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: <20140128000529.GA31410@openwall.com>
Date: Tue, 28 Jan 2014 04:05:29 +0400
From: Solar Designer <solar@...nwall.com>
To: discussions@...sword-hashing.net
Subject: Re: [PHC] Combining sliding window and bit-reversal

On Mon, Jan 27, 2014 at 11:49:37PM +0000, Samuel Neves wrote:
> You can get decent bit reversal speed using PSHUFB, by reversing up to
> 32 4-bit chunks per cycle.

Yes, but unfortunately this adds overhead of going from scalar to vector
and back (we need to bit-reverse a single 32-bit or 64-bit value), and
SIMD instructions generally have lower throughput than scalar ones.  Yet
it might be the fastest approach.  Will need to benchmark.

> Considering other architectures have native
> bit reversal instructions, it may be an acceptable solution compared to
> an ad hoc permutation.

I agree.

Alexander