From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: Strange packet drops with heavy firewalling Date: Fri, 09 Apr 2010 15:29:22 +0200 Message-ID: <1270819762.2623.97.camel@edumazet-laptop> References: <1270813662.2623.85.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: Benny Amorsen Return-path: Received: from mail-bw0-f209.google.com ([209.85.218.209]:61217 "EHLO mail-bw0-f209.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751798Ab0DIN31 (ORCPT ); Fri, 9 Apr 2010 09:29:27 -0400 Received: by bwz1 with SMTP id 1so2542144bwz.21 for ; Fri, 09 Apr 2010 06:29:26 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Le vendredi 09 avril 2010 =C3=A0 14:33 +0200, Benny Amorsen a =C3=A9cri= t : > Thank you very much for the help! I will report back whether it was t= he > hash buckets. OK You could try : ethtool -C eth0 tx-usecs 200 tx-frames 100 tx-frames-irq 100 ethtool -C eth1 tx-usecs 200 tx-frames 100 tx-frames-irq 100 (to reduce tx completion irqs) Before buying multiqueue devices, you also could try net-next-2.6 kerne= l, because RPS (Remote Packet Steering) is in. In your setup, this might help a bit, distribute the packets to all cpu= s, with appropriate cache handling.