Re: kexec_load(2) bypasses signature verification

From: "Theodore Ts'o" <tytso@mit.edu>
To: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>,
	Josh Boyer <jwboyer@fedoraproject.org>,
	David Howells <dhowells@redhat.com>,
	kexec <kexec@lists.infradead.org>,
	"Linux-Kernel@Vger. Kernel. Org" <linux-kernel@vger.kernel.org>,
	Dave Young <dyoung@redhat.com>, Petr Tesarik <ptesarik@suse.cz>
Subject: Re: kexec_load(2) bypasses signature verification
Date: Tue, 16 Jun 2015 23:26:07 -0400	[thread overview]
Message-ID: <20150617032607.GC4076@thunk.org> (raw)
In-Reply-To: <20150616202757.GB14943@redhat.com>

On Tue, Jun 16, 2015 at 04:27:57PM -0400, Vivek Goyal wrote:
> 
> So looks like you are looking for a system/option where you just want to
> always make use of kexec_file_load() and disable kexec_load(). This sounds
> like you want a kernel where kexec_load() is compiled out and you want
> only kexec_file_load() in.

Either compiled out or disabled via some flag (similar to how signed
moduled verification can be required via a flag that can be set, but
not unset once it is set), yes.

> Right now one can't do that becase kexec_file_load() depends on
> CONFIG_KEXEC option.
> 
> I am wondering that how about making CONFIG_KEXEC_FILE_LOAD independent
> of CONFIG_KEXEC. That way one can set CONFIG_KEXEC_VERIFY_SIG=y, and
> only signed kernel can be kexeced on that system.

That would certianly also be a workable strategy.

> This should gel well with long term strategy of deprecating kexec_load()
> at some point of time when kexec_file_load() is ready to completely
> replace it.

Well, note that Debian and Ubuntu are still using kexec-tools 2.0.7
(even in their latest development/unstable releases), which doesn't
have support for kexec_file_load().  So we need to get Debian to
upgrade its kexec-tools as part of this.  I'll try to file a
nag-o-gram to the Debian BTS.

						- Ted