From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: Reproducible corruption with SG offloading Date: Wed, 13 Jan 2010 21:06:27 +0100 Message-ID: <4B4E27C3.90003@gmail.com> References: <4B4B181B.3020208@rabbit.us> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, mcarlson@broadcom.com, mchan@broadcom.com To: Peter Rabbitson Return-path: Received: from mail-fx0-f225.google.com ([209.85.220.225]:34455 "EHLO mail-fx0-f225.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754901Ab0AMUGc (ORCPT ); Wed, 13 Jan 2010 15:06:32 -0500 Received: by fxm25 with SMTP id 25so584888fxm.21 for ; Wed, 13 Jan 2010 12:06:31 -0800 (PST) In-Reply-To: <4B4B181B.3020208@rabbit.us> Sender: netdev-owner@vger.kernel.org List-ID: Peter Rabbitson wrote, On 01/11/2010 01:22 PM: > I have the following card: > > 02:02.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5701 Gigabit Ethernet (rev 15) ... > Kernel driver in use: tg3 > > When confronted with very specific repeatable data streams the card eventually > starts setting garbage down the wire. If I do ethtool -k sg off the > problem disappears. Let me know if ou are interested in tracking this down, > (whether software or hardware) otherwise I'll throw the card away and go with > something else. > > Please CC me as I am not subscribed to netdev@ > > This is my minimal test case with relevant data: Interested or not, it would be useful to save more data on this, e.g. into bugzilla report (plus some link here); especially the kernel version, these tcpdumps as row files, ifconfigs, and as much as possible according to the linux-2.6/REPORTING_BUGS. Thanks, Jarek P.