phc-discussions - Re: [PHC] Combining sliding window and bit-reversal

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: <52E6F091.70703@dei.uc.pt>
Date: Mon, 27 Jan 2014 23:49:37 +0000
From: Samuel Neves <sneves@....uc.pt>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] Combining sliding window and bit-reversal

On 27-01-2014 23:22, Solar Designer wrote:
> Speaking of bit reversal instructions, it appears that we still don't
> have them on x86, not even with planned extensions.  Yet there are such
> instructions e.g. on ARM, Epiphany, AMD GCN (both scalar and vector).
>
> To be more x86-friendly, something crazy like a variable block size
> Feistel cipher using CRC-32 (one instruction on modern x86) as the F
> function might work better.  e.g. 2 rounds might be enough.  Drawback:
> complexity.
You can get decent bit reversal speed using PSHUFB, by reversing up to
32 4-bit chunks per cycle. Considering other architectures have native
bit reversal instructions, it may be an acceptable solution compared to
an ad hoc permutation.