search

fcelda / nsec5-draft

Working Copy of the NSEC5 Specification
15 stars 2 forks source link

Leo wants to hash the PK as input to ECVRF_hash_to_curve and in MGF1 for the RSA-FDH-VRF #14

Closed goldbe closed 7 years ago

goldbe commented 7 years ago

"otherwise you get into weird issues with the RO proof of pseudorandomness not composing for multiple copies of the same VRF with different PKs, because you have to program the RO in different ways that may contradict each other. Generally, hashing PK is good hygiene -- like salting hashes. Aside from proof issues, it simply makes attacker's job harder because you have to attack per PK, and precompute some crazy table once and for all."

goldbe commented 7 years ago

Added this as a comment to VRF draft but did not actually implement the change. Waiting for someone else to do this so I don't screw up any EC point to octet conversion.

goldbe commented 7 years ago

Here is the commit: https://github.com/fcelda/nsec5-draft/commit/d77815bc11d20620de412d134c6c657a35c0d777

fcelda commented 7 years ago

Fixed in 455440a304457df6a5b3a64ca4bb1cf75621b57a