From mboxrd@z Thu Jan 1 00:00:00 1970 From: Al Viro Subject: Re: fs: lockup on rename_mutex in fs/dcache.c:1035 Date: Sun, 26 Oct 2014 03:06:08 +0000 Message-ID: <20141026030608.GN7996@ZenIV.linux.org.uk> References: <544C50CB.4090408@oracle.com> <87siibttyi.fsf@x220.int.ebiederm.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Sasha Levin , linux-fsdevel , LKML , Dave Jones To: "Eric W. Biederman" Return-path: Content-Disposition: inline In-Reply-To: <87siibttyi.fsf@x220.int.ebiederm.org> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Sat, Oct 25, 2014 at 08:01:41PM -0700, Eric W. Biederman wrote: > Sasha Levin writes: > > > Hi all, > > > > While fuzzing with trinity inside a KVM tools guest running the latest -next > > kernel, I've stumbled on the following spew: > > Weird. I took a quick look and I don't see any changes in d_walk that > in Al's tree or in Linus's kernel for years. > > Has read_seqbegin_or_lock changed somewhere? > > >From a quick reading of the code it simply isn't possible for d_walk to > take the lock twice short of memory corruption. And the fact that the > code has not changed in years seems to suggest it isn't the obvious > cause of d_walk talking the rename_lock twice. It is a fairly obvious case of d_walk() forgetting to drop rename_lock. See upthread for analysis and (hopefully) a fix.