Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754532Ab3CZRhx (ORCPT ); Tue, 26 Mar 2013 13:37:53 -0400 Received: from zeniv.linux.org.uk ([195.92.253.2]:38079 "EHLO ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754438Ab3CZRhv (ORCPT ); Tue, 26 Mar 2013 13:37:51 -0400 Date: Tue, 26 Mar 2013 17:37:47 +0000 From: Al Viro To: "J. Bruce Fields" Cc: Toralf F??rster , Linux NFS mailing list , Linux Kernel Subject: Re: kernel 3.8.4 : kernel BUG at fs/locks.c:2093! part #2 Message-ID: <20130326173747.GV21522@ZenIV.linux.org.uk> References: <514F31AF.3080709@gmx.de> <20130325220143.GD10887@fieldses.org> <20130326144640.GB3353@fieldses.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130326144640.GB3353@fieldses.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1331 Lines: 31 On Tue, Mar 26, 2013 at 10:46:40AM -0400, J. Bruce Fields wrote: > > fs/locks.c:2093 says that we did the final fput of a file that still had > > posix locks held on it. > > > > I can't see how that would happen, but admittedly the nfsd4 code here is > > much too complicated for its own good. > > > > If it were possible it would be useful to know what lead up to this. A > > network trace (tcpdump -s0 -wtmp.pcap, then send me tmp.pcap) would be > > useful for that, but I fear it's probably too huge by the time you hit > > the bug. > > > > (The following warning ("rcu_sched self-detected stall") looks like the > > result of BUGing while holding file_lock_lock. That BUG should probably > > be downgraded to a WARN_ON_ONCE.) > > Can you run with test patches? > > This just makes nfsd's fput calls synchronous so that we see in the > backtrace who called them. > > (But assuming it does turn out to be one of these callers, we may then > need some more printk's in the nfsd code...). Might also be a good idea to dump the offending struct file_lock, while we are at it... -- 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/