2/8.11.6/d: outer.mc,v 1.51 2002/09/23 20:43:23 dmccart Exp $) with ESMTP id h31LFfD05037
	for <netdev@oss.sgi.com>; Tue, 1 Apr 2003 21:15:42 GMT
Received: from fmsmsxv040-1.fm.intel.com (fmsmsxvs040.fm.intel.com [132.233.42.124])
	by talaria.fm.intel.com (8.11.6p2/8.11.6/d: inner.mc,v 1.28 2003/01/13 19:44:39 dmccart Exp $) with SMTP id h31LNkc12960
	for <netdev@oss.sgi.com>; Tue, 1 Apr 2003 21:23:46 GMT
Received: from fmsmsx26.fm.intel.com ([132.233.42.26])
 by fmsmsxv040-1.fm.intel.com (NAVGW 2.5.2.11) with SMTP id M2003040113224111981
 ; Tue, 01 Apr 2003 13:22:41 -0800
Received: by fmsmsx26.fm.intel.com with Internet Mail Service (5.5.2653.19)
	id <HGZ4AFM2>; Tue, 1 Apr 2003 13:22:28 -0800
Message-ID: <C6F5CF431189FA4CBAEC9E7DD5441E010107D318@orsmsx402.jf.intel.com>
From: "Feldman, Scott" <scott.feldman@intel.com>
To: Jeff Garzik <jgarzik@pobox.com>
Cc: Robert Olsson <Robert.Olsson@data.slu.se>,
   Jason Lunz
	 <lunz@falooley.org>, netdev@oss.sgi.com
Subject: RE: [Fwd: [E1000] NAPI re-insertion w/ changes]
Date: Tue, 1 Apr 2003 13:22:15 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
content-class: urn:content-classes:message
Content-Type: text/plain;
	charset="iso-8859-1"
X-Spam-Checker-Version: SpamAssassin 2.50 (1.173-2003-02-20-exp)
X-archive-position: 2134
X-ecartis-version: Ecartis v1.0.0
Sender: netdev-bounce@oss.sgi.com
Errors-to: netdev-bounce@oss.sgi.com
X-original-sender: scott.feldman@intel.com
Precedence: bulk
X-list: netdev
Content-Length: 323
Lines: 14

> > Just curious, why give priority to Tx over Rx?
> 
> TX frees skbs, RX allocates skbs.
> 
> You lessen the chance of allocation failure or "additional 
> work" being performed by the allocator when you do it in this order.

At the expense of Rx latency.  I suppose the best would be:

rx_clean
tx_clean
rx_alloc

-scott