Re: [PATCH v4 00/12] Enroll kernel keys thru MOK

From: Mimi Zohar
Date: Thu Aug 19 2021 - 09:11:47 EST


On Thu, 2021-08-19 at 14:38 +0300, Jarkko Sakkinen wrote:
> On Wed, 2021-08-18 at 20:20 -0400, Eric Snowberg wrote:
> > Many UEFI Linux distributions boot using shim. The UEFI shim provides
> > what is called Machine Owner Keys (MOK). Shim uses both the UEFI Secure
> > Boot DB and MOK keys to validate the next step in the boot chain. The
> > MOK facility can be used to import user generated keys. These keys can
> > be used to sign an end-user development kernel build. When Linux boots,
> > pre-boot keys (both UEFI Secure Boot DB and MOK keys) get loaded in the
> > Linux .platform keyring.
> >
> > Currently, pre-boot keys are not trusted within the Linux trust boundary
> > [1]. These platform keys can only be used for kexec. If an end-user
> > wants to use their own key within the Linux trust boundary, they must
> > either compile it into the kernel themselves or use the insert-sys-cert
> > script. Both options present a problem. Many end-users do not want to
> > compile their own kernels. With the insert-sys-cert option, there are
> > missing upstream changes [2]. Also, with the insert-sys-cert option,
> > the end-user must re-sign their kernel again with their own key, and
> > then insert that key into the MOK db. Another problem with
> > insert-sys-cert is that only a single key can be inserted into a
> > compressed kernel.
> >
> > Having the ability to insert a key into the Linux trust boundary opens
> > up various possibilities. The end-user can use a pre-built kernel and
> > sign their own kernel modules. It also opens up the ability for an
> > end-user to more easily use digital signature based IMA-appraisal. To
> > get a key into the ima keyring, it must be signed by a key within the
> > Linux trust boundary.
>
> As of today, I can use a prebuilt kernel, crate my own MOK key and sign
> modules. What will be different?

The UEFI db and MOK keys are being loaded onto the .platform keyring,
which is suppose to be limited to verifying the kexec kernel image
signature. With a downstream patch, kernel modules are being verified
as well.

Initially Patrick Uiterwijk's "[PATCH 0/3] Load keys from TPM2 NV Index
on IMA keyring" patch set attempted to define a new root of trust based
on a key stored in the TPM. This patch set is similarly attempting to
define a new root of trust based on CA keys stored in the MOK db.

The purpose of this patch set is to define a new, safe trust source
parallel to the builtin keyring, without relying on a downstream patch.
With the new root of trust, the end user could sign his own kernel
modules, sign third party keys, and load keys onto the IMA keyring,
which can be used for signing the IMA policy and other files.

>
> > Downstream Linux distros try to have a single signed kernel for each
> > architecture. Each end-user may use this kernel in entirely different
> > ways. Some downstream kernels have chosen to always trust platform keys
> > within the Linux trust boundary for kernel module signing. These
> > kernels have no way of using digital signature base IMA appraisal.
> >
> > This series introduces a new Linux kernel keyring containing the Machine
> > Owner Keys (MOK) called .mok. It also adds a new MOK variable to shim.
>
> I would name it as ".machine" because it is more "re-usable" name, e.g.
> could be used for similar things as MOK. ".mok" is a bad name because
> it binds directly to a single piece of user space software.

Nayna previously said,
"I believe the underlying source from where CA keys are loaded might vary
based on the architecture (".mok" is UEFI specific.). The key part is
that this new keyring should contain only CA keys which can be later
used to vouch for user keys loaded onto IMA or secondary keyring at
runtime. It would be good to have a "ca" in the name, like .xxxx-ca,
where xxxx can be machine, owner, or system. I prefer .system-ca."

The CA keys on the MOK db is simply the first root of trust being
defined, but other roots of trust are sure to follow. For this reason,
I agree naming the new keyring "mok" should be avoided.

thanks,

Mimi