From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-io0-f177.google.com ([209.85.223.177]:34488 "EHLO mail-io0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751616AbbKQR0k (ORCPT ); Tue, 17 Nov 2015 12:26:40 -0500 Received: by ioir85 with SMTP id r85so25919532ioi.1 for ; Tue, 17 Nov 2015 09:26:39 -0800 (PST) Date: Tue, 17 Nov 2015 11:25:51 -0600 From: Seth Forshee To: Al Viro Cc: "Eric W. Biederman" , linux-bcache@vger.kernel.org, dm-devel@redhat.com, linux-raid@vger.kernel.org, linux-mtd@lists.infradead.org, linux-fsdevel@vger.kernel.org, linux-security-module@vger.kernel.org, selinux@tycho.nsa.gov, Serge Hallyn , Andy Lutomirski , linux-kernel@vger.kernel.org, Theodore Ts'o Subject: Re: [PATCH v3 0/7] User namespace mount updates Message-ID: <20151117172551.GA108807@ubuntu-hedt> References: <1447778351-118699-1-git-send-email-seth.forshee@canonical.com> <20151117170556.GV22011@ZenIV.linux.org.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20151117170556.GV22011@ZenIV.linux.org.uk> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tue, Nov 17, 2015 at 05:05:56PM +0000, Al Viro wrote: > On Tue, Nov 17, 2015 at 10:39:03AM -0600, Seth Forshee wrote: > > Hi Eric, > > > > Here's another update to my patches for user namespace mounts, based on > > your for-testing branch. These patches add safeguards necessary to allow > > unprivileged mounts and update SELinux and Smack to safely handle > > device-backed mounts from unprivileged users. > > > > The v2 posting received very little in the way of feedback, so changes > > are minimal. I've made a trivial style change to the Smack changes at > > Casey's request, and I've added Stephen's ack for the SELinux changes. > > Would you mind explaining which filesystem types do you plan to allow? > SELinux and the rest of Linux S&M bunch do fuck-all for attacks via > handcrafted fs image fed to the code in fs driver that does not expect > a given kind of inconsistencies. > > As it is, validation of on-disk metadata is not particularly strong; > what's more, protection against concurrent malicious *changes* of > fs image (via direct writes by root) is simply inexistent. > > So what is that about? The first target is fuse, which won't be vulnerable to those attacks. Shortly after that I plan to follow with support for ext4. I've been fuzzing ext4 for a while now and it has held up well, and I'm currently working on hand-crafted attacks. Ted has commented privately (to others, not to me personally) that he will fix bugs for such attacks, though I haven't seen any public comments to that effect. Seth