From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michael Breuer Subject: Re: sky2 panic in 2.6.32.1 under load Date: Thu, 24 Dec 2009 11:26:49 -0500 Message-ID: <4B339649.3020007@majjas.com> References: <4B300A2A.8040305@gmail.com> <20091223225855.4a7d00af.akpm@linux-foundation.org> <4B339290.60909@majjas.com> <4B3393C3.7050408@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7BIT Cc: Andrew Morton , "linux-kernel@vger.kernel.org" , netdev@vger.kernel.org, Stephen Hemminger To: Berck Nash Return-path: Received: from mta1.srv.hcvlny.cv.net ([167.206.4.196]:61608 "EHLO mta1.srv.hcvlny.cv.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750959AbZLXQ0k (ORCPT ); Thu, 24 Dec 2009 11:26:40 -0500 In-reply-to: <4B3393C3.7050408@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On 12/24/2009 11:16 AM, Berck Nash wrote: > Michael Breuer wrote: > >>>> Since 2.6.32, I've been getting kernel panics under heavy network load >>>> (bittorrent usage). >>>> >>>> >>> Let's cc the right list and developer. >>> >>> This is a 2.6.31->2.6.32 regression? >>> >>> >> I don't think so. I saw this on 2.6.31 - first kernel I put on this >> system. At the time, I was having issues with intel_iommu and thought >> that the culprit. >> The initial reply to this thread from Daniel Hazelton stated that this >> has been happening since 2.6.25. >> > Your oops appears to be different than mine. Hazelton did not provide > any logs, and he said "a version of this" so I don't know if his is the > same or not. > Actually, I have two different oops, and a couple of crashes without any oops at all. All were produced under the same conditions, all pointing at sky2. I'd guess at memory corruption under load with subsequent divergent manifestations. The real question is whether the bug is in the kernel or the NIC firmware. There were similar reports online of this behavior affecting Vista, however those were resolved by moving to later firmware. I've recreated the crash on the later firmware however. Summary of conditions that appear to be required for the lockup/oops/panic(s): 1. Running at 1Gb. 2. High traffic into the system (out seems ok). 3. For me, the crash only occurs when smbd is the application generating the traffic. > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ >