From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754473Ab3AKVPW (ORCPT ); Fri, 11 Jan 2013 16:15:22 -0500 Received: from terminus.zytor.com ([198.137.202.10]:52208 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753467Ab3AKVPV (ORCPT ); Fri, 11 Jan 2013 16:15:21 -0500 Message-ID: <50F080C6.5070206@zytor.com> Date: Fri, 11 Jan 2013 13:14:46 -0800 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/17.0 Thunderbird/17.0 MIME-Version: 1.0 To: Vivek Goyal CC: "Eric W. Biederman" , Konrad Rzeszutek Wilk , Daniel Kiper , "xen-devel@lists.xensource.com" , Andrew Cooper , "x86@kernel.org" , "kexec@lists.infradead.org" , "linux-kernel@vger.kernel.org" , "virtualization@lists.linux-foundation.org" , "mingo@redhat.com" , Jan Beulich , "maxim.uvarov@oracle.com" , "tglx@linutronix.de" , David Howells Subject: Re: [Xen-devel] [PATCH v3 00/11] xen: Initial kexec/kdump implementation References: <50E6F81D02000078000B3245@nat28.tlf.novell.com> <20130104170751.GB3472@host-192-168-1-59.local.net-space.pl> <20130104191146.GC6721@phenom.dumpdata.com> <20130107123404.GA2927@host-192-168-1-59.local.net-space.pl> <20130107162018.GJ3219@phenom.dumpdata.com> <87ehhsqrpr.fsf@xmission.com> <20130111165506.GD25620@phenom.dumpdata.com> <87k3rjtqi7.fsf@xmission.com> <20130111205232.GC17126@redhat.com> <50F07E2D.6010602@zytor.com> <20130111210801.GC12019@redhat.com> In-Reply-To: <20130111210801.GC12019@redhat.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/11/2013 01:08 PM, Vivek Goyal wrote: >> >> A signed /sbin/kexec would realistically have to be statically linked, >> at least in the short term; otherwise the libraries and ld.so would need >> verification as well. > > Yes. That's the expectation. Sign only statically linked exeutables which > don't do any of dlopen() stuff either. > > In fact in the patch, I fail the exec() if signed executable has > interpreter. > As I said, though (and possibly not for kexec, that depends): in the long term we probably want a way to be able to sign all kinds binaries in the system. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.