From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264592AbUENXVu (ORCPT ); Fri, 14 May 2004 19:21:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264335AbUENXSd (ORCPT ); Fri, 14 May 2004 19:18:33 -0400 Received: from mail.kroah.org ([65.200.24.183]:41440 "EHLO perch.kroah.org") by vger.kernel.org with ESMTP id S264588AbUENXSC (ORCPT ); Fri, 14 May 2004 19:18:02 -0400 Date: Fri, 14 May 2004 16:17:12 -0700 From: Greg KH To: Andrew Morton Cc: "R. J. Wysocki" , ak@muc.de, linux-kernel@vger.kernel.org Subject: Re: 2.6.6-mm2(-mm1,-bk1): SMP bug on dual AMD64 Message-ID: <20040514231712.GA18108@kroah.com> References: <200405142328.59471.rjwysocki@sisk.pl> <20040514160524.57dd9de4.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040514160524.57dd9de4.akpm@osdl.org> User-Agent: Mutt/1.5.6i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, May 14, 2004 at 04:05:24PM -0700, Andrew Morton wrote: > "R. J. Wysocki" wrote: > > > > I've just compiled the 2.6.6-mm2 kernel (with gcc-3.4) and tried to run it. > > Well, it generally works, but there seems to be an SMP bug that may be > > triggered with the help of a USB storage device (please see the attached > > log). The device works fine in spite of it, though. > > > > This bug seems to be present in the 2.6.6-mm1 and 2.6.6-bk1, but it is not > > present in the 2.6.6, apparently. > > > > My system is a dual Opteron and I use an add-on USB 2.0 card based on the NEC > > chipset (the .config is attached). > > The WARN_ON() backtrace is unrelated to USB - it is due to SCSI calling > vmalloc/vfree with interrupts disabled. People are working on that. > > As for this: > > May 14 23:57:09 chimera kernel: usb usb2: string descriptor 0 read error: -108 > May 14 23:57:09 chimera last message repeated 2 times > > It has been reported before and I assume people are working it, but it > would be nice to have confirmation of that... It should be fixed in the patches I just sent to Linus. thanks, greg k-h