From owner-freebsd-current@FreeBSD.ORG Thu Aug 26 15:52:46 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 25F2316A4CE for ; Thu, 26 Aug 2004 15:52:46 +0000 (GMT) Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id C7D8943D6A for ; Thu, 26 Aug 2004 15:52:45 +0000 (GMT) (envelope-from robert@fledge.watson.org) Received: from fledge.watson.org (localhost [127.0.0.1]) by fledge.watson.org (8.12.11/8.12.11) with ESMTP id i7QFoQ7S082554; Thu, 26 Aug 2004 11:50:26 -0400 (EDT) (envelope-from robert@fledge.watson.org) Received: from localhost (robert@localhost)i7QFoQQ3082551; Thu, 26 Aug 2004 11:50:26 -0400 (EDT) (envelope-from robert@fledge.watson.org) Date: Thu, 26 Aug 2004 11:50:26 -0400 (EDT) From: Robert Watson X-Sender: robert@fledge.watson.org To: Daniel Eriksson In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: 'Craig Boston' cc: freebsd-current@freebsd.org Subject: RE: PLEASE TEST: IPI deadlock avoidance patch X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 26 Aug 2004 15:52:46 -0000 On Thu, 26 Aug 2004, Daniel Eriksson wrote: > Craig Boston wrote: > > > Just a follow-up to my off-list message and another data point, with > > this patch I no longer get deadlocks, however I now get random data > > corruption. > > Me too! > > I've observed very strange data corruption on a filesystem exported via > Samba from a dual P3/700 running with the IPI deadlock patch. About 10% > is all files saved to the share while I had SMP enabled seems to be > corrupt in one way or another. Could you be more specific about "one way" or "another"? Specifically, are you observing files that contain unexpected garbage, unexpected regions of ASCII 0, unexpected contents of another file, etc? If it's possible to narrow it down further, using hexdump, is it possible to determine whether corruption begins and ends on page of file block boundaries? Also, is it possible for you to test whether you can trigger the corruption solely through remote access via Samba, and not through the use of other local programs and applications on the file system? Robert N M Watson FreeBSD Core Team, TrustedBSD Projects robert@fledge.watson.org Principal Research Scientist, McAfee Research