New CPace draft is out there, Feedback would be appreciated.

I would like to draw your attention to the latest version of the ID regarding CPace.

https://datatracker.ietf.org/doc/draft-irtf-cfrg-cpace/

There are a couple of minor changes. Specifically,

- we were requested to slightly change the encoding of the protocol messages as to allow for associated-data fields.
- secondly, we have received feedback that we should prepend the length of any field in the final hashes as to rule-out the possibility of length-extension-type attacks on Merkle-Damgard hashes.

We would appreciate your feedback and review of the draft. (Best by sending a message to the CFRG list, or otherwise also by posting an "issue" at https://github.com/cfrg/draft-irtf-cfrg-cpace .

Specifically, we are wondering how to proceed regarding the prepend-length function that was introduced. Currently we suggest to encode the lengths as utf-8 (which is simple for lengths below 128 bytes, but quite complex otherwise). It might also be an option to just reserve two bytes and use little-endian encodings of the lengths or use a base128 encoding (https://developers.google.com/protocol-buffers/docs/encoding).

Yours,

Björn.

Start by reading the linked CPace draft and the issue's discussion of message encoding and length prefixes. Review the proposed alternatives and the protocol implications, then post feedback to the CFRG list or this issue. Done means the encoding approach has a clear, documented decision.

Written by the indexing model from the issue text.