Re: kexec_load(2) bypasses signature verification

From: "Theodore Ts'o" <tytso@mit.edu>
To: Josh Boyer <jwboyer@fedoraproject.org>
Cc: Eric Biederman <ebiederm@xmission.com>,
	David Howells <dhowells@redhat.com>,
	kexec <kexec@lists.infradead.org>,
	"Linux-Kernel@Vger. Kernel. Org" <linux-kernel@vger.kernel.org>
Subject: Re: kexec_load(2) bypasses signature verification
Date: Mon, 15 Jun 2015 16:01:15 -0400	[thread overview]
Message-ID: <20150615200115.GG5003@thunk.org> (raw)
In-Reply-To: <CA+5PVA6YKR_=uVpM2rW4et3YCJLh9c+uuxt2koH1Fy2ZuR7W-g@mail.gmail.com>

On Mon, Jun 15, 2015 at 09:37:05AM -0400, Josh Boyer wrote:
> The bits that actually read Secure Boot state out of the UEFI
> variables, and apply protections to the machine to avoid compromise
> under the SB threat model.  Things like disabling the old kexec...

I don't have any real interest in using Secure Boot, but I *am*
interested in using CONFIG_KEXEC_VERIFY_SIG[1].  So perhaps we need to
have something similar to what we have with signed modules in terms of
CONFIG_MODULE_SIG_FORCE and module/sig_enforce, but for
KEXEC_VERIFY_SIG.  This would mean creating a separate flag
independent of the one Linus suggested for Secure Boot, but since we
have one for signed modules, we do have precedent for this sort of
thing.

						- Ted

[1] Yes, it doesn't buy all that much, since if the system is rooted
the adversary can just replace the kernel in /boot and force a normal,
slower reboot, but the same could be said for signed modules --- the
adversary could just replace all of /boot/vmlinux-<kver> and
/lib/modules/<kver>.  But both measures make it a tad more bit
difficult, especially for the adversary to do this replacement without
being noticed (for example linode will send me e-mail if the system
reboots normally, but not with a kexec-mediated reboot), and for cloud
systems where we don't have secure boot anyway, it's about the best we
can do.