Re: [PATCH v3 0/1] dt-bindings: Add device-perms property description

[Rob Herring]

On Thu, Jun 02, 2022 at 06:19:00PM -0700, Stefano Stabellini wrote:
> On Wed, 1 Jun 2022, Rob Herring wrote:
> > On Thu, May 05, 2022 at 11:23:50AM +0000, Oleksii Moisieiev wrote:
> > > Introduce device-perms property which is intended to set the device
> > > permissions for the System Management interfaces.
> > > An example of this interface is SCMI (System Control and Management
> > > Interface) which controls clocks/power-domains/resets etc from the
> > > Firmware. This property sets the device_id to set the device permissions
> > > for the Fimware using BASE_SET_DEVICE_PERMISSIONS message (see 4.2.2.10 of [0]).
> > 
> > Is that an exhaustive list of controls? Seems like there would be a 
> > GET_DEVICE_PERMISSIONS.
> > 
> > > Device permissions management described in DEN 0056, Section 4.2.2.10 [0].
> > > Given parameter should set the device_id, needed to set device
> > > permissions in the Firmware.
> > > This property is used by trusted Agent to set permissions for the devices,
> > > passed-through to the non-trusted Agents. Trusted Agent will use device-perms to
> > > set the Device permissions for the Firmware (See Section 4.2.2.10 [0]
> > > for details).
> > > Agents concept is described in Section 4.2.1 [0].
> > 
> > As I said on the call discussing this, this looks very similar to other 
> > proposals wanting to control or check permissions on devices handled by 
> > some provider. While the consumer of the binding is different in various 
> > proposals, that doesn't really matter from a DT perspective. DT is just 
> > describing some type of connection between nodes. So I'm looking for 
> > collaboration here with folks that have made prior proposals. To put it 
> > another way, for a new common binding like this, I want to see more than 
> > one user. 
> 
> Do you have a pointer to another similar proposal or the name of someone
> that might be interested and might be having a second use-case for this?

ST folks who were on the call... IIRC from earlier SystemDT calls, that 
Xilinx had a similar need? Here's the various proposals I found:

https://lore.kernel.org/all/20200701132523.32533-1-benjamin.gaignard@xxxxxx/
https://lore.kernel.org/all/20190318100605.29120-1-benjamin.gaignard@xxxxxx/
https://lore.kernel.org/all/20180227140926.22996-1-benjamin.gaignard@xxxxxx/

The h/w in question is the ETZPC or TZPC. I would guess the SCMI 
interface was designed with this h/w in mind.

Rob