<![CDATA[wolfSSL - Embedded SSL Library — Why do we have to use SHAKE-256 with ed25519 and ed448?]]> https://www.wolfssl.com/forums/topic2203-why-do-we-have-to-use-shake256-with-ed25519-and-ed448.html Wed, 23 Oct 2024 13:45:18 +0000 PunBB <![CDATA[Re: Why do we have to use SHAKE-256 with ed25519 and ed448?]]> https://www.wolfssl.com/forums/post7971.html#p7971 Hi Gabriel,

Thanks for pointing out this issue with the documentation. I've opened a PR here:
https://github.com/wolfSSL/wolfssl/pull/8102

Kind regards,
Eric - wolfSSL Support

]]> Wed, 23 Oct 2024 13:45:18 +0000 https://www.wolfssl.com/forums/post7971.html#p7971 <![CDATA[Why do we have to use SHAKE-256 with ed25519 and ed448?]]> https://www.wolfssl.com/forums/post7969.html#p7969 Reading the WolfSSL docs for both the functions wc_ed448ph_sign_hash and wc_ed25519ph_sign_hash, it says that

The hash algorithm used to create message digest must be SHAKE-256

Why is that? I tried signing messages hashed with SHA-512 and SHA3-256 and it seems to work with no errors.

Besides, looking at the ed25519.c file, the implementation for wc_ed25519_sign_msg_ex, which already hashes the message internally, the hash algorithm used is SHA-512, contradicting what the documentation says.

]]> Tue, 22 Oct 2024 23:38:04 +0000 https://www.wolfssl.com/forums/post7969.html#p7969