From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1763835AbYDOQG6 (ORCPT ); Tue, 15 Apr 2008 12:06:58 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753150AbYDOQGs (ORCPT ); Tue, 15 Apr 2008 12:06:48 -0400 Received: from host64.cybernetics.com ([70.169.137.4]:4582 "EHLO mail.cybernetics.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750958AbYDOQGr (ORCPT ); Tue, 15 Apr 2008 12:06:47 -0400 X-Greylist: delayed 1638 seconds by postgrey-1.27 at vger.kernel.org; Tue, 15 Apr 2008 12:06:47 EDT Message-ID: <4804CC2F.6030403@cybernetics.com> Date: Tue, 15 Apr 2008 11:39:27 -0400 From: Tony Battersby User-Agent: Thunderbird 2.0.0.12 (X11/20080213) MIME-Version: 1.0 To: Matt Carlson Cc: Herbert Xu , Michael Chan , David Miller , netdev , gregkh@suse.de, linux-kernel@vger.kernel.org Subject: Re: TG3 network data corruption regression 2.6.24/2.6.23.4 References: <47BA0984.2070306@cybernetics.com> <1203381120.13495.78.camel@dell> <20080218.163554.74130592.davem@davemloft.net> <1203383046.13495.87.camel@dell> <47BB00EC.3010607@cybernetics.com> <1203448265.13495.95.camel@dell> <47BB54C2.6090501@cybernetics.com> <20080220034515.GC22703@gondor.apana.org.au> <47BC44E2.9060301@cybernetics.com> <20080415001207.GA11852@localdomain> In-Reply-To: <20080415001207.GA11852@localdomain> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Matt Carlson wrote: > Hi Tony. Sorry for the radio silence. > > Michael and I have discussed this problem a bit. Another possibility is > that the chip may be having difficulty with non-dword aligned TX buffers. > Since we already know the RX side has the same problem, it isn't so > far-fetched to think that perhaps it can affect the TX side too. Can > you give the following patch a try and see if the corruption still > happens? > > Thanks, your patch fixes the problem (tested on 2.6.24.4). However, I had to change "(skb->mac_header & 3)" in your patch to "((long) skb->mac_header & 3)" since mac_header is a pointer rather than an int on 32-bit systems. Tested-by: Tony Battersby