From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755683Ab2GMRuh (ORCPT ); Fri, 13 Jul 2012 13:50:37 -0400 Received: from mx1.redhat.com ([209.132.183.28]:4024 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751623Ab2GMRug (ORCPT ); Fri, 13 Jul 2012 13:50:36 -0400 Date: Fri, 13 Jul 2012 13:50:31 -0400 From: Dave Jones To: Linus Torvalds Cc: Linux Kernel , "J. Bruce Fields" Subject: Re: Remove easily user-triggerable BUG from generic_setlease Message-ID: <20120713175030.GA8911@redhat.com> Mail-Followup-To: Dave Jones , Linus Torvalds , Linux Kernel , "J. Bruce Fields" References: <20120713173536.GB25432@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: 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 On Fri, Jul 13, 2012 at 10:47:43AM -0700, Linus Torvalds wrote: > On Fri, Jul 13, 2012 at 10:35 AM, Dave Jones wrote: > > This can be trivially triggered from userspace by passing in something unexpected. > > Argh. It looks like it would be harmless (apart from the noise), > except we hold file_lock_lock. Which turns the BUG_ON() into not just > "noise and kill the process", but "noise and kill the process and > leave a nasty lock held". Yeah, box wedged shortly afterwards. Not cool. > This seems to go back to 3.2, so stable should be cc'd, no? yeah, definitly. my bad. Dave