Re: [PATCHv7 00/13] kexec: Use BPF lskel to enable kexec to load PE format boot image

[Jason Gunthorpe <jgg@nvidia.com>]

On Fri, Oct 16, 2026 at 05:56:13PM +0200, Ard Biesheuvel wrote:
> Interesting. I'm not sure this is a reliable workaround, though, given
> that the return from ExitBootServices() occurs as a result of another
> syscall(), and the kthread is long gone by then.
>
> I think the reliable approach here is taking down all the other cores
> before doing the kexec.

Yeah, both parts need this.

The DRTM spec requires invoking the launch SMC on the boot CPU and all
other CPUs must be powered off. The firmware is supposed to check this.

> In other news, I've added some other features to the prototype: x86_64
> support, support for UKIs including sidecars etc (the kexec image loader
> now accepts a cpio archive which can be presented as the ESP) and some
> other improvements. This is highly in flux so let's compare notes often
> so things do not diverge unnecessarily.

This seems great, please ping again and we can retest it. I also was
just told the UKI flow was demo'd through there too including DRTM. It
seems like a little change is needed on the DRTM series for UKI, I
ill remark separately.

I like the idea of cpio for of all the parts, that seems
flexible. Presumably we can do a signing check of each part in the
kernel for those that want that?

Also when you are ready to post it I can ask here for some rust
reviewers, we have developed quite a deep bench of experience.

Thanks,
Jason