[PATCH] init/main.c: Initialize early LSMs after arch code

[Paul Moore]

On Thu, Aug 8, 2024 at 12:43 PM Guenter Roeck <linux at roeck-us.net> wrote:
>
> Also, there is a backtrace on ppc (also see below), but that is unrelated
> to your patches and only seen now because I enabled the security modules
> on that architecture. I'll bring that up with ppc maintainers.

Thanks for all your help testing this Guenter.  I see you've also
already submitted an AppArmor fix for the endian issue, that's very
helpful and I'm sure John will be happy to see it.

Beyond this work testing the static call patches from KP, would you be
willing to add a LSM configuration to your normal testing?  While most
of the LSM subsystem should be architecture agnostic, there are
definitely bits and pieces that can vary (as you've seen), and I think
it would be great to get more testing across a broad range of
supported arches, even if it is just some simple "does it boot" tests.

Out of curiosity, do you have your test setup documented anywhere?  It
sounds fairly impressive and I'd be curious to learn more about it.

-- 
paul-moore.com