Re: [RFC PATCH 0/2] Add examples of secure- prefixed property in documentation

[Etienne Carriere]

On Fri, 24 Apr 2020 at 23:59, Rob Herring <robh+dt@xxxxxxxxxx> wrote:
>
> On Thu, Apr 23, 2020 at 1:49 AM Etienne Carriere
> <etienne.carriere@xxxxxxxxxx> wrote:
> >
> > Hello,
> >
> > This is a proposal for adding a bit of description in the DT bindings
> > documentation of how secure- property prefix can be used for. The
> > changes in this patch series describe that for clocks and resets properties.
> >
> > Documentation file arm/secure.txt already states that secure- prefix can be
> > used for any property hence the description proposed here are not mandated.
> > However it may be useful as explicit examples of such usage.
>
> It may say that, but any new property has to be documented and
> reviewed still. I'm not sure that anything in secure.txt has actually
> gotten used.

Looking at Linux kernel, U-Boot, Qemu, EDK II, that's right :)
I guess that applies to the so-called non-secure world.

>
> You should participate in the System DT discussions in Linaro where
> how to represent different CPUs and CPU execution environments (like
> secure world) is being worked on.

Fair, I'll get information there.
Thank you for your prompt feedback.

In the same scope, I am to post a change in the Linux DTS files.
A change to define a new attribute mostly of interest for the secure
world description.
I will still post it to the LKML to get feedback about it.
Such new bindings should still be discussed in the Linux DT ML, right?

Regards,
Etienne

>
> Rob