From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Greear Subject: Re: [PATCH net-next-2.6] pktgen: speedup fragmented skbs Date: Tue, 25 Jan 2011 09:22:07 -0800 Message-ID: <4D3F06BF.7010903@candelatech.com> References: <1295975635.3588.337.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: David Miller , netdev To: Eric Dumazet Return-path: Received: from mail.candelatech.com ([208.74.158.172]:39858 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753255Ab1AYRWL (ORCPT ); Tue, 25 Jan 2011 12:22:11 -0500 In-Reply-To: <1295975635.3588.337.camel@edumazet-laptop> Sender: netdev-owner@vger.kernel.org List-ID: On 01/25/2011 09:13 AM, Eric Dumazet wrote: > We spend lot of time clearing pages in pktgen. > (Or not clearing them on ipv6 and leaking kernel memory) > > Since we dont modify them, we can use one zeroed page, and get > references on it. This page can use NUMA affinity as well. > > Define pktgen_finalize_skb() helper, used both in ipv4 and ipv6 Some devices, like vlans, can change the skb, but perhaps they will not actually mess with the paged data? Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com