From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from verein.lst.de ([213.95.11.211]:39282 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750735AbdEBPRS (ORCPT ); Tue, 2 May 2017 11:17:18 -0400 Date: Tue, 2 May 2017 17:17:17 +0200 From: Christoph Hellwig Subject: Re: [PATCH v2] xfs: publish UUID in struct super_block Message-ID: <20170502151717.GB22963@lst.de> References: <1493388001-15879-1-git-send-email-amir73il@gmail.com> <20170502073019.GD11582@lst.de> <20170502141751.GA21564@lst.de> <20170502144730.GA22578@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Miklos Szeredi Cc: Christoph Hellwig , Amir Goldstein , "Darrick J . Wong" , Vivek Goyal , Al Viro , linux-xfs , "linux-unionfs@vger.kernel.org" , linux-fsdevel On Tue, May 02, 2017 at 05:08:26PM +0200, Miklos Szeredi wrote: > Two separate issues: > > a) is s_uuid unique across all currently mounted filesystems on this system > b) is s_uuid unique for all different filesystems that have been > mounted at one time on some system > > We can check (a) but not (b). But failing (b) could have equally bad > consequences as failing (a). While (b) is harmful it's not anywhere near as harmful as (a). And while we can trivially protect against (a) protecting against the full scope of (b) (e.g. including reboots) would be very hard.