Skip to content

CoRIM SFR Profile and Example#49

Open
attzonko wants to merge 12 commits intoopencomputeproject:mainfrom
attzonko:corim
Open

CoRIM SFR Profile and Example#49
attzonko wants to merge 12 commits intoopencomputeproject:mainfrom
attzonko:corim

Conversation

@attzonko
Copy link
Collaborator

@attzonko attzonko commented Aug 7, 2025

No description provided.

Copy link

@thomas-fossati thomas-fossati left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have checked the extension CDDL and the associated example: LGTM!

A couple of suggestions:

  1. Having both checked for correctness automatically by the CI would make it more robust in the face of change, especially if the change is accidental.

  2. The CoRIM CDDL is released as an artefact of the CoRIM pipeline at each version drop (e.g.: https://github.com/ietf-rats-wg/draft-ietf-rats-corim/releases/tag/cddl-draft-ietf-rats-corim-08). While it’s under development, you might want to simplify tracking the alignment somewhat automatically (see for example what we do in CoSERV: https://github.com/rats-endorsements-distribution/draft-howard-rats-coserv/tree/main/cddl/comid.mk)

BTW, it’s great to see CoRIM being used more widely 👍

Copy link

@thomas-fossati thomas-fossati left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nice!

I left a couple of (FYI) comments. Feel free to ignore them.

@attzonko
Copy link
Collaborator Author

I have checked the extension CDDL and the associated example: LGTM!

A couple of suggestions:

  1. Having both checked for correctness automatically by the CI would make it more robust in the face of change, especially if the change is accidental.
  2. The CoRIM CDDL is released as an artefact of the CoRIM pipeline at each version drop (e.g.: https://github.com/ietf-rats-wg/draft-ietf-rats-corim/releases/tag/cddl-draft-ietf-rats-corim-08). While it’s under development, you might want to simplify tracking the alignment somewhat automatically (see for example what we do in CoSERV: https://github.com/rats-endorsements-distribution/draft-howard-rats-coserv/tree/main/cddl/comid.mk)

BTW, it’s great to see CoRIM being used more widely 👍

Thanks for lookin this over @thomas-fossati I have addressed your two points by adding a simple workflow, once the repo admins enable GitHub Actions with some Runners...we should be able to test it out. In the meantime could you take a quick look and sanity check the workflow?

@attzonko attzonko requested a review from amd-isaac September 17, 2025 21:04
@fdamato
Copy link

fdamato commented Oct 2, 2025

  • We evaluated the need for IANA registration of the CoRIM SFR extension and determined it is not required if the extension is only used within our specific CoRIM profile.

  • We confirmed that, similar to EAT, we can register an OID profile for CORIM SAFE through OCP and use a private (negative) number for the SFR extension, scoped to our SFR CoRIM profile.

  • We can control the OCP registry for our profile and update the extension value accordingly. @ericeilertson @attzonko

This commit squashes all the intermediate commits from PR opencomputeproject#49

Signed-off-by: Alex Tzonkov <[email protected]>
@attzonko
Copy link
Collaborator Author

@ericeilertson @nickhummel from my side this is ready to get squashed and merged. If we need additional work on the CoRIM profile specification we can open a new PR for that.

Thank you all for the help getting this done!

Copy link
Contributor

@rob-tetrel rob-tetrel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1

@thomas-fossati
Copy link

thomas-fossati commented Oct 13, 2025

Great stuff!

It’d be super if someone could provide the SFR serialisation/deserialisation code for veraison/corim (and/or veraison/corim-rs). Then, one could use the stock cocli to parse SRF CoRIMs, which would be awesome! 😄

- Enhancing the human readable script to handle signed CBOR and display
  all fields in human readable form.
- Addressing fix from Rob

Signed-off-by: Alex Tzonkov <[email protected]>
@attzonko
Copy link
Collaborator Author

attzonko commented Nov 7, 2025

Here is one possible schema for JIL, can we get some JIL experts input?

jil-scheme = {
  // --- Identification Phase Factors  ---
  // (Effort to create the attack and demonstrate it)
  &(elapsed-time-id: 0) => tstr
  &(expertise-id: 1) => tstr
  &(knowledge-toe-id: 2) => tstr
  &(access-toe-id: 3) => tstr
  &(equipment-id: 4) => tstr
  &(open-samples-id: 5) => tstr

  // --- Exploitation Phase Factors  ---
  // (Effort to achieve the attack on another TOE instance) 
  &(elapsed-time-ex: 6) => tstr
  &(expertise-ex: 7) => tstr
  &(knowledge-toe-ex: 8) => tstr
  &(access-toe-ex: 9) => tstr
  &(equipment-ex: 10) => tstr

  // --- Optional Factor Adjustments ---
  // (Extra points for package removal effort, added to Access to TOE) 
  ? &(package-removal-id: 11) => tstr
  ? &(package-removal-ex: 12) => tstr
  // (Extra point for Multiple Specialized equipment)
  ? &(multiple-specialized-id: 13) => tstr
  ? &(multiple-specialized-ex: 14) => tstr

  // --- Calculated Results ---
  &(identification-subtotal: 15) => tstr
  &(exploitation-subtotal: 16) => tstr
  // (Final score = identification + exploitation) 
  &(attack-potential-total: 17) => tstr
  // (e.g., "Basic", "Moderate", "High") 
  &(resistance-rating: 18) => tstr

  // --- Metadata ---
  // (e.g., "3.2.1") 
  ? &(jil-version: 19) => tstr
}

Removed device-category from the CDDL schema, addressed review comments,
and fixed test failures. Updated docs and examples to match.

Signed-off-by: Alex Tzonkov <[email protected]>
@attzonko
Copy link
Collaborator Author

@ericeilertson @nickhummel I propose we go forward with the current latest version of the profile without adding the JIL assessment, if we need to add it later we can mint a new OID to be able to easily distinguish the profiles.

Copy link

@fdamato fdamato left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants