From owner-freebsd-net@FreeBSD.ORG Sun Mar 23 06:11:57 2008 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 50C811065673 for ; Sun, 23 Mar 2008 06:11:57 +0000 (UTC) (envelope-from pyunyh@gmail.com) Received: from wf-out-1314.google.com (wf-out-1314.google.com [209.85.200.175]) by mx1.freebsd.org (Postfix) with ESMTP id 16ABC8FC1B for ; Sun, 23 Mar 2008 06:11:57 +0000 (UTC) (envelope-from pyunyh@gmail.com) Received: by wf-out-1314.google.com with SMTP id 25so2444615wfa.7 for ; Sat, 22 Mar 2008 23:11:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:received:received:date:from:to:cc:subject:message-id:reply-to:references:mime-version:content-type:content-disposition:in-reply-to:user-agent; bh=bC3qd98wiQYgAf7I+kPb+18hbhg1sOR04ZemRfO8gVo=; b=nKu48IdWpxBGJlgaL7tHG8s3Vz/+NI4MS0X8gyDxrcUtx2cfguefO+dpy9/6Z0p+fDnn/z2uXX2QOd6x9v1+Rd7Y1nb3ZsXvjJ1cp2wTovCPyYKxDX9stYeaJT7q1Sn3cTZ83Fc1qOm0f95WaeZ0U8pfq0Lc+N+is7MzrNbPZe4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=date:from:to:cc:subject:message-id:reply-to:references:mime-version:content-type:content-disposition:in-reply-to:user-agent; b=rLYTT90WZMr44rYCTTPDOOWij4msAHVcds713AqICrNCSzUwUOo01a2mCgsiGqULf4vpPakEAHRodmZczX4mfRQxpXwHYXA4jlZifMJlQwgmGBHYNtLJqENxpz5wJ75JxwhXlcNWqz/PmvB47j31ShTcN5CxhdGuk5BUt135Sms= Received: by 10.142.232.20 with SMTP id e20mr3542271wfh.198.1206252716820; Sat, 22 Mar 2008 23:11:56 -0700 (PDT) Received: from michelle.cdnetworks.co.kr ( [211.53.35.84]) by mx.google.com with ESMTPS id 24sm11906500wff.10.2008.03.22.23.11.52 (version=TLSv1/SSLv3 cipher=OTHER); Sat, 22 Mar 2008 23:11:54 -0700 (PDT) Received: from michelle.cdnetworks.co.kr (localhost.cdnetworks.co.kr [127.0.0.1]) by michelle.cdnetworks.co.kr (8.13.5/8.13.5) with ESMTP id m2N6BmIl079795 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sun, 23 Mar 2008 15:11:48 +0900 (KST) (envelope-from pyunyh@gmail.com) Received: (from yongari@localhost) by michelle.cdnetworks.co.kr (8.13.5/8.13.5/Submit) id m2N6BkYa079794; Sun, 23 Mar 2008 15:11:46 +0900 (KST) (envelope-from pyunyh@gmail.com) Date: Sun, 23 Mar 2008 15:11:46 +0900 From: Pyun YongHyeon To: Artem Belevich Message-ID: <20080323061146.GA79693@cdnetworks.co.kr> References: <47E57C8F.4090602@citrin.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i Cc: freebsd-net@freebsd.org, Anton Yuzhaninov Subject: Re: re TSO: data corruption X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: pyunyh@gmail.com List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 23 Mar 2008 06:11:57 -0000 On Sat, Mar 22, 2008 at 04:01:50PM -0700, Artem Belevich wrote: > TSO does seem to be broken on Realtek adapters. I ran into it on > Windows, went googling and found that I'm not the only one suffering: > > http://blogs.zdnet.com/Ou/?p=663 > http://www.mail-archive.com/netdev@vger.kernel.org/msg56764.html > Long time ago, I've read this article but the article is not clear to me. I know RealTek chips have several bugs not mentioned in datasheet and Tx checksum offload is one of the most serious bug in the hardware. But I didn't encounter any TSO related issues on my box. It could also be related checksum offload and padding but I have no clear evidence yet. Today, I commited pending changes in my local tree. This will address VLAN and unstability issues on PCIe based hardwares. But the stock version still does not have one important fix that is under actively testing. To rule out other possible bugs in stock driver, please download the follodwing files and set tunable hw.re.msi_disable="1" in /boot/loader.conf and let me know the result. http://people.freebsd.org/~yongari/re/if_re.c http://people.freebsd.org/~yongari/re/if_rlreg.h The above version has changed a way to work-around checksum offload bug of the hardware. -- Regards, Pyun YongHyeon