phc-discussions - [PHC] Argon2i in Monocypher

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [day] [month] [year] [list]

Date: Sat, 22 Jul 2017 12:48:20 +0200
From: Solar Designer <solar@...nwall.com>
To: discussions@...sword-hashing.net
Cc: l@...p-vaillant.fr
Subject: [PHC] Argon2i in Monocypher

Hi,

"Monocypher is an easy to use crypto library inspired by libsodium and
TweetNaCl" and it includes its own reimplementation of Argon2i in C.
Per the project's website, that implementation is unexpectedly fast:

| Blake2b     :  43% slower than Libsodium
| Argon2i     :  26% faster than Libsodium

|     Argon2i appears to be notably faster. This was unexpected. I'm
| inclined to blame the complexity of the reference implementation, and
| the fact that Monocypher has less features (mono-threaded Argon2i only).
| A word of warning, though: Argon2 is very sensitive to optimisation
| options. It's easy to make an unfair comparison by accident.

As I understand, this is a comparison of two pure C (thus, non-SIMD)
implementations, and both are slower than PHC's SIMD implementation.

http://loup-vaillant.fr/projects/monocypher/
https://github.com/LoupVaillant/Monocypher/blob/2c0e803efbd4c75f8da2ce1e56bb221f545833dc/src/monocypher.c#L478

Alexander