From: Andrew Morton Subject: Re: [PATCH 2/3] fs/binfmts: Better handling of binfmt loops Date: Tue, 30 Jul 2013 16:26:11 -0700 Message-ID: <20130730162611.205debb7fab539a034166552@linux-foundation.org> References: <1374766845-13565-1-git-send-email-zml@linux.vnet.ibm.com> <1374766845-13565-2-git-send-email-zml@linux.vnet.ibm.com> <20130730140446.1a4982e21ac8e183c65e9ed1@linux-foundation.org> <20130730161651.Horde.PkwqiJir309R_EljyDbxlqA@imap.linux.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, viro@zeniv.linux.org.uk, Zach Levis To: Zach Levis Return-path: In-Reply-To: <20130730161651.Horde.PkwqiJir309R_EljyDbxlqA@imap.linux.ibm.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Tue, 30 Jul 2013 16:16:51 -0700 Zach Levis wrote: > > Quoting Andrew Morton : > > > On Thu, 25 Jul 2013 08:40:44 -0700 Zach Levis wrote: > > > >> With these changes, when a binfmt loop is encountered, > >> the ELOOP will propogate back to the 0 depth. At this point the > >> argv and argc values will be reset to what they were originally and an > >> attempt is made to continue with the following binfmt handlers. > > > > hm, why? What problem does this fix? What value does the change offer > > to our users? > > This is used when the binfmt_misc,script,etc options are configured in > a way that would previously prevent executables from launching that > could be executed with a different binfmt but don't because of a loop > in a prior binfmt. > > Example: a qemu is configured to run 64-bit ELFs on an otherwise > 32-bit system. The system's owner switches to running with 64-bit > executables, but forgets to disable the binfmt_misc option that > redirects 64bit ELFs to qemu. Since the qemu executable is a 64-bit > ELF now, binfmt_misc keeps on matching it with the qemu rule, > preventing the execution of any 64-bit binary. So the admin can unforget to make that change and no longer has a problem. > With this patch, an error is printed and search_binary_handler() > continues on to the next handler, allowing the original executable to > run normally so the user can (hopefully) fix their misconfiguration > more easily. Is all this really worth changing the kernel for? It sounds a bit marginal.