NACK: [HIRSUTE][PATCH 0/5] Built-in Revocation certificates

Tim Gardner tim.gardner at canonical.com
Mon Aug 9 12:19:11 UTC 2021



On 8/5/21 8:59 AM, Dimitri John Ledkov wrote:
> In Impish, support was added to load revoked certificates from mokx
> (submitted upstream, revied, not accepted yet) into blacklist keyring.
> 
> Also in Impish, from upstream, there is now support to have built-in
> revoked keys. And we have 2012 UEFI key revoked by default (as also
> revoked globally via uefi dbx update).
> 
> Backport both of the above things to Hirsute, such that our kernels
> honor mokx revocations, and also have the 2012 key revoked always
> (when booted with or without working shim).
> 
> This patch series was test built and tested using the revocations list
> test case that is proposed for RT ubuntu_boot test. See
> https://lists.ubuntu.com/archives/kernel-team/2021-August/122986.html
> 
> BugLink: https://bugs.launchpad.net/bugs/1928679
> BugLink: https://bugs.launchpad.net/bugs/1932029
> 
> Dimitri John Ledkov (5):
>    UBUNTU: SAUCE: integrity: Load mokx certs from the EFI MOK config
>      table
>    UBUNTU: SAUCE: integrity: add informational messages when revoking
>      certs
>    UBUNTU: [Packaging] build canonical-revoked-certs.pem from branch/arch
>      certs
>    UBUNTU: [Packaging] Revoke 2012 UEFI signing certificate as built-in
>    UBUNTU: [Config] Configure CONFIG_SYSTEM_REVOCATION_KEYS with revoked
>      keys
> 
>   certs/blacklist.c                             |  3 +
>   debian.master/config/annotations              |  1 +
>   debian.master/config/config.common.ubuntu     |  2 +-
>   .../revoked-certs/canonical-uefi-2012-all.pem | 86 +++++++++++++++++++
>   debian/rules                                  | 14 ++-
>   .../platform_certs/keyring_handler.c          |  1 +
>   security/integrity/platform_certs/load_uefi.c | 74 ++++++++--------
>   7 files changed, 145 insertions(+), 36 deletions(-)
>   create mode 100644 debian/revoked-certs/canonical-uefi-2012-all.pem
> 

None of the git SHA1 commit IDs appear to be valid in upstream linux or 
even linux-next.

rtg
-----------
Tim Gardner
Canonical, Inc



More information about the kernel-team mailing list