From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next-2.6] inetpeer: lower false sharing effect Date: Fri, 10 Jun 2011 06:47:50 +0200 Message-ID: <1307681270.3210.9.camel@edumazet-laptop> References: <1307600810.3980.85.camel@edumazet-laptop> <1307664235.17300.44.camel@schen9-DESK> <20110609.204330.2090335955971650557.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: tim.c.chen@linux.intel.com, netdev@vger.kernel.org, andi@firstfloor.org To: David Miller Return-path: Received: from mail-wy0-f174.google.com ([74.125.82.174]:32872 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750733Ab1FJErz (ORCPT ); Fri, 10 Jun 2011 00:47:55 -0400 Received: by wya21 with SMTP id 21so1590110wya.19 for ; Thu, 09 Jun 2011 21:47:54 -0700 (PDT) In-Reply-To: <20110609.204330.2090335955971650557.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: Le jeudi 09 juin 2011 =C3=A0 20:43 -0700, David Miller a =C3=A9crit : > From: Tim Chen > Date: Thu, 09 Jun 2011 17:03:55 -0700 >=20 > > When I retest with original 3.0-rc2 kernel, inet_putpeer no longer = shows > > up, wonder if dst->peer was not set for some reason.=20 >=20 > The overhead will only show up if an inetpeer entry exists for > the destination IP address. >=20 > You can force one to be created, for example, by making a TCP > connection to that destination. Or sending big frames, to force ip fragmentation (we store the id generator in inetpeer cache)