[Defect]: KMAC (SP 800-108r1) unclear

[jvdsn]

Describe the defect

The "KMAC" family is used to describe the SP 800-108r1 KDF based on KMAC. On its own, this is already slightly confusing, but still unambiguous. However, the variants currently (almost) clash with the variants from the "SHA-3" family:

• KMAC: KMAC[-(128|256)]
• SHA-3: KMAC(128|256)`

The only difference is in the dash. Is this intended?

Secondly, for the "KMAC" family, it is indicated that the primitive is (mac), but would it not make more sense to use (kdf) as the primitive? This seems like a typo?

[stevespringett]

cc: @bhess

[bhess]

Good catch, @jvdsn.

You’re right. the first definition was intended to represent a KDF based on KMAC (per NIST SP 800-108r1), while the second refers to the Keccak-based MAC itself (per NIST SP 800-185).

To keep things clean, I’ve removed the redundant first definition and merged the logic into the existing SP 800-108r1 KDF definitions in PR #844.

Does this structure look correct to you now?