phc-discussions - RE: [PHC] A must read...

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

Message-ID: <3eaf7fdd647c40f5b2f7cf809585237f@BY2PR03MB074.namprd03.prod.outlook.com>
Date: Tue, 14 Jan 2014 22:45:09 +0000
From: Marsh Ray <maray@...rosoft.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: RE: [PHC] A must read...

> v[0] = Hash(password)
> hash = v[0]
> for i = 1 .. memoryLength - 1
>     addr = Rand(i) % i
>     hash = MultHash(hash, v[addr], prevV)
>     v[i] = hash
>     prevV = v[addr]
>
> function MultHash(v1, v2, prevV2):
>     return v1*(v2 | 1) + prevV2
>
> Actually, the % operation would dominate, but I hash a page at a time... so assume it costs 0.1 clocks.

Since Rand(i) % i doesn't depend on anything specific to the candidate password, couldn't it be precomputed? Or computed and fed to a large number of processors performing the rest of the loop in parallel.

It looks like all memory accesses are predictable, therefore a custom external memory subsystem could be constructed which accepts streaming data as fast as you can you can move it on and off the chip (IO pin count * clock rate * DDR perhaps). But presumably the defender would not be able to utilize such an architecture, so it would represent an advantage to the attacker.

- Marsh