Re: [Suit] MCUboot/Zephyr involvement

Hello David,

thank you for brining that up! Actually, at the last SUIT Hackathon, 
follow-up Virtual Interim and then Design-Team meeting, there was a 
strong tendency to allow for more than one "payload" (and corresponding 
metadata) in one "manifest" (with its own corresponding metadata). The 
drafts do not reflect that yet, but I am under the impression that on 
one hand, this will be incorporated soon, and on the other hand, every 
"user story" that a set of corresponding requirements can be derived 
from is welcomed in the context of the SUIT information model.

Dear WG, please bash my answer, if I am writing anything misleading here :)

Viele Grüße,

Henk

On 06/15/2018 11:10 AM, David Brown wrote:
> This week, Linaro held a SPRINT where among other things, we discussed 
> our future direction for firmware upgrade, especially in regards to two 
> configurations: a Secure/Non-secure dual image system on a single CPU, 
> and a similar dual image running on two separate CPUs.
> 
> At this point, management of multiple images is not deeply discussed by 
> the proposed documents, and I’m wondering if this kind of support would 
> best be left to additional documentation.
> 
> One thing I’d like to consider is how multiple images could be supported 
> by a single manifest. There are possibilities, such as concatenating the 
> manifests, or just having to track the multiple manifests separately. 
> Most of the fields in the manifest allow multiple instances of that 
> field, except for the payload, where the manifest only describes a 
> single payload. What would be the ramifications of allowing 1 or more 
> payload blocks in the manifest?
> 
> Thanks,
> 
> David__
> 
> 
> 
> _______________________________________________
> Suit mailing list
> Suit@ietf.org
> https://www.ietf.org/mailman/listinfo/suit
> 