phc-discussions - Argon V1

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: Mon, 7 Apr 2014 14:39:03 +0200
From: Dmitry Khovratovich <khovratovich@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Argon V1

Hi all,

we have found a bug in the function that produced Reed-Muller codewords for
the SubGroups transformation in Argon, which resulted in two duplicate
entries in the matrix L.

>From the security perspective, the bug does not affect collision/preimage
resistance or avalanche properties, but allows an adversary to save 12.5%
of memory in all tradeoff attacks with the same penalty factor. For
instance, when you attack the 128-MB computation with 56 MB of RAM (instead
of expected 64 MB), you get the penalty factor of 160.  Apart from that, no
cryptanalytic vector is identified.

We have fixed it, produced a new matrix, and made a new reference document
titled Argon v1:
https://www.cryptolux.org/images/0/0c/Argon-v1.pdf

The reference implementation can be found at
https://github.com/khovratovich/Argon/tree/master/Argon-Reference

All the relevant documents have been also sent to the Committee.

If you have been cryptanalysing Argon v0, it is quite likely that your
results apply to Argon v1 as well.

-- 
Best regards,
Dmitry Khovratovich

Content of type "text/html" skipped